+ All Categories
Home > Documents > Confidential: For Review Only - BMJ · 2017-03-14 · Confidential: For Review Only Dear Dr Merino...

Confidential: For Review Only - BMJ · 2017-03-14 · Confidential: For Review Only Dear Dr Merino...

Date post: 07-Jul-2020
Category:
Upload: others
View: 3 times
Download: 0 times
Share this document with a friend
213
Confidential: For Review Only Supporting insulin initiation in type 2 diabetes in primary care: Results of the Stepping Up pragmatic cluster randomised controlled clinical trial Journal: BMJ Manuscript ID BMJ.2016.034563.R2 Article Type: Research BMJ Journal: BMJ Date Submitted by the Author: 19-Nov-2016 Complete List of Authors: Furler, John; University of Melbourne, Department of General Practice O'Neal, David; University of Melbourne, Department of Medicine Speight, Jane; School of Psychology, Deakin University; The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria Manski-Nankervis, Jo-Anne; University of Melbourne, Department of General Practice Gorelik, Alexandra; Melbourne EpiCenter, the University of Melbourne Holmes-Truscott, Elizabeth; School of Psychology, Deakin University; The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria Ginnivan, LOuise; University of Melbourne, Department of General Practice Young, Doris; University of Melbourne, Department of General Practice Best, James; Lee Kong Chian School of Medicine, Nanyang Technological University Patterson, Elizabeth; University of Melbourne, Nursing Liew, Danny; School of Public Health and Preventive Medicine, Monash University Segal, Leonie; University of South Australia, Health Economics & Social Policy Group May, Carl; University of Southampton, Faculty of Health Sciences Blackberry, Irene; Australian Institute for Primary Care & Ageing, College of Science, Health and Engineering, La Trobe University; University of Melbourne, Department of General Practice Keywords: clinical inertia, insulin, primary care, randomised trial, practice nurse, type 2 diabetes https://mc.manuscriptcentral.com/bmj BMJ
Transcript

Confidential: For Review O

nly

Supporting insulin initiation in type 2 diabetes in primary

care: Results of the Stepping Up pragmatic cluster randomised controlled clinical trial

Journal: BMJ

Manuscript ID BMJ.2016.034563.R2

Article Type: Research

BMJ Journal: BMJ

Date Submitted by the Author: 19-Nov-2016

Complete List of Authors: Furler, John; University of Melbourne, Department of General Practice O'Neal, David; University of Melbourne, Department of Medicine Speight, Jane; School of Psychology, Deakin University; The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria Manski-Nankervis, Jo-Anne; University of Melbourne, Department of General Practice Gorelik, Alexandra; Melbourne EpiCenter, the University of Melbourne Holmes-Truscott, Elizabeth; School of Psychology, Deakin University; The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria Ginnivan, LOuise; University of Melbourne, Department of General Practice Young, Doris; University of Melbourne, Department of General Practice Best, James; Lee Kong Chian School of Medicine, Nanyang Technological

University Patterson, Elizabeth; University of Melbourne, Nursing Liew, Danny; School of Public Health and Preventive Medicine, Monash University Segal, Leonie; University of South Australia, Health Economics & Social Policy Group May, Carl; University of Southampton, Faculty of Health Sciences Blackberry, Irene; Australian Institute for Primary Care & Ageing, College of Science, Health and Engineering, La Trobe University; University of Melbourne, Department of General Practice

Keywords: clinical inertia, insulin, primary care, randomised trial, practice nurse, type 2 diabetes

https://mc.manuscriptcentral.com/bmj

BMJ

Confidential: For Review O

nlyDear Dr Merino

Many thanks for your correspondence of 4 November offering provisional acceptance of our

manuscript, and for the opportunity to respond further to editorial and reviewer comments.

In response to reviewer three's comments I have prepared and uploaded 2 versions of our paper,

one (3a Marked-up and 3a clean) without adjustment for baseline measures and one (3b Marked-up

and 3b clean) with adjustment for baseline measures.

As you will see from our uploaded table of responses, we think there is a case to be made for leaving

the analysis as unadjusted as we feel this is consistent with what was specified in the protocol paper.

Nevertheless we are also happy for the adjusted data and paper to be considered the final version.

This has meant some minor nuancing in the reporting of the psycho-social outcomes, with no change

to the overall message of our paper (as well as changes to the text of the analysis section of course).

All other responses are noted in our uploaded Table of responses.

We hope these changes meet with your approval and of course remain always available to discuss

any further changes or to respond to any further queries you may have.

I look forward to hearing back from you

Regards

John Furler

Page 1 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

Manuscript ID BMJ.2016.034563.R2

Responses to reviewer comments on Manuscript ID BMJ.2016.034563.R1

Comment Response

Editorial

1. Please include attachment 3a, revision attachments 4 and 5

in the final paper.

These attachments are the Intervention description, Training manual for intervention

practices and patient booklet.

We have inserted text as below at the end of the section describing our intervention

on page 9 “(Appendix 2 – Intervention Description; Appendix 3 – Training Manual for

Intervention Practices; Appendix 4 – Patient participant booklet).”

These are uploaded as appropriately named Appendices

2. Provide the data about insulin type in the manuscript as an

appendix.

We have added the following text to page 13 “Further data on insulin use in

participants is available in Appendix 5.”

We have saved the data previously provided in our response to reviewer comments as

the Appendix.

3. The table reconciling outcomes is very helpful. It should be

included as an appendix.

4. The manuscript MUST report the registered outcome. This

should not be relegated to the appendix. You may also report

other outcomes in the manuscript as long as they are

identified as not-registered outcomes.

Reference is made to this as Appendix 1 on pages 7 and 11

The following text has been added to page 7 “We registered our primary outcome as

“an absolute HbA1c reduction of 0.5% in the intervention group compared with the

control group”.”

The following text has been added to pages 12-13 “This is consistent with achieving our

registered primary outcome of “an absolute HbA1c reduction of 0.5% in the intervention

group compared with the control group”.”

5. The explanation that you provide to item 13 from the editors

should be included in the manuscript.

We have added the text below to page 16, in the section discussing the implications of

our study for clinical practice.

“The issue around personalising glycaemic targets and treatments is an important and

emerging consideration in care of people T2D. At the time of our trial start up there was

Page 2 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

vigorous debate about the need for caution in setting lower targets. In addition our

exclusion criteria ruled out participants where a higher target would definitely be

considered. Collecting reliable data on duration of CVD, severe hypoglycaemia and

hypoglycaemia unawareness did not prove feasible in this setting. Therefore we made a

pragmatic decision to use the general target of 7%.”

6. Include an explanation of the sample size discrepancies also

in the manuscript.

The following text has been added to the existing sentence on page 7 “and what was an

achievable sample size”. i.e. the explanatory note from the Appendix is now included in

the text, and includes reference or Appendix 1

7. Please report the magnitude of the treatment effects (e.g.,

odds ratio for dichotomous variables and mean difference for

continuous outcomes) and corresponding 95% confidence

intervals (CIs) when presenting the P value of each outcomes

in abstract and results.

This has been done except for one p value in the text relating to a non-parametric test

8. I could not find you responses to items 11, 12 and 13 from

reviewer #1 from the last decision letter.

In the first review round reviewer 1 comments included:

11. Glargine and glusuline are used in this study as insulin

treatments. It would be worth noting that NPH remains widely

used and that the added costs of analogue insulins where NPH

can be used without problem are widely debated.

12. Competing interests: “DNO, DL, JMN had relationship with

pharma outside the submitted work" It is unclear what this

means. It would be helpful to clarify whether this includes

relationships with any of the pharmaceutical companies

marketing insulin or blood glucose monitoring products.

13. As noted above, this is an important study addressing a

widely recognized issue. I think this pragmatic study can provide

valuable information to inform clinical practice.

We have added the following text to the discussion section on page 18: “It is worth

noting that NPH Insulin remains widely used and that the added costs of analogue

insulins where NPH can be used without problem is still subject to debate.”

Clarification and further details of competing interests have been added to the section

at the end of the MS.

We have made no explicit repose to this comment

Reviewer 1

Page 3 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

1. The authors have addressed the comments from the

reviewers. The intervention and its impact is now clearly

described. It would be important to have the health

economic analysis to judge whether the intervention is

scalable, but the difference in change in HbA1c is clinically

important.

We already include text “We also collected data on healthcare utilisation and costs, to

be reported elsewhere.” On page 11

The clinically important effect we identified is discussed in the paper.

Reviewer 2

1. In my original review, I stated the following:

This trial employed cluster randomization, involving 74 practices

with average cluster size of 3. The cluster size seemed

suboptimal, particularly that some practices might recruit even

just one or two patients, in which case the treatment outcome

may not represent the practice of that particular practice.

The authors then responded below:

We acknowledge the concern expressed by the reviewer,

however we do not believe that cluster sizes of one to two is

suboptimal. While we expect variation in the cluster size, in fact

more clusters of smaller size is generally preferable in a cluster

trial design. We analysed our data at the patient level and we

took into account the clustering effect in our analyses.

Although the authors correctly pointed out that smaller cluster

size is probably preferable in cluster-randomized trial, the

relatively large variation in the cluster sizes (1-8) may result in

the dealing with clustering effect less effective, in particular

when the number of clusters is not that large, as was this case. I

suggest that the authors consider this a limitation.

We have added the following text to the limitations section page 16

“Secondly, while a smaller cluster size is generally preferable in a cluster-randomised

trial, the relatively large variation in the cluster sizes in our study may make statistical

adjustments for clustering less effective, in particular when the number of clusters is not

large.”

2. The authors have now discussed the limitation regarding

partial reporting of episodes of hypoglycemia. They stated

the possibility of underestimating hypoglycemia in the study.

However, severe hypoglycemia is relatively rare, but clearly

We have added the text following to the limitations section, page 16 “In particular,

severe hypoglycemia is serious but relatively rare, and may not have been detected in

our study, given our sample size.”

Page 4 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

very serious. Given the current sample size, the current

finding could potentially cause overlook of the adverse effect

and the overuse of insulin in the primary health providers.

The authors may make this issue clearer.

We believe this adds sufficient cautionary weight to allow the reader to make their own

judgement about the dangers of overuse of insulin, which are well known to clinicians.

3. Two minor issues about the abstract for authors to consider,

as below.

The authors stated that secondary endpoints included proportion

who achieved target HbA1c, diabetes-specific distress (PAID) and

generic health status (AQoL-8D) in abstract. Any possibility of

reporting concise results in the abstract as well?

These have been added to the abstract as requested (now 484 words)

4. Please consider reporting treatment effect estimates when

reporting the P value of each outcome (e.g., proportion of

participants who transitioned to insulin) in abstract and

results.

See response above to Editorial comment 7

Reviewer 3

The authors have added p-values in Table 2 for the between

group baseline measurements – I know this was done as a

response to a reviewer (reviewer 1, comment 11), but baseline

measurements should be similar due to the randomisation – any

statistically different baseline measurements are by chance

alone. These baseline p-values should be removed.

These baseline p values have been removed

The protocol also says the mixed effects regression will adjust for

baseline values – so even if there were any differences they

would be incorporated into the analysis. However, the analysis as

reported in the paper, didn’t adjust for baseline values because

the authors said there were no imbalance in confounders

between the groups – so now they deviated from the planned

analysis strategy. The authors should stick to the planned

analysis, and adjust for baseline values.

In our protocol paper we stated: "Mixed-effects linear regression will be used to

compare the means between study arms for continuous outcomes, adjusting for

baseline outcome measure. Stratification variables will be included as fixed effects.

Analysis will be conducted on an intention to treat basis, with any imbalance of

confounders between study groups adjusted for in the regression analysis."

Given that there was no difference in the baseline measures there was no imbalance of

confounders and so these were not adjusted for in the final analysis. We acknowledge

that this could have been better described in our protocol paper, however we believe

the analysis we have undertaken is consistent with what was stated in the protocol

Page 5 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

paper.

We did repeat the analysis of the primary outcome measure with adjustment for both

baseline HbA1c and clustering. This showed a treatment effect (95% CI) of -0.6 (-0.9 to -

0.3); p<0.001 which is still a clinically and statistically significant result. As such, we do

not believe that our analysis method impacts on the results that have been reported.

Nevertheless, in response to Reviewer 3 comments, we have also provided an

alternative version of the paper (Version 3b clean and with mark-up) including an

alternative Table 2, adjusting for baseline measure. This has led to some minor changes

in the reporting of the psychosocial measures, although no change in the overall

message of the paper.

We are happy for either 3a or 3b to be considered the final, based on the editorial view

of our comments.

Page 6 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

1

Supporting insulin initiation in type 2 diabetes in primary care: Results of the

Stepping Up pragmatic cluster randomised controlled clinical trial

JS Furler, DN O’Neal, J Speight, J Manski-Nankervis, A Gorelik, E Holmes-Truscott, L Ginnivan, D

Young, J Best, E Patterson, D Liew, L Segal, CR May, I Blackberry,

JS Furler, Associate Professor, Department of General Practice, University of Melbourne, DN O’Neal,

Associate Professor, Department of Medicine, St Vincent’s Hospital, University of Melbourne, J

Speight, Professor, School of Psychology, Deakin University; Director, The Australian Centre for

Behavioural Research in Diabetes, Diabetes Victoria; AHP Research, United Kingdom, J Manski-

Nankervis, Lecturer, Department of General Practice, University of Melbourne, A Gorelik, Senior

Statistician, Melbourne EpiCentre, the University of Melbourne, E Holmes-Truscott, Research Fellow,

BPSc (Hons), School of Psychology, Deakin University; The Australian Centre for Behavioural

Research in Diabetes, Diabetes Victoria; L Ginnivan, Research Nurse and Credentialed Diabetes

Educator, of General Practice, University of Melbourne, D Young, Professor and Chair of General

Practice, and Associate Dean (Academic), Melbourne Medical School, University of Melbourne,

Australia, J Best, Professor and Dean, Lee Kong Chian School of Medicine, Nanyang Technological

University, Singapore, E Patterson, Professor, School of Nursing, University of Melbourne, Australia,

D Liew, Professor, School of Public Health and Preventive Medicine, Monash University, Australia, L

Segal, Professor, Health Economics and Social Policy Group, Division of Health Sciences, University of

South Australia, CR May, Professor of Health Care Innovation, Faculty of Health Sciences, University

of Southampton, UK., I Blackberry, Associate Professor and Director, John Richards Initiative,

Australian Institute for Primary Care & Ageing, College of Science, Health and Engineering, La Trobe

University, Australia,

Correspondence to: John Furler, Department of General Practice, University of Melbourne, 200

Berkeley St, Carlton, VIC, 3053, Austraila, [email protected]

Page 7 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

2

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of

all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats

and media (whether known now or created in the future), to i) publish, reproduce, distribute, display

and store the Contribution, ii) translate the Contribution into other languages, create adaptations,

reprints, include within collections and create summaries, extracts and/or, abstracts of the

Contribution, iii) create any other derivative work(s) based on the Contribution, iv) to exploit all

subsidiary rights in the Contribution, v) the inclusion of electronic links from the Contribution to

third party material where-ever it may be located; and, vi) licence any third party to do any or all of

the above.

Word count:

Abstract 467

Main paper: 4418

Page 8 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

3

Abstract

Objective

Optimising glycated haemoglobin to minimise risk of long-term complications among people with

type 2 diabetes is a global priority. Supporting timely insulin initiation through innovation in primary

care could have a major impact. Our aim was to test the effectiveness of the ‘Stepping Up’ model of

care compared to usual primary care in normalising insulin initiation as part of routine primary care

practice for type 2 diabetes, leading to improved HbA1c levels. Our hypothesis was that HbA1c

would improve among participants in intervention arm practices, facilitated through timely insulin

initiation, compared to the control arm.

Design

A two-arm, 12-month cluster-randomised controlled trial of the Stepping Up model of care.

Setting

Primary care practices in Victoria, Australia with a Practice Nurse and at least one consenting eligible

patient (HbA1c ≥7.5% on maximal oral therapy).

Participants

74 practices and 266 patients participated: mean (range) cluster size 4 (1 to 8) patients.

Intervention

The Stepping Up model of care intervention involved theory-based practice system change and re-

orientation in the roles of the health professionals in the primary care diabetes team. The core

component was an enhanced role for the Practice Nurse in leading insulin initiation, and mentoring

by a Registered Nurse-Credentialed Diabetes Educator.

Main outcome and measures

Our clinically meaningful primary endpoint was change in glycated hemoglobin (HbA1c). Secondary

endpoints included the proportion of participants who transitioned to insulin, proportion who

Page 9 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

4

achieved target HbA1c as well as change in depressive symptoms (PHQ-9), diabetes-specific distress

(PAID) and generic health status (AQoL-8D).

Results

HbA1c improved in both arms, with a clinically significant between-arm difference (mean difference

-0.7%; 95%CI -1.1 to -0.4%), favouring the intervention. At 12 months, in intervention practices,

105/151 (69.5%) adults with type 2 diabetes had commenced insulin (102 remained on insulin at 12

months); in control practices 25/115 (21.7%) commenced insulin (and 24 remained on insulin at 12

months) (OR 8.3, 95%CI 4.5 to 15.4, p<0.001). Target HbA1c (≤7% (53mmol/mol)) was achieved by

54 (35.8%) intervention participants (32 of whom were using insulin at 12 months) and 22 (19%) of

the control arm (two of whom had commenced insulin), (OR 2.2, 95%CI 1.2 to 4.3, p=0.02).

Depressive symptoms did not worsen at 12 months (PHQ-9: -1.1 (3.5) versus -0.1 (2.9), p=0.02

favouring intervention). There was no statistically significant difference between arms in the mean

(SD) change in mental health (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13)), physical health (AQoL

PCS: 0.03 (0.16) versus 0.02 (0.13)) or diabetes-specific distress (PAID: 5.6 (15.5) versus -2.4 (15.4)).

No severe hypoglycaemia events were reported.

Conclusions

Our novel model of care increased insulin initiation rates in primary care, improving glycated

hemoglobin without worsening emotional well-being. Delays in starting insulin treatment can be

overcome, enabling timely treatment intensification to improve important health outcomes.

Trial registration Australian and New Zealand Clinical Trials Registry (ACTRN12612001028897)

http://www.anzctr.org.au/

What this paper adds

What is already known on this subject?

Page 10 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

5

• Achieving and maintaining glycaemic targets early in type 2 diabetes improves long term

outcomes.

• There are barriers to early stepwise progressive treatment intensification to achieve glycaemic

targets, particularly in relation to insulin initiation in primary care.

• Interventions have had limited success in overcoming this delay in starting insulin treatment and

changing clinical practice, in part because system level barriers are not addressed.

What this study adds

• Our model of care intervention changed clinical practice, with more patients in intervention arm

practices commencing insulin, with an overall benefit in terms of HbA1c reduction, achieved

without serious adverse events, or any worsening in depressive symptoms.

• The theoretical base and flexible implementation are important characteristics of our

intervention.

• Our model of care used existing resources to work smarter and improve outcomes and, thus, has

important implications for policy makers, funders and practitioners seeking innovative ways to

provide best care for people with type 2 diabetes in primary care.

Page 11 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

6

Introduction

Nearly 600 million people will have type 2 diabetes type 2 diabetes worldwide by 2030.1 Innovation

in delivering effective clinical care to these people is an urgent global priority. To reduce the risk of

long-term macro- and microvascular complications,2 UK, European and US guidelines recommend

early adoption of insulin as a part of step-wise treatment intensification to bring glycated

hemoglobin (HbA1c) below a general target of 7% (53mmol/mol).3-6

However, insulin initiation is

frequently delayed, particularly in primary care,7 where despite being recommended as part of

routine clinical management of type 2 diabetes6, implementation is not widespread. The mean

HbA1c of people with type 2 diabetes prior to starting insulin is typically 1.5-2.0% above target: 9.3%

(78mmol/mol) in the UK8 8.6% (70mmol/mol) in a study in the US

9, 8.9% (74mmol/mol) in a large,

multi-country primary care study10

, and 9.4% (79mmol/mol) in a community study in Australia (after

a median diabetes duration of 8.1 years)11

.

Delay in treatment intensification by healthcare professionals, despite evidence that intensification

is warranted and effective12

is a major barrier to initiating insulin in type 2 diabetes13

. It can be due

to health professional factors14

(e.g. concerns about hypoglycaemia risk, lack of confidence/skills in

insulin initiation/titration), health system factors15

(e.g. competing priorities in busy, reactive

primary care settings), and patient-related factors (e.g. psychological insulin resistance16

).

Supporting and embedding insulin initiation as an element of routine primary care practice is an

important first step and building block in potentially reducing referrals to costly secondary care, and

supporting timely, early optimisation of therapy and achievement of glycaemic targets.

Our trial investigated the effectiveness of a new model of care designed to support primary care-

based insulin initiation among people with type 2 diabetes for whom it is clinically indicated. The

Stepping Up model of care is built around an enhanced, reconfigured role for the primary care

Practice Nurse.17

In Australia, clinical care of type 2 diabetes is predominantly undertaken in primary

care, largely made up of General Practitioner private group practices where Practice Nurses are a

Page 12 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

7

rapidly growing section of the workforce, taking on an increasing role in chronic condition

management. In this setting, our approach to professional behaviour change was informed by

Normalisation Process Theory18

and developed through pilot studies.17 19

Our aim was to address

both clinician and system level barriers to timely insulin initiation, and to normalise insulin initiation

as part of standard primary care practice. We hypothesised that HbA1c would improve among

participants in intervention arm practices, facilitated through timely insulin initiation, compared with

the control arm. Further, based on previous research 20 21

, we expected no significant negative

impact on participants’ general emotional wellbeing (depressive symptoms).

Methods

Study design and participants

The study design and protocol have been described previously.22

In summary, we conducted a 12-

month, two-arm, non-blinded cluster-randomised controlled trial (RCT), consistent with CONSORT

guidelines,23

to investigate the effectiveness of the Stepping Up model of care versus usual care.

General Practices in Victoria, Australia were eligible if they had at least one consenting General

Practitioner (GP) and Practice Nurse and could identify at least one eligible patient participant:

adults with type 2 diabetes with above target HbA1c (≥7.5% (58mmol/mol)) in the past six months

who were already prescribed maximum oral therapy (i.e. at least two oral hypoglycaemia agents

(OHAs) at maximum doses) or if their GP judged that insulin would be clinically appropriate. Patients

were ineligible if they were >80 years old, already using insulin, had an eGFR <30 mL/min/1.73m2,

unable to give informed consent or had a complex debilitating medical condition, e.g., severe mental

illness, end-stage cancer, unstable cardiovascular disease.

Our original protocol was based on 58 practices and an average cluster size of 5. Based on

experience in the field and what was an achievable sample size this was subsequently revised to 74

practices and an average cluster size of 3 (see Appendix 1).

Page 13 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

8

Randomisation

The unit of randomisation was the primary care practice. The study statistician (AG) computer-

generated stratified block randomisation sequences with varying block sizes (4, 6 and 8) prior to

recruitment. Practices were stratified by size (≤2 vs >2 full-time equivalent GPs), setting (private

practice vs community health centre), and participation (or not) in type 2 diabetes quality

improvement programs (the Australian Primary Care Collaborative). After providing consent and

recruiting at least one eligible patient, practices were randomised to intervention or usual care. We

used this index case method24

in all practices as our previous experience suggested that delaying

randomisation of a cluster until all patients have been recruited risks loss of engagement of GPs25

.

The research team then assisted practices to continue to identify and recruit patient participants

(through searching the practice medical record database). This meant that allocation concealment

after the index case was recruited was not possible for the GP and Practice Nurse. However, in order

to minimise potential bias, participating patients were not informed of their study allocation until

after they had provided consent. Blinding of GP, Practice Nurse and patient was also not possible

given the pragmatic nature of the intervention.

Intervention

The Stepping Up model of care, described elsewhere, 17 22

involved a re-orientation of existing

resources. It included: a) an enhanced role for the Practice Nurse in leading the discussion with

patients about intensifying treatment through insulin initiation and titration, b) simple insulin

initiation and up-titration clinical protocols, and c) a re-oriented role for the specialist Registered

Nurse-Credentialed Diabetes Educator in mentoring the Practice Nurse rather than providing direct

patient care. The training and mentoring support was designed to enhance the primary care team’s

knowledge, skills and confidence in discussing and implementing insulin initiation within the practice

as a part of routine care. Intervention practices had an in-practice briefing and training session for

GPs and Practice Nurses of approximately 60-90 minutes, following which patients with confirmed

Page 14 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

9

eligibility and completed baseline data were invited to consult their GP for an assessment to discuss

treatment intensification and referral to the Practice Nurse. Practice Nurses did not prescribe insulin,

or manage insulin dosing without liaison with the GP, based on the legal scope of practice for

generalist Practice Nurses in Australia.

Our model of care involved the acknowledgment and discussion of advantages and disadvantages of

starting insulin treatment with patients, including weight gain. We modelled shared decision making

as a part of the intervention training, drawing on the principles of motivational interviewing. This set

the scene for encouraging practitioners in intervention practices to approach participating patients

with equipoise in relation to starting insulin. While the intervention was necessarily brief in this

pragmatic trial, we included guidance and checklists for GPs and Practice Nurses to discuss the pros

and cons of insulin therapy and elicit patient concerns and expectations, while also openly

acknowledging and accepting that some patients may choose not to start insulin.

The role of Registered Nurse-Credentialed Diabetes Educator in supporting and mentoring the

Practice Nurse, and of the Practice Nurse in leading the discussion and implementation of insulin

therapy with the patient in liaison with the GP, is outlined elsewhere.15,20

Titration protocols were

based on fasting blood glucose and use of a 3-day, 7-point blood glucose profile26

to identify the

meal with the largest postprandial excursion (see appendix material). We gave no additional

instructions, so the GP had clinical autonomy regarding the management of OHAs. Practice Nurses

and GPs were encouraged to see patients as frequently as was felt to be clinically appropriate over a

period of up to 12 months, drawing as needed upon the study Registered Nurse-Credentialed

Diabetes Educator for mentoring and support, even if the patient remained undecided about, or had

decided against, starting insulin. Further details about the intervention can be found in the

referenced papers and the Appendices (Appendix 2 – Intervention Description; Appendix 3 –

Training Manual for Intervention Practices; Appendix 4 – Patient participant booklet).

Page 15 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

10

Control arm practices were given a copy of Australian type 2 diabetes management guidelines27

.

Control arm practices were offered training in the Stepping Up model of care after 12-month follow

up of patient participants was complete.

Patient involvement

Participant feedback was sought after conducting an intervention pilot study15

and this feedback was

used to refine the model of care. Participant feedback was also sought in a pilot of the data

collection forms. Throughout the main trial we maintained communication with patient participants

through a regular newsletter which included aggregate data about study progress and opportunities

to provide feedback to the study team. We assessed the burden of the intervention on patients

through interviews conducted at the end of the trial, as part of process evaluation (to be reported

elsewhere). This evaluation was led by the Chronic Illness Alliance, a consumer advocacy

organization that has been a long term collaborator of our research group. We have thanked all

participants for their involvement in the trial, and will provide, at a later date, a final summary report

of the trial outcomes. Participants have access to the study website where all published results will

be publicly available.

Endpoints and data collection

While our intervention targeted a process of care (i.e. insulin initiation), we chose a clinically

meaningful disease outcome at an individual patient level as primary endpoint: change (from

baseline to 12 months) in HbA1c, measured as a continuous variable. We registered our primary

outcome as “an absolute HbA1c reduction of 0.5% in the intervention group compared with the

control group”. HbA1c was performed at DCCT-aligned pathology laboratories and communicated to

clinicians and patients as part of usual clinical care; researchers retrieved these data from medical

records or directly form pathology laboratories. Secondary endpoints included the proportion of

participants who transitioned to insulin (this was amended from the original protocol where rate of

insulin initiation was used, which proved impractical given the small cluster size in the study); the

Page 16 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

11

proportion who achieved a target HbA1c of ≤7.0% (53mmol/mol) at 12 months; and change (from

baseline to 12 months) in depressive symptoms (PHQ-9),28

diabetes-specific distress (PAID),29

and

generic health status (AQoL-8D).30

Differences between the registered outcomes reported here and

those registered in the trial registry and justification and explanation for any changes made can be

found in the Appendix 1. We also collected data on healthcare utilisation and costs, to be reported

elsewhere.

All participants were provided with a blood glucose meter (Performa NanoTM

; Roche Diagnostics)

and instructed on its use. Subsidised low-cost blood glucose testing strips were available through the

National Diabetes Service Scheme31

. Data were uploaded from the meter at 6 and 12 months to a

secure server.

Statistical analysis

Our statistical analysis plan has been published elsewhere.22

In brief, our sample size of 224 patients

from 74 general practices (averaging 3 patients per practice) allowed us to detect an absolute 0.5%

mean HbA1c difference over 12 months between control and intervention arms with 80% power and

a standard deviation (SD) of 1 using two-sided alpha of 0.05. Data were analysed using Stata 13

(StataCorp, TX, USA). Descriptive statistics were used to summarise GP, Practice Nurse, and patient

characteristics for the two study arms and appropriate univariate tests were conducted to check for

any significant difference in potential confounders. Parametric data are reported as mean (SD) and

non-parametric data as median (IQR). Categorical data are reported as n (%). T-test for proportions

was used to compare baseline and 12-month data within each study arm for categorical data, while

paired t-tests were for used for continuous parametric data and Wilcoxon matched pairs signed rank

test for continuous non-parametric data. The individual patient was the unit of analysis and the

analytical methods allow for clustering of patients within the practices. Marginal logistic modelling

using generalised estimating equations with robust standard errors and adjustment for clustering

were used to compare binary outcomes between the two study arms. Mixed-effects linear

Page 17 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

12

regression was used to determine predictors for continuous outcomes, adjusting for clustering at the

practice level. As no imbalance in confounders was identified between study groups no adjustment

for confounders was made in the regression analysis. Analyses were conducted on an intention-to-

treat basis.

Study oversight

The study protocol was approved by the University of Melbourne Health Sciences Human Research

Ethics Sub-committee (ID 123740) and registered with the Australian and New Zealand Clinical Trials

Registry (ACTRN12612001028897). All participants gave informed consent before enrolment.

Results

Participating practices and patients

Between October 2012 and January 2014, 93 primary care practices expressed interest and

identified 521 potentially eligible patients (Figure 1). Subsequently, 19 practices did not consent any

eligible patients, leaving 74 participating practices for randomisation. Two hundred and fifty five of

the potentially eligible patients were subsequently found to be ineligible at screening (n=156) or did

not respond to the invitation letter (n=99). By April 2014, the 74 practices had identified and

consented 266 eligible patient participants (73% of potentially eligible patients identified).

Baseline characteristics of practices, GPs, Practice Nurses and participants with type 2 diabetes are

shown in Table 1. Of the total sample, 248 (93.2%) completed 12-month follow-up for the primary

endpoint. No differences in baseline characteristics were observed between study completers and

non-completers, except for a higher proportion of women (n=11) not completing than men (n=7).

Primary and secondary endpoints

Primary and secondary endpoints are shown in Table 2. At 12 months, there was a statistically and

clinically significant difference between study arms in terms of change in HbA1c (mean

difference: -0.7%; 95% CI -1.1 to -0.4%, p<0.001), favouring the intervention. This is consistent with

Page 18 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

13

achieving our registered primary outcome of “an absolute HbA1c reduction of 0.5% in the

intervention group compared with the control group”. The majority of this change in HbA1c seen in

both arms was achieved by six months (See Figure 2).

In the intervention arm 105/151 (69.5%) of patients commenced insulin (and 102 remained on

insulin at 12 months), while in the control arm, 25/115 (21.7%) commenced insulin (and 24

remained on insulin at 12 months). Median (IQR) number of days from baseline assessment to

insulin initiation in intervention and control group patients who started insulin was 32 (11.5, 134.5)

days and 85 (63, 191) days respectively (statistically significant difference: Two-sample Wilcoxon

rank-sum test; p=0.005).In the intervention arm, 17 (11.22%) patients had commenced rapid-acting

insulin at 12 months, compared with one patient in the control arm (OR 14.2 95%CI 1.8 to 109.8,

p=0.011 ). Further data on insulin use in participants is available in Appendix 5.

Target HbA1c (≤7% (53mmol/mol)) was achieved by 54 (35.8%) intervention participants (32 of

whom were using insulin at 12 months) and 22 (19%) of the control arm (two of whom had

commenced insulin), (OR 2.2 (1.2 to 4.3) p=0.02). Twenty-two (14.7%) intervention arm participants

and 20 (17%) control arm participants achieved target HbA1c without commencing insulin.

At 12 months, there was a statistically significant difference between arms in the mean (SD) change

in depressive symptoms (PHQ-9: -1.1 (3.5) versus -0.1 (2.9), favouring the intervention). There was

no significant difference in mental health (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13), physical

health (AQoL PCS: 0.03 (0.16) versus 0.02 (0.13)) or diabetes-specific distress (PAID: -5.6 (15.5)

versus -2.4 (15.4)). There was no significant difference between arms in the proportion of

participants experiencing moderate-to-severe depressive symptoms or severe diabetes-specific

distress (Table 2), nor was there any difference by insulin initiation (Wilcoxon rank sum test; p=0.98).

No significant difference was found in PHQ-9, AQoL or PAID scores when comparing participants

initiating insulin with those who did not (data not shown).

Page 19 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

14

At 12 months, there was an average weight gain in the intervention arm and an average weight loss

in the control group (1.7 (5.2) kg versus -1.1 (5.1) kg, mean (95% CI) difference 2.8 (1.6 to 4.1) kg,

p<0.001). There were no significant differences in blood pressure nor other biochemical measures

between arms at follow-up with the exception of triglycerides, which remained higher in the control

group.

At baseline, participants were using a mean (SD) of 2.0 (0.6) classes of non-insulin hypoglycaemic

agents, with no significant difference between arms (t test, p=0.89). The majority of patients were

prescribed metformin (93% across both arms) and sulfonylureas (63% across both arms). There was

no significant difference in the prescription of individual medication classes by study arm. The mean

(SD) number of classes of non-insulin hypoglycaemic agents being used at 12 months was higher in

the control compared to the intervention arm (2.3 (0.1) vs 1.9 (0.1); p=0.01). There were a higher

proportion of people in the control group using DPP4 inhibitors than the intervention group at 12

months (Table 3).

Practices in the intervention arm received a total of 183 mentoring support visits from the study

Registered Nurse-Credentialed Diabetes Educator (mean (range) visits per practice: 5.2 (1-8)). Thirty-

two percent (48/151) of participants in intervention practices completed at least one 3-day, 7-point

structured blood glucose monitoring profile over the 12-month study. Practice nurses estimated the

time they spent on the study (both clinical interactions with participating patients and on research

tasks). On a per-practice basis, 23 control practices and 27 intervention practices reported a median

(IQR) of 1.5 (0, 3.6) and 18 (9, 20.9) hours respectively.

58.3% of people in the control group were recruited prior to practice randomisation compared to

45.0% in the intervention group (statistically significant difference p=0.033). Sensitivity analysis was

conducted to explore whether there was any difference at baseline between patients who were

recruited before and after practice randomisation. No statistically significant differences were found.

Mixed-effects linear regression was used to determine the impact of the intervention, adjusting for

Page 20 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

15

clustering at the practice level for each of these groups. Treatment effect in terms of the primary

outcome remained significant in both groups (treatment effect: -0.57 95%CI: -1.1, -0.05 p value =

0.03 and -0.98 95%CI -1.49, -0.48 p value <0.001 for patients assessed prior to and after

randomisation respectively).

Adverse events

No severe hypoglycaemic events (i.e. requiring third-party assistance for recovery) or other adverse

events were reported in either study arm.

Discussion

Principal findings

Our model of care changed clinical practice, with the majority in the intervention arm commencing

insulin, producing a clinically and statistically significant improvement in glycaemic control among

adults with type 2 diabetes managed in primary care. This was despite a higher patient-to-GP ratio in

intervention practices, and was achieved safely, with no severe hypoglycaemic events, and without

deterioration in emotional wellbeing or health status. Our results indicate that, with appropriate

support and practice-system redesign, insulin initiation can become part of routine primary care

diabetes management, obviating the need to refer to specialist services with geographical, cost and

accessibility barriers.

Strengths and limitations of study

A strength of our study is the robust theoretical and empirical base to our intervention. Our

pragmatic trial of a complex intervention addressed a number of known barriers to overcoming

delay in starting insulin therapy. For example, an intentional component of our system redesign was

re-orienting the Practice Nurse and Registered Nurse-Credentialed Diabetes Educator roles, allowing

additional time to be spent with patients, within existing resources. Other strengths include the

Page 21 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

16

cluster-randomised design, minimising the risk of contamination, and our excellent participant

retention (93%).

Our study had limitations. First, practices were randomised after the first consenting patient was

identified, raising the possibility of selection bias. However, the balance in key patient characteristics

between the study arms means that any such bias was minimal. Secondly, while a smaller cluster

size is generally preferable in a cluster-randomised trial, the relatively large variation in the cluster

sizes in our study may make statistical adjustments for clustering less effective, in particular when

the number of clusters is not large. Thirdly, our sample may not have been fully representative of

the broader population of adults with type 2 diabetes managed in primary care for whom insulin is

clinically indicated. Overall, less than 15% of our sample had severe diabetes-specific distress or

moderate-to-severe depressive symptoms, a lower rate than in a recent national Australian

sample.32

We will explore implementation fidelity and variation in more detail through a qualitative

process evaluation in a subsequent paper. Finally, our medications and hypoglycaemia data were

derived from GP records and subject to the same accuracy limitations of any routinely collected

clinical dataset. Hypoglycaemia is typically under-estimated 33

and is likely to be under-reported in

routine medical records. 34

In particular, severe hypoglycemia is serious but relatively rare, and may

not have been detected in our study, given our sample size.

Comparison with other studies

Only two other trials have tested interventions to change clinical practice in this way. The AIM@GP

trial showed no improvement in insulin prescribing rates or glycated hemoglobin.35

It provided

scheduled and ongoing telephone support from a specialist diabetes educator as well as the option

to refer patients to a community pharmacist off-site for a one-hour insulin initiation session. Our

intervention differed in that it was based completely in the familiar environs of the patients’ own

primary care practice, built on existing relationships and resources (with the practice-based Practice

Nurses) and provided an immediate pathway for the GP to delegate this clinical task. A UK study,

Page 22 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

17

using a pre-post evaluation design, showed improved HbA1c (-1.4%) at 6 months in patients who

initiated insulin,36

similar in magnitude to the improvement in our study. That intervention combined

education with face-to-face and telephone specialist diabetes nurse support for GPs and Practice

Nurses and involved a full-day, off-site training for both GP and Practice Nurse. In contrast, our

intervention used a brief (60-90 minutes) on-site training incorporated into the daily running of the

practice with flexible Registered Nurse-Credentialed Diabetes Educator support as required.

Consistent with previous research,37

our study participants did not report worse psychological

outcomes at follow up, suggesting that insulin therapy can be initiated in primary care for people

with type 2 diabetes without impairing their emotional wellbeing. In fact, at 12 months, the

reduction in depressive symptoms in intervention arm participants was significantly greater than

that seen in the control arm. Insulin initiation (across both arms) did not impact on depressive

symptoms, diabetes-specific distress or generic health status at 12 months.

Conclusions and policy and practice implications

Our pragmatic trial findings have important implications for the organisation of healthcare and for

health policy. Our model of care is based on an enhanced role for a Practice Nurse and would not be

feasible where primary care doctors work in solo practice without access to a Practice Nurse.

Nevertheless, even in high-resource settings, where the move to multidisciplinary primary health

care teams is growing, for example through the growth of the Medical Home movement , our study

suggests that simply having access to a Practice Nurse will not increase appropriate insulin initiation.

To make the best use of resources, primary care workforce models need to be developed and

implemented to reorient the way specialists (Registered Nurse-Credentialed Diabetes Educators and

endocrinologists) offer support to primary care teams that include well-supported and resourced

primary care nurses. Registered Nurse-Credentialed Diabetes Educator Change is occurring in some

health systems,38

but our study provides evidence of the effectiveness and safety of such models of

care. Rather than waiting for referrals, specialist services need to offer pro-active, tailored secondary

consultation, liaison and mentoring services that are flexible and supportive of the needs of primary

Page 23 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

18

care practitioners and patients. Scaling up the model of care in metropolitan centres would require

engaging with hospitals and other health services in reorienting the role of Registered Nurse-

Credentialed Diabetes Educators currently employed in direct patient care. Investigation of e-health

modalities (e.g. online training, support and video consultations) may support implementation of

this model of care in more distant rural/remote settings.

Our trial findings also have implications for clinical practice. The issue around personalising

glycaemic targets and treatments is an important and emerging consideration in the care of people

with T2D. At the time our trial started, there was vigorous debate about the need for caution in

setting lower targets. In addition, our exclusion criteria ruled out participants for whom a higher

target would definitely be considered. Collecting reliable data on duration of CVD, severe

hypoglycaemia and hypoglycaemia unawareness did not prove feasible in this setting. Therefore we

made a pragmatic decision to use the general target of 7%. Debate continues about the advantages

and disadvantages of intensifying treatment for people with type 2 diabetes at the HbA1c levels

mandated in our study. While our findings suggest the that Stepping Up model of care builds clinical

capacity within GP and Practice Nurse teams to undertake the work of insulin initiation, the model of

care did not mandate a dogmatic approach to such therapy changes. It is worth noting the major

change in HbA1c was achieved at 6 months and that even in intervention practices, only 35% of

participating people with type 2 diabetes achieved the general HbA1c target of <7% (53 mmol/mol),

suggesting that that practitioners and patients were judicious in the way they approached

progressive treatment intensification, within the new model of care.

Our clinical protocols and algorithms were focused solely on insulin. It is worth noting that NPH

Insulin remains widely used and that the added costs of analogue insulins where NPH can be used

without problem is still subject to debate. While we did not pre-specify weight gain as a secondary

outcome, this is an adverse effect of insulin therapy and we have chosen to report it. While insulin is

still regarded as an essential therapy option, as the range of glycaemic therapies grows clinical

algorithms become more complex5. Future research needs to explore the capacity to generate real-

Page 24 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

19

time personalised treatment intensification recommendations that incorporate this increasing

complexity,39

to be used as part of the Practice Nurse-led model of care. Future research could also

address the extent to which improvements in glycaemic levels are maintained, and the extent to

which the model of care is sustained in routine clinical practice. In particular, future research could

explore use of the model of care to specifically support early adoption of insulin therapy to achieve

glycaemic targets early in people with recently diagnosed type 2 diabetes.

The global epidemic of type 2 diabetes demands innovation in care delivery. Delaying insulin

initiation when clinically indicated is neither ethical nor effective. Furthermore, health systems will

not cope with demand if insulin initiation remains anchored in specialist centres, nor will they be

able to respond to the imperative to achieve glycaemic targets early in people with recently

diagnosed type 2 diabetes. Thus our pragmatic, translational study has important implications across

health systems globally for the organisation of care for people with type 2 diabetes. Our effective

model of care has the potential to improve outcomes in people with type 2 diabetes while making

better use of scarce healthcare resources.

Page 25 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

20

Figure 1: Consort Diagram

Assessed for eligibility (93 practices, patients n=521)

Excluded: Practices - failed to recruit patients – n = 19 Patients (n=255) ♦ Ineligible (n=156); ♦ Non-responder (n=99)

Intention to treat analysis

Clusters: 38 practices, 115 patients

Allocated to control arm (n=38 practices) Received allocated control

• 38 practices, patients (n=115)

• mean cluster size 3, range 1-6

Allocated to intervention arm (n=36 practices) Received allocated intervention

• 36 practices, patients (n= 151)

• mean cluster size 4, range 1-8

Intention to treat analysis

Clusters: 36 practices, 151 patients

Allocation

Analysis

Follow-Up

Enrolment

Randomized (74 practices (n=74) with eligible patients (n = 266)

Did not complete follow-up (n= 9)

Withdrawn (n=5)

• no reason given (n=4)

• patient developed cancer (n=1)

Patient deceased (n=1) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=3)

Did not complete follow-up (n=9)

Withdrawn (n=2)

• no reason given (n=1)

• patient pregnant (n=1) Patient deceased (n=2) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=5)

Page 26 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

21

Table 1: Baseline characteristics of participating practices, health professionals and adults

with type 2 diabetes1

Intervention Control

Primary care practices 36 (49) 38 (51)

Type of practice

Private practice . 27 (75.0) 31 (81.6)

Corporate practice . 7 (19.4) 5 (13.2)

Community health centre . 2 (5.6) 2 (5.3)

Location of practice

Major city . 26 (72.2) 21 (55.3)

Inner regional area . 9 (25.0) 13 (34.2)

Outer regional area . 1 (2.8) 4 (10.5)

Physicians per practice (median (IQR)) 5 (4, 9.5) 5 (4, 9)

Practice Nurses per practice (median (IQR)) 2.5 (2, 3.5) 2 (1, 4)

Registered Nurse-Credentialed Diabetes

Educator on site: yes

. 12 (33.3) 14 (36.8)

Patients per Full-Time Equivalent GP

(median (IQR))1*

1738 (1176, 2727) 1316 (911, 1726

General Practitioners 83 (51.2) 79 (48.8)

Age, years2

. 48.8 (9.9) 49.7 (11.2)

Female . 34 (41.0) 27 (34.2)

Working hours/week2 . 36.6 (10.5) 37.3 (11.6)

Years of experience (median (IQR)) 19 (8, 26) 20 (7, 30)

Experience with insulin initiation in the

preceding 12 months3

. 48 (60.0) 36 (46.2)

Practice Nurses 48 (46.6) 55 (53.4)

Age, years4

. 44.7 (10.2) 46.0 (9.9)

Female 48 (100) 55 (100)

Diabetes educator training . 6 (12.5) 7 (12.7)

Experience with insulin initiation in the

preceding 12 months

. 16 (33.3) 16 (29.1)

Adults with type 2 diabetes 151 (56.8) 115 (43.2)

Age, years . 61.7 (9.7) 62.0 (10.6)

Female . 62 (41.1) 41 (35.7)

Highest level of education

Page 27 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

22

Primary or less . 14 (9.3) 12 (10.4)

Secondary or trade . 101 (66.9) 83 (72.2)

Tertiary . 36 (23.8) 20 (17.4)

Employed . 67 (44.4) 50 (43.5)

Health care card holder . 75 (49.7) 62 (53.9)

Diabetes duration, years (median (IQR)) . 8 (5, 12) 9 (5, 14)

HbA1c % (median (IQR)

mmol/mol (median (IQR))

. 8.7 (8.1, 9.7)

72 (65, 83)

8.5 (8, 9.6)

69 (64, 81)

Number of medical conditions (median

(IQR))

3 (2, 5) 3 (2, 5)

Number of medications (median (IQR)) 6 (5, 10) 7 (5, 10)

Medication adherence rating scale (median

IQR))5

29 (26, 30) 29 (27, 30)

Diabetes complications6

Microvascular . 17 (11.3) 16 (13.9)

Macrovascular . 22 (14.6) 21 (18.3)

Total cholesterol, mmol/L7

. 4.3 (1.0) 4.2 (1.1)

Triglycerides, mmol/L7* . 1.9 (0.1) 2.3 (1.4)

LDL cholesterol, mmol/L8 . 2.3 (0.9) 2.1 (0.9)

HDL cholesterol, mmol/L9 . 1.2 (0.3) 1.1 (0.3)

eGFR10

. 79.4 (14.4) 78.8 (14.6)

Blood pressure

Systolic . 134.6 (15.7) 133.5 (15.2)

Diastolic . 79.6 (11.1) 78.5 (9.5)

Data are mean (SD) or n (%) unless otherwise stated

1 Data available for 67 practices (33 intervention, 34 control)

2 Data available for 161 GPs (82 intervention, 79 control)

3 Data available for 158 GPs (84 intervention, 74 control)

4 Data available for 100 Practice Nurses (46 intervention, 54 control)

5 Data available for 261 patients (149 intervention, 112 control)

6 n(%) with at least one complication

7 Data available for 256 patients (144 intervention, 112 control)

8 Data available for 222 patients (130 intervention, 92 control)

9 Data available for 233 patients (134 intervention, 99 control)

10 Data available for 261 patients (147 intervention, 114 control)

Page 28 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

23

*Statistically significant difference between control and intervention groups

Registered Nurse-Credentialed Diabetes Educator: Registered Nurse-Credentialed Diabetes Educator

Page 29 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only24

Table 2: Primary and secondary endpoints: Biochemical, clinical and psychological outcomes

Endpoints Intervention Control Adjusted data for clustering

Treatment effect (95% CI) p

HbA1c %

Baseline 8.7 (8.1, 9.7) 8.5 (8, 9.6)

Follow-up 7.4 (6.9, 8.2) 8 (7.1, 9)

Change -1.3(1.4) -0.6(1.5) -0.7 9 (-1.1 to -0.4) <0.001

Participants using insulin+

Follow-up 105(69.5) 25 (21.7) 8.3* (4.5 to 15.4) <0.001

Participants with HbA1c ≤53mmol/mol (7%)+ Follow-up 54 (35.8) 24 (20.9) 2.2* (1.2 to 4.3) 0.02

Depressive symptoms (PHQ-9)1 Baseline 3(1,7) 2(1, 6.5)

Follow-up

2(0, 5) 2(0, 5)

Change -1.1(3.5) -0.1(2.9) -1.0 (-1.8 to -0.1) 0.02

Moderate-to-severe depressive symptoms (PHQ-9 total: ≥10)2+

Baseline 22(15.1) 15(13.5)

Follow-up 19(12.8) 15 (13.3) 1.0* (0.7 to 2.0) 0.92

Diabetes-specific distress (PAID)3 Baseline 15(6.3, 31.3) 12.5(5, 23.8)

Follow-up 8.8(3.8, 22.5) 10(2.5, 23.8)

Page 30 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only25

Change -5.6(15.5) -2.4(15.4) -3.3 (-7.2 to 0.6) 0.10

Severe diabetes-specific distress (PAID total: ≥40)2+

Baseline 25 (16.8) 14(12.4)

Follow-up 18 (12.1) 12(10.4) 0.2* (-0.6 to 0.9) 0.68

Health status (AQoL-8D) – Physical Component Score4

Baseline 0.63(0.20) 0.610.21

Follow-up 0.66(0.21) 0.640.21 0.52

Change 0.03(0.15) 0.02(0.13) 0.01 (-0.03 to 0.04) 0.75

Health status (AQoL-8D) – Mental Component Score4

Baseline 0.45(0.20) 0.45(0.22)

Follow-up 0.48(0.21) 0.45(0.22)

Change 0.04(0.16) -0.002(0.13) 0.04 (0.001 to 0.08) 0.05

Weight, kg Baseline 90.8(19.6) 94.6(18.9)

Follow-up 92.5(20.1) 93.5(18.9)

Change 1.7 (5.2) -1.1 (5.1) 2.8 (1.6 to 4.1) <0.001

Data are mean (SD) and median (IQR) unless otherwise indicated

1 PHQ-9: Patient Health Questionnaire 9. Range of possible scores: 0-27. A total score of ≥10 indicates at least moderate depressive symptoms. Data available for 261

patients at baseline (149 intervention, 112 control) and 263 at 12 months (149 intervention, 114 control; ITT)

2 Data available for 257 patients at baseline (146 intervention, 111 control) and 261 at 12 months (148 intervention, 113 control; ITT).

Page 31 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only26

3 PAID: Problem Areas In Diabetes. Range of possible scores: 0-100. A score of ≥40 indicates severe diabetes-related distress. Data available for 262 patients at baseline

(149 intervention, 113 control) and 264 at 12 months (149 intervention, 115 control; ITT)

4 AQoL-8D: Assessment of Quality Of Life. Maximum possible score is 1. Higher scores indicate better generic health status. Data available for 262 patients at baseline (149

intervention, 113 control) and 263 at 12 months (149 intervention, 114 control; ITT)

*Odds Ratio

+n (%)

Page 32 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only27

Figure 2: Change in primary endpoint at 6 and 12 months

Table 3: Classes of non-insulin medications at 12 months

Intervention n(%) Control n(%) P value

N 146 108

Metformin 133 (91.1) 96 (88.9) 0.56

Sulphonylurea 75 (51.4) 64 (59.3) 0.21

Acarbose 3 (2.1) 2 (1.9) 0.91

DPP4 inhibitors* 25 (17.1) 38 (35.2) 0.001

Glitazones 6 (4.1) 5 (4.6) 0.8411

Page 33 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only28

SGLT2 inhibitors 2 (1.4) 2 (1.9) 0.76

GLP1 agonists 9 (6.2) 7 (6.5) 0.92

(t test of proportions)

* At the time of the trial DPP4 inhibitors were not subsidized for use with insulin

Page 34 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

29

Authorship: JF, IB, DY and JB conceptualised the original study proposal and secured funding. JF had

overall responsibility for the study. JF, DON, JS, JMN, EHT, LG, DY, JB, EP, DL, LS, CM and IB drafted

the protocol. LG also contributed to data collection. JF, DON, JS, JMN, AG, EHT, DL and IB contributed

to the statistical analysis plan and JMN and AG led the data analysis. JF wrote the original draft of

the report. All authors contributed to data interpretation and approved the final report. As

corresponding author and Principal Investigator, JF had full access to all the data in the study and

takes responsibility for the integrity of the data and the accuracy of the data analysis. JF had final

responsibility for the decision to submit for publication and is guarantor for the study affirming that

the manuscript is an honest, accurate, and transparent account of the study being reported; that no

important aspects of the study have been omitted; and that any discrepancies from the study as

planned have been explained.

Competing interests: All authors have completed the ICMJE uniform disclosure form at

www.icmje.org/coi_disclosure.pdf and declare: We acknowledge funding from the Australian

National Health and Medical Research Council (Project Grant Application: APP1023738). The study

was also supported by an educational/research grant by Roche Diabetes Care Pty Ltd, the RACGP

Foundation RACGP/Independent Practitioner Network Pty Ltd (IPN) Grant and received in-kind

support from Sanofi. Xclinical hosted the BG data. JF was supported by a National Health and

Medical Research Council Career Development Fellowship. JS is supported by core funding to The

Australian Centre for Behavioural Research in Diabetes from Diabetes Victoria and Deakin University.

JMN was supported by a National Health and Medical Research Council postgraduate scholarship.

EHT is supported by an Australian Postgraduate Award Deakin University PhD scholarship.

JF has received unrestricted educational grants for research support from Roche Diabetes Care,

Sanofi and Medtronic; JS is a member of the Accu-Check Advisory Board (Roche Diabetes Care). Her

research group (ACBRD) has received unrestricted educational grants from Medtronic and Sanofi

Page 35 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

30

Diabetes; sponsorship to host or attend educational meetings from Lilly, Medtronic, MSD, Novo

Nordisk, Roche Diabetes Care, and Sanofi Diabetes; consultancy income from Abbott Diabetes Care,

Astra Zeneca, Roche Diabetes Care and Sanofi Diabetes; DNO, DL and JMN had various financial

relationships with pharmaceutical industries outside the submitted work including consultancies,

grants, lectures, educational activities and travel. DNO has received research and travel support and

honoraria from Sanofi, Roche and Novo and is an advisory board member to Sanofi, Novo and

Abbott. JMN has no financial relationships with companies marketing blood glucose monitoring

devices, but has received payment from Sanofi who funded the control practice training at end of

study. DL has received honoraria and research grants from Sanofi Australia. The study sponsors had

no role in the design and conduct of the study; collection, management, analysis, and interpretation

of the data; and preparation, review, or approval of the manuscript.

Funding

The study was funded by the Australian National Health and Medical Research Council (ID 1023738)

and was supported by an educational/research grant by Roche Diagnostics Australia Pty Ltd.

We thank the GPs, Practice Nurses and people with type 2 diabetes who generously participated in

the study.

Data sharing: Anonymised patient level data are available on reasonable request from the authors.

Page 36 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

31

References

1. Zimmet PZ, Magliano DJ, Herman WH, et al. Diabetes: a 21st century challenge. The Lancet

Diabetes & Endocrinology 2014;2(1):56-64.

2. Holman RR, Paul SK, Bethel MA, et al. 10-Year Follow-up of Intensive Glucose Control in Type 2

Diabetes. N Engl J Med 2008;359(15):1577-89.

3. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Patient-Centered Approach: Position Statement of the American Diabetes Association (ADA)

and the European Association for the Study of Diabetes (EASD). Diabetes Care

2012;35(6):1364-79.

4. Nathan DM, Buse JB, Davidson MB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Consensus Algorithm for the Initiation and Adjustment of Therapy: Update regarding

thiazolidinediones: a consensus statement from the American Diabetes Association and the

European Association for the Study of Diabetes. Diabetes Care 2008;31(1):173-75.

5. American Diabetes Association. Standards of Medical Care in Diabetes—2016 Abridged for

Primary Care Providers. Clinical Diabetes 2016;34(1):3-21.

6. McGuire H, Longson D, Adler A, et al. Management of type 2 diabetes in adults: summary of

updated NICE guidance. BMJ 2016;353.

7. Shah BR, Hux JE, Laupacis A, et al. Clinical Inertia in Response to Inadequate Glycemic Control: Do

specialists differ from primary care physicians? Diabetes Care 2005;28(3):600-06.

8. Blak BT, Smith HT, Hards M, et al. A retrospective database study of insulin initiation in patients

with Type 2 diabetes in UK primary care. Diabetic Medicine 2012;29(8):e191-e98.

9. Curtis B, Lage MJ. Glycemic control among patients with type 2 diabetes who initiate basal insulin:

a retrospective cohort study. Journal of Medical Economics 2014;17(1):21-31.

10. Khunti K, Damci T, Meneghini L, et al. Study of Once Daily Levemir (SOLVE™): insights into the

timing of insulin initiation in people with poorly controlled type 2 diabetes in routine clinical

practice. Diabetes, Obesity and Metabolism 2012;14(7):654-61.

11. Davis TME, Davis WA, Bruce DG. Glycaemic levels triggering intensification of therapy in type 2

diabetes in the community: the Fremantle Diabetes Study. Medical Journal of Australia

2006;184(7):325-8.

12. Phillips LS, Branch JWT, Cook CB, et al. Clinical Inertia. Annals of Internal Medicine

2001;135(9):825-34.

13. Khunti S, Davies M, Khunti K. Clinical inertia in the management of type 2 diabetes mellitus: a

focused literature review. The British Journal of Diabetes and Vascular Disease

2015;15(2):65-69.

14. Kunt T, Snoek FJ. Barriers to insulin initiation and intensification and how to overcome them.

International Journal of Clinical Practice 2009;63:6-10.

15. Furler J, Spitzer O, Young D, et al. Barriers and enablers to timely initiation of insulin in General

Practice" Australian Family Physician 2011;40(8):617-23.

16. Polonsky WH, Fisher L, Guzman S, et al. Psychological Insulin Resistance in Patients With Type 2

Diabetes: The scope of the problem. Diabetes Care 2005;28(10):2543-45.

17. Furler JS, Blackberry ID, Walker C, et al. Stepping up: a nurse-led model of care for insulin

initiation for people with type 2 diabetes. Family Practice 2014;31(3):349-56.

18. May C, Mair F, Finch T, et al. Development of a theory of implementation and integration:

Normalization Process Theory. Implementation Science 2009;4(1):29.

19. Blackberry ID, Furler JS, Ginnivan LE, et al. An exploratory trial of basal and prandial insulin

initiation and titration for type 2 diabetes in primary care with adjunct retrospective

continuous glucose monitoring: INITIATION study. Diabetes Research and Clinical Practice

2014;106(2):247-55.

Page 37 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

32

20. Hajos TRS, Pouwer F, de Grooth R, et al. Initiation of insulin glargine in patients with Type 2

diabetes in suboptimal glycaemic control positively impacts health-related quality of life. A

prospective cohort study in primary care. Diabetic Medicine 2011;28(9):1096-102.

21. Dzida G, Karnieli E, Svendsen AL, et al. Depressive symptoms prior to and following insulin

initiation in patients with type 2 diabetes mellitus: Prevalence, risk factors and effect on

physician resource utilisation. Primary Care Diabetes;9(5):346-53.

22. Furler J, Young D, Best J, et al. Can primary care team-based transition to insulin improve

outcomes in adults with type 2 diabetes: the stepping up to insulin cluster randomized

controlled trial protocol. Implementation Science 2014;9(1):20.

23. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting

parallel group randomised trials. BMJ 2010;340:698-702.

24. Giraudeau B, Ravaud P. Preventing Bias in Cluster Randomised Trials. PLoS Medicine

2009;6(5):e1000065.

25. Blackberry ID, Furler JS, Best JD, et al. Effectiveness of general practice based, practice nurse led

telephone coaching on glycaemic control of type 2 diabetes: the Patient Engagement And

Coaching for Health (PEACH) pragmatic cluster randomised controlled trial. BMJ 2013;347.

26. Polonsky WH, Fisher L, Schikman CH, et al. Structured self-monitoring of blood glucose

significantly reduces A1C levels in poorly controlled, noninsulin-treated type 2 diabetes:

results from the Structured Testing Program study. Diabetes Care 2011;34(2):262-7.

27. The Royal Australian College of General Practitioners and Diabetes Australia. General practice

management of type 2 diabetes – 2014–15. Melbourne, , 2014.

28. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. Validity of a Brief Depression Severity Measure.

Journal of General Internal Medicine 2001;16(9):606-13.

29. Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in Diabetes Scale - An evaluation of

its clinical utility. Diabetes Care 1997;20(5):760-66.

30. Richardson J, Iezzi A, Khan M, et al. Validity and Reliability of the Assessment of Quality of Life

(AQoL)-8D Multi-Attribute Utility Instrument. Patient 2014;7(1):85-96.

31. The National Diabetes Service Scheme. Secondary The National Diabetes Service Scheme.

https://www.ndss.com.au/the-ndss.

32. Speight J, Browne JL, Holmes-Truscott E, et al. Diabetes MILES - Australia 2011 Survey Report.

Canberra: Diabetes Australia, 2011.

33. Elliott L, Fidler C, Ditchfield A, et al. Hypoglycemia Event Rates: A Comparison Between Real-

World Data and Randomized Controlled Trial Populations in Insulin-Treated Diabetes.

Diabetes Therapy 2016;7(1):45-60.

34. Frier BM, Jensen MM, Chubb BD. Hypoglycaemia in adults with insulin-treated diabetes in the

UK: self-reported frequency and effects. Diabetic Medicine 2016;33(8):1125-32.

35. Harris S, Gerstein H, Yale J-F, et al. Can community retail pharmacist and diabetes expert support

facilitate insulin initiation by family physicians? Results of the AIM@GP randomized

controlled trial. BMC Health Services Research 2013;13(1):71.

36. Dale J, Martin S, Gadsby R. Insulin initiation in primary care for patients with type 2 diabetes: 3-

Year follow-up study. Primary Care Diabetes 2010;4(2):85-89.

37. Pouwer F, Hermanns N. Insulin therapy and quality of life. A review. Diabetes-Metabolism

Research and Reviews 2009;25:S4-S10.

38. Gucciardi E, Espin S, Morganti A, et al. Exploring interprofessional collaboration during the

integration of diabetes teams into primary care. BMC Family Practice 2016;17(1):1-14.

39. O'Connor PJ, Desai JR, Butler JC, et al. Current Status and Future Prospects for Electronic Point-

of-Care Clinical Decision Support in Diabetes Care. Current Diabetes Reports 2013;13(2):172-

76.

Page 38 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

1

Supporting insulin initiation in type 2 diabetes in primary care: Results of the

Stepping Up pragmatic cluster randomised controlled clinical trial

JS Furler, DN O’Neal, J Speight, J Manski-Nankervis, A Gorelik, E Holmes-Truscott, L Ginnivan, D

Young, J Best, E Patterson, D Liew, L Segal, CR May, I Blackberry,

JS Furler, Associate Professor, Department of General Practice, University of Melbourne, DN O’Neal,

Associate Professor, Department of Medicine, St Vincent’s Hospital, University of Melbourne, J

Speight, Professor, School of Psychology, Deakin University; Director, The Australian Centre for

Behavioural Research in Diabetes, Diabetes Victoria; AHP Research, United Kingdom, J Manski-

Nankervis, Lecturer, Department of General Practice, University of Melbourne, A Gorelik, Senior

Statistician, Melbourne EpiCentre, the University of Melbourne, E Holmes-Truscott, Research Fellow,

BPSc (Hons), School of Psychology, Deakin University; The Australian Centre for Behavioural

Research in Diabetes, Diabetes Victoria; L Ginnivan, Research Nurse and Credentialed Diabetes

Educator, of General Practice, University of Melbourne, D Young, Professor and Chair of General

Practice, and Associate Dean (Academic), Melbourne Medical School, University of Melbourne,

Australia, J Best, Professor and Dean, Lee Kong Chian School of Medicine, Nanyang Technological

University, Singapore, E Patterson, Professor, School of Nursing, University of Melbourne, Australia,

D Liew, Professor, School of Public Health and Preventive Medicine, Monash University, Australia, L

Segal, Professor, Health Economics and Social Policy Group, Division of Health Sciences, University of

South Australia, CR May, Professor of Health Care Innovation, Faculty of Health Sciences, University

of Southampton, UK., I Blackberry, Associate Professor and Director, John Richards Initiative,

Australian Institute for Primary Care & Ageing, College of Science, Health and Engineering, La Trobe

University, Australia,

Correspondence to: John Furler, Department of General Practice, University of Melbourne, 200

Berkeley St, Carlton, VIC, 3053, Austraila, [email protected]

Page 39 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

2

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of

all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats

and media (whether known now or created in the future), to i) publish, reproduce, distribute, display

and store the Contribution, ii) translate the Contribution into other languages, create adaptations,

reprints, include within collections and create summaries, extracts and/or, abstracts of the

Contribution, iii) create any other derivative work(s) based on the Contribution, iv) to exploit all

subsidiary rights in the Contribution, v) the inclusion of electronic links from the Contribution to

third party material where-ever it may be located; and, vi) licence any third party to do any or all of

the above.

Word count: 4142

Page 40 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

3

Abstract

Objective

Optimising glycated haemoglobin to minimise risk of long-term complications among people with

type 2 diabetes is a global priority. Supporting timely insulin initiation through innovation in primary

care could have a major impact. Our aim was to test the effectiveness of the ‘Stepping Up’ model of

care compared to usual primary care in normalising insulin initiation as part of routine primary care

practice for type 2 diabetes, leading to improved HbA1c levels. Our hypothesis was that HbA1c

would improve among participants in intervention arm practices, facilitated through timely insulin

initiation, compared to the control arm.

Design

A two-arm, 12-month cluster-randomised controlled trial of the Stepping Up model of care.

Setting

Primary care practices in Victoria, Australia with a Practice Nurse and at least one consenting eligible

patient (HbA1c ≥7.5% on maximal oral therapy).

Participants

74 practices and 266 patients participated: mean (range) cluster size 4 (1 to 8) patients.

Intervention

The Stepping Up model of care intervention involved theory-based practice system change and re-

orientation in the roles of the health professionals in the primary care diabetes team. The core

component was an enhanced role for the Practice Nurse Practice Nurse in leading insulin initiation,

and mentoring by a Registered Nurse-Credentialed Diabetes Educator.

Main outcome and measures

Our clinically meaningful primary endpoint was change in glycated hemoglobin (HbA1c). Secondary

endpoints included the proportion of participants who transitioned to insulin, proportion who

Page 41 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

4

achieved target HbA1c as well as change in depressive symptoms (PHQ-9), diabetes-specific distress

(PAID) and generic health status (AQoL-8D).

Results

HbA1c improved in both arms, with a clinically significant between-arm difference (mean difference

-0.7%; 95%CI -1.1 to -0.4%), favouring the intervention. At 12 months, in intervention practices,

105/151 (69.5%) adults with type 2 diabetes had commenced insulin (102 remained on insulin at 12

months); in control practices 25/115 (21.7%) commenced insulin (and 24 remained on insulin at 12

months) (OR 8.3, 95%CI (4.5 to 15.4), p<0.001). Target HbA1c (≤7% (53mmol/mol)) was achieved by

54 (35.8%) intervention participants (32 of whom were using insulin at 12 months) and 22 (19%) of

the control arm (two of whom had commenced insulin), (OR 2.2, 95%CI (1.2 to 4.3), p=0.0082).

Depressive symptoms did not worsen at 12 months in the intervention arm (mean difference in

PHQ-9:= -1.1 (3.5) versus -0.1 (2.9), 95%CI -1.8 to -0.1, p=0.021 favouring intervention). There was

noa statistically significant difference between arms in the mean (SD) change in mental health

component of the AQoL (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13)), p=0.05), favoring the

intervention, but no significant difference in diabetes-specific distress (PAID: 5.6 (15.5) versus -2.4

(15.4)) nor physical health (AQoL PCS: 0.03 (0.16) versus 0.02 (0.13)) or diabetes-specific distress

(PAID: 5.6 (15.5) versus -2.4 (15.4)). No severe hypoglycaemia events were reported.

Conclusions

Our novel model of care increased insulin initiation rates in primary care, improving glycated

hemoglobin without worsening emotional well-being. Delays in starting insulin treatment can be

overcome, enabling timely treatment intensification to improve important health outcomes.

Trial registration Australian and New Zealand Clinical Trials Registry (ACTRN12612001028897)

http://www.anzctr.org.au/

Page 42 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

5

What this paper adds

What is already known on this subject?

• Achieving and maintaining glycaemic targets early in type 2 diabetes improves long term

outcomes.

• There are barriers to early stepwise progressive treatment intensification to achieve glycaemic

targets, particularly in relation to insulin initiation in primary care.

• Interventions have had limited success in overcoming this delay in starting insulin treatment and

changing clinical practice, in part because system level barriers are not addressed.

What this study adds

• Our model of care intervention changed clinical practice, with more patients in intervention arm

practices commencing insulin, with an overall benefit in terms of HbA1c reduction, achieved

without serious adverse events, or any worsening in depressive symptoms.

• The theoretical base and flexible implementation are important characteristics of our

intervention.

• Our model of care used existing resources to work smarter and improve outcomes and, thus, has

important implications for policy makers, funders and practitioners seeking innovative ways to

provide best care for people with type 2 diabetes in primary care.

Page 43 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

6

Introduction

Nearly 600 million people will have type 2 diabetes type 2 diabetes worldwide by 2030.1 Innovation

in delivering effective clinical care to these people is an urgent global priority. To reduce the risk of

long-term macro- and microvascular complications,2 UK, European and US guidelines recommend

early adoption of insulin as a part of step-wise treatment intensification to bring glycated

hemoglobin (HbA1c) below a general target of 7% (53mmol/mol).3-6

However, insulin initiation is

frequently delayed, particularly in primary care,7 where despite being recommended as part of

routine clinical management of type 2 diabetes6, implementation is not widespread. The mean

HbA1c of people with type 2 diabetes prior to starting insulin is typically 1.5-2.0% above target: 9.3%

(78mmol/mol) in the UK8 8.6% (70mmol/mol) in a study in the US

9, 8.9% (74mmol/mol) in a large,

multi-country primary care study10

, and 9.4% (79mmol/mol) in a community study in Australia (after

a median diabetes duration of 8.1 years)11

.

Delay in treatment intensification by healthcare professionals, despite evidence that intensification

is warranted and effective12

is a major barrier to initiating insulin in type 2 diabetes13

. It can be due

to health professional factors14

(e.g. concerns about hypoglycaemia risk, lack of confidence/skills in

insulin initiation/titration), health system factors15

(e.g. competing priorities in busy, reactive

primary care settings), and patient-related factors (e.g. psychological insulin resistance16

).

Supporting and embedding insulin initiation as an element of routine primary care practice is an

important first step and building block in potentially reducing referrals to costly secondary care, and

supporting timely, early optimisation of therapy and achievement of glycaemic targets.

Our trial investigated the effectiveness of a new model of care designed to support primary care-

based insulin initiation among people with type 2 diabetes for whom it is clinically indicated. The

Stepping Up model of care is built around an enhanced, reconfigured role for the primary care

Practice Nurse Practice Nurse.17

In Australia, clinical care of type 2 diabetes is predominantly

undertaken in primary care, largely made up of General Practitioner private group practices where

Page 44 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

7

Practice Nurses are a rapidly growing section of the workforce, taking on an increasing role in

chronic condition management. In this setting, our approach to professional behaviour change was

informed by Normalisation Process Theory18

and developed through pilot studies.17 19

Our aim was to

address both clinician and system level barriers to timely insulin initiation, and to normalise insulin

initiation as part of standard primary care practice. We hypothesised that HbA1c would improve

among participants in intervention arm practices, facilitated through timely insulin initiation,

compared to with the control arm. Further, based on previous research 20 21

, we expected no

significant negative impact on participants’ general emotional wellbeing (depressive symptoms).

Methods

Study design and participants

The study design and protocol have been described previously.22

In summary, we conducted a 12-

month, two-arm, non-blinded cluster-randomised controlled trial (RCT), consistent with CONSORT

guidelines,23

to investigate the effectiveness of the Stepping Up model of care versus usual care.

General Practices in Victoria, Australia were eligible if they had at least one consenting General

Practitioner (GP) and Practice Nurse and could identify at least one eligible patient participant:

adults with type 2 diabetes with above target HbA1c (≥7.5% (58mmol/mol)) in the past six months

who were already prescribed maximum oral therapy (i.e. at least two oral hypoglycaemia agents

(OHAs) at maximum doses) or if their GP judged that insulin would be clinically appropriate. Patients

were ineligible if they were >80 years old, already using insulin, had an eGFR <30 mL/min/1.73m2,

unable to give informed consent or had a complex debilitating medical condition, e.g., severe mental

illness, end-stage cancer, unstable cardiovascular disease.

Our original protocol was based on 58 practices and an average cluster size of 5. Based on

experience in the field and what was an achievable sample size this was subsequently revised to 74

practices and an average cluster size of 3 (see Appendix 1).

Page 45 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

8

Randomisation

The unit of randomisation was the primary care practice. The study statistician (AG) computer-

generated stratified block randomisation sequences with varying block sizes (4, 6 and 8) prior to

recruitment. Practices were stratified by size (≤2 vs >2 full-time equivalent GPs), setting (private

practice vs community health centre), and participation (or not) in type 2 diabetes quality

improvement programs (the Australian Primary Care Collaborative). After providing consent and

recruiting at least one eligible patient, practices were randomised to intervention or usual care. We

used this index case method24

in all practices as our previous experience suggested that delaying

randomisation of a cluster until all patients have been recruited risks loss of engagement of GPs25

.

The research team then assisted practices to continue to identify and recruit patient participants

(through searching the practice medical record database). This meant that allocation concealment

after the index case was recruited was not possible for the GP and Practice Nurse. However, in order

to minimise potential bias, participating patients were not informed of their study allocation until

after they had provided consent. Blinding of GP, Practice Nurse and patient was also not possible

given the pragmatic nature of the intervention.

Intervention

The Stepping Up model of care, described elsewhere, 17 22

involved a re-orientation of existing

resources. It included: a) an enhanced role for the Practice Nurse in leading the discussion with

patients about intensifying treatment through insulin initiation and titration, b) simple insulin

initiation and up-titration clinical protocols, and c) a re-oriented role for the specialist Registered

Nurse-Credentialed Diabetes Educator in mentoring the Practice Nurse rather than providing direct

patient care. The training and mentoring support was designed to enhance the primary care team’s

knowledge, skills and confidence in discussing and implementing insulin initiation within the practice

as a part of routine care. Intervention practices had an in-practice briefing and training session for

GPs and Practice Nurses of approximately 60-90 minutes, following which patients with confirmed

Page 46 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

9

eligibility and completed baseline data were invited to consult their GP for an assessment to discuss

treatment intensification and referral to the Practice Nurse. Practice Nurses did not prescribe insulin,

or manage insulin dosing without liaison with the GP, based on the legal scope of practice for

generalist Practice Nurses in Australia.

Our model of care involved the acknowledgment and discussion of advantages and disadvantages of

starting insulin treatment with patients, including weight gain. We modelled shared decision making

as a part of the intervention training, drawing on the principles of motivational interviewing. This set

the scene for encouraging practitioners in intervention practices to approach participating patients

with equipoise in relation to starting insulin. While the intervention was necessarily brief in this

pragmatic trial, we included guidance and checklists for GPs and Practice Nurses to discuss the pros

and cons of insulin therapy and elicit patient concerns and expectations, while also openly

acknowledging and accepting that some patients may choose not to start insulin.

The role of Registered Nurse-Credentialed Diabetes Educator in supporting and mentoring the

Practice Nurse, and of the Practice Nurse in leading the discussion and implementation of insulin

therapy with the patient in liaison with the GP, is outlined elsewhere.15,20

Titration protocols were

based on fasting blood glucose and use of a 3-day, 7-point blood glucose profile26

to identify the

meal with the largest postprandial excursion (see appendix material). We gave no additional

instructions, so the GP had clinical autonomy regarding the management of OADsOHAs. Practice

Nurses and GPs were encouraged to see patients as frequently as was felt to be clinically appropriate

over a period of up to 12 months, drawing as needed upon the study Registered Nurse-Credentialed

Diabetes Educator for mentoring and support, even if the patient remained undecided about, or had

decided against, starting insulin. Further details about the intervention can be found in the

referenced papers and the Appendices (Appendix 2 – Intervention Description; Appendix 3 –

Training Manual for Intervention Practices; Appendix 4 – Patient participant booklet).

Page 47 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

10

Control arm practices were given a copy of Australian type 2 diabetes management guidelines27

.

Control arm practices were offered training in the Stepping Up model of care after 12-month follow

up of patient participants was complete.

Patient involvement

Participant feedback was sought after conducting an intervention pilot study15

and this feedback was

used to refine the model of care. Participant feedback was also sought in a pilot of the data

collection forms. Throughout the main trial we maintained communication with patient participants

through a regular newsletter which included aggregate data about study progress and opportunities

to provide feedback to the study team. We assessed the burden of the intervention on patients

through interviews conducted at the end of the trial, as part of process evaluation (to be reported

elsewhere). This evaluation was led by the Chronic Illness Alliance, a consumer advocacy

organization that has been a long term collaborator of our research group. We have thanked all

participants for their involvement in the trial, and will provide, at a later date, a final summary report

of the trial outcomes. Participants have access to the study website where all published results will

be publicly available.

Endpoints and data collection

While our intervention targeted a process of care (i.e. insulin initiation), we chose a clinically

meaningful disease outcome at an individual patient level as primary endpoint: change (from

baseline to 12 months) in HbA1c, measured as a continuous variable. We registered our primary

outcome as “an absolute HbA1c reduction of 0.5% in the intervention group compared with the

control group”. HbA1c was performed at DCCT-aligned pathology laboratories and communicated to

clinicians and patients as part of usual clinical care; researchers retrieved these data from medical

records or directly form pathology laboratories. Secondary endpoints included the proportion of

participants who transitioned to insulin (this was amended from the original protocol where rate of

insulin initiation was used, which proved impractical given the small cluster size in the study); the

Page 48 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

11

proportion who achieved a target HbA1c of ≤7.0% (53mmol/mol) at 12 months; and change (from

baseline to 12 months) in depressive symptoms (PHQ-9),28

diabetes-specific distress (PAID),29

and

generic health status (AQoL-8D).30

Differences between the registered outcomes reported here and

those registered in the trial registry and justification and explanation for any changes made can be

found in the AppendicesAppendix 1. We also collected data on healthcare utilisation and costs, to be

reported elsewhere.

All participants were provided with a blood glucose meter (Performa NanoTM

; Roche Diagnostics)

and instructed on its use. Subsidised low-cost blood glucose testing strips were available through the

National Diabetes Service Scheme31

. Data were uploaded from the meter at 6 and 12 months to a

secure server.

Statistical analysis

Our statistical analysis plan has been published elsewhere.22

In brief, our sample size of 224 patients

from 74 general practices (averaging 3 patients per practice) allowed us to detect an absolute 0.5%

mean HbA1c difference over 12 months between control and intervention arms with 80% power and

a standard deviation (SD) of 1 using two-sided alpha of 0.05. Data were analysed using Stata 13

(StataCorp, TX, USA). Descriptive statistics were used to summarise GP, Practice Nurse, and patient

characteristics for the two study arms and appropriate univariate tests were conducted to check for

any significant difference in potential confounders. Parametric data are reported as mean (SD) and

non-parametric data as median (IQR). Categorical data are reported as n (%). T-test for proportions

was used to compare baseline and 12-month data within each study arm for categorical data, while

paired t-tests were for used for continuous parametric data and Wilcoxon matched pairs signed rank

test for continuous non-parametric data. The individual patient was the unit of analysis and the

analytical methods allow for clustering of patients within the practices. Marginal logistic modelling

using generalised estimating equations with robust standard errors and adjustment for clustering

were used to compare binary outcomes between the two study arms. Mixed-effects linear

Page 49 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

12

regression was used to determine predictors for continuous outcomes, adjusting for clustering at the

practice level. As no imbalance in confounders was identified between study groups no adjustment

for confounders was made in the regression analysis. Analyses were conducted on an intention-to-

treat basis.

Study oversight

The study protocol was approved by the University of Melbourne Health Sciences Human Research

Ethics Sub-committee (ID 123740) and registered with the Australian and New Zealand Clinical Trials

Registry (ACTRN12612001028897). All participants gave informed consent before enrolment.

Results

Participating practices and patients

Between October 2012 and January 2014, 93 primary care practices expressed interest and

identified 521 potentially eligible patients (Figure 1). Subsequently, 19 practices did not consent any

eligible patients, leaving 74 participating practices for randomisation. Two hundred and fifty five of

the potentially eligible patients were subsequently found to be ineligible at screening (n=156) or did

not respond to the invitation letter (n=99). By April 2014, the 74 practices had identified and

consented 266 eligible patient participants (73% of potentially eligible patients identified).

Baseline characteristics of practices, GPs, Practice Nurses and participants with type 2 diabetes are

shown in Table 1. Of the total sample, 248 (93.2%) completed 12-month follow-up for the primary

endpoint. No differences in baseline characteristics were observed between study completers and

non-completers, except for a higher proportion of women (n=11) not completing than men (n=7).

Primary and secondary endpoints

Primary and secondary endpoints are shown in Table 2. At 12 months, there was a statistically and

clinically significant difference between study arms in terms of change in HbA1c (mean

difference: -0.7%; 95% CI -1.1 to -0.4%, p<0.001), favouring the intervention. This is consistent with

Page 50 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

13

achieving our registered primary outcome of “an absolute HbA1c reduction of 0.5% in the

intervention group compared with the control group”. The majority of this change in HbA1c seen in

both arms was achieved by six months (See Figure 2).

In the intervention arm 105/151 (69.5%) of patients commenced insulin (and 102 remained on

insulin at 12 months), while in the control arm, 25/115 (21.7%) commenced insulin (and 24

remained on insulin at 12 months). Median (IQR) number of days from baseline assessment to

insulin initiation in intervention and control group patients who started insulin was 32 (11.5, 134.5)

days and 85 (63, 191) days respectively (statistically significant difference: Two-sample Wilcoxon

rank-sum test; p=0.005).In the intervention arm, 17 (11.22%) patients had commenced rapid-acting

insulin at 12 months, compared with one patient in the control arm (OR 14.2 95%CI 1.8 to 109.8,

p=0.011 (p<0.001). Further data on insulin use in participants is available in Appendix 5.

Target HbA1c (≤7% (53mmol/mol)) was achieved by 54 (35.8%) intervention participants (32 of

whom were using insulin at 12 months) and 22 (19%) of the control arm (two of whom had

commenced insulin), (OR 2.2 (1.2 to 4.3) p=0.00802). Twenty-two (14.7%) intervention arm

participants and 20 (17%) control arm participants achieved target HbA1c without commencing

insulin.

At 12 months, there was a statistically significant difference between arms in the mean (SD) change

in depressive symptoms (PHQ-9: -1.1 (3.5) versus -0.1 (2.9), p=0.02,) and mental health (AQoL MCS:

0.04 (0.16) versus -0.002 (0.13), p=0.05), favouring the intervention)., but There was no significant

difference in mental health (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13), diabetes-specific distress

(PAID: -5.6 (15.5) versus -2.4 (15.4)) nor physical health (AQoL PCS: 0.03 (0.16) versus 0.02 (0.13)) or

diabetes-specific distress (PAID: -5.6 (15.5) versus -2.4 (15.4)). There was no significant difference

between arms in the proportion of participants experiencing moderate-to-severe depressive

symptoms or severe diabetes-specific distress (Table 2), nor was there any difference by insulin

initiation (Wilcoxon rank sum test; p=0.98). No significant difference was found in PHQ-9, PAID or

Page 51 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

14

AQoL or PAID scores when comparing participants initiating insulin with those who did not (data not

shown).

At 12 months, there was an average weight gain in the intervention arm and an average weight loss

in the control group (1.7 (5.2) kg versus -1.1 (5.1) kg, mean (95% CI) difference 2.8 (1.6 to 4.1) kg,

p<0.001). There were no significant differences in blood pressure nor other biochemical measures

between arms at follow-up with the exception of triglycerides, which remained higher in the control

group.

At baseline, participants were using a mean (SD) of 2.0 (0.6) classes of non-insulin hypoglycaemic

agents, with no significant difference between arms (t test, p=0.89). The majority of patients were

prescribed metformin (93% across both arms) and sulfonylureas (63% across both arms). There was

no significant difference in the prescription of individual medication classes by study arm. The mean

(SD) number of classes of non-insulin hypoglycaemic agents being used at 12 months was higher in

the control compared to the intervention arm (2.3 (0.1) vs 1.9 (0.1); p=0.01). There was were a

higher proportion of people in the control group using DPP4 inhibitors than the intervention group

at 12 months (Table 3).

Practices in the intervention arm received a total of 183 mentoring support visits from the study

Registered Nurse-Credentialed Diabetes Educator (mean (range) visits per practice: 5.2 (1-8)). Thirty-

two percent (48/151) of participants in intervention practices completed at least one 3-day, 7-point

structured blood glucose monitoring profile over the 12-month study. Practice nurses estimated the

time they spent on the study (both clinical interactions with participating patients and on research

tasks). On a per-practice basis, 23 control practices and 27 intervention practices reported a median

(IQR) of 1.5 (0, 3.6) and 18 (9, 20.9) hours respectively.

58.3% of people in the control group were recruited prior to practice randomisation in the control

group andcompared to 45.0% in the intervention group (statistically significant difference p=0.033).

Sensitivity analysis was conducted to explore whether there was any difference at baseline between

Page 52 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

15

patients who were recruited before and after practice randomisation. No statistically significant

differences were found. Mixed-effects linear regression was used to determine the impact of the

intervention, adjusting for clustering at the practice level for each of these groups. Treatment effect

in terms of the primary outcome remained significant in both groups (treatment effect: -0.57 95%CI:

-1.1, -0.05 p value = 0.03 and -0.98 95%CI -1.49, -0.48 p value <0.001 for patients assessed prior to

and after randomisation respectively).

Adverse events

No severe hypoglycaemic events (i.e. requiring third-party assistance for recovery) or other adverse

events were reported in either study arm.

Discussion

Principal findings

Our model of care changed clinical practice, with the majority in the intervention arm commencing

insulin, producing a clinically and statistically significant improvement in glycaemic control among

adults with type 2 diabetes managed in primary care. This was despite a higher patient-to-GP ratio in

intervention practices, and was achieved safely, with no severe hypoglycaemic events, and without

deterioration in emotional wellbeing nor health status. Our results indicate that, with appropriate

support and practice-system redesign, insulin initiation can become part of routine primary care

diabetes management, obviating the need to refer to specialist services with geographical, cost and

accessibility barriers.

Strengths and limitations of study

A strength of our study is the robust theoretical and empirical base to our intervention. Our

pragmatic trial of a complex intervention addressed a number of known barriers to overcoming

delay in starting insulin therapy. For example, an intentional component of our system redesign was

re-orienting the Practice Nurse and Registered Nurse-Credentialed Diabetes Educator roles, allowing

Page 53 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

16

additional time to be spent with patients, within existing resources. Other strengths include the

cluster-randomised design, minimising the risk of contamination, and our excellent participant

retention (93%).

Our study had limitations. First, practices were randomised after the first consenting patient was

identified, raising the possibility of selection bias. However, the balance in key patient characteristics

between the study arms means that any such bias was minimal. Secondly, while a smaller cluster

size is generally preferable in a cluster-randomised trial, the relatively large variation in the cluster

sizes in our study may make statistical adjustments for clustering less effective, in particular when

the number of clusters is not large. SecondlyThirdly, our sample may not have been fully

representative of the broader population of adults with type 2 diabetes managed in primary care for

whom insulin is clinically indicated. Overall, less than 15% of our sample had severe diabetes-specific

distress or moderate-to-severe depressive symptoms, a lower rate than in a recent national

Australian sample.32

We will explore implementation fidelity and variation in more detail through a

qualitative process evaluation in a subsequent paper. Finally, our medications and hypoglycaemia

data were derived from GP records and subject to the same accuracy limitations of any routinely

collected clinical dataset. Hypoglycaemia is typically under-estimated 33

and is likely to be under-

reported in routine medical records. 34

In particular, severe hypoglycemia is serious but relatively

rare, and may not have been detected in our study, given our sample size.

Comparison with other studies

Only two other trials have tested interventions to change clinical practice in this way. The AIM@GP

trial showed no improvement in insulin prescribing rates or glycated hemoglobin.35

It provided

scheduled and ongoing telephone support from a specialist diabetes educator as well as the option

to refer patients to a community pharmacist off-site for a one-hour insulin initiation session. Our

intervention differed in that it was based completely in the familiar environs of the patients’ own

primary care practice, built on existing relationships and resources (with the practice-based Practice

Page 54 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

17

Nurses) and provided an immediate pathway for the GP to delegate this clinical task. A UK study,

using a pre-post evaluation design, showed improved HbA1c (-1.4%) at 6 months in patients who

initiated insulin,36

similar in magnitude to the improvement in our study. That intervention combined

education with face-to-face and telephone specialist diabetes nurse support for GPs and Practice

Nurses and involved a full-day, off-site training for both GP and Practice Nurse. In contrast, our

intervention used a brief (60-90 minutes) on-site training incorporated into the daily running of the

practice with flexible Registered Nurse-Credentialed Diabetes Educator support as required.

Consistent with previous research,37

our study participants did not report worse psychological

outcomes at follow up, suggesting that insulin therapy can be initiated in primary care for people

with type 2 diabetes without impairing their emotional wellbeing. In fact, at 12 months, the

reduction in depressive symptoms in intervention arm participants was significantly greater than

that seen in the control arm. Insulin initiation (across both arms) did not impact on depressive

symptoms, diabetes-specific distress or generic health status at 12 months.

Conclusions and policy and practice implications

Our pragmatic trial findings have important implications for the organisation of healthcare and for

health policy. Our model of care is based on an enhanced role for a Practice Nurse and would not be

feasible where primary care doctors work in solo practice without access to a Practice Nurse.

Nevertheless, even in high-resource settings, where the move to multidisciplinary primary health

care teams is growing, for example through the growth of the Medical Home movement 38, our

study suggests that simply having access to a Practice Nurse will not increase appropriate insulin

initiation. To make the best use of resources, primary care workforce models need to be developed

and implemented to reorient the way specialists (Registered Nurse-Credentialed Diabetes Educators

and endocrinologists) offer support to primary care teams that include well-supported and

resourced primary care nurses. Registered Nurse-Credentialed Diabetes Educator Change is

occurring in some health systems,38

but our study provides evidence of the effectiveness and safety

of such models of care. Rather than waiting for referrals, specialist services need to offer pro-active,

Page 55 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

18

tailored secondary consultation, liaison and mentoring services that are flexible and supportive of

the needs of primary care practitioners and patients. Scaling up the model of care in metropolitan

centres would require engaging with hospitals and other health services in reorienting the role of

Registered Nurse-Credentialed Diabetes Educators currently employed in direct patient care.

Investigation of e-health modalities (e.g. online training, support and video consultations) may

support implementation of this model of care in more distant rural/remote settings. Registered

Nurse-Credentialed Diabetes Educator

Our trial findings also have implications for clinical practice. The issue around personalising

glycaemic targets and treatments is an important and emerging consideration in the care of people

with T2D. At the time of our trial started up, there was vigorous debate about the need for caution

in setting lower targets. In addition, our exclusion criteria ruled out participants wherefor whom a

higher target would definitely be considered. Collecting reliable data on duration of CVD, severe

hypoglycaemia and hypoglycaemia unawareness did not prove feasible in this setting. Therefore we

made a pragmatic decision to use the general target of 7%. Debate continues about the advantages

and disadvantages of intensifying treatment for people with type 2 diabetes at the HbA1c levels

mandated in our study. While our findings suggest the that Stepping Up model of care builds clinical

capacity within GP and Practice Nurse teams to undertake the work of insulin initiation, the model of

care did not mandate a dogmatic approach to such therapy changes. It is worth noting the major

change in HbA1c was achieved at 6 months and that even in intervention practices, only 35% of

participating people with type 2 diabetes achieved the general HbA1c target of <7% (53 mmol/mol),

suggesting that that practitioners and patients were judicious in the way they approached

progressive treatment intensification, within the new model of care.

Our clinical protocols and algorithms were focused solely on insulin. It is worth noting that NPH

Insulin remains widely used and that the added costs of analogue insulins where NPH can be used

without problem is still subject to debate. While we did not pre-specify weight gain as a secondary

outcome, this is an adverse effect of insulin therapy and we have chosen to report it. While insulin is

Page 56 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

19

still regarded as an essential therapy option, as the range of glycaemic therapies grows clinical

algorithms become more complex5. Future research needs to explore the capacity to generate real-

time personalised treatment intensification recommendations that incorporate this increasing

complexity,39

to be used as part of the Practice Nurse-led model of care. Future research could also

address the extent to which improvements in glycaemic levels are maintained, and the extent to

which the model of care is sustained in routine clinical practice. In particular, future research could

explore use of the model of care to specifically support early adoption of insulin therapy to achieve

glycaemic targets early in people with recently diagnosed type 2 diabetes.

The global epidemic of type 2 diabetes demands innovation in care delivery. Delaying insulin

initiation when clinically indicated is neither ethical nor effective. Furthermore, health systems will

not cope with demand if insulin initiation remains anchored in specialist centres, nor will they be

able to respond to the imperative to achieve glycaemic targets early in people with recently

diagnosed type 2 diabetes. Thus our pragmatic, translational study has important implications across

health systems globally for the organisation of care for people with type 2 diabetes. Our effective

model of care has the potential to improve outcomes in people with type 2 diabetes while making

better use of scarce healthcare resources.

Page 57 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

20

Figure 1: Consort Diagram

Assessed for eligibility (93 practices, patients n=521)

Excluded: Practices - failed to recruit patients – n = 19 Patients (n=255) ♦ Ineligible (n=156); ♦ Non-responder (n=99)

Intention to treat analysis

Clusters: 38 practices, 115 patients

Allocated to control arm (n=38 practices) Received allocated control

• 38 practices, patients (n=115)

• mean cluster size 3, range 1-6

Allocated to intervention arm (n=36 practices) Received allocated intervention

• 36 practices, patients (n= 151)

• mean cluster size 4, range 1-8

Intention to treat analysis

Clusters: 36 practices, 151 patients

Allocation

Analysis

Follow-Up

Enrolment

Randomized (74 practices (n=74) with eligible patients (n = 266)

Did not complete follow-up (n= 9)

Withdrawn (n=5)

• no reason given (n=4)

• patient developed cancer (n=1)

Patient deceased (n=1) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=3)

Did not complete follow-up (n=9)

Withdrawn (n=2)

• no reason given (n=1)

• patient pregnant (n=1) Patient deceased (n=2) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=5)

Page 58 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

21

Table 1: Baseline characteristics of participating practices, health professionals and adults

with type 2 diabetes1

Intervention Control

Primary care practices 36 (49) 38 (51)

Type of practice

Private practice . 27 (75.0) 31 (81.6)

Corporate practice . 7 (19.4) 5 (13.2)

Community health centre . 2 (5.6) 2 (5.3)

Location of practice

Major city . 26 (72.2) 21 (55.3)

Inner regional area . 9 (25.0) 13 (34.2)

Outer regional area . 1 (2.8) 4 (10.5)

Physicians per practice (median (IQR)) 5 (4, 9.5) 5 (4, 9)

Practice Nurses per practice (median (IQR)) 2.5 (2, 3.5) 2 (1, 4)

Registered Nurse-Credentialed Diabetes

Educator on site: yes

. 12 (33.3) 14 (36.8)

Patients per Full-Time Equivalent GP

(median (IQR))1*

1738 (1176, 2727) 1316 (911, 1726

General Practitioners 83 (51.2) 79 (48.8)

Age, years2

. 48.8 (9.9) 49.7 (11.2)

Female . 34 (41.0) 27 (34.2)

Working hours/week2 . 36.6 (10.5) 37.3 (11.6)

Years of experience (median (IQR)) 19 (8, 26) 20 (7, 30)

Experience with insulin initiation in the

preceding 12 months3

. 48 (60.0) 36 (46.2)

Practice Nurses 48 (46.6) 55 (53.4)

Age, years4

. 44.7 (10.2) 46.0 (9.9)

Female 48 (100) 55 (100)

Diabetes educator training . 6 (12.5) 7 (12.7)

Experience with insulin initiation in the

preceding 12 months

. 16 (33.3) 16 (29.1)

Adults with type 2 diabetes 151 (56.8) 115 (43.2)

Age, years . 61.7 (9.7) 62.0 (10.6)

Female . 62 (41.1) 41 (35.7)

Highest level of education

Page 59 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

22

Primary or less . 14 (9.3) 12 (10.4)

Secondary or trade . 101 (66.9) 83 (72.2)

Tertiary . 36 (23.8) 20 (17.4)

Employed . 67 (44.4) 50 (43.5)

Health care card holder . 75 (49.7) 62 (53.9)

Diabetes duration, years (median (IQR)) . 8 (5, 12) 9 (5, 14)

HbA1c % (median (IQR)

mmol/mol (median (IQR))

. 8.7 (8.1, 9.7)

72 (65, 83)

8.5 (8, 9.6)

69 (64, 81)

Number of medical conditions (median

(IQR))

3 (2, 5) 3 (2, 5)

Number of medications (median (IQR)) 6 (5, 10) 7 (5, 10)

Medication adherence rating scale (median

IQR))5

29 (26, 30) 29 (27, 30)

Diabetes complications6

Microvascular . 17 (11.3) 16 (13.9)

Macrovascular . 22 (14.6) 21 (18.3)

Total cholesterol, mmol/L7

. 4.3 (1.0) 4.2 (1.1)

Triglycerides, mmol/L7* . 1.9 (0.1) 2.3 (1.4)

LDL cholesterol, mmol/L8 . 2.3 (0.9) 2.1 (0.9)

HDL cholesterol, mmol/L9 . 1.2 (0.3) 1.1 (0.3)

eGFR10

. 79.4 (14.4) 78.8 (14.6)

Blood pressure

Systolic . 134.6 (15.7) 133.5 (15.2)

Diastolic . 79.6 (11.1) 78.5 (9.5)

Data are mean (SD) or n (%) unless otherwise stated

1 Data available for 67 practices (33 intervention, 34 control)

2 Data available for 161 GPs (82 intervention, 79 control)

3 Data available for 158 GPs (84 intervention, 74 control)

4 Data available for 100 Practice Nurses (46 intervention, 54 control)

5 Data available for 261 patients (149 intervention, 112 control)

6 n(%) with at least one complication

7 Data available for 256 patients (144 intervention, 112 control)

8 Data available for 222 patients (130 intervention, 92 control)

9 Data available for 233 patients (134 intervention, 99 control)

10 Data available for 261 patients (147 intervention, 114 control)

Page 60 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

23

*Statistically significant difference between control and intervention groups

Registered Nurse-Credentialed Diabetes Educator: Registered Nurse-Credentialed Diabetes Educator

Page 61 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only24

Table 2: Primary and secondary endpoints: Biochemical, clinical and psychological outcomes

Endpoints Intervention Control Adjusted data for clustering

Treatment effect (95% CI) p

HbA1c %

Baseline 8.7 (8.1, 9.7) 8.5 (8, 9.6) 0.37

Follow-up 7.4 (6.9, 8.2) 8 (7.1, 9)

Change -1.3(1.4) -0.6(1.5) -0.7 9 (-1.1 to -0.4) <0.001

Participants using insulin+

Follow-up 105(69.5) 25 (21.7) 8.3* (4.5 to 15.4) <0.001

Participants with HbA1c ≤53mmol/mol (7%)+ Follow-up 54 (35.8) 24 (20.9) 2.2* (1.2 to 4.3) 0.02

Depressive symptoms (PHQ-9)1 Baseline 3(1,7) 2(1, 6.5) 0.17

Follow-up

2(0, 5) 2(0, 5)

Change -1.1(3.5) -0.1(2.9) -1.0 (-1.8 to -0.1) 0.02

Moderate-to-severe depressive symptoms (PHQ-9 total: ≥10)2+

Baseline 22(15.1) 15(13.5) 0.73

Follow-up 19(12.8) 15 (13.3) 1.0* (0.7 to 2.0) 0.92

Diabetes-specific distress (PAID)3 Baseline 15(6.3, 31.3) 12.5(5, 23.8) 0.14

Follow-up 8.8(3.8, 22.5) 10(2.5, 23.8)

Page 62 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only25

Change -5.6(15.5) -2.4(15.4) -3.3 (-7.2 to 0.6) 0.10

Severe diabetes-specific distress (PAID total: ≥40)2+

Baseline 25 (16.8) 14(12.4) 0.64

Follow-up 18 (12.1) 12(10.4) 0.2* (-0.6 to 0.9) 0.68

Health status (AQoL-8D) – Physical Component Score4

Baseline 0.63(0.20) 0.610.21

Follow-up 0.66(0.21) 0.640.21 0.52

Change 0.03(0.15) 0.02(0.13) 0.01 (-0.03 to 0.04) 0.75

Health status (AQoL-8D) – Mental Component Score4

Baseline 0.45(0.20) 0.45(0.22) 0.75

Follow-up 0.48(0.21) 0.45(0.22)

Change 0.04(0.16) -0.002(0.13) 0.04 (0.001 to 0.08) 0.05

Weight, kg Baseline 90.8(19.6) 94.6(18.9) 0.11

Follow-up 92.5(20.1) 93.5(18.9)

Change 1.7 (5.2) -1.1 (5.1) 2.8 (1.6 to 4.1) <0.001

Data are mean (SD) and median (IQR) unless otherwise indicated

1 PHQ-9: Patient Health Questionnaire 9. Range of possible scores: 0-27. A total score of ≥10 indicates at least moderate depressive symptoms. Data available for 261

patients at baseline (149 intervention, 112 control) and 263 at 12 months (149 intervention, 114 control; ITT)

2 Data available for 257 patients at baseline (146 intervention, 111 control) and 261 at 12 months (148 intervention, 113 control; ITT).

Page 63 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only26

3 PAID: Problem Areas In Diabetes. Range of possible scores: 0-100. A score of ≥40 indicates severe diabetes-related distress. Data available for 262 patients at baseline

(149 intervention, 113 control) and 264 at 12 months (149 intervention, 115 control; ITT)

4 AQoL-8D: Assessment of Quality Of Life. Maximum possible score is 1. Higher scores indicate better generic health status. Data available for 262 patients at baseline (149

intervention, 113 control) and 263 at 12 months (149 intervention, 114 control; ITT)

*Odds Ratio

+n (%)

Page 64 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only27

Figure 2: Change in primary endpoint at 6 and 12 months

Table 3: Classes of non-insulin medications at 12 months

Intervention n(%) Control n(%) P value

N 146 108

Metformin 133 (91.1) 96 (88.9) 0.56

Sulphonylurea 75 (51.4) 64 (59.3) 0.21

Acarbose 3 (2.1) 2 (1.9) 0.91

DPP4 inhibitors* 25 (17.1) 38 (35.2) 0.001

Glitazones 6 (4.1) 5 (4.6) 0.8411

Page 65 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only28

SGLT2 inhibitors 2 (1.4) 2 (1.9) 0.76

GLP1 agonists 9 (6.2) 7 (6.5) 0.92

(t test of proportions)

* At the time of the trial DPP4 inhibitors were not subsidized for use with insulin

Page 66 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

29

Authorship: JF, IB, DY and JB conceptualised the original study proposal and secured funding. JF had

overall responsibility for the study. JF, DON, JS, JMN, EHT, LG, DY, JB, EP, DL, LS, CM and IB drafted

the protocol. LG also contributed to data collection. JF, DON, JS, JMN, AG, EHT, DL and IB contributed

to the statistical analysis plan and JMN and AG led the data analysis. JF wrote the original draft of

the report. All authors contributed to data interpretation and approved the final report. As

corresponding author and Principal Investigator, JF had full access to all the data in the study and

takes responsibility for the integrity of the data and the accuracy of the data analysis. JF had final

responsibility for the decision to submit for publication and is guarantor for the study affirming that

the manuscript is an honest, accurate, and transparent account of the study being reported; that no

important aspects of the study have been omitted; and that any discrepancies from the study as

planned have been explained.

Competing interests: All authors have completed the ICMJE uniform disclosure form at

www.icmje.org/coi_disclosure.pdf and declare: We acknowledge funding from the Australian

National Health and Medical Research Council (Project Grant Application: APP1023738). The study

was also supported by an educational/research grant by Roche Diabetes Care Pty Ltd, the RACGP

Foundation RACGP/Independent Practitioner Network Pty Ltd (IPN) Grant and received in-kind

support from Sanofi. Xclinical hosted the BG data. JF was supported by a National Health and

Medical Research Council Career Development Fellowship. JS is supported by core funding to The

Australian Centre for Behavioural Research in Diabetes from Diabetes Victoria and Deakin University.

JMN was supported by a National Health and Medical Research Council postgraduate scholarship.

EHT is supported by an Australian Postgraduate Award Deakin University PhD scholarship.

JF has received unrestricted educational grants for research support from Roche Diabetes Care,

Sanofi and Medtronic; JS is a member of the Accu-Check Advisory Board (Roche Diabetes Care). Her

research group (ACBRD) has received unrestricted educational grants from Medtronic and Sanofi

Page 67 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

30

Diabetes; sponsorship to host or attend educational meetings from Lilly, Medtronic, MSD, Novo

Nordisk, Roche Diabetes Care, and Sanofi Diabetes; consultancy income from Abbott Diabetes Care,

Astra Zeneca, Roche Diabetes Care and Sanofi Diabetes; DNO, DL and JMN had various financial

relationships with pharmaceutical industries outside the submitted work including consultancies,

grants, lectures, educational activities and travel. DNO has received research and travel support and

honoraria from Sanofi, Roche and Novo and is an advisory board member to Sanofi, Novo and

Abbott. JMN has no financial relationships with companies marketing blood glucose monitoring

devices, but has received payment from Sanofi who funded the control practice training at end of

study. DL has received honoraria and research grants from Sanofi Australia. The study sponsors had

no role in the design and conduct of the study; collection, management, analysis, and interpretation

of the data; and preparation, review, or approval of the manuscript.

Funding

The study was funded by the Australian National Health and Medical Research Council (ID 1023738)

and was supported by an educational/research grant by Roche Diagnostics Australia Pty Ltd.

We thank the GPs, Practice Nurses and people with type 2 diabetes who generously participated in

the study.

Data sharing: Anonymised patient level data are available on reasonable request from the authors.

Page 68 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

31

References

1. Zimmet PZ, Magliano DJ, Herman WH, et al. Diabetes: a 21st century challenge. The Lancet

Diabetes & Endocrinology 2014;2(1):56-64.

2. Holman RR, Paul SK, Bethel MA, et al. 10-Year Follow-up of Intensive Glucose Control in Type 2

Diabetes. N Engl J Med 2008;359(15):1577-89.

3. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Patient-Centered Approach: Position Statement of the American Diabetes Association (ADA)

and the European Association for the Study of Diabetes (EASD). Diabetes Care

2012;35(6):1364-79.

4. Nathan DM, Buse JB, Davidson MB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Consensus Algorithm for the Initiation and Adjustment of Therapy: Update regarding

thiazolidinediones: a consensus statement from the American Diabetes Association and the

European Association for the Study of Diabetes. Diabetes Care 2008;31(1):173-75.

5. American Diabetes Association. Standards of Medical Care in Diabetes—2016 Abridged for

Primary Care Providers. Clinical Diabetes 2016;34(1):3-21.

6. McGuire H, Longson D, Adler A, et al. Management of type 2 diabetes in adults: summary of

updated NICE guidance. BMJ 2016;353.

7. Shah BR, Hux JE, Laupacis A, et al. Clinical Inertia in Response to Inadequate Glycemic Control: Do

specialists differ from primary care physicians? Diabetes Care 2005;28(3):600-06.

8. Blak BT, Smith HT, Hards M, et al. A retrospective database study of insulin initiation in patients

with Type 2 diabetes in UK primary care. Diabetic Medicine 2012;29(8):e191-e98.

9. Curtis B, Lage MJ. Glycemic control among patients with type 2 diabetes who initiate basal insulin:

a retrospective cohort study. Journal of Medical Economics 2014;17(1):21-31.

10. Khunti K, Damci T, Meneghini L, et al. Study of Once Daily Levemir (SOLVE™): insights into the

timing of insulin initiation in people with poorly controlled type 2 diabetes in routine clinical

practice. Diabetes, Obesity and Metabolism 2012;14(7):654-61.

11. Davis TME, Davis WA, Bruce DG. Glycaemic levels triggering intensification of therapy in type 2

diabetes in the community: the Fremantle Diabetes Study. Medical Journal of Australia

2006;184(7):325-8.

12. Phillips LS, Branch JWT, Cook CB, et al. Clinical Inertia. Annals of Internal Medicine

2001;135(9):825-34.

13. Khunti S, Davies M, Khunti K. Clinical inertia in the management of type 2 diabetes mellitus: a

focused literature review. The British Journal of Diabetes and Vascular Disease

2015;15(2):65-69.

14. Kunt T, Snoek FJ. Barriers to insulin initiation and intensification and how to overcome them.

International Journal of Clinical Practice 2009;63:6-10.

15. Furler J, Spitzer O, Young D, et al. Barriers and enablers to timely initiation of insulin in General

Practice" Australian Family Physician 2011;40(8):617-23.

16. Polonsky WH, Fisher L, Guzman S, et al. Psychological Insulin Resistance in Patients With Type 2

Diabetes: The scope of the problem. Diabetes Care 2005;28(10):2543-45.

17. Furler JS, Blackberry ID, Walker C, et al. Stepping up: a nurse-led model of care for insulin

initiation for people with type 2 diabetes. Family Practice 2014;31(3):349-56.

18. May C, Mair F, Finch T, et al. Development of a theory of implementation and integration:

Normalization Process Theory. Implementation Science 2009;4(1):29.

19. Blackberry ID, Furler JS, Ginnivan LE, et al. An exploratory trial of basal and prandial insulin

initiation and titration for type 2 diabetes in primary care with adjunct retrospective

continuous glucose monitoring: INITIATION study. Diabetes Research and Clinical Practice

2014;106(2):247-55.

Page 69 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

32

20. Hajos TRS, Pouwer F, de Grooth R, et al. Initiation of insulin glargine in patients with Type 2

diabetes in suboptimal glycaemic control positively impacts health-related quality of life. A

prospective cohort study in primary care. Diabetic Medicine 2011;28(9):1096-102.

21. Dzida G, Karnieli E, Svendsen AL, et al. Depressive symptoms prior to and following insulin

initiation in patients with type 2 diabetes mellitus: Prevalence, risk factors and effect on

physician resource utilisation. Primary Care Diabetes;9(5):346-53.

22. Furler J, Young D, Best J, et al. Can primary care team-based transition to insulin improve

outcomes in adults with type 2 diabetes: the stepping up to insulin cluster randomized

controlled trial protocol. Implementation Science 2014;9(1):20.

23. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting

parallel group randomised trials. BMJ 2010;340:698-702.

24. Giraudeau B, Ravaud P. Preventing Bias in Cluster Randomised Trials. PLoS Medicine

2009;6(5):e1000065.

25. Blackberry ID, Furler JS, Best JD, et al. Effectiveness of general practice based, practice nurse led

telephone coaching on glycaemic control of type 2 diabetes: the Patient Engagement And

Coaching for Health (PEACH) pragmatic cluster randomised controlled trial. BMJ 2013;347.

26. Polonsky WH, Fisher L, Schikman CH, et al. Structured self-monitoring of blood glucose

significantly reduces A1C levels in poorly controlled, noninsulin-treated type 2 diabetes:

results from the Structured Testing Program study. Diabetes Care 2011;34(2):262-7.

27. The Royal Australian College of General Practitioners and Diabetes Australia. General practice

management of type 2 diabetes – 2014–15. Melbourne, , 2014.

28. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. Validity of a Brief Depression Severity Measure.

Journal of General Internal Medicine 2001;16(9):606-13.

29. Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in Diabetes Scale - An evaluation of

its clinical utility. Diabetes Care 1997;20(5):760-66.

30. Richardson J, Iezzi A, Khan M, et al. Validity and Reliability of the Assessment of Quality of Life

(AQoL)-8D Multi-Attribute Utility Instrument. Patient 2014;7(1):85-96.

31. The National Diabetes Service Scheme. Secondary The National Diabetes Service Scheme.

https://www.ndss.com.au/the-ndss.

32. Speight J, Browne JL, Holmes-Truscott E, et al. Diabetes MILES - Australia 2011 Survey Report.

Canberra: Diabetes Australia, 2011.

33. Elliott L, Fidler C, Ditchfield A, et al. Hypoglycemia Event Rates: A Comparison Between Real-

World Data and Randomized Controlled Trial Populations in Insulin-Treated Diabetes.

Diabetes Therapy 2016;7(1):45-60.

34. Frier BM, Jensen MM, Chubb BD. Hypoglycaemia in adults with insulin-treated diabetes in the

UK: self-reported frequency and effects. Diabetic Medicine 2016;33(8):1125-32.

35. Harris S, Gerstein H, Yale J-F, et al. Can community retail pharmacist and diabetes expert support

facilitate insulin initiation by family physicians? Results of the AIM@GP randomized

controlled trial. BMC Health Services Research 2013;13(1):71.

36. Dale J, Martin S, Gadsby R. Insulin initiation in primary care for patients with type 2 diabetes: 3-

Year follow-up study. Primary Care Diabetes 2010;4(2):85-89.

37. Pouwer F, Hermanns N. Insulin therapy and quality of life. A review. Diabetes-Metabolism

Research and Reviews 2009;25:S4-S10.

38. Gucciardi E, Espin S, Morganti A, et al. Exploring interprofessional collaboration during the

integration of diabetes teams into primary care. BMC Family Practice 2016;17(1):1-14.

39. O'Connor PJ, Desai JR, Butler JC, et al. Current Status and Future Prospects for Electronic Point-

of-Care Clinical Decision Support in Diabetes Care. Current Diabetes Reports 2013;13(2):172-

76.

Page 70 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

1

Supporting insulin initiation in type 2 diabetes in primary care: Results of the

Stepping Up pragmatic cluster randomised controlled clinical trial

JS Furler, DN O’Neal, J Speight, J Manski-Nankervis, A Gorelik, E Holmes-Truscott, L Ginnivan, D

Young, J Best, E Patterson, D Liew, L Segal, CR May, I Blackberry,

JS Furler, Associate Professor, Department of General Practice, University of Melbourne, DN O’Neal,

Associate Professor, Department of Medicine, St Vincent’s Hospital, University of Melbourne, J

Speight, Professor, School of Psychology, Deakin University; Director, The Australian Centre for

Behavioural Research in Diabetes, Diabetes Victoria; AHP Research, United Kingdom, J Manski-

Nankervis, Lecturer, Department of General Practice, University of Melbourne, A Gorelik, Senior

Statistician, Melbourne EpiCentre, the University of Melbourne, E Holmes-Truscott, Research Fellow,

BPSc (Hons), School of Psychology, Deakin University; The Australian Centre for Behavioural

Research in Diabetes, Diabetes Victoria; L Ginnivan, Research Nurse and Credentialed Diabetes

Educator, of General Practice, University of Melbourne, D Young, Professor and Chair of General

Practice, and Associate Dean (Academic), Melbourne Medical School, University of Melbourne,

Australia, J Best, Professor and Dean, Lee Kong Chian School of Medicine, Nanyang Technological

University, Singapore, E Patterson, Professor, School of Nursing, University of Melbourne, Australia,

D Liew, Professor, School of Public Health and Preventive Medicine, Monash University, Australia, L

Segal, Professor, Health Economics and Social Policy Group, Division of Health Sciences, University of

South Australia, CR May, Professor of Health Care Innovation, Faculty of Health Sciences, University

of Southampton, UK., I Blackberry, Associate Professor and Director, John Richards Initiative,

Australian Institute for Primary Care & Ageing, College of Science, Health and Engineering, La Trobe

University, Australia,

Correspondence to: John Furler, Department of General Practice, University of Melbourne, 200

Berkeley St, Carlton, VIC, 3053, Austraila, [email protected]

Page 71 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

2

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of

all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats

and media (whether known now or created in the future), to i) publish, reproduce, distribute, display

and store the Contribution, ii) translate the Contribution into other languages, create adaptations,

reprints, include within collections and create summaries, extracts and/or, abstracts of the

Contribution, iii) create any other derivative work(s) based on the Contribution, iv) to exploit all

subsidiary rights in the Contribution, v) the inclusion of electronic links from the Contribution to

third party material where-ever it may be located; and, vi) licence any third party to do any or all of

the above.

Word count:

Abstract 482

Main paper: 4407

Page 72 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

3

Abstract

Objective

Optimising glycated haemoglobin to minimise risk of long-term complications among people with

type 2 diabetes is a global priority. Supporting timely insulin initiation through innovation in primary

care could have a major impact. Our aim was to test the effectiveness of the ‘Stepping Up’ model of

care compared to usual primary care in normalising insulin initiation as part of routine primary care

practice for type 2 diabetes, leading to improved HbA1c levels. Our hypothesis was that HbA1c

would improve among participants in intervention arm practices, facilitated through timely insulin

initiation, compared to the control arm.

Design

A two-arm, 12-month cluster-randomised controlled trial of the Stepping Up model of care.

Setting

Primary care practices in Victoria, Australia with a Practice Nurse and at least one consenting eligible

patient (HbA1c ≥7.5% on maximal oral therapy).

Participants

74 practices and 266 patients participated: mean (range) cluster size 4 (1 to 8) patients.

Intervention

The Stepping Up model of care intervention involved theory-based practice system change and re-

orientation in the roles of the health professionals in the primary care diabetes team. The core

component was an enhanced role for the Practice Nurse in leading insulin initiation, and mentoring

by a Registered Nurse-Credentialed Diabetes Educator.

Main outcome and measures

Our clinically meaningful primary endpoint was change in glycated hemoglobin (HbA1c). Secondary

endpoints included the proportion of participants who transitioned to insulin, proportion who

Page 73 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

4

achieved target HbA1c as well as change in depressive symptoms (PHQ-9), diabetes-specific distress

(PAID) and generic health status (AQoL-8D).

Results

HbA1c improved in both arms, with a clinically significant between-arm difference (mean difference

-0.6%; 95%CI -0.9 to -0.3%), favouring the intervention. At 12 months, in intervention practices,

105/151 (69.5%) adults with type 2 diabetes had commenced insulin (102 remained on insulin at 12

months); in control practices 25/115 (21.7%) commenced insulin (and 24 remained on insulin at 12

months) (OR 8.3, 95% CI 4.5 to 15.4, p<0.001). Target HbA1c (≤7% (53mmol/mol)) was achieved by

54 (35.8%) intervention participants (32 of whom were using insulin at 12 months) and 22 (19%) of

the control arm (two of whom had commenced insulin), (OR 2.2, 95% CI1.2 to 4.3, p=0.02).

Depressive symptoms did not worsen at 12 months (PHQ-9: -1.1 (3.5) versus -0.1 (2.9), p=0.05).

There was a statistically significant difference between arms in the mean (SD) change in mental

health (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13), mean difference (95%CI) 0.04 (0.002 to 0.08),

p=0.04), favouring the intervention, but no significant difference in physical health (AQoL PCS: 0.03

(0.15) versus 0.02 (0.13)) nor diabetes-specific distress (PAID: 5.6 (15.5) versus -2.4 (15.4)). No

severe hypoglycaemia events were reported.

Conclusions

Our novel model of care increased insulin initiation rates in primary care, improving glycated

hemoglobin without worsening emotional well-being. Delays in starting insulin treatment can be

overcome, enabling timely treatment intensification to improve important health outcomes.

Trial registration Australian and New Zealand Clinical Trials Registry (ACTRN12612001028897)

http://www.anzctr.org.au/

What this paper adds

Page 74 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

5

What is already known on this subject?

• Achieving and maintaining glycaemic targets early in type 2 diabetes improves long term

outcomes.

• There are barriers to early stepwise progressive treatment intensification to achieve glycaemic

targets, particularly in relation to insulin initiation in primary care.

• Interventions have had limited success in overcoming this delay in starting insulin treatment and

changing clinical practice, in part because system level barriers are not addressed.

What this study adds

• Our model of care intervention changed clinical practice, with more patients in intervention arm

practices commencing insulin, with an overall benefit in terms of HbA1c reduction, achieved

without serious adverse events, or any worsening in depressive symptoms.

• The theoretical base and flexible implementation are important characteristics of our

intervention.

• Our model of care used existing resources to work smarter and improve outcomes and, thus, has

important implications for policy makers, funders and practitioners seeking innovative ways to

provide best care for people with type 2 diabetes in primary care.

Page 75 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

6

Introduction

Nearly 600 million people will have type 2 diabetes type 2 diabetes worldwide by 2030.1 Innovation

in delivering effective clinical care to these people is an urgent global priority. To reduce the risk of

long-term macro- and microvascular complications,2 UK, European and US guidelines recommend

early adoption of insulin as a part of step-wise treatment intensification to bring glycated

hemoglobin (HbA1c) below a general target of 7% (53mmol/mol).3-6

However, insulin initiation is

frequently delayed, particularly in primary care,7 where despite being recommended as part of

routine clinical management of type 2 diabetes6, implementation is not widespread. The mean

HbA1c of people with type 2 diabetes prior to starting insulin is typically 1.5-2.0% above target: 9.3%

(78mmol/mol) in the UK8 8.6% (70mmol/mol) in a study in the US

9, 8.9% (74mmol/mol) in a large,

multi-country primary care study10

, and 9.4% (79mmol/mol) in a community study in Australia (after

a median diabetes duration of 8.1 years)11

.

Delay in treatment intensification by healthcare professionals, despite evidence that intensification

is warranted and effective12

is a major barrier to initiating insulin in type 2 diabetes13

. It can be due

to health professional factors14

(e.g. concerns about hypoglycaemia risk, lack of confidence/skills in

insulin initiation/titration), health system factors15

(e.g. competing priorities in busy, reactive

primary care settings), and patient-related factors (e.g. psychological insulin resistance16

).

Supporting and embedding insulin initiation as an element of routine primary care practice is an

important first step and building block in potentially reducing referrals to costly secondary care, and

supporting timely, early optimisation of therapy and achievement of glycaemic targets.

Our trial investigated the effectiveness of a new model of care designed to support primary care-

based insulin initiation among people with type 2 diabetes for whom it is clinically indicated. The

Stepping Up model of care is built around an enhanced, reconfigured role for the primary care

Practice Nurse.17

In Australia, clinical care of type 2 diabetes is predominantly undertaken in primary

care, largely made up of General Practitioner private group practices where Practice Nurses are a

Page 76 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

7

rapidly growing section of the workforce, taking on an increasing role in chronic condition

management. In this setting, our approach to professional behaviour change was informed by

Normalisation Process Theory18

and developed through pilot studies.17 19

Our aim was to address

both clinician and system level barriers to timely insulin initiation, and to normalise insulin initiation

as part of standard primary care practice. We hypothesised that HbA1c would improve among

participants in intervention arm practices, facilitated through timely insulin initiation, compared with

the control arm. Further, based on previous research 20 21

, we expected no significant negative

impact on participants’ general emotional wellbeing (depressive symptoms).

Methods

Study design and participants

The study design and protocol have been described previously.22

In summary, we conducted a 12-

month, two-arm, non-blinded cluster-randomised controlled trial (RCT), consistent with CONSORT

guidelines,23

to investigate the effectiveness of the Stepping Up model of care versus usual care.

General Practices in Victoria, Australia were eligible if they had at least one consenting General

Practitioner (GP) and Practice Nurse and could identify at least one eligible patient participant:

adults with type 2 diabetes with above target HbA1c (≥7.5% (58mmol/mol)) in the past six months

who were already prescribed maximum oral therapy (i.e. at least two oral hypoglycaemia agents

(OHAs) at maximum doses) or if their GP judged that insulin would be clinically appropriate. Patients

were ineligible if they were >80 years old, already using insulin, had an eGFR <30 mL/min/1.73m2,

unable to give informed consent or had a complex debilitating medical condition, e.g., severe mental

illness, end-stage cancer, unstable cardiovascular disease.

Our original protocol was based on 58 practices and an average cluster size of 5. Based on

experience in the field and what was an achievable sample size this was subsequently revised to 74

practices and an average cluster size of 3 (see Appendix 1).

Page 77 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

8

Randomisation

The unit of randomisation was the primary care practice. The study statistician (AG) computer-

generated stratified block randomisation sequences with varying block sizes (4, 6 and 8) prior to

recruitment. Practices were stratified by size (≤2 vs >2 full-time equivalent GPs), setting (private

practice vs community health centre), and participation (or not) in type 2 diabetes quality

improvement programs (the Australian Primary Care Collaborative). After providing consent and

recruiting at least one eligible patient, practices were randomised to intervention or usual care. We

used this index case method24

in all practices as our previous experience suggested that delaying

randomisation of a cluster until all patients have been recruited risks loss of engagement of GPs25

.

The research team then assisted practices to continue to identify and recruit patient participants

(through searching the practice medical record database). This meant that allocation concealment

after the index case was recruited was not possible for the GP and Practice Nurse. However, in order

to minimise potential bias, participating patients were not informed of their study allocation until

after they had provided consent. Blinding of GP, Practice Nurse and patient was also not possible

given the pragmatic nature of the intervention.

Intervention

The Stepping Up model of care, described elsewhere, 17 22

involved a re-orientation of existing

resources. It included: a) an enhanced role for the Practice Nurse in leading the discussion with

patients about intensifying treatment through insulin initiation and titration, b) simple insulin

initiation and up-titration clinical protocols, and c) a re-oriented role for the specialist Registered

Nurse-Credentialed Diabetes Educator in mentoring the Practice Nurse rather than providing direct

patient care. The training and mentoring support was designed to enhance the primary care team’s

knowledge, skills and confidence in discussing and implementing insulin initiation within the practice

as a part of routine care. Intervention practices had an in-practice briefing and training session for

GPs and Practice Nurses of approximately 60-90 minutes, following which patients with confirmed

Page 78 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

9

eligibility and completed baseline data were invited to consult their GP for an assessment to discuss

treatment intensification and referral to the Practice Nurse. Practice Nurses did not prescribe insulin,

or manage insulin dosing without liaison with the GP, based on the legal scope of practice for

generalist Practice Nurses in Australia.

Our model of care involved the acknowledgment and discussion of advantages and disadvantages of

starting insulin treatment with patients, including weight gain. We modelled shared decision making

as a part of the intervention training, drawing on the principles of motivational interviewing. This set

the scene for encouraging practitioners in intervention practices to approach participating patients

with equipoise in relation to starting insulin. While the intervention was necessarily brief in this

pragmatic trial, we included guidance and checklists for GPs and Practice Nurses to discuss the pros

and cons of insulin therapy and elicit patient concerns and expectations, while also openly

acknowledging and accepting that some patients may choose not to start insulin.

The role of Registered Nurse-Credentialed Diabetes Educator in supporting and mentoring the

Practice Nurse, and of the Practice Nurse in leading the discussion and implementation of insulin

therapy with the patient in liaison with the GP, is outlined elsewhere.15,20

Titration protocols were

based on fasting blood glucose and use of a 3-day, 7-point blood glucose profile26

to identify the

meal with the largest postprandial excursion (see appendix material). We gave no additional

instructions, so the GP had clinical autonomy regarding the management of OHAs. Practice Nurses

and GPs were encouraged to see patients as frequently as was felt to be clinically appropriate over a

period of up to 12 months, drawing as needed upon the study Registered Nurse-Credentialed

Diabetes Educator for mentoring and support, even if the patient remained undecided about, or had

decided against, starting insulin. Further details about the intervention can be found in the

referenced papers and the Appendices (Appendix 2 – Intervention Description; Appendix 3 –

Training Manual for Intervention Practices; Appendix 4 – Patient participant booklet).

Page 79 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

10

Control arm practices were given a copy of Australian type 2 diabetes management guidelines27

.

Control arm practices were offered training in the Stepping Up model of care after 12-month follow

up of patient participants was complete.

Patient involvement

Participant feedback was sought after conducting an intervention pilot study15

and this feedback was

used to refine the model of care. Participant feedback was also sought in a pilot of the data

collection forms. Throughout the main trial we maintained communication with patient participants

through a regular newsletter which included aggregate data about study progress and opportunities

to provide feedback to the study team. We assessed the burden of the intervention on patients

through interviews conducted at the end of the trial, as part of process evaluation (to be reported

elsewhere). This evaluation was led by the Chronic Illness Alliance, a consumer advocacy

organization that has been a long term collaborator of our research group. We have thanked all

participants for their involvement in the trial, and will provide, at a later date, a final summary report

of the trial outcomes. Participants have access to the study website where all published results will

be publicly available.

Endpoints and data collection

While our intervention targeted a process of care (i.e. insulin initiation), we chose a clinically

meaningful disease outcome at an individual patient level as primary endpoint: change (from

baseline to 12 months) in HbA1c, measured as a continuous variable. We registered our primary

outcome as “an absolute HbA1c reduction of 0.5% in the intervention group compared with the

control group”. HbA1c was performed at DCCT-aligned pathology laboratories and communicated to

clinicians and patients as part of usual clinical care; researchers retrieved these data from medical

records or directly form pathology laboratories. Secondary endpoints included the proportion of

participants who transitioned to insulin (this was amended from the original protocol where rate of

insulin initiation was used, which proved impractical given the small cluster size in the study); the

Page 80 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

11

proportion who achieved a target HbA1c of ≤7.0% (53mmol/mol) at 12 months; and change (from

baseline to 12 months) in depressive symptoms (PHQ-9),28

diabetes-specific distress (PAID),29

and

generic health status (AQoL-8D).30

Differences between the registered outcomes reported here and

those registered in the trial registry and justification and explanation for any changes made can be

found in the Appendix 1. We also collected data on healthcare utilisation and costs, to be reported

elsewhere.

All participants were provided with a blood glucose meter (Performa NanoTM

; Roche Diagnostics)

and instructed on its use. Subsidised low-cost blood glucose testing strips were available through the

National Diabetes Service Scheme31

. Data were uploaded from the meter at 6 and 12 months to a

secure server.

Statistical analysis

Our statistical analysis plan has been published elsewhere.22

In brief, our sample size of 224 patients

from 74 general practices (averaging 3 patients per practice) allowed us to detect an absolute 0.5%

mean HbA1c difference over 12 months between control and intervention arms with 80% power and

a standard deviation (SD) of 1 using two-sided alpha of 0.05. Data were analysed using Stata 13

(StataCorp, TX, USA). Descriptive statistics were used to summarise GP, Practice Nurse, and patient

characteristics for the two study arms and appropriate univariate tests were conducted to check for

any significant difference in potential confounders. Parametric data are reported as mean (SD) and

non-parametric data as median (IQR). Categorical data are reported as n (%). T-test for proportions

was used to compare baseline and 12-month data within each study arm for categorical data, while

paired t-tests were for used for continuous parametric data and Wilcoxon matched pairs signed rank

test for continuous non-parametric data. The individual patient was the unit of analysis and the

analytical methods allow for clustering of patients within the practices. Marginal logistic modelling

using generalised estimating equations with robust standard errors and adjustment for baseline

measures and clustering were used to compare binary outcomes between the two study arms.

Page 81 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

12

Mixed-effects linear regression was used to determine predictors for continuous outcomes,

adjusting for baseline measures and clustering at the practice level. Analyses were conducted on an

intention-to-treat basis.

Study oversight

The study protocol was approved by the University of Melbourne Health Sciences Human Research

Ethics Sub-committee (ID 123740) and registered with the Australian and New Zealand Clinical Trials

Registry (ACTRN12612001028897). All participants gave informed consent before enrolment.

Results

Participating practices and patients

Between October 2012 and January 2014, 93 primary care practices expressed interest and

identified 521 potentially eligible patients (Figure 1). Subsequently, 19 practices did not consent any

eligible patients, leaving 74 participating practices for randomisation. Two hundred and fifty five of

the potentially eligible patients were subsequently found to be ineligible at screening (n=156) or did

not respond to the invitation letter (n=99). By April 2014, the 74 practices had identified and

consented 266 eligible patient participants (73% of potentially eligible patients identified).

Baseline characteristics of practices, GPs, Practice Nurses and participants with type 2 diabetes are

shown in Table 1. Of the total sample, 248 (93.2%) completed 12-month follow-up for the primary

endpoint. No differences in baseline characteristics were observed between study completers and

non-completers, except for a higher proportion of women (n=11) not completing than men (n=7).

Primary and secondary endpoints

Primary and secondary endpoints are shown in Table 2. At 12 months, there was a statistically and

clinically significant difference between study arms in terms of change in HbA1c (mean

difference: -0.6%; 95% CI -0.9 to -0.3%, p<0.001), favouring the intervention. This is consistent with

achieving our registered primary outcome of “an absolute HbA1c reduction of 0.5% in the

Page 82 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

13

intervention group compared with the control group”. The majority of this change in HbA1c seen in

both arms was achieved by six months (See Figure 2).

In the intervention arm 105/151 (69.5%) of patients commenced insulin (and 102 remained on

insulin at 12 months), while in the control arm, 25/115 (21.7%) commenced insulin (and 24

remained on insulin at 12 months). Median (IQR) number of days from baseline assessment to

insulin initiation in intervention and control group patients who started insulin was 32 (11.5, 134.5)

days and 85 (63, 191) days respectively (statistically significant difference: Two-sample Wilcoxon

rank-sum test; p=0.005). In the intervention arm, 17 (11.22%) patients had commenced rapid-acting

insulin at 12 months, compared with one patient in the control arm (p<0.001). Further data on

insulin use in participants is available in Appendix 5.

Target HbA1c (≤7% (53mmol/mol)) was achieved by 54 (35.8%) intervention participants (32 of

whom were using insulin at 12 months) and 22 (19%) of the control arm (two of whom had

commenced insulin), (OR 2.2, 95%CI 1.2 to 4.3, p=0.02). Twenty-two (14.7%) intervention arm

participants and 20 (17%) control arm participants achieved target HbA1c without commencing

insulin.

At 12 months, depressive symptoms had not worsened and there was no statistically significant

difference between arms in the mean (SD) change (PHQ-9: -1.1 (3.5) versus -0.1 (2.9)). There was a

statistically significant difference in mental health (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13),

favouring the intervention, but no significant difference in physical health (AQoL PCS: 0.03 (0.15)

versus 0.02 (0.13)) or diabetes-specific distress (PAID: -5.6 (15.5) versus -2.4 (15.4)). There was no

significant difference between arms in the proportion of participants experiencing moderate-to-

severe depressive symptoms or severe diabetes-specific distress (Table 2), nor was there any

difference by insulin initiation (Wilcoxon rank sum test; p=0.98). No significant difference was found

in PHQ-9, AQoL or PAID scores when comparing participants initiating insulin with those who did not

(data not shown).

Page 83 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

14

At 12 months, there was an average weight gain in the intervention arm and an average weight loss

in the control group (1.7 (5.2) kg versus -1.1 (5.1) mean (95% CI) difference 2.8 (1.5 to 4.0) kg,

p<0.001). There were no significant differences in blood pressure nor other biochemical measures

between arms at follow-up with the exception of triglycerides, which remained higher in the control

group.

At baseline, participants were using a mean (SD) of 2.0 (0.6) classes of non-insulin hypoglycaemic

agents, with no significant difference between arms (t test, p=0.89). The majority of patients were

prescribed metformin (93% across both arms) and sulfonylureas (63% across both arms). There was

no significant difference in the prescription of individual medication classes by study arm. The mean

(SD) number of classes of non-insulin hypoglycaemic agents being used at 12 months was higher in

the control compared to the intervention arm (2.3 (0.1) vs 1.9 (0.1); p=0.01). There were a higher

proportion of people in the control group using DPP4 inhibitors than the intervention group at 12

months (Table 3).

Practices in the intervention arm received a total of 183 mentoring support visits from the study

Registered Nurse-Credentialed Diabetes Educator (mean (range) visits per practice: 5.2 (1-8)). Thirty-

two percent (48/151) of participants in intervention practices completed at least one 3-day, 7-point

structured blood glucose monitoring profile over the 12-month study. Practice nurses estimated the

time they spent on the study (both clinical interactions with participating patients and on research

tasks). On a per-practice basis, 23 control practices and 27 intervention practices reported a median

(IQR) of 1.5 (0, 3.6) and 18 (9, 20.9) hours respectively.

58.3% of people in the control group were recruited prior to practice randomisation compared to

45.0% in the intervention group (statistically significant difference p=0.033). Sensitivity analysis was

conducted to explore whether there was any difference at baseline between patients who were

recruited before and after practice randomisation. No statistically significant differences were found.

Mixed-effects linear regression was used to determine the impact of the intervention, adjusting for

Page 84 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

15

clustering at the practice level for each of these groups. Treatment effect in terms of the primary

outcome remained significant in both groups (treatment effect: -0.57 95%CI: -1.1 to -0.05 p value =

0.03 and -0.98 95%CI -1.49 to -0.48 p value <0.001 for patients assessed prior to and after

randomisation respectively).

Adverse events

No severe hypoglycaemic events (i.e. requiring third-party assistance for recovery) or other adverse

events were reported in either study arm.

Discussion

Principal findings

Our model of care changed clinical practice, with the majority in the intervention arm commencing

insulin, producing a clinically and statistically significant improvement in glycaemic control among

adults with type 2 diabetes managed in primary care. This was despite a higher patient-to-GP ratio in

intervention practices, and was achieved safely, with no severe hypoglycaemic events, and without

deterioration in emotional wellbeing or health status. Our results indicate that, with appropriate

support and practice-system redesign, insulin initiation can become part of routine primary care

diabetes management, obviating the need to refer to specialist services with geographical, cost and

accessibility barriers.

Strengths and limitations of study

A strength of our study is the robust theoretical and empirical base to our intervention. Our

pragmatic trial of a complex intervention addressed a number of known barriers to overcoming

delay in starting insulin therapy. For example, an intentional component of our system redesign was

re-orienting the Practice Nurse and Registered Nurse-Credentialed Diabetes Educator roles, allowing

additional time to be spent with patients, within existing resources. Other strengths include the

Page 85 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

16

cluster-randomised design, minimising the risk of contamination, and our excellent participant

retention (93%).

Our study had limitations. First, practices were randomised after the first consenting patient was

identified, raising the possibility of selection bias. However, the balance in key patient characteristics

between the study arms means that any such bias was minimal. Secondly, while a smaller cluster

size is generally preferable in a cluster-randomised trial, the relatively large variation in the cluster

sizes in our study may make statistical adjustments for clustering less effective, in particular when

the number of clusters is not large. Thirdly, our sample may not have been fully representative of

the broader population of adults with type 2 diabetes managed in primary care for whom insulin is

clinically indicated. Overall, less than 15% of our sample had severe diabetes-specific distress or

moderate-to-severe depressive symptoms, a lower rate than in a recent national Australian

sample.32

We will explore implementation fidelity and variation in more detail through a qualitative

process evaluation in a subsequent paper. Finally, our medications and hypoglycaemia data were

derived from GP records and subject to the same accuracy limitations of any routinely collected

clinical dataset. Hypoglycaemia is typically under-estimated33

and is likely to be under-reported in

routine medical records.34

In particular, severe hypoglycemia is serious but relatively rare, and may

not have been detected in our study, given our sample size.

Comparison with other studies

Only two other trials have tested interventions to change clinical practice in this way. The AIM@GP

trial showed no improvement in insulin prescribing rates or glycated hemoglobin.35

It provided

scheduled and ongoing telephone support from a specialist diabetes educator as well as the option

to refer patients to a community pharmacist off-site for a one-hour insulin initiation session. Our

intervention differed in that it was based completely in the familiar environs of the patients’ own

primary care practice, built on existing relationships and resources (with the practice-based Practice

Nurses) and provided an immediate pathway for the GP to delegate this clinical task. A UK study,

Page 86 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

17

using a pre-post evaluation design, showed improved HbA1c (-1.4%) at 6 months in patients who

initiated insulin,36

similar in magnitude to the improvement in our study. That intervention combined

education with face-to-face and telephone specialist diabetes nurse support for GPs and Practice

Nurses and involved a full-day, off-site training for both GP and Practice Nurse. In contrast, our

intervention used a brief (60-90 minutes) on-site training incorporated into the daily running of the

practice with flexible Registered Nurse-Credentialed Diabetes Educator support as required.

Consistent with previous research,37

our study participants did not report worse psychological

outcomes at follow up, suggesting that insulin therapy can be initiated in primary care for people

with type 2 diabetes without impairing their emotional wellbeing. In fact, at 12 months, the

reduction in depressive symptoms in intervention arm participants was significantly greater than

that seen in the control arm. Insulin initiation (across both arms) did not impact on depressive

symptoms, diabetes-specific distress or generic health status at 12 months.

Conclusions and policy and practice implications

Our pragmatic trial findings have important implications for the organisation of healthcare and for

health policy. Our model of care is based on an enhanced role for a Practice Nurse and would not be

feasible where primary care doctors work in solo practice without access to a Practice Nurse.

Nevertheless, even in high-resource settings, where the move to multidisciplinary primary health

care teams is growing, for example through the growth of the Medical Home movement 38, our

study suggests that simply having access to a Practice Nurse will not increase appropriate insulin

initiation. To make the best use of resources, primary care workforce models need to be developed

and implemented to reorient the way specialists (Registered Nurse-Credentialed Diabetes Educators

and endocrinologists) offer support to primary care teams that include well-supported and

resourced primary care nurses. Registered Nurse-Credentialed Diabetes Educator Change is

occurring in some health systems,38

but our study provides evidence of the effectiveness and safety

of such models of care. Rather than waiting for referrals, specialist services need to offer pro-active,

tailored secondary consultation, liaison and mentoring services that are flexible and supportive of

Page 87 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

18

the needs of primary care practitioners and patients. Scaling up the model of care in metropolitan

centres would require engaging with hospitals and other health services in reorienting the role of

Registered Nurse-Credentialed Diabetes Educators currently employed in direct patient care.

Investigation of e-health modalities (e.g. online training, support and video consultations) may

support implementation of this model of care in more distant rural/remote settings.

Our trial findings also have implications for clinical practice. The issue around personalising

glycaemic targets and treatments is an important and emerging consideration in the care of people

with T2D. At the time our trial started, there was vigorous debate about the need for caution in

setting lower targets. In addition our exclusion criteria ruled out participants for whom a higher

target would definitely be considered. Collecting reliable data on duration of CVD, severe

hypoglycaemia and hypoglycaemia unawareness did not prove feasible in this setting. Therefore we

made a pragmatic decision to use the general target of 7%. Debate continues about the advantages

and disadvantages of intensifying treatment for people with type 2 diabetes at the HbA1c levels

mandated in our study. While our findings suggest the that Stepping Up model of care builds clinical

capacity within GP and Practice Nurse teams to undertake the work of insulin initiation, the model of

care did not mandate a dogmatic approach to such therapy changes. It is worth noting the major

change in HbA1c was achieved at 6 months and that even in intervention practices, only 35% of

participating people with type 2 diabetes achieved the general HbA1c target of <7% (53 mmol/mol),

suggesting that that practitioners and patients were judicious in the way they approached

progressive treatment intensification, within the new model of care.

Our clinical protocols and algorithms were focused solely on insulin. It is worth noting that NPH

Insulin remains widely used and that the added costs of analogue insulins where NPH can be used

without problem is still subject to debate. While we did not pre-specify weight gain as a secondary

outcome, this is an adverse effect of insulin therapy and we have chosen to report it. While insulin is

still regarded as an essential therapy option, as the range of glycaemic therapies grows clinical

algorithms become more complex5. Future research needs to explore the capacity to generate real-

Page 88 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

19

time personalised treatment intensification recommendations that incorporate this increasing

complexity,39

to be used as part of the Practice Nurse-led model of care. Future research could also

address the extent to which improvements in glycaemic levels are maintained, and the extent to

which the model of care is sustained in routine clinical practice. In particular, future research could

explore use of the model of care to specifically support early adoption of insulin therapy to achieve

glycaemic targets early in people with recently diagnosed type 2 diabetes.

The global epidemic of type 2 diabetes demands innovation in care delivery. Delaying insulin

initiation when clinically indicated is neither ethical nor effective. Furthermore, health systems will

not cope with demand if insulin initiation remains anchored in specialist centres, nor will they be

able to respond to the imperative to achieve glycaemic targets early in people with recently

diagnosed type 2 diabetes. Thus our pragmatic, translational study has important implications across

health systems globally for the organisation of care for people with type 2 diabetes. Our effective

model of care has the potential to improve outcomes in people with type 2 diabetes while making

better use of scarce healthcare resources.

Page 89 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

20

Figure 1: Consort Diagram

Assessed for eligibility (93 practices, patients n=521)

Excluded: Practices - failed to recruit patients – n = 19 Patients (n=255) ♦ Ineligible (n=156); ♦ Non-responder (n=99)

Intention to treat analysis

Clusters: 38 practices, 115 patients

Allocated to control arm (n=38 practices) Received allocated control

• 38 practices, patients (n=115)

• mean cluster size 3, range 1-6

Allocated to intervention arm (n=36 practices) Received allocated intervention

• 36 practices, patients (n= 151)

• mean cluster size 4, range 1-8

Intention to treat analysis

Clusters: 36 practices, 151 patients

Allocation

Analysis

Follow-Up

Enrolment

Randomized (74 practices (n=74) with eligible patients (n = 266)

Did not complete follow-up (n= 9)

Withdrawn (n=5)

• no reason given (n=4)

• patient developed cancer (n=1)

Patient deceased (n=1) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=3)

Did not complete follow-up (n=9)

Withdrawn (n=2)

• no reason given (n=1)

• patient pregnant (n=1) Patient deceased (n=2) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=5)

Page 90 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

21

Table 1: Baseline characteristics of participating practices, health professionals and adults

with type 2 diabetes1

Intervention Control

Primary care practices 36 (49) 38 (51)

Type of practice

Private practice . 27 (75.0) 31 (81.6)

Corporate practice . 7 (19.4) 5 (13.2)

Community health centre . 2 (5.6) 2 (5.3)

Location of practice

Major city . 26 (72.2) 21 (55.3)

Inner regional area . 9 (25.0) 13 (34.2)

Outer regional area . 1 (2.8) 4 (10.5)

Physicians per practice (median (IQR)) 5 (4, 9.5) 5 (4, 9)

Practice Nurses per practice (median (IQR)) 2.5 (2, 3.5) 2 (1, 4)

Registered Nurse-Credentialed Diabetes

Educator on site: yes

. 12 (33.3) 14 (36.8)

Patients per Full-Time Equivalent GP

(median (IQR))1*

1738 (1176, 2727) 1316 (911, 1726

General Practitioners 83 (51.2) 79 (48.8)

Age, years2

. 48.8 (9.9) 49.7 (11.2)

Female . 34 (41.0) 27 (34.2)

Working hours/week2 . 36.6 (10.5) 37.3 (11.6)

Years of experience (median (IQR)) 19 (8, 26) 20 (7, 30)

Experience with insulin initiation in the

preceding 12 months3

. 48 (60.0) 36 (46.2)

Practice Nurses 48 (46.6) 55 (53.4)

Age, years4

. 44.7 (10.2) 46.0 (9.9)

Female 48 (100) 55 (100)

Diabetes educator training . 6 (12.5) 7 (12.7)

Experience with insulin initiation in the

preceding 12 months

. 16 (33.3) 16 (29.1)

Adults with type 2 diabetes 151 (56.8) 115 (43.2)

Age, years . 61.7 (9.7) 62.0 (10.6)

Female . 62 (41.1) 41 (35.7)

Highest level of education

Page 91 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

22

Primary or less . 14 (9.3) 12 (10.4)

Secondary or trade . 101 (66.9) 83 (72.2)

Tertiary . 36 (23.8) 20 (17.4)

Employed . 67 (44.4) 50 (43.5)

Health care card holder . 75 (49.7) 62 (53.9)

Diabetes duration, years (median (IQR)) . 8 (5, 12) 9 (5, 14)

HbA1c % (median (IQR)

mmol/mol (median (IQR))

. 8.7 (8.1, 9.7)

72 (65, 83)

8.5 (8, 9.6)

69 (64, 81)

Number of medical conditions (median

(IQR))

3 (2, 5) 3 (2, 5)

Number of medications (median (IQR)) 6 (5, 10) 7 (5, 10)

Medication adherence rating scale (median

IQR))5

29 (26, 30) 29 (27, 30)

Diabetes complications6

Microvascular . 17 (11.3) 16 (13.9)

Macrovascular . 22 (14.6) 21 (18.3)

Total cholesterol, mmol/L7

. 4.3 (1.0) 4.2 (1.1)

Triglycerides, mmol/L7* . 1.9 (0.1) 2.3 (1.4)

LDL cholesterol, mmol/L8 . 2.3 (0.9) 2.1 (0.9)

HDL cholesterol, mmol/L9 . 1.2 (0.3) 1.1 (0.3)

eGFR10

. 79.4 (14.4) 78.8 (14.6)

Blood pressure

Systolic . 134.6 (15.7) 133.5 (15.2)

Diastolic . 79.6 (11.1) 78.5 (9.5)

Data are mean (SD) or n (%) unless otherwise stated

1 Data available for 67 practices (33 intervention, 34 control)

2 Data available for 161 GPs (82 intervention, 79 control)

3 Data available for 158 GPs (84 intervention, 74 control)

4 Data available for 100 Practice Nurses (46 intervention, 54 control)

5 Data available for 261 patients (149 intervention, 112 control)

6 n(%) with at least one complication

7 Data available for 256 patients (144 intervention, 112 control)

8 Data available for 222 patients (130 intervention, 92 control)

9 Data available for 233 patients (134 intervention, 99 control)

10 Data available for 261 patients (147 intervention, 114 control)

Page 92 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

23

*Statistically significant difference between control and intervention groups

Registered Nurse-Credentialed Diabetes Educator: Registered Nurse-Credentialed Diabetes Educator

Page 93 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only24

Table 2: Primary and secondary endpoints: Biochemical, clinical and psychological outcomes

Endpoints Intervention Control Adjusted data for baseline measure

and clustering

Treatment effect (95% CI) p

HbA1c %

Baseline 8.7 (8.1, 9.7) 8.5 (8, 9.6)

Follow-up 7.4 (6.9, 8.2) 8 (7.1, 9)

Change -1.3(1.4) -0.6(1.5) -0.6 (-0.9 to -0.3) <0.001

Participants using insulin+ Follow-up 105(69.5) 25 (21.7) 8.3* (4.5 to 15.4) <0.001

Participants with HbA1c ≤53mmol/mol (7%)+ Follow-up 54 (35.8) 24 (20.9) 2.2* (1.2 to 4.3) 0.02

Depressive symptoms (PHQ-9)1 Baseline

3(1,7) 2(1, 6.5)

Follow-up

2(0, 5) 2(0, 5)

Change -1.1(3.5) -0.1(2.9) -0.8 (-1.6 to -0.01) 0.047

Moderate-to-severe depressive symptoms (PHQ-9 total: ≥10)2+

Baseline 22(15.1) 15(13.5)

Follow-up 19(12.8) 15 (13.3) 0.82+* (0.3 to 2.2) 0.69

Diabetes-specific distress (PAID)3 Baseline 15(6.3, 31.3) 12.5(5, 23.8) 0.14

Page 94 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only25

Follow-up 8.8(3.8, 22.5) 10(2.5, 23.8)

Change -5.6(15.5) -2.4(15.4) -1.9 (--5.1to 1.3) 0.24

Severe diabetes-specific distress (PAID total: ≥40)2+

Baseline 25 (16.8) 14(12.4)

Follow-up 18 (12.1) 12(10.4) 1.0* (0.4 to 2.3) 0.93

Health status (AQoL-8D) – Physical Component Score4 Baseline 0.63(0.20) 0.61 (0.21)

Follow-up 0.66(0.21) 0.64(0.21) 0.52

Change 0.03(0.15) 0.02(0.13) 0.01 (-0.03 to 0.04) 0.64

Health status (AQoL-8D) – Mental Component Score4

Baseline 0.45(0.20) 0.45(0.22)

Follow-up 0.48(0.21) 0.45(0.22)

Change 0.04(0.16) -0.002(0.13) 0.04 (0.002 to 0.08) 0.04

Weight, kg Baseline 90.8(19.6) 94.6(18.9)

Follow-up 92.5(20.1) 93.5(18.9)

Change 1.7 (5.2) -1.1(5.1) 2.8 (1.5 to 4.0) <0.001

Page 95 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only26

Page 96 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only27

Data are mean (SD) and median (IQR) unless otherwise indicated

Page 97 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only28

1 PHQ-9: Patient Health Questionnaire 9. Range of possible scores: 0-27. A total score of ≥10 indicates at least moderate depressive symptoms. Data available for 261

patients at baseline (149 intervention, 112 control) and 263 at 12 months (149 intervention, 114 control; ITT)

2 Data available for 257 patients at baseline (146 intervention, 111 control) and 261 at 12 months (148 intervention, 113 control; ITT).

3 PAID: Problem Areas In Diabetes. Range of possible scores: 0-100. A score of ≥40 indicates severe diabetes-related distress. Data available for 262 patients at baseline

(149 intervention, 113 control) and 264 at 12 months (149 intervention, 115 control; ITT)

4 AQoL-8D: Assessment of Quality Of Life. Maximum possible score is 1. Higher scores indicate better generic health status. Data available for 262 patients at baseline (149

intervention, 113 control) and 263 at 12 months (149 intervention, 114 control; ITT)

*Odds Ratio

+n (%)

Page 98 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only29

Figure 2: Change in primary endpoint at 6 and 12 months

Page 99 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only30

Table 3: Classes of non-insulin medications at 12 months

Intervention n(%) Control n(%) P value

N 146 108

Metformin 133 (91.1) 96 (88.9) 0.56

Sulphonylurea 75 (51.4) 64 (59.3) 0.21

Acarbose 3 (2.1) 2 (1.9) 0.91

DPP4 inhibitors* 25 (17.1) 38 (35.2) 0.001

Glitazones 6 (4.1) 5 (4.6) 0.8411

SGLT2 inhibitors 2 (1.4) 2 (1.9) 0.76

GLP1 agonists 9 (6.2) 7 (6.5) 0.92

(t test of proportions)

* At the time of the trial DPP4 inhibitors were not subsidized for use with insulin

Page 100 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

31

Authorship: JF, IB, DY and JB conceptualised the original study proposal and secured funding. JF had

overall responsibility for the study. JF, DON, JS, JMN, EHT, LG, DY, JB, EP, DL, LS, CM and IB drafted

the protocol. LG also contributed to data collection. JF, DON, JS, JMN, AG, EHT, DL and IB contributed

to the statistical analysis plan and JMN and AG led the data analysis. JF wrote the original draft of

the report. All authors contributed to data interpretation and approved the final report. As

corresponding author and Principal Investigator, JF had full access to all the data in the study and

takes responsibility for the integrity of the data and the accuracy of the data analysis. JF had final

responsibility for the decision to submit for publication and is guarantor for the study affirming that

the manuscript is an honest, accurate, and transparent account of the study being reported; that no

important aspects of the study have been omitted; and that any discrepancies from the study as

planned have been explained.

Competing interests: All authors have completed the ICMJE uniform disclosure form at

www.icmje.org/coi_disclosure.pdf and declare: We acknowledge funding from the Australian

National Health and Medical Research Council (Project Grant Application: APP1023738). The study

was also supported by an educational/research grant by Roche Diabetes Care Pty Ltd, the RACGP

Foundation RACGP/Independent Practitioner Network Pty Ltd (IPN) Grant and received in-kind

support from Sanofi. Xclinical hosted the BG data. JF was supported by a National Health and

Medical Research Council Career Development Fellowship. JS is supported by core funding to The

Australian Centre for Behavioural Research in Diabetes from Diabetes Victoria and Deakin University.

JMN was supported by a National Health and Medical Research Council postgraduate scholarship.

EHT is supported by an Australian Postgraduate Award Deakin University PhD scholarship.

JF has received unrestricted educational grants for research support from Roche Diabetes Care,

Sanofi and Medtronic; JS is a member of the Accu-Check Advisory Board (Roche Diabetes Care). Her

research group (ACBRD) has received unrestricted educational grants from Medtronic and Sanofi

Page 101 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

32

Diabetes; sponsorship to host or attend educational meetings from Lilly, Medtronic, MSD, Novo

Nordisk, Roche Diabetes Care, and Sanofi Diabetes; consultancy income from Abbott Diabetes Care,

Astra Zeneca, Roche Diabetes Care and Sanofi Diabetes; DNO, DL and JMN had various financial

relationships with pharmaceutical industries outside the submitted work including consultancies,

grants, lectures, educational activities and travel. DNO has received research and travel support and

honoraria from Sanofi, Roche and Novo and is an advisory board member to Sanofi, Novo and

Abbott. JMN has no financial relationships with companies marketing blood glucose monitoring

devices, but has received payment from Sanofi who funded the control practice training at end of

study. DL has received honoraria and research grants from Sanofi Australia. The study sponsors had

no role in the design and conduct of the study; collection, management, analysis, and interpretation

of the data; and preparation, review, or approval of the manuscript.

Funding

The study was funded by the Australian National Health and Medical Research Council (ID 1023738)

and was supported by an educational/research grant by Roche Diagnostics Australia Pty Ltd.

We thank the GPs, Practice Nurses and people with type 2 diabetes who generously participated in

the study.

Data sharing: Anonymised patient level data are available on reasonable request from the authors.

Page 102 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

33

References

1. Zimmet PZ, Magliano DJ, Herman WH, et al. Diabetes: a 21st century challenge. The Lancet

Diabetes & Endocrinology 2014;2(1):56-64.

2. Holman RR, Paul SK, Bethel MA, et al. 10-Year Follow-up of Intensive Glucose Control in Type 2

Diabetes. N Engl J Med 2008;359(15):1577-89.

3. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Patient-Centered Approach: Position Statement of the American Diabetes Association (ADA)

and the European Association for the Study of Diabetes (EASD). Diabetes Care

2012;35(6):1364-79.

4. Nathan DM, Buse JB, Davidson MB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Consensus Algorithm for the Initiation and Adjustment of Therapy: Update regarding

thiazolidinediones: a consensus statement from the American Diabetes Association and the

European Association for the Study of Diabetes. Diabetes Care 2008;31(1):173-75.

5. American Diabetes Association. Standards of Medical Care in Diabetes—2016 Abridged for

Primary Care Providers. Clinical Diabetes 2016;34(1):3-21.

6. McGuire H, Longson D, Adler A, et al. Management of type 2 diabetes in adults: summary of

updated NICE guidance. BMJ 2016;353.

7. Shah BR, Hux JE, Laupacis A, et al. Clinical Inertia in Response to Inadequate Glycemic Control: Do

specialists differ from primary care physicians? Diabetes Care 2005;28(3):600-06.

8. Blak BT, Smith HT, Hards M, et al. A retrospective database study of insulin initiation in patients

with Type 2 diabetes in UK primary care. Diabetic Medicine 2012;29(8):e191-e98.

9. Curtis B, Lage MJ. Glycemic control among patients with type 2 diabetes who initiate basal insulin:

a retrospective cohort study. Journal of Medical Economics 2014;17(1):21-31.

10. Khunti K, Damci T, Meneghini L, et al. Study of Once Daily Levemir (SOLVE™): insights into the

timing of insulin initiation in people with poorly controlled type 2 diabetes in routine clinical

practice. Diabetes, Obesity and Metabolism 2012;14(7):654-61.

11. Davis TME, Davis WA, Bruce DG. Glycaemic levels triggering intensification of therapy in type 2

diabetes in the community: the Fremantle Diabetes Study. Medical Journal of Australia

2006;184(7):325-8.

12. Phillips LS, Branch JWT, Cook CB, et al. Clinical Inertia. Annals of Internal Medicine

2001;135(9):825-34.

13. Khunti S, Davies M, Khunti K. Clinical inertia in the management of type 2 diabetes mellitus: a

focused literature review. The British Journal of Diabetes and Vascular Disease

2015;15(2):65-69.

14. Kunt T, Snoek FJ. Barriers to insulin initiation and intensification and how to overcome them.

International Journal of Clinical Practice 2009;63:6-10.

15. Furler J, Spitzer O, Young D, et al. Barriers and enablers to timely initiation of insulin in General

Practice" Australian Family Physician 2011;40(8):617-23.

16. Polonsky WH, Fisher L, Guzman S, et al. Psychological Insulin Resistance in Patients With Type 2

Diabetes: The scope of the problem. Diabetes Care 2005;28(10):2543-45.

17. Furler JS, Blackberry ID, Walker C, et al. Stepping up: a nurse-led model of care for insulin

initiation for people with type 2 diabetes. Family Practice 2014;31(3):349-56.

18. May C, Mair F, Finch T, et al. Development of a theory of implementation and integration:

Normalization Process Theory. Implementation Science 2009;4(1):29.

19. Blackberry ID, Furler JS, Ginnivan LE, et al. An exploratory trial of basal and prandial insulin

initiation and titration for type 2 diabetes in primary care with adjunct retrospective

continuous glucose monitoring: INITIATION study. Diabetes Research and Clinical Practice

2014;106(2):247-55.

Page 103 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

34

20. Hajos TRS, Pouwer F, de Grooth R, et al. Initiation of insulin glargine in patients with Type 2

diabetes in suboptimal glycaemic control positively impacts health-related quality of life. A

prospective cohort study in primary care. Diabetic Medicine 2011;28(9):1096-102.

21. Dzida G, Karnieli E, Svendsen AL, et al. Depressive symptoms prior to and following insulin

initiation in patients with type 2 diabetes mellitus: Prevalence, risk factors and effect on

physician resource utilisation. Primary Care Diabetes;9(5):346-53.

22. Furler J, Young D, Best J, et al. Can primary care team-based transition to insulin improve

outcomes in adults with type 2 diabetes: the stepping up to insulin cluster randomized

controlled trial protocol. Implementation Science 2014;9(1):20.

23. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting

parallel group randomised trials. BMJ 2010;340:698-702.

24. Giraudeau B, Ravaud P. Preventing Bias in Cluster Randomised Trials. PLoS Medicine

2009;6(5):e1000065.

25. Blackberry ID, Furler JS, Best JD, et al. Effectiveness of general practice based, practice nurse led

telephone coaching on glycaemic control of type 2 diabetes: the Patient Engagement And

Coaching for Health (PEACH) pragmatic cluster randomised controlled trial. BMJ 2013;347.

26. Polonsky WH, Fisher L, Schikman CH, et al. Structured self-monitoring of blood glucose

significantly reduces A1C levels in poorly controlled, noninsulin-treated type 2 diabetes:

results from the Structured Testing Program study. Diabetes Care 2011;34(2):262-7.

27. The Royal Australian College of General Practitioners and Diabetes Australia. General practice

management of type 2 diabetes – 2014–15. Melbourne, , 2014.

28. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. Validity of a Brief Depression Severity Measure.

Journal of General Internal Medicine 2001;16(9):606-13.

29. Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in Diabetes Scale - An evaluation of

its clinical utility. Diabetes Care 1997;20(5):760-66.

30. Richardson J, Iezzi A, Khan M, et al. Validity and Reliability of the Assessment of Quality of Life

(AQoL)-8D Multi-Attribute Utility Instrument. Patient 2014;7(1):85-96.

31. The National Diabetes Service Scheme. Secondary The National Diabetes Service Scheme.

https://www.ndss.com.au/the-ndss.

32. Speight J, Browne JL, Holmes-Truscott E, et al. Diabetes MILES - Australia 2011 Survey Report.

Canberra: Diabetes Australia, 2011.

33. Elliott L, Fidler C, Ditchfield A, et al. Hypoglycemia Event Rates: A Comparison Between Real-

World Data and Randomized Controlled Trial Populations in Insulin-Treated Diabetes.

Diabetes Therapy 2016;7(1):45-60.

34. Frier BM, Jensen MM, Chubb BD. Hypoglycaemia in adults with insulin-treated diabetes in the

UK: self-reported frequency and effects. Diabetic Medicine 2016;33(8):1125-32.

35. Harris S, Gerstein H, Yale J-F, et al. Can community retail pharmacist and diabetes expert support

facilitate insulin initiation by family physicians? Results of the AIM@GP randomized

controlled trial. BMC Health Services Research 2013;13(1):71.

36. Dale J, Martin S, Gadsby R. Insulin initiation in primary care for patients with type 2 diabetes: 3-

Year follow-up study. Primary Care Diabetes 2010;4(2):85-89.

37. Pouwer F, Hermanns N. Insulin therapy and quality of life. A review. Diabetes-Metabolism

Research and Reviews 2009;25:S4-S10.

38. Gucciardi E, Espin S, Morganti A, et al. Exploring interprofessional collaboration during the

integration of diabetes teams into primary care. BMC Family Practice 2016;17(1):1-14.

39. O'Connor PJ, Desai JR, Butler JC, et al. Current Status and Future Prospects for Electronic Point-

of-Care Clinical Decision Support in Diabetes Care. Current Diabetes Reports 2013;13(2):172-

76.

Page 104 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

1

Supporting insulin initiation in type 2 diabetes in primary care: Results of the

Stepping Up pragmatic cluster randomised controlled clinical trial

JS Furler, DN O’Neal, J Speight, J Manski-Nankervis, A Gorelik, E Holmes-Truscott, L Ginnivan, D

Young, J Best, E Patterson, D Liew, L Segal, CR May, I Blackberry,

JS Furler, Associate Professor, Department of General Practice, University of Melbourne, DN O’Neal,

Associate Professor, Department of Medicine, St Vincent’s Hospital, University of Melbourne, J

Speight, Professor, School of Psychology, Deakin University; Director, The Australian Centre for

Behavioural Research in Diabetes, Diabetes Victoria; AHP Research, United Kingdom, J Manski-

Nankervis, Lecturer, Department of General Practice, University of Melbourne, A Gorelik, Senior

Statistician, Melbourne EpiCentre, the University of Melbourne, E Holmes-Truscott, Research Fellow,

BPSc (Hons), School of Psychology, Deakin University; The Australian Centre for Behavioural

Research in Diabetes, Diabetes Victoria; L Ginnivan, Research Nurse and Credentialed Diabetes

Educator, of General Practice, University of Melbourne, D Young, Professor and Chair of General

Practice, and Associate Dean (Academic), Melbourne Medical School, University of Melbourne,

Australia, J Best, Professor and Dean, Lee Kong Chian School of Medicine, Nanyang Technological

University, Singapore, E Patterson, Professor, School of Nursing, University of Melbourne, Australia,

D Liew, Professor, School of Public Health and Preventive Medicine, Monash University, Australia, L

Segal, Professor, Health Economics and Social Policy Group, Division of Health Sciences, University of

South Australia, CR May, Professor of Health Care Innovation, Faculty of Health Sciences, University

of Southampton, UK., I Blackberry, Associate Professor and Director, John Richards Initiative,

Australian Institute for Primary Care & Ageing, College of Science, Health and Engineering, La Trobe

University, Australia,

Correspondence to: John Furler, Department of General Practice, University of Melbourne, 200

Berkeley St, Carlton, VIC, 3053, Austraila, [email protected]

Page 105 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

2

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of

all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats

and media (whether known now or created in the future), to i) publish, reproduce, distribute, display

and store the Contribution, ii) translate the Contribution into other languages, create adaptations,

reprints, include within collections and create summaries, extracts and/or, abstracts of the

Contribution, iii) create any other derivative work(s) based on the Contribution, iv) to exploit all

subsidiary rights in the Contribution, v) the inclusion of electronic links from the Contribution to

third party material where-ever it may be located; and, vi) licence any third party to do any or all of

the above.

Word count: 4142

Page 106 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

3

Abstract

Objective

Optimising glycated haemoglobin to minimise risk of long-term complications among people with

type 2 diabetes is a global priority. Supporting timely insulin initiation through innovation in primary

care could have a major impact. Our aim was to test the effectiveness of the ‘Stepping Up’ model of

care compared to usual primary care in normalising insulin initiation as part of routine primary care

practice for type 2 diabetes, leading to improved HbA1c levels. Our hypothesis was that HbA1c

would improve among participants in intervention arm practices, facilitated through timely insulin

initiation, compared to the control arm.

Design

A two-arm, 12-month cluster-randomised controlled trial of the Stepping Up model of care.

Setting

Primary care practices in Victoria, Australia with a Practice Nurse Practice Nurse and at least one

consenting eligible patient (HbA1c ≥7.5% on maximal oral therapy).

Participants

74 practices and 266 patients participated: mean (range) cluster size 4 (1 to 8) patients.

Intervention

The Stepping Up model of care intervention involved theory-based practice system change and re-

orientation in the roles of the health professionals in the primary care diabetes team. The core

component was an enhanced role for the Practice Nurse in leading insulin initiation, and mentoring

by a Registered Nurse-Credentialed Diabetes Educator.

Main outcome and measures

Our clinically meaningful primary endpoint was change in glycated hemoglobin (HbA1c). Secondary

endpoints included the proportion of participants who transitioned to insulin, proportion who

Page 107 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

4

achieved target HbA1c as well as change in depressive symptoms (PHQ-9), diabetes-specific distress

(PAID) and generic health status (AQoL-8D).

Results

HbA1c improved in both arms, with a clinically significant between-arm difference (mean difference

-0.76%; 95%CI -1.10.9 to -0.43%), favouring the intervention. At 12 months, in intervention practices,

105/151 (69.5%) adults with type 2 diabetes had commenced insulin (102 remained on insulin at 12

months); in control practices 25/115 (21.7%) commenced insulin (and 24 remained on insulin at 12

months) (OR 8.3, 95% CI (4.5 to 15.4), p<0.001). Target HbA1c (≤7% (53mmol/mol)) was achieved by

54 (35.8%) intervention participants (32 of whom were using insulin at 12 months) and 22 (19%) of

the control arm (two of whom had commenced insulin), (OR 2.2, 95% CI (1.2 to 4.3), p=0.0082).

Depressive symptoms did not worsen in the intervention armat 12 months (mean difference in PHQ-

9:= -1.1 (3.5) versus -0.1 (2.9), 95%CI -1.8 6 to -0.01, p=0.021 05 favoring intervention). There was a

statistically significant difference between arms in the mean (SD) change in the mental health

component of the AQoL (AQoL MCS: 0.04 (0.16) versus -0.002 (0.13), mean difference (95%CI) 0.04

(0.002 to 0.08), p=0.04), favouring the intervention, but no significant difference in diabetes-specific

distress (PAID: 5.6 (15.5) versus -2.4 (15.4)) nor physical health (AQoL PCS: 0.03 (0.15) versus 0.02

(0.13)) nor diabetes-specific distress (PAID: 5.6 (15.5) versus -2.4 (15.4)). No severe hypoglycaemia

events were reported.

Conclusions

Our novel model of care increased insulin initiation rates in primary care, improving glycated

hemoglobin without worsening emotional well-being. Delays in starting insulin treatment can be

overcome, enabling timely treatment intensification to improve important health outcomes.

Trial registration Australian and New Zealand Clinical Trials Registry (ACTRN12612001028897)

http://www.anzctr.org.au/

Page 108 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

5

What this paper adds

What is already known on this subject?

• Achieving and maintaining glycaemic targets early in type 2 diabetes improves long term

outcomes.

• There are barriers to early stepwise progressive treatment intensification to achieve glycaemic

targets, particularly in relation to insulin initiation in primary care.

• Interventions have had limited success in overcoming this delay in starting insulin treatment and

changing clinical practice, in part because system level barriers are not addressed.

What this study adds

• Our model of care intervention changed clinical practice, with more patients in intervention arm

practices commencing insulin, with an overall benefit in terms of HbA1c reduction, achieved

without serious adverse events, or any worsening in depressive symptoms.

• The theoretical base and flexible implementation are important characteristics of our

intervention.

• Our model of care used existing resources to work smarter and improve outcomes and, thus, has

important implications for policy makers, funders and practitioners seeking innovative ways to

provide best care for people with type 2 diabetes in primary care.

Page 109 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

6

Introduction

Nearly 600 million people will have type 2 diabetes type 2 diabetes worldwide by 2030.1 Innovation

in delivering effective clinical care to these people is an urgent global priority. To reduce the risk of

long-term macro- and microvascular complications,2 UK, European and US guidelines recommend

early adoption of insulin as a part of step-wise treatment intensification to bring glycated

hemoglobin (HbA1c) below a general target of 7% (53mmol/mol).3-6

However, insulin initiation is

frequently delayed, particularly in primary care,7 where despite being recommended as part of

routine clinical management of type 2 diabetes6, implementation is not widespread. The mean

HbA1c of people with type 2 diabetes prior to starting insulin is typically 1.5-2.0% above target: 9.3%

(78mmol/mol) in the UK8 8.6% (70mmol/mol) in a study in the US

9, 8.9% (74mmol/mol) in a large,

multi-country primary care study10

, and 9.4% (79mmol/mol) in a community study in Australia (after

a median diabetes duration of 8.1 years)11

.

Delay in treatment intensification by healthcare professionals, despite evidence that intensification

is warranted and effective12

is a major barrier to initiating insulin in type 2 diabetes13

. It can be due

to health professional factors14

(e.g. concerns about hypoglycaemia risk, lack of confidence/skills in

insulin initiation/titration), health system factors15

(e.g. competing priorities in busy, reactive

primary care settings), and patient-related factors (e.g. psychological insulin resistance16

).

Supporting and embedding insulin initiation as an element of routine primary care practice is an

important first step and building block in potentially reducing referrals to costly secondary care, and

supporting timely, early optimisation of therapy and achievement of glycaemic targets.

Our trial investigated the effectiveness of a new model of care designed to support primary care-

based insulin initiation among people with type 2 diabetes for whom it is clinically indicated. The

Stepping Up model of care is built around an enhanced, reconfigured role for the primary care

Practice Nurse Practice Nurse.17

In Australia, clinical care of type 2 diabetes is predominantly

undertaken in primary care, largely made up of General Practitioner private group practices where

Page 110 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

7

Practice Nurses are a rapidly growing section of the workforce, taking on an increasing role in

chronic condition management. In this setting, our approach to professional behaviour change was

informed by Normalisation Process Theory18

and developed through pilot studies.17 19

Our aim was to

address both clinician and system level barriers to timely insulin initiation, and to normalise insulin

initiation as part of standard primary care practice. We hypothesised that HbA1c would improve

among participants in intervention arm practices, facilitated through timely insulin initiation,

compared to with the control arm. Further, based on previous research 20 21

, we expected no

significant negative impact on participants’ general emotional wellbeing (depressive symptoms).

Methods

Study design and participants

The study design and protocol have been described previously.22

In summary, we conducted a 12-

month, two-arm, non-blinded cluster-randomised controlled trial (RCT), consistent with CONSORT

guidelines,23

to investigate the effectiveness of the Stepping Up model of care versus usual care.

General Practices in Victoria, Australia were eligible if they had at least one consenting General

Practitioner (GP) and Practice Nurse and could identify at least one eligible patient participant:

adults with type 2 diabetes with above target HbA1c (≥7.5% (58mmol/mol)) in the past six months

who were already prescribed maximum oral therapy (i.e. at least two oral hypoglycaemia agents

(OHAs) at maximum doses) or if their GP judged that insulin would be clinically appropriate. Patients

were ineligible if they were >80 years old, already using insulin, had an eGFR <30 mL/min/1.73m2,

unable to give informed consent or had a complex debilitating medical condition, e.g., severe mental

illness, end-stage cancer, unstable cardiovascular disease.

Our original protocol was based on 58 practices and an average cluster size of 5. Based on

experience in the field and what was an achievable sample size this was subsequently revised to 74

practices and an average cluster size of 3 (see Appendix 1).

Page 111 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

8

Randomisation

The unit of randomisation was the primary care practice. The study statistician (AG) computer-

generated stratified block randomisation sequences with varying block sizes (4, 6 and 8) prior to

recruitment. Practices were stratified by size (≤2 vs >2 full-time equivalent GPs), setting (private

practice vs community health centre), and participation (or not) in type 2 diabetes quality

improvement programs (the Australian Primary Care Collaborative). After providing consent and

recruiting at least one eligible patient, practices were randomised to intervention or usual care. We

used this index case method24

in all practices as our previous experience suggested that delaying

randomisation of a cluster until all patients have been recruited risks loss of engagement of GPs25

.

The research team then assisted practices to continue to identify and recruit patient participants

(through searching the practice medical record database). This meant that allocation concealment

after the index case was recruited was not possible for the GP and Practice Nurse. However, in order

to minimise potential bias, participating patients were not informed of their study allocation until

after they had provided consent. Blinding of GP, Practice Nurse and patient was also not possible

given the pragmatic nature of the intervention.

Intervention

The Stepping Up model of care, described elsewhere, 17 22

involved a re-orientation of existing

resources. It included: a) an enhanced role for the Practice Nurse in leading the discussion with

patients about intensifying treatment through insulin initiation and titration, b) simple insulin

initiation and up-titration clinical protocols, and c) a re-oriented role for the specialist Registered

Nurse-Credentialed Diabetes Educator in mentoring the Practice Nurse rather than providing direct

patient care. The training and mentoring support was designed to enhance the primary care team’s

knowledge, skills and confidence in discussing and implementing insulin initiation within the practice

as a part of routine care. Intervention practices had an in-practice briefing and training session for

GPs and Practice Nurses of approximately 60-90 minutes, following which patients with confirmed

Page 112 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

9

eligibility and completed baseline data were invited to consult their GP for an assessment to discuss

treatment intensification and referral to the Practice Nurse. Practice Nurses did not prescribe insulin,

or manage insulin dosing without liaison with the GP, based on the legal scope of practice for

generalist Practice Nurses in Australia.

Our model of care involved the acknowledgment and discussion of advantages and disadvantages of

starting insulin treatment with patients, including weight gain. We modelled shared decision making

as a part of the intervention training, drawing on the principles of motivational interviewing. This set

the scene for encouraging practitioners in intervention practices to approach participating patients

with equipoise in relation to starting insulin. While the intervention was necessarily brief in this

pragmatic trial, we included guidance and checklists for GPs and Practice Nurses to discuss the pros

and cons of insulin therapy and elicit patient concerns and expectations, while also openly

acknowledging and accepting that some patients may choose not to start insulin.

The role of Registered Nurse-Credentialed Diabetes Educator in supporting and mentoring the

Practice Nurse, and of the Practice Nurse in leading the discussion and implementation of insulin

therapy with the patient in liaison with the GP, is outlined elsewhere.15,20

Titration protocols were

based on fasting blood glucose and use of a 3-day, 7-point blood glucose profile26

to identify the

meal with the largest postprandial excursion (see appendix material). We gave no additional

instructions, so the GP had clinical autonomy regarding the management of OADsOHAs. Practice

Nurses and GPs were encouraged to see patients as frequently as was felt to be clinically appropriate

over a period of up to 12 months, drawing as needed upon the study Registered Nurse-Credentialed

Diabetes Educator for mentoring and support, even if the patient remained undecided about, or had

decided against, starting insulin. Further details about the intervention can be found in the

referenced papers and the Appendices (Appendix 2 – Intervention Description; Appendix 3 –

Training Manual for Intervention Practices; Appendix 4 – Patient participant booklet).

Page 113 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

10

Control arm practices were given a copy of Australian type 2 diabetes management guidelines27

.

Control arm practices were offered training in the Stepping Up model of care after 12-month follow

up of patient participants was complete.

Patient involvement

Participant feedback was sought after conducting an intervention pilot study15

and this feedback was

used to refine the model of care. Participant feedback was also sought in a pilot of the data

collection forms. Throughout the main trial we maintained communication with patient participants

through a regular newsletter which included aggregate data about study progress and opportunities

to provide feedback to the study team. We assessed the burden of the intervention on patients

through interviews conducted at the end of the trial, as part of process evaluation (to be reported

elsewhere). This evaluation was led by the Chronic Illness Alliance, a consumer advocacy

organization that has been a long term collaborator of our research group. We have thanked all

participants for their involvement in the trial, and will provide, at a later date, a final summary report

of the trial outcomes. Participants have access to the study website where all published results will

be publicly available.

Endpoints and data collection

While our intervention targeted a process of care (i.e. insulin initiation), we chose a clinically

meaningful disease outcome at an individual patient level as primary endpoint: change (from

baseline to 12 months) in HbA1c, measured as a continuous variable. We registered our primary

outcome as “an absolute HbA1c reduction of 0.5% in the intervention group compared with the

control group”. HbA1c was performed at DCCT-aligned pathology laboratories and communicated to

clinicians and patients as part of usual clinical care; researchers retrieved these data from medical

records or directly form pathology laboratories. Secondary endpoints included the proportion of

participants who transitioned to insulin (this was amended from the original protocol where rate of

insulin initiation was used, which proved impractical given the small cluster size in the study); the

Page 114 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

11

proportion who achieved a target HbA1c of ≤7.0% (53mmol/mol) at 12 months; and change (from

baseline to 12 months) in depressive symptoms (PHQ-9),28

diabetes-specific distress (PAID),29

and

generic health status (AQoL-8D).30

Differences between the registered outcomes reported here and

those registered in the trial registry and justification and explanation for any changes made can be

found in the AppendicesAppendix 1. We also collected data on healthcare utilisation and costs, to be

reported elsewhere.

All participants were provided with a blood glucose meter (Performa NanoTM

; Roche Diagnostics)

and instructed on its use. Subsidised low-cost blood glucose testing strips were available through the

National Diabetes Service Scheme31

. Data were uploaded from the meter at 6 and 12 months to a

secure server.

Statistical analysis

Our statistical analysis plan has been published elsewhere.22

In brief, our sample size of 224 patients

from 74 general practices (averaging 3 patients per practice) allowed us to detect an absolute 0.5%

mean HbA1c difference over 12 months between control and intervention arms with 80% power and

a standard deviation (SD) of 1 using two-sided alpha of 0.05. Data were analysed using Stata 13

(StataCorp, TX, USA). Descriptive statistics were used to summarise GP, Practice Nurse, and patient

characteristics for the two study arms and appropriate univariate tests were conducted to check for

any significant difference in potential confounders. Parametric data are reported as mean (SD) and

non-parametric data as median (IQR). Categorical data are reported as n (%). T-test for proportions

was used to compare baseline and 12-month data within each study arm for categorical data, while

paired t-tests were for used for continuous parametric data and Wilcoxon matched pairs signed rank

test for continuous non-parametric data. The individual patient was the unit of analysis and the

analytical methods allow for clustering of patients within the practices. Marginal logistic modelling

using generalised estimating equations with robust standard errors and adjustment for baseline

measures and clustering were used to compare binary outcomes between the two study arms.

Page 115 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

12

Mixed-effects linear regression was used to determine predictors for continuous outcomes,

adjusting for baseline measures and clustering at the practice level. As no imbalance in confounders

was identified between study groups no adjustment for confounders was made in the regression

analysis. Analyses were conducted on an intention-to-treat basis.

Study oversight

The study protocol was approved by the University of Melbourne Health Sciences Human Research

Ethics Sub-committee (ID 123740) and registered with the Australian and New Zealand Clinical Trials

Registry (ACTRN12612001028897). All participants gave informed consent before enrolment.

Results

Participating practices and patients

Between October 2012 and January 2014, 93 primary care practices expressed interest and

identified 521 potentially eligible patients (Figure 1). Subsequently, 19 practices did not consent any

eligible patients, leaving 74 participating practices for randomisation. Two hundred and fifty five of

the potentially eligible patients were subsequently found to be ineligible at screening (n=156) or did

not respond to the invitation letter (n=99). By April 2014, the 74 practices had identified and

consented 266 eligible patient participants (73% of potentially eligible patients identified).

Baseline characteristics of practices, GPs, Practice Nurses and participants with type 2 diabetes are

shown in Table 1. Of the total sample, 248 (93.2%) completed 12-month follow-up for the primary

endpoint. No differences in baseline characteristics were observed between study completers and

non-completers, except for a higher proportion of women (n=11) not completing than men (n=7).

Primary and secondary endpoints

Primary and secondary endpoints are shown in Table 2. At 12 months, there was a statistically and

clinically significant difference between study arms in terms of change in HbA1c (mean

difference: -0.76%; 95% CI -1.10.9 to -0.43%, p<0.001), favouring the intervention. This is consistent

Page 116 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

13

with achieving our registered primary outcome of “an absolute HbA1c reduction of 0.5% in the

intervention group compared with the control group”. The majority of this change in HbA1c seen in

both arms was achieved by six months (See Figure 2).

In the intervention arm 105/151 (69.5%) of patients commenced insulin (and 102 remained on

insulin at 12 months), while in the control arm, 25/115 (21.7%) commenced insulin (and 24

remained on insulin at 12 months). Median (IQR) number of days from baseline assessment to

insulin initiation in intervention and control group patients who started insulin was 32 (11.5, 134.5)

days and 85 (63, 191) days respectively (statistically significant difference: Two-sample Wilcoxon

rank-sum test; p=0.005). In the intervention arm, 17 (11.22%) patients had commenced rapid-acting

insulin at 12 months, compared with one patient in the control arm (p<0.001). Further data on

insulin use in participants is available in Appendix 5.

Target HbA1c (≤7% (53mmol/mol)) was achieved by 54 (35.8%) intervention participants (32 of

whom were using insulin at 12 months) and 22 (19%) of the control arm (two of whom had

commenced insulin), (OR 2.2, 95%CI (1.2 to 4.3,) p=0.00802). Twenty-two (14.7%) intervention arm

participants and 20 (17%) control arm participants achieved target HbA1c without commencing

insulin.

At 12 months, depressive symptoms had not worsened and there was a no statistically significant

difference between arms in the mean (SD) change in depressive symptoms (PHQ-9: -1.1 (3.5) versus

-0.1 (2.9), p=0.02). There was a and statistically significant difference in mental health (AQoL MCS:

0.04 (0.16) versus -0.002 (0.13), p=0.054), favouring the intervention, but no significant difference

significant difference in diabetes-specific distress (PAID: -5.6 (15.5) versus -2.4 (15.4)) nor in physical

health (AQoL PCS: 0.03 (0.165) versus 0.02 (0.13)) or diabetes-specific distress (PAID: -5.6 (15.5)

versus -2.4 (15.4)). There was no significant difference between arms in the proportion of

participants experiencing moderate-to-severe depressive symptoms or severe diabetes-specific

distress (Table 2), nor was there any difference by insulin initiation (Wilcoxon rank sum test; p=0.98).

Page 117 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

14

No significant difference was found in PHQ-9, PAID or AQoL or PAID scores when comparing

participants initiating insulin with those who did not (data not shown).

At 12 months, there was an average weight gain in the intervention arm and an average weight loss

in the control group (1.7 (5.2) kg versus -1.1 (5.1) mean (95% CI) difference 2.8 (1.5 to 4.0) kg,

p<0.001). There were no significant differences in blood pressure nor other biochemical measures

between arms at follow-up with the exception of triglycerides, which remained higher in the control

group.

At baseline, participants were using a mean (SD) of 2.0 (0.6) classes of non-insulin hypoglycaemic

agents, with no significant difference between arms (t test, p=0.89). The majority of patients were

prescribed metformin (93% across both arms) and sulfonylureas (63% across both arms). There was

no significant difference in the prescription of individual medication classes by study arm. The mean

(SD) number of classes of non-insulin hypoglycaemic agents being used at 12 months was higher in

the control compared to the intervention arm (2.3 (0.1) vs 1.9 (0.1); p=0.01). There was were a

higher proportion of people in the control group using DPP4 inhibitors than the intervention group

at 12 months (Table 3).

Practices in the intervention arm received a total of 183 mentoring support visits from the study

Registered Nurse-Credentialed Diabetes Educator (mean (range) visits per practice: 5.2 (1-8)). Thirty-

two percent (48/151) of participants in intervention practices completed at least one 3-day, 7-point

structured blood glucose monitoring profile over the 12-month study. Practice nurses estimated the

time they spent on the study (both clinical interactions with participating patients and on research

tasks). On a per-practice basis, 23 control practices and 27 intervention practices reported a median

(IQR) of 1.5 (0, 3.6) and 18 (9, 20.9) hours respectively.

58.3% of people in the control group were recruited prior to practice randomisation in the control

group and compared to 45.0% in the intervention group (statistically significant difference p=0.033).

Sensitivity analysis was conducted to explore whether there was any difference at baseline between

Page 118 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

15

patients who were recruited before and after practice randomisation. No statistically significant

differences were found. Mixed-effects linear regression was used to determine the impact of the

intervention, adjusting for clustering at the practice level for each of these groups. Treatment effect

in terms of the primary outcome remained significant in both groups (treatment effect: -0.57 95%CI:

-1.1 to, -0.05 p value = 0.03 and -0.98 95%CI -1.49 to, -0.48 p value <0.001 for patients assessed

prior to and after randomisation respectively).

Adverse events

No severe hypoglycaemic events (i.e. requiring third-party assistance for recovery) or other adverse

events were reported in either study arm.

Discussion

Principal findings

Our model of care changed clinical practice, with the majority in the intervention arm commencing

insulin, producing a clinically and statistically significant improvement in glycaemic control among

adults with type 2 diabetes managed in primary care. This was despite a higher patient-to-GP ratio in

intervention practices, and was achieved safely, with no severe hypoglycaemic events, and without

deterioration in emotional wellbeing nor health status. Our results indicate that, with appropriate

support and practice-system redesign, insulin initiation can become part of routine primary care

diabetes management, obviating the need to refer to specialist services with geographical, cost and

accessibility barriers.

Strengths and limitations of study

A strength of our study is the robust theoretical and empirical base to our intervention. Our

pragmatic trial of a complex intervention addressed a number of known barriers to overcoming

delay in starting insulin therapy. For example, an intentional component of our system redesign was

re-orienting the Practice Nurse and Registered Nurse-Credentialed Diabetes Educator roles, allowing

Page 119 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

16

additional time to be spent with patients, within existing resources. Other strengths include the

cluster-randomised design, minimising the risk of contamination, and our excellent participant

retention (93%).

Our study had limitations. First, practices were randomised after the first consenting patient was

identified, raising the possibility of selection bias. However, the balance in key patient characteristics

between the study arms means that any such bias was minimal. Secondly, while a smaller cluster

size is generally preferable in a cluster-randomised trial, the relatively large variation in the cluster

sizes in our study may make statistical adjustments for clustering less effective, in particular when

the number of clusters is not large. SecondlyThirdly, our sample may not have been fully

representative of the broader population of adults with type 2 diabetes managed in primary care for

whom insulin is clinically indicated. Overall, less than 15% of our sample had severe diabetes-specific

distress or moderate-to-severe depressive symptoms, a lower rate than in a recent national

Australian sample.32

We will explore implementation fidelity and variation in more detail through a

qualitative process evaluation in a subsequent paper. Finally, our medications and hypoglycaemia

data were derived from GP records and subject to the same accuracy limitations of any routinely

collected clinical dataset. Hypoglycaemia is typically under-estimated 33

and is likely to be under-

reported in routine medical records. 34

In particular, severe hypoglycemia is serious but relatively

rare, and may not have been detected in our study, given our sample size.

Comparison with other studies

Only two other trials have tested interventions to change clinical practice in this way. The AIM@GP

trial showed no improvement in insulin prescribing rates or glycated hemoglobin.35

It provided

scheduled and ongoing telephone support from a specialist diabetes educator as well as the option

to refer patients to a community pharmacist off-site for a one-hour insulin initiation session. Our

intervention differed in that it was based completely in the familiar environs of the patients’ own

primary care practice, built on existing relationships and resources (with the practice-based Practice

Page 120 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

17

Nurses) and provided an immediate pathway for the GP to delegate this clinical task. A UK study,

using a pre-post evaluation design, showed improved HbA1c (-1.4%) at 6 months in patients who

initiated insulin,36

similar in magnitude to the improvement in our study. That intervention combined

education with face-to-face and telephone specialist diabetes nurse support for GPs and Practice

Nurses and involved a full-day, off-site training for both GP and Practice Nurse. In contrast, our

intervention used a brief (60-90 minutes) on-site training incorporated into the daily running of the

practice with flexible Registered Nurse-Credentialed Diabetes Educator support as required.

Consistent with previous research,37

our study participants did not report worse psychological

outcomes at follow up, suggesting that insulin therapy can be initiated in primary care for people

with type 2 diabetes without impairing their emotional wellbeing. In fact, at 12 months, the

reduction in depressive symptoms in intervention arm participants was significantly greater than

that seen in the control arm. Insulin initiation (across both arms) did not impact on depressive

symptoms, diabetes-specific distress or generic health status at 12 months.

Conclusions and policy and practice implications

Our pragmatic trial findings have important implications for the organisation of healthcare and for

health policy. Our model of care is based on an enhanced role for a Practice Nurse and would not be

feasible where primary care doctors work in solo practice without access to a Practice Nurse.

Nevertheless, even in high-resource settings, where the move to multidisciplinary primary health

care teams is growing, for example through the growth of the Medical Home movement 38, our

study suggests that simply having access to a Practice Nurse will not increase appropriate insulin

initiation. To make the best use of resources, primary care workforce models need to be developed

and implemented to reorient the way specialists (Registered Nurse-Credentialed Diabetes Educators

and endocrinologists) offer support to primary care teams that include well-supported and

resourced primary care nurses. Registered Nurse-Credentialed Diabetes Educator Change is

occurring in some health systems,38

but our study provides evidence of the effectiveness and safety

of such models of care. Rather than waiting for referrals, specialist services need to offer pro-active,

Page 121 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

18

tailored secondary consultation, liaison and mentoring services that are flexible and supportive of

the needs of primary care practitioners and patients. Scaling up the model of care in metropolitan

centres would require engaging with hospitals and other health services in reorienting the role of

Registered Nurse-Credentialed Diabetes Educators currently employed in direct patient care.

Investigation of e-health modalities (e.g. online training, support and video consultations) may

support implementation of this model of care in more distant rural/remote settings. Registered

Nurse-Credentialed Diabetes Educator

Our trial findings also have implications for clinical practice. The issue around personalising

glycaemic targets and treatments is an important and emerging consideration in the care of people

with T2D. At the time of our trial started, up there was vigorous debate about the need for caution

in setting lower targets. In addition our exclusion criteria ruled out participants wherefor whom a

higher target would definitely be considered. Collecting reliable data on duration of CVD, severe

hypoglycaemia and hypoglycaemia unawareness did not prove feasible in this setting. Therefore we

made a pragmatic decision to use the general target of 7%. Debate continues about the advantages

and disadvantages of intensifying treatment for people with type 2 diabetes at the HbA1c levels

mandated in our study. While our findings suggest the that Stepping Up model of care builds clinical

capacity within GP and Practice Nurse teams to undertake the work of insulin initiation, the model of

care did not mandate a dogmatic approach to such therapy changes. It is worth noting the major

change in HbA1c was achieved at 6 months and that even in intervention practices, only 35% of

participating people with type 2 diabetes achieved the general HbA1c target of <7% (53 mmol/mol),

suggesting that that practitioners and patients were judicious in the way they approached

progressive treatment intensification, within the new model of care.

Our clinical protocols and algorithms were focused solely on insulin. It is worth noting that NPH

Insulin remains widely used and that the added costs of analogue insulins where NPH can be used

without problem is still subject to debate. While we did not pre-specify weight gain as a secondary

outcome, this is an adverse effect of insulin therapy and we have chosen to report it. While insulin is

Page 122 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

19

still regarded as an essential therapy option, as the range of glycaemic therapies grows clinical

algorithms become more complex5. Future research needs to explore the capacity to generate real-

time personalised treatment intensification recommendations that incorporate this increasing

complexity,39

to be used as part of the Practice Nurse-led model of care. Future research could also

address the extent to which improvements in glycaemic levels are maintained, and the extent to

which the model of care is sustained in routine clinical practice. In particular, future research could

explore use of the model of care to specifically support early adoption of insulin therapy to achieve

glycaemic targets early in people with recently diagnosed type 2 diabetes.

The global epidemic of type 2 diabetes demands innovation in care delivery. Delaying insulin

initiation when clinically indicated is neither ethical nor effective. Furthermore, health systems will

not cope with demand if insulin initiation remains anchored in specialist centres, nor will they be

able to respond to the imperative to achieve glycaemic targets early in people with recently

diagnosed type 2 diabetes. Thus our pragmatic, translational study has important implications across

health systems globally for the organisation of care for people with type 2 diabetes. Our effective

model of care has the potential to improve outcomes in people with type 2 diabetes while making

better use of scarce healthcare resources.

Page 123 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

20

Figure 1: Consort Diagram

Assessed for eligibility (93 practices, patients n=521)

Excluded: Practices - failed to recruit patients – n = 19 Patients (n=255) ♦ Ineligible (n=156); ♦ Non-responder (n=99)

Intention to treat analysis

Clusters: 38 practices, 115 patients

Allocated to control arm (n=38 practices) Received allocated control

• 38 practices, patients (n=115)

• mean cluster size 3, range 1-6

Allocated to intervention arm (n=36 practices) Received allocated intervention

• 36 practices, patients (n= 151)

• mean cluster size 4, range 1-8

Intention to treat analysis

Clusters: 36 practices, 151 patients

Allocation

Analysis

Follow-Up

Enrolment

Randomized (74 practices (n=74) with eligible patients (n = 266)

Did not complete follow-up (n= 9)

Withdrawn (n=5)

• no reason given (n=4)

• patient developed cancer (n=1)

Patient deceased (n=1) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=3)

Did not complete follow-up (n=9)

Withdrawn (n=2)

• no reason given (n=1)

• patient pregnant (n=1) Patient deceased (n=2) Patient lost to follow up/not contactable/unable to attend for 12 month timepoint (n=5)

Page 124 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

21

Table 1: Baseline characteristics of participating practices, health professionals and adults

with type 2 diabetes1

Intervention Control

Primary care practices 36 (49) 38 (51)

Type of practice

Private practice . 27 (75.0) 31 (81.6)

Corporate practice . 7 (19.4) 5 (13.2)

Community health centre . 2 (5.6) 2 (5.3)

Location of practice

Major city . 26 (72.2) 21 (55.3)

Inner regional area . 9 (25.0) 13 (34.2)

Outer regional area . 1 (2.8) 4 (10.5)

Physicians per practice (median (IQR)) 5 (4, 9.5) 5 (4, 9)

Practice Nurses per practice (median (IQR)) 2.5 (2, 3.5) 2 (1, 4)

Registered Nurse-Credentialed Diabetes

Educator on site: yes

. 12 (33.3) 14 (36.8)

Patients per Full-Time Equivalent GP

(median (IQR))1*

1738 (1176, 2727) 1316 (911, 1726

General Practitioners 83 (51.2) 79 (48.8)

Age, years2

. 48.8 (9.9) 49.7 (11.2)

Female . 34 (41.0) 27 (34.2)

Working hours/week2 . 36.6 (10.5) 37.3 (11.6)

Years of experience (median (IQR)) 19 (8, 26) 20 (7, 30)

Experience with insulin initiation in the

preceding 12 months3

. 48 (60.0) 36 (46.2)

Practice Nurses 48 (46.6) 55 (53.4)

Age, years4

. 44.7 (10.2) 46.0 (9.9)

Female 48 (100) 55 (100)

Diabetes educator training . 6 (12.5) 7 (12.7)

Experience with insulin initiation in the

preceding 12 months

. 16 (33.3) 16 (29.1)

Adults with type 2 diabetes 151 (56.8) 115 (43.2)

Age, years . 61.7 (9.7) 62.0 (10.6)

Female . 62 (41.1) 41 (35.7)

Highest level of education

Page 125 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

22

Primary or less . 14 (9.3) 12 (10.4)

Secondary or trade . 101 (66.9) 83 (72.2)

Tertiary . 36 (23.8) 20 (17.4)

Employed . 67 (44.4) 50 (43.5)

Health care card holder . 75 (49.7) 62 (53.9)

Diabetes duration, years (median (IQR)) . 8 (5, 12) 9 (5, 14)

HbA1c % (median (IQR)

mmol/mol (median (IQR))

. 8.7 (8.1, 9.7)

72 (65, 83)

8.5 (8, 9.6)

69 (64, 81)

Number of medical conditions (median

(IQR))

3 (2, 5) 3 (2, 5)

Number of medications (median (IQR)) 6 (5, 10) 7 (5, 10)

Medication adherence rating scale (median

IQR))5

29 (26, 30) 29 (27, 30)

Diabetes complications6

Microvascular . 17 (11.3) 16 (13.9)

Macrovascular . 22 (14.6) 21 (18.3)

Total cholesterol, mmol/L7

. 4.3 (1.0) 4.2 (1.1)

Triglycerides, mmol/L7* . 1.9 (0.1) 2.3 (1.4)

LDL cholesterol, mmol/L8 . 2.3 (0.9) 2.1 (0.9)

HDL cholesterol, mmol/L9 . 1.2 (0.3) 1.1 (0.3)

eGFR10

. 79.4 (14.4) 78.8 (14.6)

Blood pressure

Systolic . 134.6 (15.7) 133.5 (15.2)

Diastolic . 79.6 (11.1) 78.5 (9.5)

Data are mean (SD) or n (%) unless otherwise stated

1 Data available for 67 practices (33 intervention, 34 control)

2 Data available for 161 GPs (82 intervention, 79 control)

3 Data available for 158 GPs (84 intervention, 74 control)

4 Data available for 100 Practice Nurses (46 intervention, 54 control)

5 Data available for 261 patients (149 intervention, 112 control)

6 n(%) with at least one complication

7 Data available for 256 patients (144 intervention, 112 control)

8 Data available for 222 patients (130 intervention, 92 control)

9 Data available for 233 patients (134 intervention, 99 control)

10 Data available for 261 patients (147 intervention, 114 control)

Page 126 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

23

*Statistically significant difference between control and intervention groups

Registered Nurse-Credentialed Diabetes Educator: Registered Nurse-Credentialed Diabetes Educator

Page 127 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only24

Table 2: Primary and secondary endpoints: Biochemical, clinical and psychological outcomes

Endpoints Intervention Control Adjusted data for baseline measure

and clustering

Treatment effect (95% CI) p

HbA1c %

Baseline 8.7 (8.1, 9.7) 8.5 (8, 9.6)

Follow-up 7.4 (6.9, 8.2) 8 (7.1, 9)

Change -1.3(1.4) -0.6(1.5) -0.6 (-0.9 to -0.3) <0.001

Participants using insulin+ Follow-up 105(69.5) 25 (21.7) 8.3* (4.5 to 15.4) <0.001

Participants with HbA1c ≤53mmol/mol (7%)+ Follow-up 54 (35.8) 24 (20.9) 2.2* (1.2 to 4.3) 0.02

Depressive symptoms (PHQ-9)1 Baseline

3(1,7) 2(1, 6.5)

Follow-up

2(0, 5) 2(0, 5)

Change -1.1(3.5) -0.1(2.9) -0.8 (-1.6 to -0.01) 0.047

Moderate-to-severe depressive symptoms (PHQ-9 total: ≥10)2+

Baseline 22(15.1) 15(13.5)

Follow-up 19(12.8) 15 (13.3) 0.82+* (0.3 to 2.2) 0.69

Diabetes-specific distress (PAID)3 Baseline 15(6.3, 31.3) 12.5(5, 23.8) 0.14

Page 128 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only25

Follow-up 8.8(3.8, 22.5) 10(2.5, 23.8)

Change -5.6(15.5) -2.4(15.4) -1.9 (--5.1to 1.3) 0.24

Severe diabetes-specific distress (PAID total: ≥40)2+

Baseline 25 (16.8) 14(12.4)

Follow-up 18 (12.1) 12(10.4) 1.0* (0.4 to 2.3) 0.93

Health status (AQoL-8D) – Physical Component Score4 Baseline 0.63(0.20) 0.61 (0.21)

Follow-up 0.66(0.21) 0.64(0.21) 0.52

Change 0.03(0.15) 0.02(0.13) 0.01 (-0.03 to 0.04) 0.64

Health status (AQoL-8D) – Mental Component Score4

Baseline 0.45(0.20) 0.45(0.22)

Follow-up 0.48(0.21) 0.45(0.22)

Change 0.04(0.16) -0.002(0.13) 0.04 (0.002 to 0.08) 0.04

Weight, kg Baseline 90.8(19.6) 94.6(18.9)

Follow-up 92.5(20.1) 93.5(18.9)

Change 1.7 (5.2) -1.1(5.1) 2.8 (1.5 to 4.0) <0.001

Page 129 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only26

Endpoints Intervention Control Adjusted data for clustering

Treatment effect (95% CI) p

HbA1c %

Baseline 8.7 (8.1, 9.7) 8.5 (8, 9.6) 0.37

Follow-up 7.4 (6.9, 8.2) 8 (7.1, 9)

Change -1.3(1.4) -0.6(1.5) -0.7 9 (-1.1 to -0.4) <0.001

Participants using insulin+ Follow-up 105(69.5) 25 (21.7) 8.3* (4.5 to 15.4) <0.001

Participants with HbA1c ≤53mmol/mol (7%)+ Follow-up 54 (35.8) 24 (20.9) 2.2* (1.2 to 4.3) 0.02

Depressive symptoms (PHQ-9)1 Baseline

3(1,7) 2(1, 6.5) 0.17

Follow-up

2(0, 5) 2(0, 5)

Change -1.1(3.5) -0.1(2.9) -1.0 (-1.8 to -0.1) 0.02

Moderate-to-severe depressive symptoms (PHQ-9 total: ≥10)2+

Baseline 22(15.1) 15(13.5) 0.73

Follow-up 19(12.8) 15 (13.3) 1.0* (0.7 to 2.0) 0.92

Diabetes-specific distress (PAID)3 Baseline 15(6.3, 31.3) 12.5(5, 23.8) 0.14

Follow-up 8.8(3.8, 22.5) 10(2.5, 23.8)

Change -5.6(15.5) -2.4(15.4) -3.3 (-7.2 to 0.6) 0.10

Page 130 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only27

Severe diabetes-specific distress (PAID total: ≥40)2+ Baseline 25 (16.8) 14(12.4) 0.64

Follow-up 18 (12.1) 12(10.4) 0.2* (-0.6 to 0.9) 0.68

Health status (AQoL-8D) – Multidimensional4

Baseline 0.30(0.19) 0.31(0.22) 0.81

Follow-up 0.34(0.21) 0.32(0.21)

Change 0.04(0.15) 0.01(0.13) 0.04 (-0.001 to 0.07) 0.06

Health status (AQoL-8D) – Physical Component Score4

Baseline 0.63(0.20) 0.610.21

Follow-up 0.66(0.21) 0.640.21 0.52

Change 0.03(0.15) 0.02(0.13) 0.01 (-0.03 to 0.04) 0.75

Health status (AQoL-8D) – Mental Component Score4 Baseline 0.45(0.20) 0.45(0.22) 0.75

Follow-up 0.48(0.21) 0.450.22

Change 0.04(0.16) -0.0020.13 0.04 (0.001 to 0.08) 0.05

Weight, kg Baseline 90.8(19.6) 94.618.9 0.11

Follow-up 92.5(20.1) 93.518.9

Change 1.7 (5.2) -1.15.1 2.8 (1.6 to 4.1) <0.001

Data are mean (SD) and median (IQR) unless otherwise indicated

Page 131 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only28

1 PHQ-9: Patient Health Questionnaire 9. Range of possible scores: 0-27. A total score of ≥10 indicates at least moderate depressive symptoms. Data available for 261

patients at baseline (149 intervention, 112 control) and 263 at 12 months (149 intervention, 114 control; ITT)

2 Data available for 257 patients at baseline (146 intervention, 111 control) and 261 at 12 months (148 intervention, 113 control; ITT).

3 PAID: Problem Areas In Diabetes. Range of possible scores: 0-100. A score of ≥40 indicates severe diabetes-related distress. Data available for 262 patients at baseline

(149 intervention, 113 control) and 264 at 12 months (149 intervention, 115 control; ITT)

4 AQoL-8D: Assessment of Quality Of Life. Maximum possible score is 1. Higher scores indicate better generic health status. Data available for 262 patients at baseline (149

intervention, 113 control) and 263 at 12 months (149 intervention, 114 control; ITT)

*Odds Ratio

+n (%)

Page 132 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only29

Figure 2: Change in primary endpoint at 6 and 12 months

Page 133 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only30

Table 3: Classes of non-insulin medications at 12 months

Intervention n(%) Control n(%) P value

N 146 108

Metformin 133 (91.1) 96 (88.9) 0.56

Sulphonylurea 75 (51.4) 64 (59.3) 0.21

Acarbose 3 (2.1) 2 (1.9) 0.91

DPP4 inhibitors* 25 (17.1) 38 (35.2) 0.001

Glitazones 6 (4.1) 5 (4.6) 0.8411

SGLT2 inhibitors 2 (1.4) 2 (1.9) 0.76

GLP1 agonists 9 (6.2) 7 (6.5) 0.92

(t test of proportions)

* At the time of the trial DPP4 inhibitors were not subsidized for use with insulin

Page 134 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

31

Authorship: JF, IB, DY and JB conceptualised the original study proposal and secured funding. JF had

overall responsibility for the study. JF, DON, JS, JMN, EHT, LG, DY, JB, EP, DL, LS, CM and IB drafted

the protocol. LG also contributed to data collection. JF, DON, JS, JMN, AG, EHT, DL and IB contributed

to the statistical analysis plan and JMN and AG led the data analysis. JF wrote the original draft of

the report. All authors contributed to data interpretation and approved the final report. As

corresponding author and Principal Investigator, JF had full access to all the data in the study and

takes responsibility for the integrity of the data and the accuracy of the data analysis. JF had final

responsibility for the decision to submit for publication and is guarantor for the study affirming that

the manuscript is an honest, accurate, and transparent account of the study being reported; that no

important aspects of the study have been omitted; and that any discrepancies from the study as

planned have been explained.

Competing interests: All authors have completed the ICMJE uniform disclosure form at

www.icmje.org/coi_disclosure.pdf and declare: We acknowledge funding from the Australian

National Health and Medical Research Council (Project Grant Application: APP1023738). The study

was also supported by an educational/research grant by Roche Diabetes Care Pty Ltd, the RACGP

Foundation RACGP/Independent Practitioner Network Pty Ltd (IPN) Grant and received in-kind

support from Sanofi. Xclinical hosted the BG data. JF was supported by a National Health and

Medical Research Council Career Development Fellowship. JS is supported by core funding to The

Australian Centre for Behavioural Research in Diabetes from Diabetes Victoria and Deakin University.

JMN was supported by a National Health and Medical Research Council postgraduate scholarship.

EHT is supported by an Australian Postgraduate Award Deakin University PhD scholarship.

JF has received unrestricted educational grants for research support from Roche Diabetes Care,

Sanofi and Medtronic; JS is a member of the Accu-Check Advisory Board (Roche Diabetes Care). Her

research group (ACBRD) has received unrestricted educational grants from Medtronic and Sanofi

Page 135 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

32

Diabetes; sponsorship to host or attend educational meetings from Lilly, Medtronic, MSD, Novo

Nordisk, Roche Diabetes Care, and Sanofi Diabetes; consultancy income from Abbott Diabetes Care,

Astra Zeneca, Roche Diabetes Care and Sanofi Diabetes; DNO, DL and JMN had various financial

relationships with pharmaceutical industries outside the submitted work including consultancies,

grants, lectures, educational activities and travel. DNO has received research and travel support and

honoraria from Sanofi, Roche and Novo and is an advisory board member to Sanofi, Novo and

Abbott. JMN has no financial relationships with companies marketing blood glucose monitoring

devices, but has received payment from Sanofi who funded the control practice training at end of

study. DL has received honoraria and research grants from Sanofi Australia. The study sponsors had

no role in the design and conduct of the study; collection, management, analysis, and interpretation

of the data; and preparation, review, or approval of the manuscript.

Funding

The study was funded by the Australian National Health and Medical Research Council (ID 1023738)

and was supported by an educational/research grant by Roche Diagnostics Australia Pty Ltd.

We thank the GPs, Practice Nurses and people with type 2 diabetes who generously participated in

the study.

Data sharing: Anonymised patient level data are available on reasonable request from the authors.

Page 136 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

33

References

1. Zimmet PZ, Magliano DJ, Herman WH, et al. Diabetes: a 21st century challenge. The Lancet

Diabetes & Endocrinology 2014;2(1):56-64.

2. Holman RR, Paul SK, Bethel MA, et al. 10-Year Follow-up of Intensive Glucose Control in Type 2

Diabetes. N Engl J Med 2008;359(15):1577-89.

3. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Patient-Centered Approach: Position Statement of the American Diabetes Association (ADA)

and the European Association for the Study of Diabetes (EASD). Diabetes Care

2012;35(6):1364-79.

4. Nathan DM, Buse JB, Davidson MB, et al. Management of Hyperglycemia in Type 2 Diabetes: A

Consensus Algorithm for the Initiation and Adjustment of Therapy: Update regarding

thiazolidinediones: a consensus statement from the American Diabetes Association and the

European Association for the Study of Diabetes. Diabetes Care 2008;31(1):173-75.

5. American Diabetes Association. Standards of Medical Care in Diabetes—2016 Abridged for

Primary Care Providers. Clinical Diabetes 2016;34(1):3-21.

6. McGuire H, Longson D, Adler A, et al. Management of type 2 diabetes in adults: summary of

updated NICE guidance. BMJ 2016;353.

7. Shah BR, Hux JE, Laupacis A, et al. Clinical Inertia in Response to Inadequate Glycemic Control: Do

specialists differ from primary care physicians? Diabetes Care 2005;28(3):600-06.

8. Blak BT, Smith HT, Hards M, et al. A retrospective database study of insulin initiation in patients

with Type 2 diabetes in UK primary care. Diabetic Medicine 2012;29(8):e191-e98.

9. Curtis B, Lage MJ. Glycemic control among patients with type 2 diabetes who initiate basal insulin:

a retrospective cohort study. Journal of Medical Economics 2014;17(1):21-31.

10. Khunti K, Damci T, Meneghini L, et al. Study of Once Daily Levemir (SOLVE™): insights into the

timing of insulin initiation in people with poorly controlled type 2 diabetes in routine clinical

practice. Diabetes, Obesity and Metabolism 2012;14(7):654-61.

11. Davis TME, Davis WA, Bruce DG. Glycaemic levels triggering intensification of therapy in type 2

diabetes in the community: the Fremantle Diabetes Study. Medical Journal of Australia

2006;184(7):325-8.

12. Phillips LS, Branch JWT, Cook CB, et al. Clinical Inertia. Annals of Internal Medicine

2001;135(9):825-34.

13. Khunti S, Davies M, Khunti K. Clinical inertia in the management of type 2 diabetes mellitus: a

focused literature review. The British Journal of Diabetes and Vascular Disease

2015;15(2):65-69.

14. Kunt T, Snoek FJ. Barriers to insulin initiation and intensification and how to overcome them.

International Journal of Clinical Practice 2009;63:6-10.

15. Furler J, Spitzer O, Young D, et al. Barriers and enablers to timely initiation of insulin in General

Practice" Australian Family Physician 2011;40(8):617-23.

16. Polonsky WH, Fisher L, Guzman S, et al. Psychological Insulin Resistance in Patients With Type 2

Diabetes: The scope of the problem. Diabetes Care 2005;28(10):2543-45.

17. Furler JS, Blackberry ID, Walker C, et al. Stepping up: a nurse-led model of care for insulin

initiation for people with type 2 diabetes. Family Practice 2014;31(3):349-56.

18. May C, Mair F, Finch T, et al. Development of a theory of implementation and integration:

Normalization Process Theory. Implementation Science 2009;4(1):29.

19. Blackberry ID, Furler JS, Ginnivan LE, et al. An exploratory trial of basal and prandial insulin

initiation and titration for type 2 diabetes in primary care with adjunct retrospective

continuous glucose monitoring: INITIATION study. Diabetes Research and Clinical Practice

2014;106(2):247-55.

Page 137 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

34

20. Hajos TRS, Pouwer F, de Grooth R, et al. Initiation of insulin glargine in patients with Type 2

diabetes in suboptimal glycaemic control positively impacts health-related quality of life. A

prospective cohort study in primary care. Diabetic Medicine 2011;28(9):1096-102.

21. Dzida G, Karnieli E, Svendsen AL, et al. Depressive symptoms prior to and following insulin

initiation in patients with type 2 diabetes mellitus: Prevalence, risk factors and effect on

physician resource utilisation. Primary Care Diabetes;9(5):346-53.

22. Furler J, Young D, Best J, et al. Can primary care team-based transition to insulin improve

outcomes in adults with type 2 diabetes: the stepping up to insulin cluster randomized

controlled trial protocol. Implementation Science 2014;9(1):20.

23. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting

parallel group randomised trials. BMJ 2010;340:698-702.

24. Giraudeau B, Ravaud P. Preventing Bias in Cluster Randomised Trials. PLoS Medicine

2009;6(5):e1000065.

25. Blackberry ID, Furler JS, Best JD, et al. Effectiveness of general practice based, practice nurse led

telephone coaching on glycaemic control of type 2 diabetes: the Patient Engagement And

Coaching for Health (PEACH) pragmatic cluster randomised controlled trial. BMJ 2013;347.

26. Polonsky WH, Fisher L, Schikman CH, et al. Structured self-monitoring of blood glucose

significantly reduces A1C levels in poorly controlled, noninsulin-treated type 2 diabetes:

results from the Structured Testing Program study. Diabetes Care 2011;34(2):262-7.

27. The Royal Australian College of General Practitioners and Diabetes Australia. General practice

management of type 2 diabetes – 2014–15. Melbourne, , 2014.

28. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. Validity of a Brief Depression Severity Measure.

Journal of General Internal Medicine 2001;16(9):606-13.

29. Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in Diabetes Scale - An evaluation of

its clinical utility. Diabetes Care 1997;20(5):760-66.

30. Richardson J, Iezzi A, Khan M, et al. Validity and Reliability of the Assessment of Quality of Life

(AQoL)-8D Multi-Attribute Utility Instrument. Patient 2014;7(1):85-96.

31. The National Diabetes Service Scheme. Secondary The National Diabetes Service Scheme.

https://www.ndss.com.au/the-ndss.

32. Speight J, Browne JL, Holmes-Truscott E, et al. Diabetes MILES - Australia 2011 Survey Report.

Canberra: Diabetes Australia, 2011.

33. Elliott L, Fidler C, Ditchfield A, et al. Hypoglycemia Event Rates: A Comparison Between Real-

World Data and Randomized Controlled Trial Populations in Insulin-Treated Diabetes.

Diabetes Therapy 2016;7(1):45-60.

34. Frier BM, Jensen MM, Chubb BD. Hypoglycaemia in adults with insulin-treated diabetes in the

UK: self-reported frequency and effects. Diabetic Medicine 2016;33(8):1125-32.

35. Harris S, Gerstein H, Yale J-F, et al. Can community retail pharmacist and diabetes expert support

facilitate insulin initiation by family physicians? Results of the AIM@GP randomized

controlled trial. BMC Health Services Research 2013;13(1):71.

36. Dale J, Martin S, Gadsby R. Insulin initiation in primary care for patients with type 2 diabetes: 3-

Year follow-up study. Primary Care Diabetes 2010;4(2):85-89.

37. Pouwer F, Hermanns N. Insulin therapy and quality of life. A review. Diabetes-Metabolism

Research and Reviews 2009;25:S4-S10.

38. Gucciardi E, Espin S, Morganti A, et al. Exploring interprofessional collaboration during the

integration of diabetes teams into primary care. BMC Family Practice 2016;17(1):1-14.

39. O'Connor PJ, Desai JR, Butler JC, et al. Current Status and Future Prospects for Electronic Point-

of-Care Clinical Decision Support in Diabetes Care. Current Diabetes Reports 2013;13(2):172-

76.

Page 138 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

Appendix 1: Table of changes made to outcome measures and sample size with explanatory notes

Registered Protocol document Published protocol paper Explanatory notes

Primary

An absolute HbA1c reduction of

0.5% in the intervention group

compared with the control

group

The primary outcome of the

study is change in HbA1c

(measured as a continuous

measure)

Change (from baseline to 12

months) in HbA1c, measured as

a continuous variable

The registered primary outcome

was not well worded to capture

and describe our intention, ie to

identify an absolute difference

of 0.5% in the change in the

mean HbA1c between baseline

and 12 months in the

intervention and control groups.

Therefore in the protocol

document, final version Sept

2013, we wrote “The primary

outcome of the study is change

in HbA1c (measured as a

continuous measure)” and

hypothesised that “Participants

in the intervention arm will

achieve an absolute HbA1c

mean difference of 0.5% lower

than control group participants

at 12 months (primary

outcome)”

Secondary

1. Proportion of participants

transferring to insulin at 12

months post baseline;

1. The rate of insulin initiation in

participating patients

1. The proportion of participants

who transitioned to insulin (this

was amended from the original

protocol where rate of insulin

initiation was used, which

proved impractical given the

small cluster size in the study);

Because of small cluster sizes,

reporting rates of insulin

initiation was not feasible, as

explained in the text.

Page 139 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

2. The level of HbA1c at which

insulin initiation occurs at 12

months post baseline;

2. The level of HbA1c at which

insulin initiation occurs

We did not have access to data

to be able to report the level of

HbA1c at which insulin initiation

occurs at 12 months post

baseline

3. Proportion of participants

achieving adequate glycaemic

control (HbA1c <7%) at 12

months post baseline;

4. Proportion of participants

achieving individualised HbA1c

targets according to the

Australian Diabetes Society

guidelines at 12 months post

baseline;

3. The proportion of patients

who achieve glycaemic control

(HbA1c<7.0%)

4. The proportion of patients

who achieve individualized

HbA1c targets according to

Australian Diabetes Society

guidelines

2. The proportion who achieved

a target HbA1c of ≤7·0%

(53mmol/mol) at 12 months;

Not reported

We amended <7% to <7% to be

consistent with guidelines

The issue around personalising

glycaemic targets and

treatments is an important and

emerging consideration in care

of people T2D. At the time of

our trial start up there was

vigorous debate about the need

for caution in setting lower

targets. In addition our exclusion

criteria ruled out participants

where a higher target would

definitely be considered.

Page 140 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

Collecting reliable data on

duration of CVD, severe

hypoglycaemia and

hypoglycaemia unawareness did

not prove feasible in this setting.

Therefore we made a pragmatic

decision to use the general

target of 7%. Our manuscript

includes text about the apparent

judicious approach practitioners

were adopting in relation to

pursuing glycaemic targets and

we believe this realistically

addresses the underlying issue

of personalised care on the basis

of data available to us.

5. Impact on psychometric

scores as assessed using the

Assessment of Quality of Life

(AQoL­8D), Patient Health

Questionnaire (PHQ­9) and

Problem Areas in Diabetes

(PAID) at 12 months post

baseline;

5. Changes in psychometric

scores including AQoL­8D, PHQ­

9 and PAID

3. Change (from baseline to 12

months) in depressive symptoms

(PHQ­9), diabetes­specific

distress (PAID),generic health

status (AQoL­8D).

No change

6. Reduction in net healthcare

utilization and costs as assessed

using Medicare Benefits Scheme

and Pharmaceutical Benefits

Scheme and Victorian

Department of Health data

linkage (subject to approval)

at12 months post baseline.

6. Reduced net health care

utilization and costs

Not reported We will report economic

analyses in a separate paper.

Page 141 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

Sample

size

290 patients from 58 general

practices (5 patients per

practice)

290 patients from 58 general

practices (5 patients per

practice)

Our sample size of 224 patients

from 74 general practices

(averaging 3 patients per

practice) allowed us to detect an

absolute 0·5% mean HbA1c

difference over 12 months

between control and

intervention arms with 80%

power and a standard deviation

(SD) of 1 using two­sided alpha

of 0·05.

Our original protocol was based

on 58 practices and an average

cluster size of 5. Based on

experience in the field and what

was an achievable sample size

this was subsequently revised to

74 practices and an average

cluster size of 3.

Page 142 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyBriefing visit

GPs and PNs participated in a practice briefing visit led by the study DNE. At this visit, the Stepping

Up Program resources were detailed to GPs and PNs. Roles and responsibilities of GP, PN, DNE and

other health professionals in initiating insulin were discussed. Practices were encouraged to identify

local DNEs, dieticians and endocrinologists for support and/or referral in ongoing management of

patients.

Training session

GPs and PNs attended an in-practice training session of 1.5 hours. This was facilitated by a GP with

skills in diabetes management and a study DNE. This session focused on the rationale for use of

insulin, further clarification of role of PN and GP, the clinical protocol for starting and titrating doses

and addressing common patient-level barriers (including motivational interviewing and goal setting

strategies based on case presentations). A paper-based tool on which patients completed 3-day

fasting blood glucose levels was presented and discussed. The tool was designed to facilitate dose

adjustment in discussion between patient and PN. There was opportunity for hands on

familiarisation with insulin delivery systems.

GP and PN working together

Following the training session, practices called participating patients in for a GP review consultation,

followed by an appointment with the PN on the same day. GPs reviewed the patients’ diabetes

management and if appropriate made a recommendation for starting insulin. The patient then saw

the PN for further discussion and an insulin initiation assessment, during which the PN worked

through the resources in the ‘patient pack’ with the patient and a checklist to guide discussion about

the advantages and disadvantages of insulin therapy. If the patient agreed, at that visit the PN gave

the first dose of glargine insulin 10 units. The PN reviewed the patient the following day, guided by a

checklist and observed the patient self-administer a second dose of glargine insulin 10 units.

Subsequently, the patients called the PN for review by telephone every 3 days and were supported

to follow the simple protocol for dose titration, done in liaison with the GP (although patients had

the option of moving onto self-titration using the protocol when confident). If the patient decided

not to start insulin at that visit, a review visit was arranged in one month at which the topic was

discussed again.

GPs and PNs were asked to see participating patients as often as they deemed clinically necessary

over the follow-up period of 3 months. Consultation and referral to DNE and endocrinologist was at

the clinical discretion of the GP. The study DNE and endocrinologist were also available for

consultation as needed.

Page 143 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyPatient Booklet

Name:

Practice Name:

Ste

ppin

g U

p S

tud

yPractice ID:Patient ID:

Patient_Booklet_v3.indd 1 10/10/2012 3:06:37 PM

Page 144 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

Thank you for agreeing to participate in the Stepping Up Study.

This booklet contains some information regarding type 2 diabetes and insulin which you can refer to during your participation in this study. Please bring this book to all of your appointments, together with your blood glucose meter which we will give to you as part of this study.

Friends or family members may also like to read this information to learn more about your diabetes and its management.

Please remember to return this Booklet to your Practice Nurse at the end of the Study.

Other helpful sources of information include:

> Diabetes Australia (www.diabetesaustralia.com.au, Phone 1300 136 588)

> Local diabetes support groups

Contents

General Practice Contact Details ........................................................................................................... 1

My Appointments .................................................................................................................................... 2

General Information for People with Type 2 Diabetes ........................................................................ 3

How much exercise should I do and why is it important for people with diabetes? ....................... 3

Some tips for healthy food choices for people with diabetes .......................................................... 7

Blood glucose monitoring – when should I do it and why? And how can I tell if my diabetes is well controlled? ........................................................................................................................................ 15

What is a ‘hypo’ and what do I do if I have one? ............................................................................. 21

What should I do if I get sick? .......................................................................................................... 25

Frequently Asked Questions about Insulin ........................................................................................... 27

A ‘How to’ Guide to Using Solostar Insulin Pens....................................................... Inside back cover

Patient_Booklet_v3.indd 2 10/10/2012 3:06:37 PM

Page 145 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyGeneral Practice Contact Details

GenerAl PrACTICe

Address:

Phone:

Fax:

Opening Hours:

If this book has been lost, please return it to:

Dr Irene Blackberry

Department of General Practice

University of Melbourne

200 Berkeley St

Carlton Vic 3053

| 1

PrACTICe nUrSe

Name:

Days / time of work:

GenerAl PrACTITIoner

Name:

Days / time of work:

Patient_Booklet_v3.indd 3 10/10/2012 3:06:38 PM

Page 146 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

2 |

Stepping Up StudyMy Appointments

Please fill in this appointment page with the assistance of the Practice Nurse

Date Time

Screening Visit with the Practice Nurse

Baseline GP Visit

Baseline Practice Nurse Visit 1

Baseline Practice Nurse Visit 2

Practice Nurse Phone Consultations

Practice Nurse Visits at the Clinic

Insulin Review Visit with the GP

Insulin Review Visit with the Practice Nurse

Patient 6 month Data Collection Visit

Final (12 month) GP visit

Final (12 month) Practice Nurse Visit

Additional appointments

Reason Date Time

 

Patient_Booklet_v3.indd 4 10/10/2012 3:06:38 PM

Page 147 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

| 3

Stepping Up StudyGeneral Information for People with Type 2 Diabetes

How much exercise should I do and why is it important for people with diabetes?

Getting enough regular physical activity is important for maintaining good health and ensuring good diabetes management. While you may be thinking ‘that’s easier said than done’, you may be surprised to learn that exercising isn’t about ‘no pain-no gain’. Regular physical activity can become an enjoyable part of your day with long-term benefits to your diabetes and your overall health.

Why do it?Physical activity is essential to everyone to stay healthy. For people with diabetes, being regularly physically active has even greater benefits.

For the person with diabetes, physical activity helps to:Improve the body’s response to insulin which can lower blood glucose levels•Lower blood pressure and cholesterol levels, reducing the risk of heart disease•Control weight•Reduce the risk of developing diabetes complications•

Other positives include:Stronger bones • Improved mood•Increased energy levels •Reduced stress and tension•Improved sleep•

What activities are recommended?Aerobic activities that get your large muscles moving such as walking, swimming or cycling are all recommended. Not everyone finds activity enjoyable, so choose activities you enjoy doing and get you moving.

Strength training activities that make you use your muscles against a resistance, such as squats or lifting weights, are also recommended.

Ideas to help you to ‘get moving’:Walking is easy, cheap and doesn’t take any special skill – just a good pair of walking •shoes. Up the pace and distance covered as you get fitter.Be creative and try something different – perhaps ballroom dancing, water aerobics, water •walking or Tai Chi. Check with your local community centre about free programs.

Physical activity benefits everyone in many ways but for people with diabetes, being regularly physically active, eating well and not smoking are very important to their future health.

Talking diabetes No.27

physical activity & type 2 diabetes

Continued over...

Revised August 2009 A diabetes information series from State / Territory organisations of Diabetes Australia – Copyright© 2009

Patient_Booklet_v3.indd 5 10/10/2012 3:06:38 PM

Page 148 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

aim for at least 30 minutes each day of 'moderate-intensity' physical activity“4 |

physical activity & diabetesStand and move about while on the phone rather than sitting.•

Consider buying a pedometer (step counter) and count your daily steps, aiming to •increase to a level decided by you and your doctor.

Think about things you do using a machine. Could you do them in a more active way?•

How much is enough?The length of time you spend will depend on the type of activity you choose and whether you need to lose weight.

Moderate intensity exercise

Aim to do at least 30 minutes of moderate intensity physical activity such as brisk walking or swimming every day. This can be two 15-minute sessions or even three 10-minute sessions. To achieve a level of moderate intensity physical activity, you need to notice your breathing and heart rate speeding up and perhaps a light sweat. If you are gasping and unable to talk, you are probably working too hard.

If you’re trying to lose weight, you may need to aim for 60 to 90 minutes every day. This will depend on how active you are already and other things such as the food you eat. Discuss this with your doctor, dietitian and exercise physiologist.

Vigorous intensity exercise

As an alternative to moderate intensity exercise, you may choose to do three 20-minute sessions per week of vigorous intensity exercise such as jogging, running, aerobics classes or strenuous gardening.

Strength training

Aim to include strength training twice a week in addition to your moderate intensity or vigorous intensity exercise. Perform 8–10 different exercises using all the major muscle groups. Repeat each exercise 8–12 times, completing two sets for each exercise. Lift a weightthatyoucanlift8–12timesbutfinddifficulttoliftonthelastfewrepetitions.

Strength training activities include body weight exercises such as wall pushups or sitting and standing from a chair, machine based exercises or free weight exercises such as lifting dumb bells. If you are unsure how to do strength training exercises safely, consult an exercise physiologist, physiotherapist or other appropriate health professional.

What should I do before starting an activity program?Diabetes can put you at risk of certain conditions that could be affected by physical activity. This check list will help you to ‘get moving’ with safety.

Before starting any new activity program, check with your doctor who will consider your •blood glucose levels, any diabetes related complications and the condition of your heart and blood vessels.

Your doctor may advise you to have a stress test as a precaution if you:•> are over 35> have had type 2 diabetes for more than 10 years

aim for at least 30 minutes each day of ‘moderate-intensity’ physical activity

Revised August 2009

2

Patient_Booklet_v3.indd 6 10/10/2012 3:06:38 PM

Page 149 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 5

aim for at least 30 minutes each day of 'moderate-intensity' physical activity”

> have high blood pressure> have/have had heart problems.

As most physical activity involves using your feet, consider seeing a podiatrist before •you start your program for advice on suitable footwear and other helpful information.

Is there anything I need to do before and during my physical activity session?

Set yourself goals to stay motivated – and when you achieve your goals, •reward yourself!

Foryourfirstfewsessions,itisagoodideatotestyourbloodglucoselevelbefore,•during and after exercise, especially if you’re on insulin or certain diabetes tablets that can lower blood glucose levels. Always carry quickly absorbed glucose such as jellybeans or glucose tablets in case your blood glucose level drops too low. For more information refer to the Hypoglycaemia and Diabetes information sheet.

Weargoodquality,wellfitting,closed-infootwearasrecommendedbyyourpodiatrist.•

Start slowly, gradually increasing the pace and length of each session. •

Aim to do your activity sessions at regular times and on set days.•

Do not be physically active if you are unwell.•

Don’t get dehydrated. Drink enough water to avoid thirst and remember you will need •a bit more than usual while being active.

Take short breaks along the way if being active for long periods.•

Weardiabetesidentification(eg:MedicAlert• ®).

Believe that physical activity is as vital to your health as the air you breathe. Each time •you set out on your activity session, make a mental commitment: “This is forever”.

Doing your activity sessions with a friend or family member or as part of a regular •group can help you to stay motivated and make it more fun.

Wear sunscreen, protect your head and layer your clothing so you can add or remove •clothes as needed.

Are there times when I should stop my activity session?Stop and rest if you experience chest, abdominal, neck or arm pain or tightness, or •even vague discomfort. Stop and rest if you feel breathless, faint or lightheaded or have any other unusual symptoms while exercising. These symptoms could mean heart trouble that requires urgent treatment.

If these symptoms – any symptoms – do not settle within 10 minutes, you or someone with you MUST call an ambulance to take you to the nearest hospital emergency department immediately. If the symptoms settle in less than 10 minutes, you should go to your doctor as soon as possible for a checkup. This must be done before you do more exercise.

aim for at least 30 minutes each day of ‘moderate-intensity’ physical activity

Revised August 2009

3

Patient_Booklet_v3.indd 7 10/10/2012 3:06:38 PM

Page 150 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

6 |

physical activity & diabetes If you experience leg pain, stop until the pain goes away then resume your activity. Make •sure your doctor knows about the leg pain. Gradually you should be able to exercise longer without pain but treatment may be required.

If you are experiencing symptoms of a hypo, stop, test blood glucose levels and •treat your hypo. Wait 10 to 15 minutes, test again and follow up with longer acting carbohydrate such as a sandwich, glass of milk or two biscuits. Do not continue to exercise until your symptoms have disappeared. For more information refer to the Hypoglycaemia and Diabetes information sheet.

Is there anything I need to do after my activity session?Check your feet after exercise or at least once a day looking for signs of redness, •blisters, cracks and calluses. If your feet perspire, change your socks after activity.

Physical activity can lower your blood glucose level for up to 48 hours afterwards, so do a •test. You might notice a temporary rise after activity. This rise varies between individuals and is due to the release of hormones during periods of intense muscle activity.

Each time you reach your goal, reward yourself with a movie, a new shirt or a low fat •latte, then set new goals to stay motivated.

Know how your body respondsAs everybody reacts differently, it is important to know your own blood glucose response to activity. Many of the early signs of a hypo (eg: sweating, feeling faint and weakness) are also feelings you may have during physical activity and can therefore go unnoticed. Your diabetes educator, dietitian or doctor will advise if your medication, insulin or eating plan needs adjusting.

If you have concerns about any activity program, talk to your doctor or diabetes educator or contact your State or Territory Diabetes Organisation on 1300 136 588.

Would you like to join Australia’s leading diabetes organisation?> Dietary services > Free magazines > Children’s services> Educational literature > Product discounts > Support groups

For more information phone 1300 136 588 or visit your State/Territory Organisation’s website:

ACT www.diabetes-act.com.au NSW www.diabetesnsw.com.auNT www.healthylivingnt.org.au QLD www.diabetesqld.org.auSA www.diabetessa.com.au TAS www.diabetestas.com.auVIC www.diabetesvic.org.au WA www.diabeteswa.com.au

The design, content and production of this diabetes information sheet has been undertaken by:

> Diabetes Australia – NSW > Diabetes Australia – Victoria > Diabetes Australia – Queensland > Diabetes Australia – Tasmania > Diabetes ACT > Diabetes SA > Diabetes WA > Healthy Living NTThe original medical and educational content of this information sheet has been reviewed by the Health Care and Education Committee of Diabetes Australia Ltd. Photocopying this publication in its original form is permitted for educational purposes only. Reproduction in any other form by third parties is prohibited. For any matters relating to this information sheet, please contact National Publications at [email protected] or phone 02 9527 1951.

Health professionals: For bulk copies of this resource, contact Diabetes Australia in your State/Territory.

Revised August 2009 A diabetes information series from State / Territory organisations of Diabetes Australia – Copyright© 2009

Patient_Booklet_v3.indd 8 10/10/2012 3:06:38 PM

Page 151 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 7

Some tips for healthy food choices for people with diabetes

1

Healthy eating, along with regular physical activity, can help you to manage your blood glucose levels, reduce your blood fats (cholesterol and triglycerides) and maintain a healthy weight. Refer to the Physical Activity and Type 2 Diabetes information sheet for more about how to be active every day.

What should I eat?To help manage your diabetes, your meals need to be:

> An appropriate size – not too large> Regular and spread evenly throughout the day> Lower in fat, particularly saturated fat> Based on high fibre carbohydrate foods such as wholegrain breads and cereals,

dried beans, lentils, starchy vegetables and fruits.���

On the following pages, we give information about different types of foods and their effect on our health:

Fats page 2Carbohydrates page 3Sugar/Alternative sweeteners page 5Protein page 5Alcohol page 6

A typical one day meal plan:

Breakfast page 7Light meal page 7Main meal page 7Between meal snacks (if required) page 8

Healthy eating for people with diabetes is no different to that which is recommended for everyone. There is no need to prepare separate meals or buy special foods, so relax and enjoy healthy eating along with the whole family!

Talking diabetes No.15

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

food choices for people with diabetes

Revised August 2010

Patient_Booklet_v3.indd 9 10/10/2012 3:06:38 PM

Page 152 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

eating too much fat may make you gainit more difficult to manage

weight which in the long run may make your blood glucose levels“

8 |

2

food choices

FatFats have the highest energy (kilojoule or calorie) content of all foods.��� Eating too much fat may make you gain weight which in the long run may make it more difficult to manage your blood glucose levels.��� On the other hand small amounts of healthier fats add flavour to your food, may improve your health and reduce your risk of heart disease.��� Therefore, the type of fat you eat is important, as well as the amount.���

Saturated fat and trans fatIt is important to limit saturated and trans fats because they raise your LDL-C (‘bad’ cholesterol) levels and lower good cholesterol.��� Saturated fat is found in animal foods like fatty meat, milk, butter and cheese.��� Vegetable fats that are saturated include palm oil (found in solid cooking fats, snack foods or convenience foods) and coconut products such as copha, coconut milk or cream.��� Trans fat occurs naturally in small amounts in meat and dairy food as well as in other foods during the manufacturing process such as margarine.��� Food sources of trans fats include hard margarines (usually cooking margarines), deep fried foods and commercial food products made with shortening such as pastries.���

To reduce saturated and trans fats:

> Choose reduced or low fat milk, yoghurt, ice cream and custard.���> Choose lean meat and trim any fat off before cooking.���> Remove the skin from chicken (where possible, before cooking).���> Avoid using butter, lard, dripping, cream, sour cream, copha, coconut milk,

coconut cream and hard cooking margarines.���> Limit cheese, try reduced fat varieties.���> Limit pastries, cakes, puddings, chocolate and cream biscuits to

special occasions.���> Limit pre-packaged biscuits, savoury packet snacks, cakes, frozen and

convenience meals.���> Limit the use of processed deli meats (devon/polony/fritz/luncheon meat, chicken

loaf, salami etc) and sausages.���> Avoid fried takeaway foods such as chips, fried chicken and battered fish.��� Choose

BBQ chicken (without the skin) and grilled fish instead.���> Avoid pies, sausage rolls and pasties.���> Avoid creamy sauces or dressings.��� Choose sauces based on tomato or other low

fat ingredients and low fat dressings made from small amounts of polyunsaturated or monounsaturated fats (eg: sunflower, grapeseed, olive or canola oils).��� Some sauces and dressings can be very high in salt, even if they are low fat.��� Choose lower salt varieties or make them yourself without any added salt.���

> Limit creamy style soups.���

eating too much fat may make you gain weight which in the long run may make it more difficult to manage your blood glucose levels

Patient_Booklet_v3.indd 10 10/10/2012 3:06:39 PM

Page 153 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 9

eating too much fat may make you gainit more difficult to manage

weight which in the long run may make your blood glucose levels”

3

Polyunsaturated and monounsaturated fatSome fat is important for good health.��� Use a variety of polyunsaturated and monounsaturated types to achieve a good balance.���

Polyunsaturated fats includePolyunsaturated margarines (check the label for the word ‘polyunsaturated’)• Sunflower, safflower, soybean, corn, cottonseed, grapeseed and sesame oils• The fat found in oily fish such as herring, mackerel, sardine, salmon and tuna•

Monounsaturated fats includeCanola and olive margarines• Canola and olive oil• Avocado•

Seeds, nuts, nut spreads and peanut oil contain a combination of polyunsaturated and monounsaturated fat.���

Ideas for enjoying healthy fats:

> Stir-fry meat and vegetables in a little canola oil (or oil spray) with garlic or chilli

> Dress a salad or steamed vegetables with a little olive oil and lemon juice or vinegar

> Sprinkle sesame seeds on steamed vegetables

> Use linseed bread and spread with a little canola margarine

> Snack on a handful of unsalted nuts, or add some to a stir-fry or salad

> Spread avocado on sandwiches and toast, or add to a salad

> Eat more fish (twice a week) because it contains a special type of fat (omega 3) that is good for your heart

> Do more dry roasting, grilling, microwaving and stir-frying in a non-stick pan

Carbohydrate Carbohydrate foods are the best energy source for your body.��� When they are digested they break down to form glucose in the bloodstream.��� If you eat regular meals and spread your carbohydrate foods evenly throughout the day, you will help maintain your energy levels without causing large rises in your blood glucose levels.���If you take insulin or diabetes tablets, you may need to eat between-meal snacks.��� Discuss this with your Accredited Practising Dietitian (APD) or Credentialled Diabetes Educator (CDE).���All carbohydrate foods are digested to produce glucose.��� The amount of carbohydrate you eat will affect how high your blood glucose levels rise after a meal.��� Too large a serve can mean too large a rise.���

eating too much fat may make you gain weight which in the long run may make it more difficult to manage your blood glucose levels

Patient_Booklet_v3.indd 11 10/10/2012 3:06:39 PM

Page 154 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

10 |

4

food choicesAlthough all carbohydrate foods break down into glucose, they do so at different rates – some slow, some fast.��� The glycemic index (GI) is a way of describing how a carbohydrate-containing food affects blood glucose levels.��� The GI of foods will also affect your blood glucose response.���The best combination is to eat moderate amounts of carbohydrate and include high fibre foods that also have a low GI.��� Your dietitian (APD) can give you an idea of how much you need to eat.���

The foods listed below are high in carbohydrate and are healthy choices. Those in bold have a lower GI:

> Bread or bread rolls – especially wholegrain and wholemeal varieties such as Burgen® Breads, Tip-Top 9 Grain Original, 9 Grain Wholemeal®, 9 Grain Original Mini Loaf, Wonderwhite® Lower GI.

> High fibre breakfast cereals such as rolled oats, All-Bran®, Guardian®, Weet-Bix® or untoasted muesli.���

> Pasta, rice (Basmati or Doongara) and other grains such as barley, bulgur and couscous.���

> Legumes – baked beans, kidney beans, chick peas, lentils, 3 bean mix.> Fruit – all types such as apples, oranges, peaches, bananas, melons.��� Fruit is a

good source of fibre; try to eat the whole fruit rather than drinking the juice.��� Include at least 2 serves of fruit a day (1 serve = 1 medium piece apple, orange or pear OR 2 small pieces kiwifruit, plum).���

> Milk products or dairy alternatives – choose low fat varieties of milk, soy drink (calcium fortified), yoghurt and custard.��� Include 2–3 serves a day (1 serve = 1 cup of milk OR 200 g yoghurt).���

> Vegetables that contain a significant amount of carbohydrate – potatoes, orange sweet potato, yams, corn.��� Other vegetables (such as salad vegetables, green vegetables, and orange vegetables) are generally low in carbohydrate and therefore have little effect on your blood glucose levels.��� Include at least 5 serves of vegetables each day (1 serve = 1 cup salad vegetables OR ½ cup cooked vegetables OR 1 medium potato).���

> You also need to consider a food’s other nutritional qualities such as fat, added sugar and salt content.��� While some high fat foods and many sugary foods have a low GI, such as chocolate, ice cream and toasted muesli, they are often not suitable for everyday eating.���

> Some occasional foods (such as dry or sweet biscuits, chocolate or chips) and sugary foods (such as jam, honey or sugar) are also carbohydrate foods.��� These can be eaten in small amounts.���

the best combination is to eat moderate amounts of carbohydrate and include high fibre foods that also have a lower GIthe best combination is to eat

include high fibre foodsmoderate amounts of carbohydrate andthat also have a lower GI“

Patient_Booklet_v3.indd 12 10/10/2012 3:06:39 PM

Page 155 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 11

5

SugarA healthy eating plan for diabetes can include some sugar.��� However, it is still important to consider the nutritional value of the foods you eat.��� In particular, high energy foods such as sweets, lollies and standard soft drinks should not be consumed.���Some sugar may also be used in cooking and many recipes can be modified to use less than the amount stated or substituted with an alternative sweetener.��� Select recipes that are low in fat (particularly saturated fat) and contain some fibre.��� In general, foods with added sugars should be consumed sparingly (manufacturers sometimes use fruit juice or other sources of sugar to avoid using table sugar).��� If too much is eaten at one time, they may affect your weight, dental health and overall diabetes control.��� Discuss with your dietitian or diabetes educator about when and how frequently to include these types of foods/drinks.���

Alternative sweeteners While it is no longer necessary to always use alternative sweeteners instead of sugar, artificially sweetened products are suitable alternatives for foods and drinks that are high in added sugars, such as cordials and soft drinks.���Alternative sweeteners based on acesulphame K (950)*, aspartame (951)*, cyclamate (952)*, saccharin (954)*, sucralose (955)*, alitame (956)*, stevia (960)* or neotame (961)* are all suitable for people with diabetes.��� They don’t provide kilojoules, won’t affect blood glucose levels and are found in many low joule products.��� These have all been approved for use in Australia by Food Standards Australia New Zealand.��� However it is important to remember that many foods that use alternative sweeteners (such as soft drinks) are not everyday foods so should still be consumed in small amounts.���

Protein Most protein foods do not directly affect your blood glucose levels.��� They include lean meat, poultry without the skin, seafood, eggs (not fried), unsalted nuts and soy products such as tofu and legumes (dried beans and lentils).��� Legumes are a good source of fibre and should be included regularly.��� They are also a carbohydrate food so will affect your blood glucose levels.��� Protein foods do provide important nutrients for good health.��� However most Australians already eat enough protein and do not need to eat more.���

Other foods, condiments and drinksYou can use these foods to add flavour and variety to your meals:

> Herbs, spices, garlic, chilli, lemon juice, vinegar and other seasonings> Products labelled ‘low joule’ eg: low joule/diet soft drinks, low joule jelly> Water, soda water, plain mineral water, tea, coffee, herbal tea

the best combination is to eat moderate amounts of carbohydrate and include high fibre foods that also have a lower GI

* This number may appear on the ingredient list in place of the name.���

the best combination is to eatinclude high fibre foods

moderate amounts of carbohydrate andthat also have a lower GI”

Patient_Booklet_v3.indd 13 10/10/2012 3:06:39 PM

Page 156 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

12 |

6

food choices

AlcoholIf you enjoy alcohol, it is generally acceptable to have two standard drinks a day*.��� However, if you need to lose weight, you may need to limit your alcohol intake further.��� It’s best to drink alcohol with a meal or some carbohydrate-containing food and aim to have alcohol free days.���

One standard drink is equal to:

> 100 ml wine > 285 ml regular beer> 30 ml spirits > 60 ml fortified wine> 425 ml low-alcohol beer (less than 3% alcohol)

It is important to remember:

> That all alcoholic drinks are high in kilojoules and can contribute to weight gain.���> That low alcohol or ‘lite’ beers contain less alcohol than regular beers so a

standard drink size is larger.��� > People with diabetes do not need to have diet or low carbohydrate beers.��� These

beers may be lower in carbohydrate but they are not necessarily lower in alcohol.��� > When mixing drinks use low joule/diet mixers such as diet cola, diet ginger ale,

diet tonic water.���> That drinking a lot of alcohol can increase the risk of hypoglycaemia if you are

taking insulin or certain diabetes tablets.���> Some people may need to have less alcohol than generally recommended due to

their age, medication or the need to lose weight.��� Discuss alcohol with your doctor or dietitian and refer to the Alcohol and Diabetes information sheet.���

Weight managementBeing overweight, especially around your waistline, makes it more difficult to manage your diabetes and increases your risk of heart disease.���A small weight loss (5–10% of body weight) can make a big difference to your health when you are overweight but if you need to lose more weight and can, you should certainly do so.��� Measuring your waistline is a great way of checking your progress rather than weighing yourself regularly.��� In general, if you are of Caucasian origin, women should aim for a waist circumference less than 80 cm and men should aim for less than 94 cm.��� Appropriate measurements for other ethnic groups can be checked with your doctor.���

if you drink alcohol, it’s best to have it with a meal and try to include alcohol-free days each week

* NHMRC, Australian Guidelines to Reduce Health Risks from Drinking Alcohol (2009).���

if you drink alcohol, it's bestinclude alcohol-free

to have it with a meal and try todays each week“

Patient_Booklet_v3.indd 14 10/10/2012 3:06:39 PM

Page 157 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 13

7

If you are carrying excess weight around the middle, try to lose some of it by:Reducing your portion sizes and following a diet plan lower in kilojoules and total • fat (particularly saturated fat).��� A dietitian (APD) can help you with specific advice on adjusting your food intake to help with weight loss.���Doing regular physical activity such as walking, dancing, riding a bike or swimming.���• Seeking advice from your doctor, dietitian (APD), diabetes educator (CDE) or State or • Territory Diabetes Organisation.���

An example of a typical meal plan for one dayChoose foods you like and which satisfy you and remember to include carbohydrate foods in each meal or snack to help manage your blood glucose levels.��� Carbohydrate-containing foods are highlighted in italics in the menu below.���

Breakfast – for example…¾ cup of high fibre • breakfast cereal with low fat milk OR2 slices of • bread or toast, preferably wholegrain, wholemeal or high fibre white with thinly spread margarine, peanut butter, jam, Vegemite® or try with baked beans, grilled tomato, or sardines PLUS1 piece of • fruitTea, coffee or water•

Light meal – for example… • 1 sandwich made with 2 slices of• bread, or 1 bread roll or 4 dry biscuits – preferably wholegrain or wholemeal – with thinly spread margarineSalad vegetables• A small serve of lean meat, skinless poultry, seafood, egg, fat reduced cheese or a • more generous serve of legumes (such as beans or lentils)1 piece of • fruitWater, tea or coffee•

Main meal – for example…1 • bread roll or 2 slices of bread (preferably wholegrain or wholemeal) OR 1 cup of cooked pasta or rice OR 2 medium potatoes or 1 cup sweet potato or corn Other vegetables (include freely)• A small serve of lean meat, skinless poultry, seafood, egg, fat reduced cheese or a • more generous serve of legumes (such as beans or lentils)1 piece of • fruit OR small amount of low fat yoghurt or custard Water, tea or coffee•

You can eat your main meal at lunch or dinner, whichever you prefer.���

if you drink alcohol, it’s best to have it with a meal and try to include alcohol-free days each weekif you drink alcohol, it's bestinclude alcohol-free

to have it with a meal and try todays each week”

Patient_Booklet_v3.indd 15 10/10/2012 3:06:39 PM

Page 158 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

14 |

8

food choices Between-meal snacksPeople with diabetes on certain types of tablets or insulin may require one snack between each meal and for supper.��� However, most people aiming to control their body weight may not need a snack between every meal.��� To find out what is best for you, discuss this with your dietitian or diabetes educator.��� Good snack ideas include 1 piece of fruit, 1 tub of low fat yoghurt, 1 cup of low fat milk, 1 slice of wholegrain bread, 1 slice of fruit bread or 2 high fibre crispbreads.��� See the information sheet Healthy Snacks and Diabetes for more snack ideas.���Your State or Territory Diabetes Organisation recommends that everyone with diabetes visit a dietitian (APD) for personal advice.���

For more informationThe example menu plan on page 7 is a guide only.��� For more personalised information, an Accredited Practising Dietitian will help.���

To find an APD in your area, contact:The Dietitians Association of Australia on 1800 812 942 or www.���daa.���asn.���au• Your State or Territory Diabetes Organisation on 1300 136 588 or go to their website • as listed below.��� Dietitians are based in many local hospitals, diabetes centres and community health • centres and are also listed in the telephone directory.���

Remember – good food and regular activity will help to keep you healthy!

Would you like to join Australia’s leading diabetes organisation?> Dietary services > Free magazines > Children’s services> Educational literature > Product discounts > Support groups

For more information phone 1300 136 588 or visit your State/Territory Organisation’s website: ACT www.���diabetes-act.���com.���au NSW www.���australiandiabetescouncil.���comNT www.���healthylivingnt.���org.���au QLD www.���diabetesqueensland.���org.���auSA www.���diabetessa.���com.���au TAS www.���diabetestas.���com.���auVIC www.���diabetesvic.���org.���au WA www.���diabeteswa.���com.���au

The design, content and production of this diabetes information sheet have been undertaken by:

> ACT Diabetes ACT > NSW Australian Diabetes Council > NT Healthy Living NT > QLD Diabetes Australia – Queensland > SA Diabetes SA > TAS Diabetes Tasmania> VIC Diabetes Australia – Vic > WA Diabetes WA

The original medical and educational content of this information sheet has been reviewed by the Health Care and Education Committee of Diabetes Australia Ltd. Photocopying this publication in its original form is permitted for educational purposes only. Reproduction in any other form by third parties is prohibited. For any matters relating to this information sheet, please contact National Publications at [email protected] or phone 02 9527 1951.

Health professionals: For bulk copies of this resource, contact your Diabetes State/Territory Organisations as listed.

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

Patient_Booklet_v3.indd 16 10/10/2012 3:06:40 PM

Page 159 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 15

Blood glucose monitoring – when should I do it and why? And how can I tell if my diabetes is well controlled?

blood glucose monitoring

Self-blood glucose monitoring is a valuable diabetes management tool, which enables people to check their own blood glucose levels as often as they need to or as recommended.

Why is it so important to test my blood?Regular testing and recording of your blood glucose level can reinforce your healthy lifestyle choices as well as inform you of your response to other choices and influences.Importantly, blood glucose level pattern changes can alert you and your health care team to a possible need for a change in how your diabetes is being managed.

Testing your blood glucose levels will help you to:> Develop confidence in looking after your diabetes.

> Better understand the relationship between your blood glucose levels and the exercise you do, the food you eat and other lifestyle influences such as travel, stress and illness.

> Know how your lifestyle choices and medication, if used, are making a difference.

> Find out immediately if your blood glucose levels are too high (hyperglycaemia) or too low (hypoglycaemia), helping you to make important decisions such as eating before exercise, treating a ‘hypo’ or seeking medical advice if sick. (Refer to the individual information sheets on Physical activity and type 2 diabetes; Hypoglycaemia and diabetes; Sick days and type 1 diabetes; Sick days and type 2 diabetes).

> Know when to seek the advice of your diabetes health team about adjusting your insulin, tablets, meal or snack planning when blood glucose goals are not being met.

A diabetes health professional such as a diabetes educator can help you to choose the meter that’s best for you. Your diabetes educator will also give you all the information you need about how, where and when to test your blood glucose levels and work with you in planning a routine that works for you and the life you lead.

Talking diabetes No.04

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

Revised August 2010blood glucose monitoring

Self-blood glucose monitoring is a valuable diabetes management tool, which enables people to check their own blood glucose levels as often as they need to or as recommended.

Why is it so important to test my blood?Regular testing and recording of your blood glucose level can reinforce your healthy lifestyle choices as well as inform you of your response to other choices and influences.Importantly, blood glucose level pattern changes can alert you and your health care team to a possible need for a change in how your diabetes is being managed.

Testing your blood glucose levels will help you to:> Develop confidence in looking after your diabetes.

> Better understand the relationship between your blood glucose levels and the exercise you do, the food you eat and other lifestyle influences such as travel, stress and illness.

> Know how your lifestyle choices and medication, if used, are making a difference.

> Find out immediately if your blood glucose levels are too high (hyperglycaemia) or too low (hypoglycaemia), helping you to make important decisions such as eating before exercise, treating a ‘hypo’ or seeking medical advice if sick. (Refer to the individual information sheets on Physical activity and type 2 diabetes; Hypoglycaemia and diabetes; Sick days and type 1 diabetes; Sick days and type 2 diabetes).

> Know when to seek the advice of your diabetes health team about adjusting your insulin, tablets, meal or snack planning when blood glucose goals are not being met.

A diabetes health professional such as a diabetes educator can help you to choose the meter that’s best for you. Your diabetes educator will also give you all the information you need about how, where and when to test your blood glucose levels and work with you in planning a routine that works for you and the life you lead.

Talking diabetes No.04

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

Revised August 2010

Patient_Booklet_v3.indd 17 10/10/2012 3:06:40 PM

Page 160 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 1616 |

blood glucose monitoring

How do I test my blood?You will need a blood glucose meter, a lancet device with lancets and test strips. The finger is pricked with the lancet to obtain a very small drop of blood which is then applied to a test strip placed in the meter. The results are displayed within seconds.Blood glucose meters are usually sold as kits giving you all the equipment you need to start. There are many different types, offering different features and at different prices to meet individual needs. Most of these are available from your State or Territory Diabetes Organisation, pharmacies and some diabetes centres.

What do I aim for?Successful management of diabetes is all about aiming for a careful balance between the food you eat, how active you are and the medication you take for your diabetes. Because this is a delicate balance, it can be quite difficult to achieve ideal control all the time.For some people, the ranges will vary depending on the individual and their circumstances. While it is important to keep your blood glucose levels as close to a normal or non-diabetic state as possible to prevent complications, it is equally important to check with your diabetes educator or doctor for the range of blood glucose levels that are right and safe for you. Therefore the following information should be treated only as a general guide.

Targets for glycaemic control Target ranges may differ depending on your age, duration of diabetes, the type of medication you are taking and if you have any other medical problems. Speak with your doctor about your individual target ranges. Normal blood glucose levels are between 4.0–7.8mmol/L.

Type 1 diabetes¹

Target levels 4–6mmol/L before meals4–8mmol/L two hours after starting meals

Type 2 diabetes²

Target levels 6–8mmol/L before meals6–10mmol/L two hours after starting meals

People with type 2 diabetes who are not taking a sulphonylurea medication or insulin could aim for a blood glucose level as close to normal as possible.

Risk of hypoglycaemia for both type 1 and type 2 (low blood glucose)

Less than 4mmol/L – if insulin or certain types of tablets are used, but does not apply to all tablets or for people who do not take any tablets for diabetes. Check with your doctor what applies to you.

successful management of diabetes is all about careful balance between food, activity and your diabetes medication

1 Targets are as recommended by the American Diabetes Association. NHMRC guidelines are currently under development.

2 Targets are as recommended by the NHMRC, Blood Glucose Control in Type 2 Diabetes, (2009).2

successful management of diabetesfood, activity and

is all about careful balance betweenyour diabetes medication“

Patient_Booklet_v3.indd 18 10/10/2012 3:06:40 PM

Page 161 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 17

Who is at risk of LOW blood glucose (hypoglycaemia)?People who are using insulin or those taking diabetes tablets which increase their • own insulin production are at risk as both medications have the effect of lowering blood glucose. They can therefore cause hypoglycaemia (low blood glucose) when blood glucose levels are less than 4mmol/L†. (Note: Hypoglycaemia can occur at higher blood glucose levels in children and people who have had elevated blood glucose levels for a long time). People whose diabetes is managed by lifestyle alone or with other types of • diabetes tablets which do not increase their own insulin production, are not at risk of hypoglycaemia.

Are HIGH blood glucose levels dangerous? Sometimes you may get a higher blood glucose reading than usual and you may not be able to figure out the reason. When you are sick with a virus or flu, your blood glucose levels will nearly always go up and you may need to contact your doctor, especially if ketones are present (more likely to develop in a person with type 1 diabetes). However, it is only when blood glucose levels are consistently higher than they ought to be over weeks or months that the damage-causing complications can occur.

What causes glucose levels to go up and down?There are a number of common causes for glucose levels to increase or decrease. These include:

> Food – time eaten, type and amount of carbohydrate (eg: bread, pasta, cereals, starchy vegetables, fruit and milk)

> Exercise or physical activity > Illness and pain

> Diabetes medication > Alcohol

> Emotional stress > Other medications

> Testing techniques

successful management of diabetes is all about careful balance between food, activity and your diabetes medication

3

successful management of diabetesfood, activity and

is all about careful balance betweenyour diabetes medication”

Patient_Booklet_v3.indd 19 10/10/2012 3:06:40 PM

Page 162 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

18 |

When should I test?Your doctor or diabetes educator will help you decide how many tests are needed and the levels to aim for.You will also be advised to record all your tests. Even though your meter may have a memory, it is important to keep a record of your readings in a diary and to take this with you to all appointments with your diabetes health team. Most meters on the market have software which allow you to download your records in different formats such as graphs and charts. Even if you can do this, it is still helpful to keep a diary, not only for your tests but also details of your daily activities, the food you eat and other relevant information. This will provide both you and your diabetes health team with important information in deciding if and how your treatment may need to be adjusted.Ask your doctor or diabetes educator how you can use a diary to help you to better manage your diabetes.

General guidelinesFrequency of testing may vary depending on your treatment. Check with your doctor • or diabetes educator as to when it is suggested you test. Possible times are: > before breakfast (fasting) > two hours after a meal > before bed Testing four times a day is usually recommended for people with type 1 diabetes. • However many people test more often, such as those using an insulin pump (CSII – continuous subcutaneous insulin infusion).

Test more often when you are:> Being more physically active or less physically active

> Sick or stressed

> Experiencing changes in routine or eating habits eg: travelling

> Changing or adjusting your insulin or medication

> Experiencing symptoms of hypoglycaemia

> Experiencing symptoms of hyperglycaemia

> Experiencing night sweats or morning headaches

successful management of diabetes is all about careful balance between food, activity and your diabetes medication

4

Patient_Booklet_v3.indd 20 10/10/2012 3:06:40 PM

Page 163 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 19

What if the test result doesn’t sound right?If you’re not convinced that a result is correct, here’s a suggested check list:

> Have the strips expired?

> Is the strip the right one for the meter?

> Is there enough blood on the strip?

> Has the strip been put into the meter the right way?

> Have the strips been affected by climate, heat or light?

> Did you wash and thoroughly dry your hands before doing the test?

> Is the meter clean?

> Is the meter too hot or too cold?

> Is the calibration code correct?

> Is the battery low or flat?��

All meters will give a different result with a different drop of blood. As long as there is not a big difference (more than 2mmol/L) there is not usually cause for concern.The accuracy of all meters can be checked with meter-specific liquid drops called control solutions. These are expensive, have a short shelf life and only last a few months once opened. However, your diabetes health professional or pharmacy may be able to do this for you at no charge.

What is a glycosylated haemoglobin (HbA1c) test?The HbA1c test shows an average of your blood glucose level over the past 10–12 weeks and should be arranged by your doctor every 3–6 months.

Is the HbA1c the same as testing your own BGLs?No. The HbA1c test doesn’t show the highs and lows that your home testing shows. Therefore it does not replace the tests you do yourself but is used as an added tool in giving the overall picture of your blood glucose management.

How does it work?A glycosylated haemoglobin test is possible because red blood cells (RBC) are continuously being made by your long bones and released into your circulation. When these cells are released, they pick up glucose in the blood stream at that time.Each RBC lasts about 120 days. Therefore any blood sample will have a range of cells released over the previous 120 days with different amounts of glucose attached. The HbA1c test gives a good guide to the average.

5

Patient_Booklet_v3.indd 21 10/10/2012 3:06:40 PM

Page 164 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

20 |

blood glucose monitoring

What HbA1c do I aim for? The goal for most people with diabetes will be in the 6.5–7% (48–53mmol/mol) range however this may need to be higher for children and the old and frail. Your doctor will advise.

How is HbA1c reported?The way that HbA1c is reported is changing. HbA1c has been expressed as a percentage (%). From 2011 it will be reported in IFCC HbA1c units as mmol/mol. The new method is more accurate and consistent between laboratories. For some time both mmol/mol and % will be reported by pathology laboratories.

HbA1c % 6 7 8 9 10

HbA1c mmol/mol 42 53 64 75 86

More informationMany hospitals have a diabetes clinic where you can find out more about blood glucose monitoring. Contact your:> Local hospital for your nearest diabetes clinic or> State or Territory Diabetes Organisation on 1300 136 588

Would you like to join Australia’s leading diabetes organisation?> Dietary services > Free magazines > Children’s services> Educational literature > Product discounts > Support groups

For more information phone 1300 136 588 or visit your State/Territory Organisation’s website: ACT www.diabetes-act.com.au NSW www.australiandiabetescouncil.comNT www.healthylivingnt.org.au QLD www.diabetesqueensland.org.auSA www.diabetessa.com.au TAS www.diabetestas.com.auVIC www.diabetesvic.org.au WA www.diabeteswa.com.au

The design, content and production of this diabetes information sheet have been undertaken by:

> ACT Diabetes ACT > NSW Australian Diabetes Council > NT Healthy Living NT > QLD Diabetes Australia – Queensland > SA Diabetes SA > TAS Diabetes Tasmania > VIC Diabetes Australia – Vic > WA Diabetes WA

The original medical and educational content of this information sheet has been reviewed by the Health Care and Education Committee of Diabetes Australia Ltd. Photocopying this publication in its original form is permitted for educational purposes only. Reproduction in any other form by third parties is prohibited. For any matters relating to this information sheet, please contact National Publications at [email protected] or phone 02 9527 1951.

Health professionals: For bulk copies of this resource, contact your Diabetes State/Territory Organisation as listed.

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

Patient_Booklet_v3.indd 22 10/10/2012 3:06:40 PM

Page 165 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 21

What is a ‘hypo’ and what do I do if I have one?

Hypoglycaemia is a condition that occurs when the blood glucose level has dropped too low, usually below 4mmol/L, although this can vary. It is important to treat hypoglycaemia quickly to stop the blood glucose level from falling even lower. It is also commonly referred to as a ‘hypo’, low blood glucose or insulin reaction.

What are the main causes of hypoglycaemia?Hypoglycaemia can be caused by one or a number of events such as:

> Delaying or missing a meal > Not eating enough carbohydrate

> Unplanned physical activity > More strenuous exercise than usual

> Drinking alcohol* > Too much insulin or diabetes tablets

While these are known causes of hypoglycaemia, in many cases, no specific cause can be identified.

What are the symptoms?While symptoms vary from person to person, common feelings are:

> Weakness, trembling or shaking > Sweating

> Light headedness > Headache

> Lack of concentration/behaviour change > Dizziness

> Tearful/crying > Irritability

> Numbness around the lips and fingers > Hunger

If you feel any of these symptoms, test your blood glucose level if time and circumstances permit. If you are unable to do so, treat as a ‘hypo’ just to be sure.

Hypoglycaemia is most common in people who inject insulin or are taking certain tablets to manage their diabetes. It is not a problem for those who manage their diabetes through a healthy eating plan alone.

Talking diabetes No.23

Revised August 2009 A diabetes information series from State / Territory organisations of Diabetes Australia – Copyright© 2009

hypoglycaemia & diabetes

* The risk of hypoglycaemia increases, the more alcohol you drink. Refer to the Alcohol and Diabetes information sheet.

Patient_Booklet_v3.indd 23 10/10/2012 3:06:41 PM

Page 166 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

22 |

hypoglycaemia & diabetesHow is a ‘hypo’ treated?The first thing to do is to be sure you’re safe. For example, if you’re driving a vehicle, pull over to the side of the road. Then:

STEP 1 – Most important!

Have some easily absorbed carbohydrate, for example:

Glucose tablets equivalent to 15 grams carbohydrate • OR

6–7 jellybeans • OR

1/2 can regular soft drink (not ‘diet ’) • OR

3 teaspoons sugar or honey • OR

1/2 glass fruit juice•

Please Note: For those taking Glucobay® (Acarbose), hypoglycaemia must be treated with glucose.

If circumstances permit, re-test blood glucose levels to ensure they have risen above 4mmol/L. It may take 10–15 minutes to see a rise in blood glucose levels. If symptoms persist or your blood glucose level remains below 4 mmol/L, repeat Step 1.

STEP 2

If your next meal is more than 20 minutes away, you will need to eat some longer acting carbohydrate. This could be one of the following:

A slice of bread OR •

1 glass of milk or soy milk OR •

1 piece of fruit OR •

2–3 pieces of dried apricots, figs or other dried fruit OR •

1 tub natural low fat yoghurt •

For more individualised advice, please speak to your diabetes health professional.

it is important to treat a ‘hypo’ immediately to stop your blood glucose level from falling lower

Revised August 2009

2

it is important to treat a 'hypo'blood glucose level

immediately to stop yourfrom falling lower“

Patient_Booklet_v3.indd 24 10/10/2012 3:06:41 PM

Page 167 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 23

What happens if it’s not treated?If not treated quickly, the blood glucose level can continue to drop which may progress to:

> Loss of coordination > Slurred speech

> Confusion > Loss of consciousness/fitting

You will need the help of others!

What to do if the person is unconscious, drowsy or unable to swallow:

THIS IS AN EMERGENCY!

They must not be given any food or drink by mouth. Here’s what needs to be done:

Place the person on their side making sure their airway is clear.•

Give an injection of Glucagon if available and you are trained to give it. •

Phone for an ambulance (dial 000) stating a ‘diabetic emergency’.•

Wait with the person until the ambulance arrives.•

When they regain consciousness, the person will require carbohydrate to • maintain their blood glucose level.

Glucagon

Glucagon is a hormone which raises the blood glucose level and is injected in a similar way to insulin. Glucagon is recommended to reverse severe hypoglycaemia in people with diabetes. If you are able to treat your own ‘hypo’, you do not need Glucagon which is always given by another person. Your doctor or diabetes educator will recommend you have Glucagon on hand in case of a severe ‘hypo’ and will show you, your family and friends how to use it.

Hypoglycaemia unawareness

Some people feel no symptoms of a ‘hypo’, or only experience symptoms when the blood glucose level drops very low. This problem is more likely to occur in someone who has had diabetes for a number of years or in people who have ‘hypos’ frequently. People who have hypoglycaemia unawareness must check their blood glucose levels more frequently. It is strongly recommended they discuss their condition with a doctor or diabetes educator.

it is important to treat a ‘hypo’ immediately to stop your blood glucose level from falling lower

Revised August 2009

3

it is important to treat a 'hypo'blood glucose level

immediately to stop yourfrom falling lower”

Patient_Booklet_v3.indd 25 10/10/2012 3:06:41 PM

Page 168 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

hypoglycaemia & diabetes What else should I do?

> Wear identification that says you have diabetes.

> Make a note in your monitoring book of any ‘hypos’ you have and discuss it with your doctor or diabetes educator at your next visit.

> Make sure your family, friends, co-workers, school staff and carers know how to recognise and treat hypoglycaemia.

> Look for the cause of your ‘hypo’ so you can try to prevent the situation from occurring again.

> Contact your doctor or diabetes educator if you are having ‘hypos’ often.

> If on insulin or certain types of diabetes medication, always carry quick acting ‘hypo’ treatment with you.

> If taking medication called Acarbose (Glucobay®), carry pure glucose with you such as glucose tablets, glucose gel or Lucozade.

> Refer to the Alcohol and Diabetes information sheet for more advice about drinking alcohol and hypoglycaemia.

> Eat carbohydrates if you are drinking alcohol.

> Before driving a motor vehicle, test your blood glucose level and make sure it is above 4 mmol/L.

Would you like to join Australia’s leading diabetes organisation?> Dietary services > Free magazines > Children’s services> Educational literature > Product discounts > Support groups

For more information phone 1300 136 588 or visit your State/Territory Organisation’s website:

ACT www.diabetes-act.com.au NSW www.diabetesnsw.com.auNT www.healthylivingnt.org.au QLD www.diabetesqld.org.auSA www.diabetessa.com.au TAS www.diabetestas.com.auVIC www.diabetesvic.org.au WA www.diabeteswa.com.au

The design, content and production of this diabetes information sheet has been undertaken by:

> Diabetes Australia – NSW > Diabetes Australia – Victoria > Diabetes Australia – Queensland > Diabetes Australia – Tasmania > Diabetes ACT > Diabetes SA > Diabetes WA > Healthy Living NTThe original medical and educational content of this information sheet has been reviewed by the Health Care and Education Committee of Diabetes Australia Ltd. Photocopying this publication in its original form is permitted for educational purposes only. Reproduction in any other form by third parties is prohibited. For any matters relating to this information sheet, please contact National Publications at [email protected] or phone 02 9527 1951.

Health professionals: For bulk copies of this resource, contact Diabetes Australia in your State/Territory.

Revised August 2009 A diabetes information series from State / Territory organisations of Diabetes Australia – Copyright© 2009

24 |

Patient_Booklet_v3.indd 26 10/10/2012 3:06:41 PM

Page 169 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 25

What should I do if I get sick?

Everyday illness or infections will nearly always cause a rise in blood glucose levels whether you have type 1 or type 2 diabetes. Therefore, at the earliest sign of any form of illness such as a cold or virus, it is important for you to take action.

What to do when unwell When to call your doctor1. Tell someone If you are alone, tell someone you are unwell so they can

check on you.

If you need help and your ‘carer’ is unable to help you, ask them to call your doctor.

2. Test Check your blood glucose levels at least every 2–4 hours

(ideal targets when well are 6–8mmol/L before meals and 6–10mmol/L after meals (2 hours after starting the meal)). Refer to the Blood Glucose Monitoring information sheet for more details.

3. Keep drinking and (if possible) eatingIf you take insulin or diabetes tablets, it is important to avoid hypoglycaemia – see below.It is also important to avoid becoming dehydrated by drinking extra unsweetened fluids every hour such as water, diet soft drinks, diet cordial, weak tea, coffee, vegetable juice or broth.

• IfyouCANeatnormally Continue to eat as normal and sip an extra ½ –1 cup of unsweetened fluids (as listed above) every hour.

Call your doctor. You will probably need to keep taking your diabetes tablets or insulin and will need advice about what to do.

• IfyouCAN’Teatnormally Have some easy to manage carbohydrate drinks, snacks or small meals every 1–2 hours (see suggestions on page 2).

• IfyouCAN’Teatatallandyourbloodglucoselevelis: Call your doctor if you can’t eat at all.• More than 15mmol/L: Drink unsweetened fluids as

listed above.• Less than 15mmol/L: Drink sweetened fluids as listed on

page 2.

Call your doctor if your blood glucose level is consistently above 15mmol/L for more than 12 hours.

Call your doctor if:• Vomiting or diarrhoea continues

for more than 12 hours.• You continue to feel unwell or

become drowsy.

When unwell, test your blood glucose levels often, keep drinking and, if possible, eating – and rest. Be aware that there may be times when you will need to contact your doctor or diabetes educator.

Talking diabetes No.33

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

sick days & type 2 diabetesRevised August 2010

Patient_Booklet_v3.indd 27 10/10/2012 3:06:41 PM

Page 170 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

sick days & type 2 diabetes It is very important to keep up your carbohydrate and fluids, even when you are ill. Here are some ideas to provide 15 grams of carbohydrate per hour if your blood glucose levels are under 15mmol/L, especially if you take insulin or tablets for your diabetes.Drinks providing approximately 15 grams of carbohydrateMilk 1 cup (250ml)Milk + flavouring ¾ cup milk + 1 tablespoon of Milo®, Actavite® or Quik®

Fruit juice* ¾ cupTea or coffee Add 1 tablespoon of sugar or honeyHot lemon juice Add 1 tablespoon of sugar or honeyHerbal tea Add 1 tablespoon of sugar or honeyGastrolyte 4 sachetsOrdinary soft drink* or cordial* (not diet) ¾ cupSports drink (eg: Gatorade) 1 cupSnacks providing approximately 15 grams of carbohydrateCrackers or crispbread 3 Sao®/Ryvita® etcDry toast 1 slicePlain sweet biscuits 3 Milk Arrowroot/Morning Coffee etcMashed potato ½ cupRice ⅓ cupBreakfast cereals ½ cup Special K®, 2 Weetbix®

Porridge (made with water) ⅓ cup Ordinary jelly or custard ½ cupIce cream 3 scoops Ice blocks 1½ sticks

* Care should be taken with these fluids if diarrhoea occurs. They may need to be diluted up to 1:5 for best absorption.

Would you like to join Australia’s leading diabetes organisation?> Dietary services > Free magazines > Children’s services> Educational literature > Product discounts > Support groups

For more information phone 1300 136 588 or visit your State/Territory Organisation’s website: ACT www.diabetes-act.com.au NSW www.australiandiabetescouncil.comNT www.healthylivingnt.org.au QLD www.diabetesqueensland.org.auSA www.diabetessa.com.au TAS www.diabetestas.com.auVIC www.diabetesvic.org.au WA www.diabeteswa.com.au

The design, content and production of this diabetes information sheet have been undertaken by:

> ACT Diabetes ACT > NSW Australian Diabetes Council > NT Healthy Living NT > QLD Diabetes Australia – Queensland > SA Diabetes SA > TAS Diabetes Tasmania> VIC Diabetes Australia – Vic > WA Diabetes WA

The original medical and educational content of this information sheet has been reviewed by the Health Care and Education Committee of Diabetes Australia Ltd. Photocopying this publication in its original form is permitted for educational purposes only. Reproduction in any other form by third parties is prohibited. For any matters relating to this information sheet, please contact National Publications at [email protected] or phone 02 9527 1951.

Health professionals: For bulk copies of this resource, contact your Diabetes State/Territory Organisation as listed.

Revised August 2010 A diabetes information series from Diabetes State/Territory Organisations – Copyright© 2010

26 |

Patient_Booklet_v3.indd 28 10/10/2012 3:06:41 PM

Page 171 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

| 27

Stepping Up StudyFrequently Asked Questions about InsulinIt is normal to have questions and concerns about starting insulin.

Here we address some of the common concerns that people with diabetes have. Remember that your GP, practice nurse and other health professionals are available to discuss these with you further and answer any questions that you might have.

1. I am worried that injecting Insulin will hurt.

You will most likely be surprised to find out that the very short, fine needles in pre-filled disposable insulin pens make insulin injections quite comfortable. They are much less uncomfortable than finger-pricking for blood glucose readings.

Be sure to rotate the sites where you give your injections.

2. Taking insulin will be too inconvenient and difficult.

Pre-filled disposable insulin pens offer simple, adjustable dosing in an easy-to-carry, discrete pen-like delivery system. They do not need to be stored in the fridge, meaning you can go about your daily routine as normal and still take the insulin with you if needed. It can also allow you to adjust your insulin to your requirements giving you far more flexibility than tablets.

3. I will have a hard time remembering to take my insulin

Most people start insulin by taking a long-acting insulin at bedtime. This can be easily remembered for example by associating taking your insulin with something you already do at bedtime such as cleaning your teeth. Keeping your insulin with your toothbrush is a helpful reminder.

However, it can be taken at any time of the day – so choose a time that is best for you.

4. Taking insulin will mean I have failed to manage my diabetes.

All people with type 2 diabetes will require insulin eventually. Type 2 diabetes occurs because your pancreas can no longer produce enough insulin for your needs. Replacing this insulin is the most logical approach to treatment.

It is important to realise that type 2 diabetes is a progressive condition, and starting insulin is a normal and expected part of living well with diabetes. Most people who have type 2 diabetes – even if they follow their diabetes care plan closely – will eventually need to advance to insulin therapy. Starting insulin helps protect your body from complications from poor glucose control and will help you to stay healthy and well and able to enjoy life. Don’t assume the change to insulin means you did something wrong. The most important thing you can do now is to integrate this change in treatment with your daily life and make insulin a part of your new daily routine

5. I’m afraid of the side effects of insulin

Insulin has been used for many years and is the longest known treatment for diabetes. Over these years it has been shown to be a safe way to treat diabetes.

Weight gain can be an issue if your HbA1c is more than 7.5%. The weight can be prevented by paying special attention to your diet and increasing your activity.

Patient_Booklet_v3.indd 29 10/10/2012 3:06:42 PM

Page 172 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

28 |

Stepping Up StudyPeople who have had diabetes for a while and whose blood glucose remains high are at a greater risk for serious health conditions such as heart disease, kidney disease, and amputations. The long-term uncontrolled blood glucose levels lead to these problems, not starting insulin. Keeping your blood glucose levels as close to normal as possible can help to reduce your risk of developing complications of diabetes. When you take insulin according to your diabetes-care plan, it is a safe and effective way of treating your diabetes.

The most common side effect of insulin therapy is hypoglycaemia or low blood glucose. Severe hypoglycaemia is rare and it is more common in people with type 1 diabetes. People with type 2 diabetes generally experience severe hypoglycaemia very rarely, even when taking insulin. This is even less likely when starting using the once daily injection of very long acting insulin. Checking your blood glucose frequently can help you avoid hypoglycaemia. Knowing the warning signs of hypoglycemia and treating it quickly can help you avoid severe hypoglycaemia.

6. I am confused about how and when to take my insulin.

Insulin is now simpler to use than ever. The plan for starting insulin in this project is extremely simple: just a once daily injection! There will be plenty of written material available to help you if needed and your practice nurse and GP are available to help you monitor and adjust your dose very easily or answer any questions. Don’t hesitate to ask questions until you understand what to do.

7. Taking insulin means I have to give up activities I enjoy.

Using insulin allows you more flexibility than tablets. If you are going to do something and you think your glucose levels will go low – you can reduce the amount of insulin that you take.

In most cases, you can continue to take part in any and all of the activities you enjoyed before you started insulin therapy. You just may need to monitor your blood glucose a little more frequently and add some snacks to make sure your blood glucose stays as close to normal as possible. Talk to your GP and Practice Nurse about any specific activities you are involved in that you would like to know more about

Source: Hayes RP, Bowman L, et. al. Understanding diabetes medications form the perspective of patients with type 2 diabetes. Prerequisite to medication concordance. The Diabetes Educator 2006;32(3):404-414..

Patient_Booklet_v3.indd 30 10/10/2012 3:06:42 PM

Page 173 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

 

remember to check you have the right insulin:

> lantus (long acting insulin) is in a GreY pen > Apidra (short acting insulin) is in a BlUe pen

Stepping Up StudyA 'How to' Guide to Using Solostar Insulin Pens

Patient_Booklet_v3.indd 31 10/10/2012 3:06:42 PM

Page 174 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyIf you require more information about the Stepping Up Study please contact the following:

Study Investigator and CoordinatorDr Irene BlackberryDepartment of General PracticeThe University of Melbourne200 Berkeley StCarlton Vic 3053T: 8344 3373F: 9347 6136E: [email protected]

Study funded by the national Health and Medical research Council, roche Diagnostics and Sanofi-Aventis

Study Investigators:

Dr John Furler, Prof Doris Young, Prof James Best, Prof Elizabeth Patterson, Dr Irene Blackberry, A/Prof David O’Neal from the University of Melbourne; Prof Danny Liew from Melbourne Health; Prof Leonie Segal from University of South Australia; Prof Jane Speight from Diabetes Australia Victoria

and Deakin University; and Prof Carl May from University of Southampton, UK

PhD Students:

Dr Jo-Anne Manski-Nankervis and Ms Elizabeth Holmes-Truscott

Study Diabetes Educator (RN):

Ms Louise Ginnivan

Patient_Booklet_v3.indd 32 10/10/2012 3:06:42 PM

Page 175 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyInsulin initiation and

intensification in the general practice setting

for adults with type 2 diabetes

A training manual for General Practitioners and

Practice Nurses

Department of General PracticeThe University of Melbourne

July 2012Ste

ppin

g U

p S

tud

yPage 176 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyStepping Up Study

Investigators and Contact Details

Study Investigators:

Dr John Furler, Prof Doris Young, Prof James Best, Prof Elizabeth Patterson, Dr Irene Blackberry, A/Prof David O’Neal from the University of Melbourne; Prof Danny Liew from Melbourne Health; Prof Leonie Segal from

University of South Australia; Prof Jane Speight from Diabetes Australia Victoria and Deakin University; and Prof Carl May from University of Southampton, UK

PhD Students:

Dr Jo-Anne Manski-Nankervis and Ms Elizabeth Holmes-Truscott

Study Diabetes Educator (RN):

Ms Louise Ginnivan

Credentialed Diabetes Educator - RNMs Louise GinnivanDepartment of General PracticeThe University of Melbourne200 Berkeley StCarlton Vic 3053Phone: 8344 3373Mobile: 0499 599 084E-mail: [email protected]

Study Investigator and GPDr John FurlerDepartment of General PracticeThe University of Melbourne200 Berkeley StCarlton Vic 3053Phone: 8344 4747Mobile: 0419 393 156E-mail: [email protected]

Study Investigator and EndocrinologistA/Prof David O’NealDepartment of MedicineThe University of MelbourneSt Vincent’s HospitalPhone: 9288 2574Mobile: 0425 731 665E-mail: [email protected]

Study Investigator and CoordinatorDr Irene BlackberryDepartment of General PracticeThe University of Melbourne200 Berkeley StCarlton Vic 3053Phone: 8344 3373Mobile: 0424 924 014E-mail: [email protected]

Page 177 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyIntroduction ............................................................................................................................................. 1

Stepping Up ............................................................................................................................................. 4

Identifying patients in your practice who may benefit from the initiation of insulin ....................... 5

Use of motivational interviewing to assist patients in making the transition to insulin ................. 5

Introduction to insulin and its administration ...................................................................................... 6

What is insulin? ................................................................................................................................ 7

How does insulin work? ................................................................................................................... 7

Why is insulin therapy required in type 2 diabetes? ......................................................................... 7

Thedifferenttypesandprofilesofinsulinsavailable ....................................................................... 9

Glargine (Lantus) ..................................................................................................................... 10

Glulisine (Apidra) ..................................................................................................................... 11

Delivery of insulin ............................................................................................................................. 12

Choosing an injection site ....................................................................................................... 12

Teaching injection technique .................................................................................................. 13

Insulin Dosing ................................................................................................................................... 15

Glargine (Lantus) ..................................................................................................................... 15

Glulisine (Apidra) ..................................................................................................................... 18

Side effects of insulin ....................................................................................................................... 19

Hypoglycaemia ........................................................................................................................ 19

Storage of insulin and sharps disposal ............................................................................................. 20

Stepping Up protocol .............................................................................................................................. 20

Blood glucose monitoring ................................................................................................................ 20

Starting glargine (Lantus) ................................................................................................................. 21

Starting glulisine (Apidra) .................................................................................................................. 22

References ................................................................................................................................................ 23

Appendix .................................................................................................................................................. 24

Motivational Interviewing Resources ............................................................................................... 24

Stepping Up StudyContents

Page 178 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyFigure 1: Diabetes Fast Facts .................................................................................................................... 1

Figure 2: Treatment algorithm (PBS approved medications) for type 2 diabetes ...................................... 3

Figure3:Intensificationoftherapy:dospecialistsdifferfromGPs? ......................................................... 4

Figure 4: Essential Components of Motivational Interviewing .................................................................. 6

Figure 5: Change talk ................................................................................................................................. 6

Figure 6: The body's physiologic insulin pattern........................................................................................ 8

Figure 7: Central obesity............................................................................................................................ 8

Figure 8: Decreasing beta cell function as part of the progression of T2DM ............................................ 8

Figure9:Timeprofileofinsulins ............................................................................................................... 9

Figure 10: Glargine (Lantus)....................................................................................................................... 11

Figure 11: Glulisine (Apidra) ....................................................................................................................... 11

Figure 12: Injection sites for insulin administration ................................................................................... 12

Figure 13: Lipohypertrophy........................................................................................................................ 12

Figure 14: Quick reference guide to SoloStar pens (Lantus and Apidra) ................................................... 14

Figure 15: Pinch up for slim people or those using injection sites without much subcutaneous fat ........ 15

Figure 16: Accu-chek®360o Insulin adjustment tool for use with basal insulin glargine (Lantus) .............. 16

Figure 17: Accu-chek®360o 3dayprofilingtool ......................................................................................... 17

Figure 18: Insulin adjustment tool for use with prandial insulin glulisine (Apidra) ..................................... 18

TablesTable 1: Targets suggested by Guidelines................................................................................................. 2

Table 2: Patient and Health Professional factors which act as barriers to insulin initiation ....................... 4

Table 3: Current insulins available ............................................................................................................. 10

Stepping Up StudyFigures

Page 179 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyIntroductionType 2 diabetes mellitus (T2DM) is a chronic medical condition arising as a result of a combination of genetic and environmental factors. It occurs most commonly in people over 40 years of age and is characterized by reduced or less effective insulin, resulting in an increased concentration of glucose in the blood [1]. The AusDiab study (1999-2000) estimated that 7.1% of Australian adults over the age of 25 years had T2DM, a portion of which were undiagnosed [1] T2DM is common, is associated with a numberofcomplicationsandhasahighfinancialcosttotheindividualandthecommunity(seeFigure1).

Figure 1: Diabetes Fast Facts

 

•Around 275 Australians develop diabetes every day•Many cases of T2DM remain undiagnosed and it is estimated that 1.7 million Australians may have T2DM

•Up to 60% of cases of T2DM can be prevented.•Australia’s indigenous population suffers the fourth highest rate of Type 2 diabetes in the world 

Prevalence

•People with diabetes are almost three times more likely to have high blood pressure, obesity or elevated blood fats e.g. cholesterol 

•They are two to three times more likely to have cardiovascular disease, e.g. heart disease and stroke •65‐80 percent of people with diabetes will die of coronary heart disease •15 percent of people with diabetes have heart disease compared to 2.5 percent without diabetes •Renal disease accounts for 8‐14 percent of deaths in people with diabetes •5 percent of people with diabetes will experience foot ulcers •Of the 3000 amputations every year in people with diabetes, most are preventable 

Complications

•Type 2 diabetes costs Australia $3 billion per year •Average cost for a person with Type 2 diabetes who has no complications is $10,900 •If there are complications this cost almost doubles to $20,525 •The 4 percent of people who have diagnosed diabetes account for 12 percent of health costs in Australia 

Financial costs

Currently the majority of care for people with T2DM occurs in the general practice setting. The BEACH program, a continuous national study of general practice activity, estimated that an average of 4.4 million type 2 diabetes encounters were managed in Australian general practice in the year 2010-11. Non-gestational diabetes was the third most frequently managed chronic medical condition in general practice behind hypertension and depressive disorder [3]. However, when insulin initiation is required it is generally not occurring in a timely manner [4, 5] and the majority of patients are referred to specialist care (endocrinologists and credentialed diabetes nurse educators (DNE))[6]. Indeed, people with diabetes represent the third most common group referred to specialists after that for malignant skin neoplasms and pregnancy [3].

| 1

Page 180 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyThere are a number of published guidelines which state the target for glycaemic control (as measured by HbA1c and capillary blood glucose levels) and the lifestyle and pharmacological means by which to achieve this. Depending on the presence of symptoms and degree of glycaemia, people with T2DM may utilise diet and exercise, oral hypoglycaemic agents (OHA), non-insulin injectibles or insulin to improve glycaemic control (see Figure 2). The target levels for HbA1c, fasting blood glucose and post prandial glucose are illustrated in Table 1. Meeting glycaemic targets early in the course of T2DM is important because of metabolic memory –the follow up to the United Kingdom Prospective Diabetes Study (UKPDS) demonstrated that there was an association between well controlled diabetes early in the course of the disease with a continued reduction in microvascular risk, myocardial infarction and death from any cause ten years later [7].

Stepping Up Study

Table 1: Targets suggested by Guidelines

  Normal  Targets for diabetes patients 

    ADS/EASD   AACE  IDF  Stepping Up study  

HbA1c  % mmol/mol 

 <6.0  <42 

 <7.0 * 

<53* <6.5  <48 

<6.5  <48 

<7.0  <53 

Fasting blood glucose (mmol/L) 

<5.5   3.9‐7.2   <6.0  <6.0  <7.0 

Post prandial blood glucose (mmol/L) 

<7.8   <10   <7.8  <8.0  <10.0 

 *This is the general ADS target for people on insulin therapy

Nevertheless, a gap exists between recommended care and real world practice. Many patients have an HbA1c over the recommended target. Less than 50% of people with T2DM have a HbA1c less than 7.5% (58mmol/mol) yet only 10-15% of people with T2DM are using insulin. In an Australian study the median duration of type 2 diabetes before commencing insulin was 8.1 years from diagnosis and the median HbA1c was 9.4% (79mmol/mol) [8].

ClinicalinertiaresultinginadelayinintensificationofpharmacologicaltherapyisanissueforbothGPsand specialists, although studies suggest that GPs may delay insulin initiation longer than specialists (seeFigure3)Manypatientandhealthprofessionalbarrierstoinsulininitiationhavebeenidentified(seeTable 2) [9-12].

2 |

Page 181 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 3

Figure 2: Treatment algorithm (PBS approved medications) for type 2 diabetes

From RACGP [2]

Page 182 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyFigure 3: Intensification of therapy: do specialists differ from GPs?

From Shah et al [13]

Table 2: Patient and Health Professional factors which act as barriers to insulin initiation

Patient factors Health Professional factors

• Belief that diabetes is not a serious illness • Fear of addiction • Belief that insulin makes patient fat • Fear of hypoglycaemia • Belief that insulin would not help • Pain associated with insulin injection • Other fears regarding injection of insulin • Pain associated with blood tests • Lack of faith in doctor • Anxiety • Concern that they can never stop insulin • Life will be restricted as a result of starting

insulin • Belief that insulin causes problems like

blindness • Non-compliance with medical appointments • Non-compliance with medications

• Belief that patient wouldn’t comply with treatment

• Fear of hypoglycaemia in a specific patient • Belief that patient couldn’t cope with pain

involved in insulin injection • Patient too old or inadequate level of

education • No experience with treatment • Not wanting to give to obese patients because

insulin would result in further weight gain • Belief that a specific patient’s diabetes is so

severe that even insulin wouldn’t help • Lack of resources in office based practice –

drug cost, staff availability, skills of staff, time • Belief that insulin initiation is complex • Lack of motivation to improve clinical practice • Belief that insulin would impair patient quality

of life • Patient co-morbidities

Stepping UpThe current challenge ahead is to develop a model of care for insulin initiation in Australian general practice which will overcome the current barriers, be acceptable to patients and health professionals and be able to be implemented into ongoing practice. Stepping Up is a new model of care which has four main components:

1. General practitioners and practice nurses working in partnership to identify patients requiring insulin

2. Simple, robust protocols and tools supporting insulin initiation in the Primary Care setting

4 |

Page 183 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 5

3. GPs and PNs working in partnership to initiate and titrate insulin in primary care

4. Endocrinologists and diabetes educators acting as support resources in a hub and spoke manner

In this training manual education will be provided which will allow you to implement this model of care in your practice. The topics covered will be:

1. Identifyingpatientsinyourpracticewhomaybenefitfromtheinitiationofinsulin

2. Use of motivational interviewing to assist patients in making the transition to insulin

3. Introduction to insulin and its administration

4. Stepping Up protocol

Identifying patients in your practice who may benefit from the initiation of insulinPotentiallyanyonewithT2Dcanbenefitfromtheadditionofinsulintotheirmanagement.Strong indications for insulin therapy include:

> HbA1cconsistently≥7.5%formorethan3months

> Maximum oral therapy yet control not ideal

> Oral hypoglycaemic treatments not tolerated or contraindicated.

You can proactively identify these patients by generating a list through your pathology provider, using the PEN Clinical Audit Tool or your medical software.

Insulin therapy should be discussed as a treatment option at the time of diagnosis. It should not be used as a threatened punishment for poor compliance. Treatment should start:

> As soon as there is evidence of deteriorating glycaemic control

> After exploring whether the person could change their lifestyle or current medication

> After full discussion of all the pros and cons of insulin therapy

Use of motivational interviewing to assist patients in making the transition to insulinAs discussed previously there are many potential barriers that may put people off starting insulintherapy.However,therearealsomanypotentialbenefitssuchassymptomcontrolandreduced risk of developing complications. Motivational interviewing is one tool that can be used to engage patients in making the decision to commence insulin.

Motivational interviewing is a patient-centred yet directive counselling style for helping people explore and resolve ambivalence about behaviour change. It is an approach designed to help patients build on their own motivation and reach a decision to change. The basic counselling skills that can be used by health professionals are: attending behaviours, focus, empathy, affirmation,reflection,openendedquestionsandsummarising.

Page 184 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyFigure 4: Essential Components of Motivational Interviewing

 

•Ambivalence is normal and powerful.•Try to understand the patient's perspective

Express empathy

•Motivation for change happens when people perceive a discrepancy between where they are and where they want to be

•Use change talk (see below)

Develop discrepancy

•Reframe patient concerns positively•Avoid confrontation•Emphasise personal choice and control

Roll with resistance

•Belief in the ability to change is an important motivator•Use confidence rulers and get patient to discuss how goals might be achieved•Value the patient as a resource for finding solutions to problems

Support self efficacy

Figure 5: Change talk

 

•What things may happen if your diabetes remains as poorly controlled as this?

Problem recognition

•What might be some of the advantages in going on to insulin?•Who are the people in your life that would support you making the change to insulin?•If we were to bump into each other in six months time, what do you think you would like to tell me about your diabetes and how you are managing it? How would you like things to turn out?

Intention to change and optimism for change

IntheappendixtothismanualtherearetwopapersbySimetalwhichyoumightfindhelpfulasaresource for motivational interviewing and assisting patients make behavioural changes.

Introduction to insulin and its administrationAt the end of this section you will understand:

> What insulin is

> How insulin works and why it is needed in type 2 diabetes

> Thedifferenttypesandprofilesofinsulinsavailable

> How insulin is delivered

6 |

Page 185 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 7

What is insulin?“We have obtained from the pancreas of animals a mysterious something which injected into totally diabetic dogs completely removes all the cardinal symptoms of the disease…If the substance works on the human, it will be a great boon to medicine”

J.B. Collip, January 8, 1922[14]

2012marks90yearssinceinsulinwasfirstextractedandusedtotreathumanswithdiabetes.Insulinis a protein hormone produced from the cleavage of a precursor (proinsulin) produced by the beta cells of the pancreas. In normal human adults, the pancreas secretes approximately 40 to 50 units of insulin per day. Basal insulin refers to the quantity of insulin secreted in the fasting state, whilst stimulated insulinsecretionoccursinresponsetoingestedmeals[15].Thefirstinsulinsproducedforusebypeoplewith diabetes were derived from beef and pork. Now human insulin is made using recombinant DNA technologyandinsulinswithdifferenttimeprofiles,fromultra-shorttolongacting,havebeenproduced.

How does insulin work?Insulin receptors occur on many cells in the body but the main metabolic effects occur in fat, liver and muscle cells. The major function of insulin is to promote the storage of ingested nutrients. In the liver insulin promotes glycogen synthesis and storage and inhibits glycogen breakdown. It promotes protein synthesis in muscle and triglyceride storage in fat cells. The net result of these actions is to remove glucose from the blood and transport it into tissues [15].

The liver releases glucose at a relatively constant rate all the time, with a slight dip during the night and a surge before dawn. A steady release of insulin is therefore needed to maintain normal blood glucose levels. After meals (post prandial) there is a burst of insulin, often called the meal-time bolus (see Figure 6). Whenever glucose is released into the bloodstream from food, a matching release of insulin is required for up to two hours in order to move the glucose into the cells. How long this increased insulin level is needed depends on the type of carbohydrate, its glycaemic index, and the fat content of the meal.

Why is insulin therapy required in type 2 diabetes?Type 2 diabetes is characterised by reduced or less effective insulin.

1. Less effective insulin (insulin resistance)

Visceral fat (accumulated in central obesity) is more metabolically active than peripheral fat and releaseslargequantitiesofnon-esterifiedfattyacids(NEFA)(seeFigure7).NEFAhaveseveralmetabolic actions that can cause insulin resistance. Insulin resistance also develops with a cluster of clinical and biochemical features known as metabolic syndrome or the insulin resistance syndrome. This consists of:

> Glucose intolerance

> Truncal obesity

> Hypertension

> Low HDL (high density lipoproteins or good cholesterol)

> High LDL ( low density lipoproteins or bad cholesterol)

> High triglycerides

Insulin resistance occurs when insulin receptors on the cells of muscle and fat tissue do not respond to the effects of insulin. Simply put, insulin cannot carry the glucose from the bloodstream into the cells resulting in a state of excess insulin and glucose in the bloodstream.

Page 186 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study2. Reduced insulin

In response to increased blood glucose levels the pancreas continues to secrete insulin to try and provide the cells with energy. Over a period of time beta cells in the pancreas become exhausted and insulin production is reduced. By the time type 2 diabetes (T2D) is diagnosed, the pancreas has lost 50% of beta cell function (see Figure 8). As T2DM progresses, oral hypoglycaemicagentsmaynotbesufficienttocontrolbloodglucoselevels.Insulintherapyisanimportant adjunct to therapy to compensate for the body’s decreased production of this hormone.

Figure 6: The body's physiologic insulin pattern

 

The body’s physiologic insulin pattern

The body’s normal insulin secretory response is biphasicWhite JR, 2003, Porte D & Kahn S, 1995

Figure 7: Central obesity

Figure 8: Decreasing beta cell function as part of the progression of T2DM

5HOMA=homeostasis model assessmentAdapted from Holman RR. Diabetes Res Clin Pract 1998;40(suppl 1):S21–5.

Decreasing -cell function as part of the progression of T2DM

Normal -cell function by HOMA (%)

Time (years)

0

20

40

60

80

100

― 10 ―8 ―6 ―4 ―2 0 2 4 6

Time of diagnosis?

Pancreatic function~50% of normal

 

8 |

Page 187 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 9

The different types and profiles of insulins availableManufactured insulin aims to mimic natural patterns. Current insulins used for the treatment of type 2 diabetes are listed in Table 3. The number of injections, type and dose of insulin varies between individuals and should be tailored to their lifestyle and eating habits and will be decided on the basis of their blood glucose pattern throughout the day and before and after meals.

Figure 9: Time profile of insulins

 

Ultra short acting insulin

Basal insulin

Pre‐mixed insulin

Page 188 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyTable 3: Current insulins available

 

From RACGP [2]

The two insulins which will be used in the Stepping Up study are glargine (Lantus) and glulisine (Apidra)

Glargine (Lantus)Glargine (Lantus) is a long acting basal analogue insulin. This insulin is peakless, and as a result is associated with less hypoglycaemic episodes. It has less impact on weight gain, has been associated withsignificantimprovementsinhealthrelatedQualityofLifemeasuresandisadministeredonceper day. This is usually given in the evening, but the time of day the injection is given can be altered depending on patient preference provided that it is given at the same time each day.

10 |

Page 189 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 11

After injection Lantus insulin clumps under the skin, and releases from there very slowly and evenly over 24 hours, therefore an increase of 4 units is spread over the 24 hours (i.e. 0.16 units increase of this insulin per hour, a very small increment!)

Glulisine (Apidra)Glulisine (Apidra) is a rapid acting insulin analogue which is given at mealtimes. It is generally commenced when target fasting blood glucose levels have been attained using glargine (Lantus) but postprandial blood glucose levels remain high (>10mmol/L).

Because of its rapid onset of action (within 10 minutes of injection) the patient needs to be advised to inject this insulin just as they are sitting to eat the meal or just after starting the meal (either eat and inject or inject and eat). Rapid acting insulins are more likely to cause hypoglycaemia and so the patient needs to be educated about the symptoms and management of this potential side effect.

If the pre-meal injection is forgotten, it is NOT to be given an hour or two later when a high glucose is detected, but rather ask them to record in their BG diary that it was missed. The following day they can return to the planned schedule and inject prior to the designated meal.

In contrast to glargine (Lantus) which acts for 24 hours, glulisine (Apidra) acts almost immediately and is inactive within 2 to 4 hours. Therefore the carbohydrate content of the actual meal is the only target of this insulin, which results in a lowered risk of hypoglycaemia compared to the extended tails of other fastactinginsulins,suchasActrapid.EveryincreaseddoseofApidrainsulincanresultinasignificanteffect on the post meal BGL.

NomatterhowhightheBGLis,thefirst(starting)doseof4unitsApidrashouldbeadheredto,untilitis established how sensitive your patient is to this insulin. In some patients one unit of Apidra may drop the blood glucose level 4 to 6mmol/l, in others 4 units may only drop them 1mmol/l. Until the effect on each individual patient is established it is important to maintain frequent contact in relation to dose adjustment of Apidra.

Figure 11: Glulisine (Apidra)

Figure 10: Glargine (Lantus)

Page 190 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyDelivery of insulinIn this section you will learn how to choose an injection site and how to teach the patient how to inject.

Choosing an injection siteIn the Stepping Up study we recommend that the abdomen is used as the site for all injections (see Figure 12)

Figure 12: Injection sites for insulin administration

Abdominal rotation pattern by quadrants (Diagram courtesy of Lourdes Saez-de Ibarra and Ruth Gaspar, Diabetes Nurses and Specialist Educators from La Paz Hospital, Madrid, Spain).

When injecting rapid and long-acting analogue insulin, each injection should be administered within a different site, even if injected at different times during the day. For the patient’s ease of memory we suggest “Lantus on the left; rapid (Apidra) on the right”. In the diagram above Lantus would be rotated within sites 2 and 3 and Apidra within sites 1 and 4. [16]

It is recommended that insulin injection sites are rotated as repeatedly injecting into the same small area results in lumps (lipohypertrophy) which hinder insulin absorption and can be unsightly. Alternate between the left and right side on a weekly basis, and rotate sites within the same area. Each injection shouldbeatleastafingersbreadthawayfromthelastone.Checkforlumpsonaregularbasis.Iflipohypertrophy (see Figure 13) is found, that area should not be used for injection until it has become soft again. This may take weeks or even months, depending on the severity of the lipohypertrophy.

 

Figure 13: Lipohypertrophy

 

12 |

Page 191 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 13

Some points to remember

> Temperature – heat also speeds up the absorption of insulin. People should therefore avoid injecting immediately before or after a hot bath or shower.

> Injecting through clothing – people sometimes feel they need to do this, for example while traveling or in social situations, but it should be discouraged especially considering the very short 4 mm needles being used.

Teaching injection techniqueTeaching insulin injection technique is a key role for the practice nurse.

Glargine (Lantus) and glulisine (Apidra) both come in a completely disposable, ready made up pen which only requires the needle to be inserted at the tip of the pen. This is the easiest way for patients to commence insulin. Both pens are used according to the instructions below (see Figure 14). The patient must always check they are injecting the correct insulin at the correct time (eg Lantus = grey pen = inject in the evening; Apidra = blue pen = inject with meals).

The key points for this task are:

1. Make sure the person attaches the needle, dials the dose and gives the injection themselves. You may need to guide them – but don’t do it for them. Advise the patient to use a new needle for every injection.

2. Do an ‘air shot’ before each injection. An air shot will make sure the plunger is connecting, and expel air from the pen.

3. Ensure hands and the skin that will receive the injection are clean. Alcohol wipes are not required. They are an astringent and can make the injection more painful.

4. To ‘pinch up’ or not to ‘pinch up’? Insulin should be injected into soft fat, not muscle. To avoid intramuscular injection, slim people, or those using injection sites without much subcutaneous fat, may need to ‘pinch up’ and /or use a shorter needle length. Pinch up is not usually required for people with type 2 diabetes (See Figure 15)

5. Insert the needle at a 90o angle and push the needle in to the hilt

6. Inject the insulin

7. After the injection, leave the needle in the skin for 5 -10 seconds to avoid leakage. With large doses, it many need to be left in for longer.

8. Occasionally, there may be bleeding after the needle is withdrawn. Reassure the person, and advise them to apply gentle pressure for a couple of minutes to minimize bruising. They should not rub the area, as this may increase the rate of absorption.

Advise the patient that needles and syringes are free for people registered with the National Diabetes Services Scheme (NDSS). Ensure patient is a member and that the NDSS is advised when they commence insulin (form is available in patient encounter pack in the practice or via the NDSS website).

Page 192 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyFigure 14: Quick reference guide to SoloStar pens (Lantus and Apidra)

 

14 |

Page 193 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 15

Figure 15: Pinch up for slim people or those using injection sites without much subcutaneous fat

Insulin Dosing

Glargine (Lantus)Allpatientsarecommencedon10unitsofglargine(Lantus).Thefirstinjectionwilloccurintheclinicwith the practice nurse. The patient can then continue insulin at a time convenient to them, provided it is at approximately the same time each day. Evening or pre-bed administration is recommended. The insulin dose is titrated every three days using the Accu-chek®360o Insulin Adjustment Tool for basal insulin glargine (see Figure 16) until the fasting blood glucose target of 4.0-7.0mmol/L is achieved.

To ensure the reliable absorption of insulin, injections must be made into the subcutaneous tissue.

 

Pinchupamoundofcleanskinbetweenthumbandindexfingerbeforegentlypushing the needle into the mound at an angle of 90o. Release the grip on the skin fold once the needle has been removed as releasing too soon can provoke an intramuscular injection.

 

Correct pinch‐up 

 

Incorrect pinch‐up   

 

There are some instances where patients can’t pinch-up, eg arthritis. If thin patients choose not to pinch-up they should inject at 45o to decrease the chances of giving an intramuscular injection.

Page 194 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyFigure 16: Accu-chek®360o Insulin adjustment tool for use with basal insulin glargine (Lantus)

Insulin Adjustment ToolFor use with insulin glargine (Lantus®)

Patient Name

Bring this form and your Accu-Chek blood glucose monitoring system to your next appointment.

# Based on O’Neal DN et al. INITIATION STUDY: a pilot study examining a model of care for initiating patients with Type 2 Diabetes on a basal +/- prandial insulin regimen in primary care with adjunct Continuous Glucose Monitoring. http://www.diabetesccre.unimelb.edu.au/professionals/documents/Sanofi_Meeting_INITIATION_ONeal_et_al_Oct-2011.pdf [accessed May 30th, 2012]** Modified according to the STEPPING UP study protocol from Davies M et al. Improvement of Glycemic Control in Subjects with Poorly Controlled Type 2 Diabetes. Diabetes Care 28:1282–1288, 2005

Oral Diabetes Medications Dose Time/Day

Healthcare Professional Name

Insulin Starting Dose (units)

Day 1 Day 1 Day 1Day 2 Day 2 Day 2Day 3 Day 3 Day 3

Period Period Period

Handling Hypoglycaemia#

If you feel Hypoglycaemic or have blood glucose below 4 mmol/L at any time since your last insulin dose:

1 Treat immediately by following instructions given to you by your healthcare professional

2 Cross out any remaining blood glucose or calculation fields in current Period

3 Reduce your daily insulin dose by -2 to -4 units and start a new Period

3 3 3

Adjustment Adjustment AdjustmentStart Date

units

mmol/L mmol/L mmol/Lmmol/L mmol/L mmol/Lmmol/L mmol/L mmol/L

units units

3-Day Total

Daily Dose (units)

Average Blood Glucose

Fasting Blood Glucose (mmol/L)finger prick Record

as newDaily Dose

on yournext sheet

NotesUse this space to record details of hypoglycaemia

Daily Dose Time

Warning: Do not use this form without first consulting your healthcare professional

CAUTION: INVESTIGATIONAL DEVICE. FOR INVESTIGATIONAL USE IN THE STEPPING UP STUDY ONLY.

Find Your Dose Adjustment Before Dosing on Day 3**If you were NOT hypoglycaemic at this dose, use this table to find your Adjustment.

Target range

Average Blood Glucose (mmol/L)

Adjustment (units)

4.0-7.0

+0*

7.1-8.0 8.1-10.0

+2 +4

Over 10

+6

* Dose Adjustment is complete when your Average Blood Glucose falls within the target range. Continue monitoring your fasting blood glucose daily as directed by your healthcare professional

Find Your Dose Adjustment Before Dosing on Day 3**If you were NOT hypoglycaemic at this dose, use this table to find your Adjustment.

Target range

Average Blood Glucose (mmol/L)

Adjustment (units)

4.0-7.0

+0*

7.1-8.0 8.1-10.0

+2 +4

Over 10

+6

* Dose Adjustment is complete when your Average Blood Glucose falls within the target range. Continue monitoring your fasting blood glucose daily as directed by your healthcare professional

Day 1 Day 1 Day 1Day 2 Day 2 Day 2Day 3 Day 3 Day 3

Period 104/03/2011

10 14 142124 -2

8.3 8.67.88.9 7.73.18.7

25.9

8.6

6.3

22.6

7.5

207/03/2011

Fasting Blood Glucose hypoglycaemiaon 08/03/2011Adjusted Daily Dose by -2

309/03/2011

3 3 3

Adjustment Adjustment Adjustment

Period Period

Start Date

units

mmol/L mmol/L mmol/Lmmol/L mmol/L mmol/Lmmol/L mmol/L mmol/L

units units

3-Day Total

Daily Dose (units)

Average Blood Glucose

Fasting Blood Glucose (mmol/L)finger prick Record

as newDaily Dose

on yournext sheet

NotesUse this space to record details of hypoglycaemia

CAUTION: INVESTIGATIONAL DEVICE. FOR INVESTIGATIONAL USE IN THE STEPPING UP STUDY ONLY.

Insulin Adjustment ToolFor use with insulin glargine (Lantus®)

Steps Examples

Why use this formYou have been diagnosed with Type 2 diabetes and your healthcare provider has recently prescribed insulin as part of your treatment. This form will help you adjust your Daily Dose of insulin to achieve an average fasting blood glucose that falls within the target range specifed by your healthcare provider.

Definition: A fasting blood glucose is one where you do not eat food or drink liquids – except water - for at least 8 hours before testing. The most convenient time to measure your fasting blood glucose is when you wake up each day—and before your first meal of the day.

Starting Doseis 10 units

Period 1• User is not hypoglycaemic• Average Blood Glucose calculated on Day 3 = 8.6 mmol/L• Adjustment is +4• New Daily Dose is 14 units

Period 2• User is hypoglycaemic on Day 2 with a Fasting Blood Glucose of 3.1 mmol/L• Adjustment is -2 and a new Period is started with Daily Dose 12 units

Period 3• User is not hypoglycaemic• Average Blood Glucose calculated on Day 3 = 7.5 mmol/L• Adjustment is +2• New Daily Dose is 14 units (carried forward as new Daily Dose on next sheet)

Fasting

• Test and record your fasting blood glucose using your Accu-Chek blood glucose meter.• If hypoglycaemic, see Handling Hypoglycaemia.

Before Dosing Day 3

• Add blood glucose values for days 1, 2 and 3 to find the 3 Day Total.• Divide the Total by 3 to find the Average Blood Glucose.• Enter your Adjustment using the Dose Adjustment table. Note: Dose Adjustment is complete when your Average Blood Glucose falls within the Target Range. Continue with the current dose of insulin according to your healthcare professional’s instructions.• Add your current Daily Dose and the Adjustment to find your new Daily Dose.

Handling Hypoglycaemia

If you feel hypoglycaemic at any time, refer tothe Handling Hypoglycaemia instructions.

Before using this form:• Read all instructions. • Complete the User Information section with your healthcare professional. • Confirm that your healthcare professional has supplied you with an Insulin Starting Dose and Daily Dose Time. • Record the Start Date for each week.Completing your Dose Adjustment:• Dose Adjustment is complete when your fasting blood glucose falls within the Target Range.• Continue monitoring your fasting blood glucose daily as directed by your healthcare professional.

Instructions and Examples

Lantus® is a trademark of Sanofi. Accu-Chek and Accu-Chek 360° are trademarks of Roche. ©2012 Roche. Reproduction Prohibited. Ver 1.0.0.0

Handling Hypoglycaemia#

If you feel Hypoglycaemic or have blood glucose below 4 mmol/L at any time since your last insulin dose:

1 Treat immediately by following instructions given to you by your healthcare professional

2 Cross out any remaining blood glucose or calculation fields in current Period

3 Reduce your daily insulin dose by -2 to -4 units and start a new Period

16 |

Page 195 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 17

Figure 17: Accu-chek®360o 3 day profiling tool

WARNING: Do not adjust your prescribed oral medication or insulin therapy without first consulting your healthcare professional.

Bring this form and your Accu-Chek blood glucose monitoring system to your next healthcare professional appointment.

BLOOD GLUCOSE

HEALTHCARE PROFESSIONAL NAME

HEALTHCARE PROFESSIONAL PHONE

WARNING: Do not adjust your prescribed oral medication or insulin therapy without first consulting your healthcare professional.

Bring this form and your Accu-Chek blood glucose monitoring system to your next healthcare professional appointment.

BLOOD GLUCOSE

HEALTHCARE PROFESSIONAL NAME

HEALTHCARE PROFESSIONAL PHONE

CAUTION: INVESTIGATIONAL DEVICE. FOR INVESTIGATIONAL USE IN THE STEPPING UP STUDY ONLY.

Accu-Chek 360° View 3-day Profiling Tool

BLOOD GLUCOSE

Accu-Chek 360° View 3-day Profiling ToolData can show you:• Trends in blood glucose levels

• The relationship between blood glucose values and

- Time of day - Meal size - Activity level - Diabetes medication (if prescribed)

FOOD DIARYUse this space to fill in what you eat and drink over 3 days.

Accu-Chek 360° View 3-day Profiling Tool

By drawing a line through the recordedresults, you can easily identify trends inblood glucose.

Out-of-range blood glucose values canindicate a need for better blood glucosecontrol, and might suggest the need toadjust and/or change therapy.

Accu-Chek and Accu-Chek 360º View are registered trademarks of Roche.© 2012 Roche. All other trademarks are the property of their respective owners.

www.accu-chekacademy.com.auRoche Diagnostics Australia Pty Ltd31 Victoria Avenue, Castle Hill NSW 2154ABN 29 003 001 205Accu-Chek Enquiry Line: 1800 251 816

Day 1

Breakfast

Snack

Lunch

Dinner

Drinks (soft drinks, hot beverages, alcohol, etc)

Day 2

Breakfast

Snack

Lunch

Dinner

Drinks (soft drinks, hot beverages, alcohol, etc)

Day 3

Breakfast

Snack

Lunch

Dinner

Drinks (soft drinks, hot beverages, alcohol, etc)

Fill in the dates for the days on which you will track your blood glucose results.

Test your blood glucose using your Accu-Chek blood glucose monitoring system at the times indicated to the left.

Enter the time of the test in the first row of the chart.

If you use insulin, enter your insulin dose (units).

Based on your normal eating habits, describe this meal size by circling Small, Medium or Large in the second row.

Rate your activity level on a scale of 1 (very low) to 5 (very high) and circle that score.

Enter your blood glucose value in the fifth row for that day.

Graph your blood glucose level (from Step 7) by placing an X in the corresponding row of the chart. Then connect the Xs. See other side for example.

Step 1

Step 3

Step 4

Step 5

Step 6

Step 7

Step 8

Step 2

Instructions to patient:Complete this form over3 consecutive days.

Page 196 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyGlulisine (Apidra)Initiation of glulisine (Apidra) is indicated when the patient has achieved the target fasting blood glucose of 4.0-7.0 mmol/L but have postprandial blood glucose readings of above 10mmol/L. If there is more thanonemealatwhichthisoccurs,targettheonewiththegreatestpost-mealexcursionfirst.The3dayprofilingtoolwillassistyouindeterminingthis(seeFigure17).ThestartingdoseforApidrais4units.

Remember: No matter how high the BGL is, the first (starting) dose of 4 units APIDRA should be adhered to, until it is established how sensitive your patient is to this insulin. In some patients one unit of Apidra may drop the blood glucose level 4 to 6mmol/l, in others 4 units may only drop them 1mmol/l. Until the effect on each individual patient is established it is important to maintain frequent contact in relation to dose adjustment of Apidra. If the pre-meal injection is forgotten, it is NOT to be given an hour or two later when a high glucose is detected, but rather ask them to record in their BG diary that it was missed. The following day they can return to the planned schedule and inject prior to the designated meal.

Glulisine (Apidra) is titrated using the Insulin adjustment tool for use with prandial insulin glulisine (Apidra) (see Figure 18) every three days until the target of 4.5 – 7.0mmol/L two hours after the meal is attained (note: for some patients the GP may aim for a target of up to 10mmol/L – this is at the discretion of the clinician).

Figure 18: Insulin adjustment tool for use with prandial insulin glulisine (Apidra)

Notes

Use this space to record details of hypoglycaemia

Average Blood Glucose

3-Day Total

Blood glucose 2 hours after meal mmol/L

Finger Prick

Start Date

Daily Dose (units)

Insulin Adjustment Tool for use with Prandial insulin

Patient Name

Healthcare Professional Name

Oral Diabetes Medications Dose Time/DayInsulin Starting Dose (units)

Warning: Do not use this form without first consulting your healthcare professional

Handling HypoglycaemiaIf you feel Hypoglycaemic or have blood glucose below 4mmol/L at any time since your last insulin dose:

1 Treat immediately by following instructions given to you by your healthcare professional

2 Cross out any remaining blood glucose or calculation fields in current Period

3 Reduce the dose of the last insulin given by 2 to 4 and start a new Period

Find Your Dose Adjustment Before dosing on Day 3If you were NOT hypoglycaemic at this dose, use this table to find your Adjustment.

Average BloodGlucose (mg/dL) More than 14

Adjustment (units) +6 to 8 units

Less than 4.5 4.5-7.0 7.1-10.0 10.1-12.0 12.1-14.0

-2-4 units 0* +0 to 2 units +2 units +4 to 6 units

*Dose Adjustment is complete when your Average Blood Glucose Falls within the target range. Continue monitoring your fasting blood glucose daily as directed by your healthcare professional.

Target range

Period

units

Day 1

mmol/L++

=

÷ 3 =

+ =

Adjustment

Day 2

mmol/L

Day 3

mmol/L

Period

units

Day 1

mmol/L++

=

÷ 3 =

Day 2

mmol/L

Day 3

mmol/L

>

Record as new

Daily Dose on your

next sheet

Tick which meal this sheet applies to:

Breakfast

Lunch

Dinner

ONLY ADJUST ONE MEAL AT A TIME

>+ =

Adjustment

Period

units

Day 1

mmol/L++

=

÷ 3 =

Day 2

mmol/L

Day 3

mmol/L

+ =

Adjustment

APIDRA

18 |

Page 197 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 19

Side effects of insulinInsulin is very safe when used with the guidance of health professionals and at the correct dose. The main side effect is hypoglycaemia (low blood glucose level). Mild hypoglycaemia can be treated very quickly by the person with diabetes without too much disruption to their day. Very low blood glucose levels can be dangerous and must not be ignored. In order to keep the patient symptomatic of hypoglycaemia at a reasonable level (around 3.5 to 3.8mmol/l) stress the importance of rapid, appropriate treatment of the hypoglycaemia as soon as symptoms are detected. Regularly delaying treatment for 10 to 15 minutes will result in hypoglycaemia unawareness at that level, and symptoms will progressively not manifest themselves until lower and lower glucose levels are reached.

Some people may experience a slight reaction where the injection was given. This usually goes away within a few days. Very rarely, a person may experience a reaction to the insulin that requires them to stop it and start another type of insulin. If they experience any side effects please discuss with the GP.

Weight gain can occur after starting insulin therapy. Often, the longer people wait before starting insulin, the more weight they gain. This can be curbed by increasing exercise and reducing energy intake. The benefitsofbetterbloodglucosecontrolwithinsulinoutweightherisksofincreasedweight.

HypoglycaemiaHypoglycaemia is the main potential side effect of insulin therapy and it is essential that the person starting insulin, and their immediate family, know what symptoms to expect, how to reduce the risks of hypos and how to treat them.

Hypoglycaemia can be caused by one of a number of events such as:

> Delaying or missing a meal

> Unplanned physical activity

> Not eating enough carbohydrate

> More strenuous exercise than usual

> Too much insulin or diabetes tablets

While symptoms vary from person to person, common feelings are:

> Weakness, trembling or shaking.

> Light headedness

> Dizziness

> Tearfulness/crying

> Hunger

> Sweating

> Headache

> Lack of concentration/behaviour change

> Irritability

> Numbnessaroundthelipsandfingers

Treatment of hypoglycaemiaThefirstthingtodoisensuresafety.Forexampleifdrivingacar,pullovertothesideoftheroad.Then,advise the patient to have some quick acting carbohydrate that is easy to consume, for example;

> ½ can regular soft drink, or

> ½ glass of fruit juice, or

> 3 teaspoons of sugar or honey, or

> 6-7 jelly beans, or glucose tablet equivalent to 15 grams of carbohydrate

Page 198 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyThe following steps are also recommended. The order and timing of these steps depends on the severity of the ‘hypo’ and time and circumstances.

> Wait 10-15 minutes. If it isn’t rising, repeat treatment with a quick acting carbohydrate as described above.

> If the next meal is more than 20 minutes away, eat some longer acting carbohydrate. This could be one of the following:

’ A sandwich or

’ 1 glass of milk or soy milk or

’ 1 piece of fruit or

’ 2-3piecesofdriedapricots,figsorotherdriedfruitor

’ 1 tub of natural low fat yoghurt or

’ 6 small dry biscuits and cheese

If consciousness is impaired oral intake of carbohydrate should not be attempted and caregivers should be advised to seek emergency medical care. Treatment will include management of hypoglycaemia and review of diabetes medications.

Storage of insulin and sharps disposalUnopened insulin should be stored in the fridge, between 2 and 8 degrees Celsius. Once opened, insulin may be kept at room temperature (between 25 to 30 degrees Celsius) for one month and then discarded.

Insulin can be damaged by extreme temperatures. It must not be left where temperatures reach over 30 degrees, e.g. in the car or in direct sunlight. Insulin should not be allowed to freeze as it will lose its potency,andmustbediscarded.Discardtheinsuliniflumpsorflakesareseenintheinsulinorontheinside of the cartridge and are not able to be dissolved by gently rotating the pen.

Used syringes, pen needles and lancets must be disposed of in an Australian Safety Standards-approved sharps container which is puncture proof and has a secure lid. These are usually yellow in colour and are available through pharmacies, your local municipal council and DA–VIC.

Stepping Up protocol

Blood glucose monitoringAll patients participating in the Stepping Up study will be supplied with a Roche Performa Nano blood glucose meter.

Patients should be directed to monitor blood glucose levels at least twice daily - before breakfast and at another time (preferably about 2 hours after a meal). The second reading can be rotated. When consideringtheadditionofApidra,thepatientwillberequestedtocompletethe3dayprofiletoolwhichrequires seven blood glucose readings per day for three days (see Figure 17, Page 17).

Times and readings are to be recorded in the blood glucose monitoring diary and tools provided. Any symptomatic hypoglycaemic episodes are to be recorded in the diary and any episodes of severe hypoglycaemia (requiring assistance from a third party) are to be reported to the study team within 24 hours.

Patients are to be advised to bring the diary and meter in with them at every visit to the clinic. The diary will be collected at the conclusion of the study. The Roche SmartPix tool will be available to clinics to upload blood glucose readings directly from the meter to a computer so that information can also be viewed visually. Patients still need to record their readings in the diary and tools provided.

20 |

Page 199 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 21

Starting glargine (Lantus)

Page 200 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyStarting glulisine (Apidra)

22 |

Page 201 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 23

References1. Australian Institute of Health and Welfare (AIHW), Diabetes: Australian facts 2008. Diabetes series

no. 8. Cat. no. CVD 40. Canberra: AIHW. 2008.

2. Diabetes Australia & Royal Australian College of General Practitioners (RACGP), Diabetes management in general practice - 17th edition, in Guidelines for Type 2 diabetes. 2011/2012, Diabetes Australia.

3. Britt, H., et al., General practice activity in Australia 2010-11. General Practice series 29. Vol. 2011. 2011: Sydney University Press.

4. Calvert, M.J., R.J. McManus, and N. Freemantle, Management of type 2 diabetes with multiple oral hypoglycaemic agents or insulin in primary care: retrospective cohort study. Br J Gen Pract, 2007. 57(539): p. 455-460.

5. Harris, S.B., et al., Clinical inertia in patients with T2DM requiring insulin in family practice. Canadian Family Physician Médecin De Famille Canadien, 2010. 56(12): p. e418-24.

6. Britt, H., G.C. Miller, J. Charles, J. Henderson, C. Bayram, L. Valenti, Y. Pan, C. Harrison, S. Fahridin, J. O'Halloran, General Practice activity in Australia 2008-09. 2009, General practice series no. 25. Cat no GEP 25. Canberra:AIHW.

7. Holman, R., et al., 10-year follow-up of intensive glucose control in type 2 diabetes. New England Journal of Medicine, 2008. 359: p. 1577-89.

8. Davis, T.M.E., W.A. Davis, and D.G. Bruce, Glycaemic levels triggering intensification of therapy in type 2 diabetes in the community: the Fremantle Diabetes Study. Medical Journal of Australia, 2006. 184: p. 325-328.

9. Spoelstra, J.A., et al., Factors associated with switching from oral hypoglycaemic agents to insulin therapy. Netherlands Journal of Medicine, 2002. 60(6): p. 243-248.

10. Nakar, S., et al., Transition to insulin in Type 2 diabetes: family physicians' misconception of patients' fears contributes to existing barriers. Journal of Diabetes and its Complications, 2007. 21(4): p. 220-226.

11. Brunton, S., et al., Type 2 diabetes: the role of insulin. The Journal Of Family Practice, 2005. 54(5): p. 445-452.

12. Kunt, T. and F.J. Snoek, Barriers to insulin initiation and intensification and how to overcome them. International journal of clinical practice. Supplement, 2009. (164): p. 6-10.

13. Shah, B.R., et al., Clinical inertia in response to inadequate glycemic control: do specialists differ from primary care physicians? Diabetes care, 2005. 28(3): p. 600-6.

14. Bliss, M., The discovery of insulin. 1996, Toronto, Ontario: McClelland & Stewart Inc.

15. Masharani, U., J.H. Karam, and M.S. German, Pancreatic hormones and diabetes mellitus, in Basic and clinical endocrinology - seventh edition, F.S. Greenspan and D.G. Gardner, Editors. 2004, McGraw-Hill: USA.

16. The Forum for Injection Technique, Diabetes Care. The First UK Injection Recommendations - 2nd edition, D. Hicks, et al., Editors. 2011.

Page 202 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyAppendix

Motivational Interviewing Resources

clinical practice

Moira G Sim MBBS, FRACGP, FAChAM, is Associate Professor, Centre for Postgraduate Medicine, Deputy Director, Systems & Innovation Research Centre for Health, Edith Cowan University, and a general practitioner, Western Australia. [email protected]

toni Wain BA, BSc, is Program Manager, Systems & Innovation Research Centre for Health, Edith Cowan University, Perth, Western Australia.

eric Khong MBBS, FRACGP, is Senior Lecturer, Centre for Postgraduate Medicine, Edith Cowan University and School of Psychiatry and Clinical Neurosciences, University of Western Australia, and a general practitioner, Western Australia.

instigating behaviour change in patients to help them achieve a healthy lifestyle is a critical component of general practice to combat rising chronic disease in the community. an individual’s health behaviour is determined by a complex interplay between their knowledge and understanding of health and disease, the personal meaning and relevance of that knowledge, their confidence in their ability to make changes, and a range of other factors acting as facilitators and barriers to change.1

Brief interventionsBrief interventions (provision of information and advice) by general practitioners are successful in promoting healthy behaviour in relation to smoking, alcohol, physical activity and nutrition.2,3 General practitioners develop therapeutic relationships with patients over time, are respected, and have multiple opportunities to provide brief advice in a range of health behaviours. Furthermore, GPs can tailor information to the individual, which is more effective than generic information.4,5 A framework that has been widely promoted for brief intervention is the 5As approach – Ask, Assess, Advise, Assist and Arrange follow up. Initially developed for managing tobacco dependence, it has subsequently been applied to many other health behaviours.6 With this approach, all patients are screened for unhealthy behaviours (ask). All those identified by screening are assessed for their readiness to change (assess) and advised to alter their behaviour (advise). Those who agree to change are provided with practical help to make the change (assist) and followed up to provide support (arrange). This can include practical strategies such as the use of a decisional balance sheet,7 which compares the pros and cons of change, setting dates and times for action and support medication. Brief interventions in general practice usually take a few minutes and are incorporated into routine care. While there is strong evidence

influencing behaviour change in general practice Part 1 – brief intervention and motivational interviewing

BackgroundBehaviour change toward achieving a healthy lifestyle is important for all Australians, and general practitioners have a key role to play in assisting patients to make these changes.

ObjectiveThis is the first of a two part series which provides the background to approaches to influencing behaviour change in general practice, from brief interventions to motivational interviewing (MI). The second part of the series will explore motivational interviewing in more detail.

Discussion General practitioners have a key role in changing their patients’ health behaviours. There are a range of tools GPs can use to help enhance their patients’ motivation to achieve their health goals.

reprinted from auStralian FaMily phySician Vol. 38, No. 11, November 2009 885

24 |

Page 203 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 25

Appendix

Motivational Interviewing Resources

influencing behaviour change in general practice – part 1 – brief intervention and motivational interviewingclinical practice

Another brief intervention approach that has a stronger emphasis on empathy and therapeutic interaction is the use of the acronym FRAMES – Feedback about risk, emphasis on personal Responsibility, Advice to change, a Menu of options, Empathy and facilitation of Self efficacy.13 While the ‘brief interventions’ for which FRAMES was developed referred to a few short term counselling sessions and not the brief interventions that occur in general practice, the components can still be relevant to the longitudinal interaction that occurs in general practice over multiple consultations.

to support brief interventions, they are sometimes insufficient to alter behaviour.7 When positive outcomes are not apparent, the sense of inability to change behaviour can be associated with frustration, a lack of confidence and negative attitudes toward attempting to change health behaviours.8,9

transtheoretical model of change

The transtheoretical model (TTM) of change, commonly known as the ‘cycle of change’ or ‘stages of change’, was originally developed by Prochaska and DiClemente in 1982 in relation to tobacco use.10,11 It has since been applied to many behaviours and helps to explain why attempts to change behaviour may not work.11 There have been many versions of this model. In its original form, it consisted of five stages:• precontemplation(notthinkingaboutchange)• contemplation(thinkingaboutchange)• preparation/decisionmaking(activelyplanningchange)• action (changing or recently changed and new behaviour not

established), and • maintenance (new behaviour established and working on

maintenance). Relapse was not considered to be a stage as it was seen as an event that terminates the action or maintenance phase, prompting a movement back into an earlier phase. Later, the authors proposed a ‘spiral of change’ which suggests that most relapsers do not revolve endlessly in circles regressing back to the beginning, but instead learn from each attempt and progress with each cycle11 (Figure 1). While the cycle of change is frequently associated with MI, Miller and Rollnick specifically point out that they are not the same thing and that MI was never based upon TTM. The transtheoretical model of change is a model of how and why changes occur, while MI is a specific clinical method to enhance motivation for change. It is not necessary to assign a stage of change in order to apply MI.12

Another model that evolved from the original stages of change is the ’contemplation ladder’,20,21 in which the clinician asks people to identify where they are on a ladder (Figure 2).

ambivalence, resistance and defence mechanisms

Ambivalence is a normal aspect of human nature and is a natural transition phase in the process of change.7 For example, a person who is dependent on a drug can have a ‘love-hate relationship’ with the drug, ie. intensely positive feelings toward the drug together with an intense dislike of the problems associated with use of the drug and the control it has over the person. Change and resistance to change are seen as two sides of a coin.7

Giving brief advice, regardless of the stage of change, may be effective; this advice may trigger consideration of, or a decision to, change. On the other hand, advice can increase resistance to change7 in the same way that someone who feels they are being nagged might become resentful and less willing to change.

Preparation

Preparation

Preparation

Relapse

Relapse

Contemplation

Contemplation

Contemplation

High threat

Emotional arousal and search for options

Low threat

No action

High coping

Behaviour change to reduce threat

Low coping

Defence mechanismand denial to reduce fear

I stopped/cut downa while ago

I have just stopped/cut down

I have made real plansto stop/cut down

I think I need to stop/cut down but I’m notsure I want to

I’m happy with whatI’m doing, I don’t feel the need to stop

Maintenance

Action

Preparation

Contemplation

Precontemplation

Precontemplation

Maintenance

Maintenance

Maintenance

Action

Action

Action

Long term success

Where do you see yourself on this ladder?

Threat appraisalPerceived severity – how bad would it be if the threat happened?Perceived vulnerability – how likely is it to happen to me?

Coping appraisalPerceived response efficiency – will changing behaviour reduce the threat?Perceived self efficacy – can I change my behaviour?

Figure 1. The spiral of change

Preparation

Preparation

Preparation

Relapse

Relapse

Contemplation

Contemplation

Contemplation

High threat

Emotional arousal and search for options

Low threat

No action

High coping

Behaviour change to reduce threat

Low coping

Defence mechanismand denial to reduce fear

I stopped/cut downa while ago

I have just stopped/cut down

I have made real plansto stop/cut down

I think I need to stop/cut down but I’m notsure I want to

I’m happy with whatI’m doing, I don’t feel the need to stop

Maintenance

Action

Preparation

Contemplation

Precontemplation

Precontemplation

Maintenance

Maintenance

Maintenance

Action

Action

Action

Long term success

Where do you see yourself on this ladder?

Threat appraisalPerceived severity – how bad would it be if the threat happened?Perceived vulnerability – how likely is it to happen to me?

Coping appraisalPerceived response efficiency – will changing behaviour reduce the threat?Perceived self efficacy – can I change my behaviour?

Figure 2. The contemplation ladder20,21

886 reprinted from auStralian FaMily phySician Vol. 38, No. 11, November 2009

Continued from previous page

Page 204 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyAppendix

Motivational Interviewing Resources

influencing behaviour change in general practice – part 1 – brief intervention and motivational interviewing clinical practice

resistance, Avoid arguments, Can do, and Express empathy. These are summarised in Table 1.15

Motivational interviewing has been applied to many aspects of behaviour change ranging from alcohol and drug dependence, smoking cessation, weight loss, physical activity, the treatment of asthma and diabetes, adherence to treatment and follow up, and criminal activity. A systematic review and meta-analysis of randomised controlled trials using MI as the intervention found 74% of trials demonstrated a positive effect of MI. Measures of clinical relevance that improved with MI included body mass index, cholesterol, blood pressure, blood alcohol concentration, blood glucose, and length of hospital stay. The number of encounters and length of follow up was more important than the length of each encounter, with 64% of studies using brief encounters of 15 minutes being effective. Both psychologists and physicians obtained an effect in 80% of the studies.16 In a trial comparing GPs who were randomised to MI training to those who were not, the GPs trained in MI evaluated it to be more effective and no more time consuming than ‘traditional advice giving’.17

changing behaviour in general practiceGeneral practitioners manage many health issues within a limited period of time in each consultation. It is relatively simple to integrate brief interventions consisting of the provision of information and advice relevant to the health issues being dealt with. This may be triggered by the presenting complaint or from routine screening for risk factors. Patients who are ready to change can be provided with further assistance and follow up. Brief advice from a respected person such as a medical practitioner may be effective regardless of the stage of change.3 However, by recognising different needs at each stage of change, approaches

Motivational interviewing

Motivational interviewing is a collaborative, person centred way of guiding the patient to elicit and strengthen motivation to change.12 The goal is to increase intrinsic motivation rather than to impose it externally. Involving a flexible blend of informing, asking and listening, it works to evoke the patient’s own values, goals, insights, motivation and resources for change.14 The approach to MI is expressed in the acronym GRACE: Generate a gap, Roll with

Table 1. Using GRACE in all stages of change

Generate a gap The goal is to increase intrinsic motivation by generating a gap (between what the patient wants and what is) rather than to impose it externallyroll with resistance If resistance is encountered, alter the strategy used. Ambivalence is viewed as normal, not pathological, and is explored openlyavoid arguments Arguing increases resistance to change. It should be the individual and not the therapist who voices the arguments for change. The goal is to encourage the patient to hear themselves say why they want to changecan do Encourage self efficacy and hope. A person may perceive a serious problem but still will not move toward change unless there is hope for successexpress empathy Listen, communicate acceptance and support and gently persuade while respecting personal views and choice

Table 2. Stages of change using brief intervention and motivational interviewing approaches

precontemplation(not considering change)

contemplation(considering change)

preparation (planning change)

action (recent change)

Maintenance(change established)

5as •Ask•Assess•Advise

•Ask•Assess•Advise•Assist

•Ask•Assess•Advise•Assist•Arrange

•Ask

•Assist•Arrange

•Ask

•Arrange

FraMeS •Feedback•Responsibility•Advice•Menu(options)•Empathy•Selfefficacy

•Feedback•Responsibility•Advice•Menu(options)•Empathy•Selfefficacy

•Feedback•Responsibility•Advice•Menu(options)•Empathy•Selfefficacy

•Feedback•Responsibility•Advice•Menu(options)•Empathy•Selfefficacy

•Feedback•Responsibility•Advice•Menu(options)•Empathy•Selfefficacy

Grace •Buildmotivation •Buildmotivationand strengthen commitment

•Strengthencommitment

•Strengthencommitment

•Strengthencommitment

reprinted from auStralian FaMily phySician Vol. 38, No. 11, November 2009 887

Continued from previous page

26 |

Page 205 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 27

Appendix

Motivational Interviewing Resources

influencing behaviour change in general practice – part 1 – brief intervention and motivational interviewingclinical practice

14. Rollnick S, Miller WR, Butler CC. Motivational interviewing in health care: Helping patients change behaviour. New York: Guilford Press, 2008.

15. Winarski J, Silver S, Kraybill A. Motivational interviewing: Applications for PATH service providers – an edited transcript of the PATH national teleconference call. Available at http://pathprogramarchive.samhsa.gov/text_only/tech_assist/Transcripts/MotivationalInterviewing_2_2003.asp.

16. Rubak S, Sandbaek A, Lauritzen T, Christensen B. Motivational interviewing: A systematic review and meta-analysis. Br J Gen Pract 2005;55:305–12.

17. Rubak S, Sandbaek A, Lauritzen T, Borch-Johnsen K, Christensen B. An education and training course in motivational interviewing influence: GPs’ professional behaviour – ADDITION Denmark. Br J Gen Pract 2006;56:429–36.

18. Evers KE, Prochaska JM, Prochaska JO, Driskell MM, Cummins CO, Velicer WF. Strengths and weaknesses of health behaviour change programs on the internet. J Health Psychol 2003;8:63–70.

19. Evers KE, Cummins CO, Prochaska JO, Prochaska JM. Online health behavior and disease management programs: Are we ready for them? Are they ready for us? J Med Internet Res 2005;7:e27.

20. Baker A, Kay-Lambkin F, Lee NK, Clare M, Jenner L. A brief cognitive behavioural intervention for regular amphetamine users. Canberra: Australian Government Department of Health and Ageing, 2003.

21. Biener L, Abrams DA. The contemplation ladder: Validation of a measure of readi-ness to consider smoking cessation. Health Psychol 1991;10:360–5.

can be tailored to the individual’s stage.18,19 This is illustrated in Table 2, which maps the stages of change against some brief intervention and MI approaches. A simple question that can help the GP to assess readiness to change and the right approach to use is: ‘How do you feel about your smoking/drinking/lackofexercise?’Theanswer to thisquestionwillquickly establish if the person is ready to be given brief advice, or needs a gentler explorative approach using MI principles. Medical practitioners are usually confident in helping people who have made a decision and commitment to change. Motivational interviewing fills the gap in providing strategies to explore ambivalence, build motivation and strengthen commitment in those who are uncertain or not yet considering change.

conclusionGeneral practitioners are respected, see the majority of the population and have continuity of care, which means they play an important role in enhancing health outcomes through changing behaviour. Each GP develops a personal communication style; while this personal style might be effective in changing behaviour in a proportion of patients, having a range of tools increases the GP’s repertoire, allowing them to facilitate more change.

Conflict of interest: none declared.

references1. Nutbeam D, Harris H. Theory in a nutshell: A practical guide to health promotion

theories. 2nd edn. Sydney: McGraw-Hill, 2004.2. SNAP: A population health guide to behavioural risk factors in general practice.

TheRACGP,2004.Availableatwww.racgp.org.au/guidelines/snap.3. Nutrition and Health Foundation. Motivational aspects of behaviour change 2007.

Dublin: NHF, 2007. Available at www.nutritionandhealth.ie [Accessed 12 April 2009].

4. Oenema A, Brug J. Feedback strategies to raise awareness of personal dietary intake: Results of a randomized controlled trial. Prev Med 2003;36:429–39.

5. Resnicow K, Davis RE, Zhang G, et al. Tailoring a fruit and vegetable intervention on novel motivational constructs: Results of a randomized study. Ann Behav Med 2008;35:159–69.

6. Fiore MC, Bailey WC, Cohen SJ, et al. Treating tobacco use and dependence. United States Department of Health and Humans Services, 2000. Available at www.surgeongeneral.gov/tobacco/treating_tobacco_use.pdf [Accessed 8 April2009].

7. Miller WR, Rollnick S. Motivational interviewing: Preparing people for change. 2nd edn. New York: The Guilford Press, 2002.

8. Wens J, Vermeire E, Royen PV, Sabbe B, Denekens J. GPs’ perspectives of type 2 diabetes patients’ adherence to treatment: A qualitative analysis of barriers and solutions. BMC Fam Pract 2005;6:20.

9. Vogt F, Hall S, Marteau TM. General practitioners’ and family physicians’ negative beliefs and attitudes towards discussing smoking cessation with patients: A sys-tematic review. Addiction 2005;100:1423–31.

10. DiClemente CC, Prochaska JO. Self change and therapy change of smoking behav-iour: A comparison of processes of cessation and maintenance. Addict Behav 1982;7:133–42.

11. Prochaska JO, DiClemente CC, Norcross JC. In search of how people change: Applications to addictive behaviours. Am Psychol 1992;47:1102–14.

12. Miller WR, Rollnick S. Ten things that motivational interviewing is not. Behav Cogn Psychother 2009;37:129–40.

13. Miller WR, Sanchez VC. Motivating young adults for treatment and lifestyle change. In: Howard G, editor. Issues in alcohol use and misuse by young adults. Notre Dame: University of Notre Dame Press, 1994. CORRESPONDENCE [email protected]

888 reprinted from auStralian FaMily phySician Vol. 38, No. 11, November 2009

Continued from previous page

Page 206 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyAppendix

Motivational Interviewing Resources

clinical practice

Motivational interviewing (Mi) is a collaborative person centred guidance strategy to elicit and strengthen motivation to change.1 it evolved from carl roger’s client centred counselling approach which focuses on the person’s interests and concerns, but differs by being consciously directive toward resolving ambivalence and moving toward change.2 the goal is to increase intrinsic motivation rather than to impose it externally.3 it was initially developed from work with problem drinkers, where in comparison with confrontational directive styles, motivational reflective styles were associated with lower levels of resistance and a higher likelihood of long term change.4

The ‘spirit’ of MI is collaborative (a partnership between the patient and the clinician), evocative (evoking from the patient’s own values, goals, insights, motivation and resources for change), and honouring patient autonomy (acceptance that the patient makes his/her own choices).5 It is particularly useful for those who are reluctant to change or ambivalent about changing behaviour.6

Motivation is seen not as a personal trait, but as an interpersonal process that results from the interaction between the practitioner and the patient. How the practitioner acts influences motivation to change. Resistance to change and denial is considered a signal to the therapist to alter strategies.3

ready, willing and able to changeFor behaviour change to occur, a person has to want to change, feel that they can change, and feel it is the right time to prioritise this action.3 Motivational interviewing can help build motivation, commitment and confidence to change. Simply knowing that change is needed is not enough. Even if a person wants to change, they need to believe that they can before they take action. According to the protection motivation theory (PMT) developed by Rogers,7 if a person believes there

Background Behaviour change toward achieving a healthy lifestyle is important for all Australians, and general practitioners have a key role to play in assisting patients to make these changes.

Objective This is the second of two articles on influencing behaviour change in general practice. This article deals with the ‘how to’ of motivational interviewing in the general practice setting.

Discussion Motivational interviewing can help build motivation, commitment and confidence to change. General practitioners can use motivational interviewing to help their patients achieve their health goals. Motivational interviewing is not about a set of techniques and questions; it is about creating a climate that facilitates change; it is more about listening than telling, evoking rather than instilling. Motivational interviewing can be done in the brief periods available in consultations over time.

influencing behaviour change in general practice Part 2 – motivational interviewing approaches

Moira G Sim MBBS, FRACGP, FAChAM, is Associate Professor, School of Nursing Midwifery and Postgraduate Medicine, Principal, Systems & Innovation Research Centre for Health, Edith Cowan University, and a general practitioner, Western Australia. [email protected]

toni Wain BA, BSc, GradDipHlthServMgt, is Program Manager, Systems & Innovation Research Centre for Health, Edith Cowan University, Perth, Western Australia.

eric Khong MBBS, GradCertHlthEcons, GradDipPHC, FRACGP, is Senior Lecturer, Centre for Postgraduate Medicine, Edith Cowan University and School of Psychiatry and Clinical Neurosciences, University of Western Australia, and a general practitioner, Western Australia.

986 reprinted from auStralian FaMily phySician Vol. 38, No. 12, December 2009

Continued from previous page

28 |

Page 207 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 29

Appendix

Motivational Interviewing Resources

over this? What do you think of that?’ These questions can evoke consideration toward change from the individual

•Avoidarguing–itshouldbetheindividualandnottheclinicianwhovoices the arguments for change.3 The goal of MI is to encourage the individual to hear themselves say why they want to change. If the doctor is perceived as challenging the patient’s position and not listening, then the patient will work harder to try to convince the doctor of the arguments for not changing. In the process of vocalising the reasons against change, they reinforce their own resistance to change since humans tend to move toward congruence between external actions (speech and action) and internal attitudes (beliefs and values).9,12 Avoid the ‘yes, but...’ arguments where you and the person argue over change. Instead try, ‘I’m wondering if you have any ideas about how you can exercise more, even when you’re very busy?’ Or, ‘It sounds like my ideas aren’t very good. Do you have any ideas?’

•Can do – a person convinced of a need to change will not movetoward change without self efficacy (belief that they can succeed).13 Without this, they are likely to adopt defensive coping (eg. rationalisation, denial) to reduce discomfort instead of behaviour change.11 There is no ‘right way’ to change, and all previous attempts and learning, as well as pharmacotherapy and psychotherapy options, can be explored

•Express empathy – listening and communicating acceptance,understanding of ambivalence and respect for the individual’s decisions.11 Active listening is encapsulated by the acronym ‘OARS’ (open ended questions, affirmations, reflective listening, summaries).3

Each clinician develops a unique style, which might involve a blend of empathy, humour, ‘straight talk’, encouragement and other personal touches which can be adapted to the individual consultation.

is a serious health threat but does not believe that anything can be done about it, the results are defence mechanisms and denial to reduce the emotional arousal associated with this knowledge of threat8 (Figure 1).

the guiding principles of motivational interviewing

The acronym ‘RULE’ summarises the principles of MI:5

•Resistingtherightingreflex–resistingtheneedtosolveproblemsortell patients what to do. This may seem at odds with the philosophy of brief intervention in which GPs are encouraged to advise patients to change their behaviour. However, repeatedly advising behaviour change in those who are not ready to change may increase their resistance and therefore be counterproductive

•Understandandexplorethepatient’sownmotivations–humanstendto believe what we hear ourselves say.5 Bem’s self perception theory suggests that what we say and do influences our own attitudes.9 Hearing someone else say something is not as powerful as hearing ourselves say it. The goal of MI is to increase intrinsic motivation rather than to impose it externally

•Listenwithempathy–inachievingbehaviourchangeakeytaskistolisten in a way that will draw out the patient’s issues so that it can be heard by both the patient and the doctor

•Empower thepatientbyencouraginghopeandoptimism–patientsknow better than anyone else how to change their own behaviour but may sometimes lack confidence in their own ability. A skilled practitioner encourages patients to vocalise why and how they intend to change during the consultation, knowing that this both reinforces the patient’s own expertise in their own actions and influences the patient’s attitudes.

Asking instead of telling draws out the patient’s thoughts, feelings, understanding and motivations. For example, ‘Your blood sugar is high. What do you think of that? Does that worry you? Why do you think it’s gone up? What can you do about that? Do you think you can manage that? How have you managed to do it before?’ The answers to questions give us insights to the person’s level of knowledge and beliefs, helping us to provide more relevant information and an appropriate response. A second acronym, which provides useful strategies during consultations, is ‘GRACE’.10

•Generate a gap – motivation for change happens when peopleperceive a discrepancy between where they are and where they want to be.11 Raising awareness of the adverse consequences of behaviours by exploring experiences, values and attitudes nonjudgmentally can help generate this gap

•Roll with resistance – if the individual perceives an attack,defensiveness ensues, which evokes resistance. The patient’s resistance is not challenged, instead MI ‘rolls with’ the momentum, viewing ambivalence as normal. In this way resistance is decreased and new perspectives are invited by exploring ambivalence openly.11 For example: ‘Oh no, you’re not going to have a go about my smoking again!’ can be met with, ‘Has someone been giving you a hard time

Figure 1. Protection motivation theory8

No action

Defence mechanism and denial to reduce fear

Emotional arousal and search for options

Behaviour change to reduce threat

High threat

High success expected Low success expected

Low threat

Threat appraisal•Perceivedseverity–howbadwoulditbeifthethreathappened?•Perceivedvulnerability–howlikelyisittohappentome?

Response success appraisal•Perceivedresponseefficiency–willchangingbehaviourreducethethreat?•Perceivedselfefficacy–canIchangemybehaviour?

reprinted from auStralian FaMily phySician Vol. 38, No. 12, December 2009 987

Continued from previous page

Page 208 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up StudyAppendix

Motivational Interviewing Resources

influencing behaviour change in general practice – part 2 – motivational interviewing approachesclinical practice

•Need–Imust,Ineed,important,gotto,reallyhaveto•Commitment–Iwill,I’mgoingto,it’stimenow.In eliciting change talk, instead of giving information, we ask questions and invite comments that draw out the patient’s expressions of what, why, how and when to change. Among the ‘change talk’ statements, ‘commitment talk’ is the most predictive of change. Figure 2 demonstrates the uphill process of exploring ambivalence and building motivation while encouraging change talk. Building commitment becomes easier as resistance decreases and motivation increases.

changing behaviour in general practicePeople are often ambivalent about health behaviours and resistance and change are two sides of a coin.1 It is important to stress that MI is not about a set of techniques and questions; it is about creating a climate that facilitates change; it is more about listening than telling, evoking rather than instilling; and communicates, ‘You have what you need, and together we will find it’.16 Motivational interviewing can be done in the brief periods available in consultations over time1 (Table 1, Case study).

relapse management

Lapses (a brief return to the earlier behaviour) and relapses (a sustained return to the earlier behaviour) are common. It is important to stress that a lapse is different to a relapse and can result in new behaviour. Both clinicians and patients will often feel a sense of failure when relapses occur. This can be reframed, eg. ‘I’m a failure’ can be viewed as a partial success in a person who knows what needs to be changed, who is motivated to keep going despite adversity, and is willing to accept

eliciting ‘change talk’Motivational interviewing can be divided into two phases: building motivation and strengthening commitment.3 Miller and Rollnick use the term ‘change talk’ to refer to statements from the individual that reinforce the movement toward change. The aim is to elicit ‘change talk’ statements through skilful questioning and reflection which express the following desire, ability, reasons, need and commitment (DARN-C)14,15:•Desire–Iwant,Isowant,Iwish•Ability–Ican,Icould,it’spossible,IknowIcan•Reasons–because,since,I’msickof,Ihateit

Table 1. Questions to facilitate discussion in motivational interviewing in general practice

Brief: the following questions can be used within a 5 minute discussionQuestion rationale‘What do you like about (your smoking or other behaviour)?’

This question is unexpected as most would be expecting a lecture. It gives an opportunity to listen and build rapport. It also gives valuable information that helps to understand the context of the behaviour. It may also be important to consider how to replace this function if the behaviour were to stop

‘What don’t you like about...?’ This question is critical as it draws out the internal motivation for change If appropriate you can add your own concerns

If you believe that adding your own concerns about the behaviour might help to tip the balance toward change and not increase resistance, this can be done

Summarise: ‘So you like... but you don’t like..., so where does that leave you?’

It is up to the patient to decide what needs to be done

Summarise, agree on a plan Aim to get commitment to a plan which might range from action to change, an agreement to return to discuss further or an agreement for the issue to be raised again later

Briefer: the following questions can be used within a 1–2 minute discussion‘Onascaleof1–10where10isalot,howmuch do you want to... (make the change)?’

The patient will usually pick a number higher than 2. Regardless of the answer, you can usually ask the next question. If the answer is 2 you can ask, ‘Why so high, why is it not 1?’

‘Why so high?’ The patient then tells you why (s)he wants to change, ie. argues for change ‘So what do you want to do about it?’ The patient then states how to move toward change. Again aim to get a commitment to a

plan which might range from action to change, an agreement to return to discuss further or an agreement for the issue to be raised again later

Figure 2. Encouraging change talk: stages and tools18

Strengthening commitment

Exploring ambivalenceBuilding motivation

harder ‘uphill’ workWorking against the status quoAmplifying ambivalenceCare not to increase resistance

GraceGenerate a gapRoll with resistanceAvoid argumentsCan doExpress empathy

OarSOpen ended questionsAffirmationsReflectionsSummaries

easier ‘downhill’ rideElicit self motivating statements, benefits and plans for change, commitments and concrete plans

encourage ‘change talk’Statement reflecting desire to, ability to, reasons for,

need for and commitment to change

988 reprinted from auStralian FaMily phySician Vol. 38, No. 12, December 2009

Continued from previous page

30 |

Page 209 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nly

Stepping Up Study

| 31

Appendix

Motivational Interviewing Resources

influencing behaviour change in general practice – part 2 – motivational interviewing approaches clinical practice

12. Festinger L. A theory of cognitive dissonance. Stanford, CA: Stanford University Press, 1957.

13. BanduraA.Self-efficacymechanisminhumanagency.AmPsychol1982;37:122–47.14. Amrhein PC, Miller WR, Yahne CE, Palmer M, Fulcher L. Client commitment lan-

guage during motivational interviewing drug use outcomes. J Consult Clin Psychol 2003;71:862–78.

15. Clark MD, Walters S, Gingerich R, Meltzer M. Motivational interviewing for proba-tionofficers:Tippingthebalancetowardschange.FederalProbation2006;70:38–44.

16. Hettema J, Steele J, Miller WR. Motivational interviewing. Ann Rev Clin Psychol 2005;1:91–111.

17. Bundy C. Changing behaviour: Using motivational interviewing techniques. J R Soc Med2004;44(Suppl):43–7.

18. National Institute of Corrections. Justice system assessment and training. 2006. Accession number 019791. Available at www.nicic.org/Library/019791 [Accessed 30 April 2009].

help to achieve this goal.17 The goal is to build self efficacy, strengthen commitment and to support. Behaviour change takes time and people may move backward and forward before achieving longer term change.

conclusionGeneral practitioners are in a strong position to make a difference to population and individual health outcomes in Australia as they provide continuing primary health care. Each GP develops a personal communication style, which is effective in changing behaviour in a proportion of patients, but having a range of tools such as MI helps achieve greater change.

resourceThe October 2009 issue of The Royal Australian College of General Practitioners’ check Program, contains case studies that illustrate the practi-cal application and complements the theoretical discussion in this series of articles. Available at www.racgp.org.au/check. Conflict of interest: none declared.

references1. Miller WR, Rollnick S. Ten things that motivational interviewing is not. Behav Cogn

Psychother2009;37:129–40.2. Rogers C. The necessary and sufficient conditions of therapeutic personality change.

JConsultPsychol1957;21:95–103.3. Miller WR, Rollnick S. Motivational interviewing: Preparing people for change, 2nd

edn. New York: The Guilford Press, 2002.4. Miller WR, Benefield RG, Tonigan JS. Enhancing motivation for change in problem

drinking: A controlled comparison of two therapist styles. J Consult Clin Pyschol 2003;61:455–61.

5. Rollnick S, Miller WR, Butler CC. Motivational interviewing in health care: Helping patients change behaviour. New York: Guilford Press, 2008.

6. Heather N, Rollnick S, Bell A, Richmond R. Effects of brief counseling among heavy drinkersidentifiedongeneralhospitalwards.DrugAlcoholRev1996;15:29–38.

7. Rogers RW. A protection motivation theory of fear appeals and attitude change. J Psychol1975;91:93–114.

8. Floyd DL, Prentice-Dunn S, Rogers RW. A meta-analysis of research on protection motivationtheory.JApplSocPsychol2000;30:407–29.

9. Bem DJ. An experimental analysis of self-persuasion. J Exp Soc Psychol 1965;1:199–218.

10. Kraybill A. Motivational interviewing: Applications for PATH service providers [an edited transcript of the PATH national teleconference call]. Available at http://path-program.samhsa.gov/pdf/Transcript_Motiv_Inter_2_11.pdf.

11. Miller WR. Motivational enhancement therapy with drug abusers. Albuquerque: The University of New Mexico, 1995.

case study Dr M is running late, trying to get on to her next appointment. Her exiting patient, Kyle, hesitates with his hand on the doorknob, and says, ‘I know you’ve been on at me forever but I think I need to stop smoking’. [Heartsink – ‘I do need to deal with this now, but I don’t have time’.] Dr M says, ‘Tell me the most important reason you need to stop smoking now’. When Kyle answers, she notes these (now understanding his internal motivation), agrees they are very important reasons (reinforcing his motivations) and says she’d like to work with him on this. She asks, ‘Are you willing to commit to coming back to talk to me next week to set up a Quit plan?’ (builds commitment). That conversation took less than 1 minute and leaves the patient with enhanced motivation and a commitment to return to discuss the change plan.Note: Providing the patient with a decisional balance sheet to consider at home can help some people to see their dilemma more clearly but the decisional balance technique differs from MI in that it gives equal weight to pros and cons, while MI deliberately aims to influence the direction of change by strengthening internal motivation for change and avoiding reinforcing reasons against change.1

CORRESPONDENCE [email protected]

reprinted from auStralian FaMily phySician Vol. 38, No. 12, December 2009 989

Continued from previous page

Page 210 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyS

tep

pin

g U

p S

tud

yPage 211 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review O

nlyAppendix 5: Details about insulin regimens

Based on data for 254 patients at 12 months

1. Rapid acting insulin: 1 patient in control group (0.9%) and 17 patients in intervention group

(11.6%); chi test p=0.001

2. Intermediate acting insulin: 1 patient in control group (0.9%) and 2 patients in intervention

group (1.4%). Chi test; p=0.75.

3. Long acting basal insulin (glargine): 23 (21.3%) in the control group and 97 (66.4%) in the

intervention group. Chi test; p<0.0001.

4. Mixed insulin: 0 patients in the control group, 3 (2.1%) in the intervention group. Chi test;

p=0.13.

Rapid Intermediate Long Mixed

Control 1 1 23 0

Intervention 17 1 97 3

Based on survey data:

• The median (IQR) total insulin dose in the control group was 16 (13, 35) units. Data available for

21 people

• The median (IQR) total insulin dose in the intervention group was 34 (20, 50) units. Data

available for 101 people.

This is significantly different (Wilcoxon rank sum test; p<0.0001)

Data on differences in the use of non-insulin agents are included as Table 3

No significant differences are seen in the use of intermediate and mixed insulin. While a greater

proportion of intervention patients did progress to a rapid acting insulin, this is an outcome of the

model of care intervention, and the change in the primary outcome reflects the effect of the model

of care rather than simply a difference in insulin regime used.

Page 212 of 212

https://mc.manuscriptcentral.com/bmj

BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Recommended