November 2019 – APSAD Conference, Hobart, Tasmania
Emma Black1,2,3, Rachel Deacon1,2, Llewellyn Mills1,2, Adrian J Dunlop4,5, Nadine Ezard3,6,7, Raimondo Bruno3,8, Anthony Shakeshaft3, Michael Farrell3, Jennifer Holmes9, Michelle Cretikos9, Mark Montebello2,3,10, David Reid11,13, Steven Childs12, Krista Siefried6,7, Kristie Mammen1 and Nicholas Lintzeris1,2,13
1 Drug and Alcohol (D&A) Services, South Eastern Sydney Local Health District (LHD), Sydney, Australia, 2 Discipline of Addiction Medicine, Sydney University, Sydney, Australia, 3 Faculty of Medicine, University of NSW, Sydney, Australia, 4 D&A Clinical Services, Hunter New England LHD, Newcastle, Australia, 5 School of Medicine and Public Health, University of Newcastle, Newcastle, Australia, 6 D&A Services, St Vincent’s Hospital, Sydney, Australia, 7 National Centre for Clinical Research into Emerging Drugs, Sydney, Australia, 8 School of Medicine, University of Tasmania, Hobart, Australia, 9 Centre for Population Health, NSW Ministry of Health, Sydney, Australia, 10 D&A Services, North Sydney LHD, Sydney, Australia, 11 D&A Services, Illawarra and Shoalhaven LHD, Wollongong, Australia, 12 D&A Services, Central Coast LHD, Gosford, Australia, 13 NSW Drug and Alcohol Clinical Research and Improvement Network (DACRIN)
Characteristics of the drug treatment population in New South Wales –focus on amphetamine type substances (ATS)A COQI Project
Acknowledgements
• Clients of participating treatment services• Staff across participating NSW Local Health Districts (LHDs),
including:• Directors and Managers of Drug and Alcohol Services, • Data Managers and Custodians• Clinicians• Project Officers, Research Officers & Administrative Staff• Research Governance and Ethics teams
• The funder: NCCRED• The COQI and MA Data Project Teams
Credit: cliparting.com
Background – what makes this project unique?
• ~2016 Introduction of EMR to AODTS: CHOC
• Presents a new research opportunity to inform treatment
• Enables us to ↑ understanding of real-life large-scale clinical data
• across client populations • at a point in time, and• over time
• Focus on public outpatient treatment • excludes detox, NGOs, private providers
CHOC data includes
• Pre-existing NSW MDS-DATS, such as • age• sex (binary)• principal drug of concern (main drug people are seeking treatment for)• main treatment type (e.g. counselling, case management, OST)
• Addition of Australian Treatment Outcomes Profile (ATOP) Ryan et al. 2014 – past 28 day
• substance use (alcohol, heroin, other opioids, cannabis, amphetamines, cocaine, benzodiazepines, tobacco, injecting)
• life situation & stressors (days work/study, homelessness/risk of eviction, caring for/living with children, arrest, violence to self/others)
• health and wellbeing (self rated psychological wellbeing, physical wellbeing, overall quality of life)
Background – why amphetamine-type substances?
• 2nd most common drug of concern in AOD services, after alcohol AIHW, 2018
• However, limited understanding of these clients as a group Bartu et al, 2004; McKetin et al., 2018
• Gaps in population level knowledge:• characteristics of people who use ATS in AOD services • their participation in health services • AOD treatment outcomes
• Topic of ongoing media and political interest• we need evidence to inform these discussions!
• Improving our understanding of this important and diverse group of people will enable us to better meet their treatment needs.
Today’s aims:
1. Provide preliminary example of the type of work that can be done
2. Describe characteristics of people who use ATS in the NSW public outpatient AOD treatment population• Currently gathering data that includes follow ups – will enable us
to look at outcomes
Specific question:
Is there a difference in the health and wellbeing of clients who have recently used ATS at entry to treatment compared to clients who have not?
Wellbeing at treatment entry
N=3,031 outpatient clients across 4 NSW Local Health Districts, Jan-Dec 2017
Wellbeing: self-ratings of physical health, psychological health & quality of life (good/poor)Image credit: www.bandt.com.au
Client descriptors (N=3,031, calendar year 2017)
Not used ATSin past 28d
N=2,388
Used ATS in past 28dN=643
Sig. Total clientsN=3,031
DemographicsAge: mean years (SD)
39 (12.3) 35 (9.7) p<0.001 38 (11.9)
% Male 68% 69% ns 68%Principal Drug of ConcernAlcohol 53% 11%
p<0.001
44%ATS 8% 49% 16%Cannabis 17% 11% 16%Opioids 22% 29% 24%
Client descriptors (N=3,031, calendar year 2017)
Not used ATSin past 28d
N=2,388
Used ATS in past 28dN=643
Sig. Total clientsN=3,031
DemographicsAge: mean years (SD)
39 (12.3) 35 (9.7) p<0.001 38 (11.9)
% Male 68% 69% ns 68%Principal Drug of ConcernAlcohol 53% 11%
p<0.001
44%ATS 8% 49% 16%Cannabis 17% 11% 16%Opioids 22% 29% 24%
At Assessment: days of ATS use, past 28 days
2388
14186 35 43 21 21 14 40 12 8 10 38 2 18 6 17 6 3 1 7 8 2 5 9 6 6 3 750
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Freq
uenc
y
Number of days used in past 28
No recent ATS use N=2388 Any recent ATS use N=643
All clients at assessment: life situation, stressors & substance use, past 28 days
17%
15%
37%
15%
21%
61%
11%
8%
11%
13%
31%
injected
non-OST opioid use
cannabis use
benzo use
ATS use
alcohol use
violence (self/other)
arrested
living with child <5y
housing stress
work/study
At Assessment: life situation and stressors, past 28 days
33%
10%12%
7%9%
24% 24%
7%
12%
17%
work/study housing stress living with child <5y arrested experiencedviolence
(self/other)
Not used ATS in past 28d Used ATS in past 28d
**** **** **** **** ****
**** p<0.001
At Assessment: substance use & injecting, past 28 days
63%
31%
12% 13%
2%6%
1%
52%
61%
28%23%
6%
58%
10%
Alcohol Cannabis Non-OSTopioids
Benzodiazepines Cocaine Injected Sharedequipment
Not used ATS in past 28d Used ATS in past 28d
**** **** **** **** **** **** ****
**** p<0.001
At Assessment: Clients reporting poor health & wellbeing
48%
38%42%
62%
45%
58%
Psychological health Physical health Quality of life
Not used ATS in past 28d Used ATS in past 28d
**** *** ****
*** p ≤ 0.001 *** p ≤ 0.0001
Analysis – binary logistic regressions
• Is amphetamine use itself significantly associated with poorer health and wellbeing at treatment entry?
• or is it better explained by other factors (e.g. age, sex, housing stress, violence, other substance use)?
• Ran 3 separate binary logistic regressions looking at:1. Psychological health2. Physical health3. Quality of life
• Full multivariate regression• Sensitivity analyses (backwards & forwards stepwise
regressions) yielded similar results
Factors associated with poor psychological health at treatment entry
0 1 2 3 4
injectedopioidscocaine
cannabisbenzos
ATSalcohol
violence (self/other)arrest
housing stresslive with child under 5y
work/studysex (male)
age✱✱✱✱
✱✱
✱✱✱✱
✱✱
✱
Predictors of poor physical health at treatment entry
0 1 2 3 4
injectedopioidscocaine
cannabisbenzos
ATSalcohol
violence (self/other)arrest
housing stresslive with child under 5y
work/studysex (male)
age✱✱✱✱
✱✱✱✱
✱
✱
✱✱
✱
Predictors of poor quality of life at treatment entry
0 1 2 3 4
injectedopioidscocaine
cannabisbenzos
ATSalcohol
violence (self/other)arrest
housing stresslive with child under 5y
work/studysex (male)
age ✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱
Regression SummaryPoor psych health Poor physical
healthPoor quality of
lifeOlder age **** ***
Sex (male) ****(women rate more poorly)
****(women rate more poorly)
****(women rate more poorly)
Work/study ****(not working = poorer)
Live with child under 5y
Housing stress ** ****Arrest
Violence (self/other) **** * ****Alcohol use **** **** ****ATS use **** * ***Benzodiazepine use **** ** ***Cannabis use *Cocaine use
Non-OST Opioid use ** *Injected
Take home messages
• Important to consider days of use, not just PDOC• ↑ ATOP completions more informed service provision hopefully better outcomes for clients!
• Huge potential for use of EMR data to inform services• See COQI symposium on Wednesday morning for more info
on this!
• ATS use at entry to treatment is associated with poorer self-ratings of health and wellbeing at that time, even when other factors accounted for
What next?
Treatment outcomes: do ATS use play a role?
• Similar, but larger, dataset• 6 LHDs• People entering AOD treatment mid-2016-mid 2019• Includes assessment and follow up ATOPs
• Combined dataset in preparation• Focus on OST
• Large % of Tx population; CHOC inc ATOP now permits us to look at use of drugs that are not principal drug of concern & outcomes (e.g. change in drug use; health & wellbeing)
Preliminary data: n=544 people entering OST with a follow up (3-9mths later), 3 LHDs
22%
27%
28%
16%
19%
7%
5%
9%
13%
20%
40%
39%
47%
23%
43%
6%
4%
8%
22%
12%
0% 10% 20% 30% 40% 50%
poor quality of life
poor phys health
poor psych health
ATS use
non-OST opioid use
violence (self/other)
arrested
living with child <5y
housing stress
work/study
Assessment Follow-up