Directorate of Governance & Assurance, December 2017 Page 1 of 18
NLG (18) 023
DATE OF MEETING 30 January 2018
REPORT FOR Trust Board of Directors – Public
REPORT FROM Dr Kate Wood, Acting Medical Director
CONTACT OFFICER Jeremy Daws, Head of Quality Assurance
SUBJECT Monthly Mortality Report
BACKGROUND DOCUMENT (IF ANY) Annual Quality Account
PURPOSE OF THE PAPER: For Information
EXECUTIVE SUMMARY (PLEASE INCLUDE A BRIEF SUMMARY OF THE PAPER, KEY POINTS & ANY RISK ISSUES AND MITIGATING ACTIONS WHERE APPROPRIATE)
Key Points identified from this report include:
Page 5 –Learning from Deaths Dashboard outlines the number of deaths and the number of those that have been reviewed using the Trust’s Structured Judgement Review process (SJR). This now includes an assessment of whether the death was avoidable or not, using a Likert-type 6 factor scale. This is the initial reviewer’s assessment from the retrospective assessment of the medical record. Any case reviews completed that identify that further understanding is needed is reviewed a second time by another clinician. This process links into the Trust’s Serious Incident Framework if necessary. It should be stressed that this data is not a reliable measure of deaths that were avoidable, rather it is designed as an indicator to support local review and learning processes with the aim of helping improve quality of care.
Page 7 – Crude mortality (all & non-elective) above statistically calculated limits, although has remained static at 1.6% (all deaths) and 3.8% (non-elective) during November 2017.
Page 7 – Higher SHMI and HSMR at DPoW, slight reduction noted during summer months (from July). Gradual increase seen at SGH.
Page 8 – Significant increase seen in the number of non-elective patients in DPoW with a zero length of stay, since the introduction of DPoW Ambulatory Care model. A reduction also seen in the average length of stay on AMU over the same time frame, reducing to an average of 1 day LOS. Non-elective LOS also reducing over the DPoW Medicine specialty from a peak of 10 days in May to 7.
Page 8 – DPoW medical outliers shows an increase in October/November to 12% compared to 1-2% at SGH.
Page 9 – Continued reduction in the number of patients waiting more than 4/12 hours in A&E at DPoW, both sites achieving 90% compliance with 4 hour target during November.
Page 13/14 – Official SHMI of 114, +5 points from previous quarterly release, higher than expected, next official SHMI released in March 2018.
Page 13 – ‘Provisional’ SHMI of 114 versus previous 106 (previous 12 months).
Page 13 – HSMR increased from 105 to 115 (+10 points).
Page 16 – Change of episode on the same ward within 24 hours of admission elevated in Medicine
Directorate of Governance & Assurance, December 2017 Page 2 of 18
HAVE STAFF SIDE BEEN CONSULTED ON THE PROPOSALS? NOT APPLICABLE
HAVE THE RELEVANT SERVICE USERS/CARERS BEEN CONSULTED ON THE PROPOSALS?
NOT APPLICABLE
ARE THERE ANY FINANCIAL CONSEQUENCES ARISING FROM THE RECOMMENDATIONS?
NO
IF YES, HAVE THESE BEEN AGREED WITH THE RELEVANT BUDGET HOLDER AND DIRECTOR OF FINANCE, AND HAVE ANY FUNDING ISSUES BEEN RESOLVED?
NOT APPLICABLE
ARE THERE ANY LEGAL IMPLICATIONS ARISING FROM THIS PAPER THAT THE BOARD NEED TO BE MADE AWARE OF?
NO
WHERE RELEVANT, HAS PROPER CONSIDERATION BEEN GIVEN TO THE NHS CONSTITUTION IN ANY DECISIONS OR ACTIONS PROPOSED?
YES
WHERE RELEVANT, HAS PROPER CONSIDERATION BEEN GIVEN TO SUSTAINABILITY IMPLICATIONS (QUALITY & FINANCIAL) & CLIMATE CHANGE?
YES
THE PROPOSALS OR ARRANGEMNTS OUTLINED IN THIS PAPER SUPPORT THE ACHIEVEMENT OF THE TRUST OBJECTIVE(S)
YES
THE PROPOSAL OR ARRANGEMENTS OUTLINED IN THIS PAPER ENSURE COMPLIANCE WITH THE REGULATORY OR GOVERNANCE REQUIREMENTS LISTED
YES
THE PROPOSALS OR ARRAGEMENTS OUTLINED IN THIS PAPER TAKE ACCOUNT OF REQUIREMENTS IN RESPECT OF EQUALITY & DIVERSITY
YES
ACTION REQUIRED BY THE BOARD
The Board is asked to note the contents of the Mortality Report and the key observations drawn from the data:
Key observations:
(1) The Trust’s mortality rates (crude, non-elective crude, SHMI and HSMR) all show increases over the last 12 months, DPoW being higher than SGH.
(2) Patient access and flow measures demonstrate that DPoW management arrangements have been different to those at SGH although some significant improvements noted in the last 3 months, since introduction of DPoW Ambulatory Care.
(3) Note the inclusion of the Learning from Deaths Dashboard.
The Board (and the Mortality Assurance & Clinical Improvement Committee) is asked to note the recommendations on page 4.
Directorate of Governance & Assurance, December 2017 Page 3 of 18
Directorate of Governance & Assurance
Board Report – Mortality Summary
December 2017
Contents
1.0 Introduction
4
2.0 Board Action 4
3.0 Recommendations 4
At a Glance – Learning from Deaths Dashboard 5
At a Glance – Mortality Dashboard 6
4.0 Mortality Indicators
10
5.0 Clinical Coding Indicators 15
6.0 Glossary 17
Directorate of Governance & Assurance, December 2017 Page 4 of 18
1.0 INTRODUCTION The monthly mortality report seeks to provide an update on the most recent information available and support the focus of reducing the Trust’s current mortality ratio.
2.0 BOARD ACTION
The Board are asked to note the key points outlined on page 1, forming part of the Board’s Front sheet,
Note the inclusion of the Learning from Deaths Dashboard and the breakdown of the reviews completed regarding whether the death was felt to have been avoidable or not. The Board should be aware that this is based on a retrospective assessment from the medical record and based on the reviewer’s views of what has been recorded. This should be interpreted with caution. This links to the Secretary of State for Health’s ambitions to publish avoidable deaths data. In previous month’s mortality reports, the avoidability data has not been reported due to the concerns that this retrospective review of the case is not a reliable measure of determining if a death was avoidable or not, and because this question was originally included as part of the methodology to support learning from case reviews with the intention of improving local care quality, not to act as a performance indicator. Work is underway to determine from other Trusts the approach taken to reporting of this data.
3.0 RECOMMENDATIONS
There are no new recommendations.
Directorate of Governance & Assurance, December 2017 Page 5 of 18
Northern Lincolnshire and Goole NHS Foundation Trust: Learning from Deaths Dashboard - November 2017-18
Time Series: Start date 2016-17 Q1 End date 2017-18 Q4
This Month This Month This Month
122 22 0
This Quarter (QTD) This Quarter (QTD) This Quarter (QTD)
275 32 0
This Year (YTD) This Year (YTD) This Year (YTD) 2
1056 128 1
Score 5
Slight evidence of avoidability Definitely not avoidable
This Month 0 0.0% This Month 0 0.0% This Month 0 0.0% This Month 0 0.0% This Month 0 0.0% This Month 0 0.0%7
This Quarter (QTD) 0 0.0% This Quarter (QTD) 0 0.0% This Quarter (QTD) 0 0.0% This Quarter (QTD) 0 0.0% This Quarter (QTD) 0 0.0% This Quarter (QTD) 37 100.0%-
This Year (YTD) 0 0.0% This Year (YTD) 0 0.0% This Year (YTD) 1 1.0% This Year (YTD) 2 1.9% This Year (YTD) 1 1.0% This Year (YTD) 99 96.1%
Time Series: Start date 2016-17 Q1 End date 2017-18 Q4
This Month This Month This Month
0 0 0
This Quarter (QTD) This Quarter (QTD) This Quarter (QTD)
0 0 0
This Year (YTD) This Year (YTD) This Year (YTD)
4 0 0
Description:
The suggested dashboard is a tool to aid the systematic recording of deaths and learning from care provided by NHS Trusts. Trusts are encouraged to use this to record relevant incidents of mortality, number of deaths reviewed and cases from which lessons can be
learnt to improve care.
Summary of total number of deaths and total number of cases reviewed under the Structured Judgement Review Methodology
18 0 0
Summary of total number of learning disability deaths and total number reviewed under the LeDeR methodology
2 0 0
Last Year Last Year Last Year
0 0 0
Last Quarter
Last Year Last Year
Last Quarter Last Quarter
Total Number of Deaths in scope Total Deaths Reviewed Through the
LeDeR Methodology (or equivalent)
Total Number of deaths considered to
have been potentially avoidable
Last Month Last Month Last Month
Total Number of Deaths, Deaths Reviewed and Deaths Deemed Avoidable for patients with
identified learning disabilities
Total Deaths Reviewed
Total Deaths Reviewed by RCP Methodology Score
Definitely avoidable Strong evidence of avoidability Probably avoidable (more than 50:50) Probably avoidable but not very likely
1647 150 -
Score 1 Score 2 Score 3 Score 4 Score 6
Last Quarter
Avoidable deaths data: This is an assessment of whether the death was felt to be avoidable or not, using a Likert-type 6 factor scale. This is the initial reviewer’s assessment from the retrospective assessment of the medical record. Any case reviews completed that
identify that further understanding is needed is reviewed a second time by another clinician. This process links into the Trust’s Serious Incident Framework if necessary. It should be stressed that this data is not a reliable measure of deaths that were avoidable,
rather it is designed as an indicator to support local review and learning processes with the aim of helping improve quality of care.
Total Number of Deaths, Deaths Reviewed and Deaths Deemed Avoidable (does not include
patients with identified learning disabilities or patients from A & E)
153 10 0
Last Quarter Last Quarter
Total Number of Deaths in Scope
Total Number of deaths considered to
have been potentially avoidable
(RCP<=3)
Last Month Last Month Last Month
348 53 0
Last Year 0
100
200
300
400
500
600
Q1 2016-17 Q2 Q3 Q4 Q1 2017-18 Q2 Q3 Q4
Mortality over time, total deaths reviewed and deaths considered to have been potentially avoidable(Note: Changes in recording or review practice may make comparison over time invalid) Total
deaths
Deathsreviewed
Deathsconsidered
likely tohave beenavoidable
0
1
2
3
4
5
6
7
8
Q1 2016-17 Q2 Q3 Q4 Q1 2017-18 Q2 Q3 Q4
Mortality over time, total deaths reviewed and deaths considered to have been potentially avoidable(Note: Changes in recording or review practice may make comparison over time invalid)
Totaldeaths
Deathsreviewed
Deathsconsidered
likely tohave beenavoidable
Directorate of Governance & Assurance, December 2017 Page 6 of 18
Mortality Dashboard
(year to Nov-2017) (year to Jun-2017) (year to Aug-2017) (year to Sep-2017) (year to Sep-2017)
(year to Nov-2016) (year to Mar-2017) (year to Aug-2016) (year to Sep-2016) (year to Sep-2016)
In and Out of Hospital HED SHMI - moving annual totals
(year to Aug-2017) 116 v 139 97 v 132 149 v 57
(year to Aug-2016) 107 v 151 93 v 130 55 v 71
Key:
HED SHMI banding status: "Higher than Expected" Top Six Diagnosis Groups: Crude Mortality Trending last 24 months (Dec-2015 to Nov-2017)
Key: Monthly Rate Linear Monthly Rate Mean UCL LCL
(year to Aug-2017)
(year to Aug-2016)
Non Elective In Hospital Crude Mortality by day of death/discharge
(year to Aug-2017)
(year to Aug-2016)
Produced by Information Services, December 2017 *Septicaemia is a sub-diagnosis of the Infection diagnosis group
HED
SHMI
109 112 117
102 101 116
HED
SHMI
113 104 116
108 142 108
105
99 v 140
Full SHMI In Hospital SHMI Out Hospital National Average
1.50% 1,590 106,055 114 106
Trust in/out
SHMI: 106 v 135 Grimsby
in/out SHMI:
Scunthorpe
in/out SHMI:
Goole in/out
SHMI:
NB We discharge fewer non-elective patients at weekends whilst the number of deaths is
fairly consistent across the week - this contributes to higher crude mortality rate at weekends
RAMI1.58% 1,696 107,063 119 114 115 99
In-Hospital Crude Rate Deaths Discharges National SHMI HED SHMI HSMR
986,
055
6,85
8
6,84
0
7,09
7
7,79
7
4,79
5
3,84
2
241259 246 227
255
235209
0.0%
1.0%
2.0%
3.0%
4.0%
5.0%
6.0%
0
2,000
4,000
6,000
8,000
10,000
12,000
Mon Tue Wed Thu Fri Sat Sun
Died In Hospital
Other Discharges
Died In Hospital Rate
Directorate of Governance & Assurance, December 2017 Page 7 of 18
Mortality A3 Dashboard - December 2017: Focus on Flow, Admissions, Discharges, Staffing and Mortality Outcomes
* The Trust's crude mortality rate is increasing overall, but saw a decrease in the latest month. * The Trust's non-elective crude mortality is driving the Trust's increased mortality rate, and is significantly higher than peer.
* The crude mortality has been above the upper control limit since January 2017. * When broken down to site level, a steeper increasing trend is seen on the DPoW site.
* The Trust's SHMI was showing an increasing trend and is currently 'higher than expected', but was decreased in the latest month. * The Trust's HSMR also shows an increasing trend and continues to be 'higher than expected', but saw a drop in the latest ocuple of months.
* The increase in Trust SHMI was mainly driven by the SHMI at DPoW, which is currently outside expected limits despite a drop in the latest month. * The HSMR at DPoW is increasing, significantly above that of SGH, and DPoW is currently outside expected limits despite a drop in the last few months.
* The SGH SHMI trend is also increasing (but to a lesser extent than DPOW) and is currently also just outside the expected limits. * The SGH HSMR performance continues to be just within the expected limits, but an increasing trend is seen.
* During the peak months of pressure seen during the early part of 2017, DPoW mortality rate peaked above 2%. * During same time frame, SGH mortality rate peaked at 2% with the number of deaths in January peaking at 80.
* The no. of discharges increased and the mortality rate decreased in March 2017 in line when the Trust had a discharge team over a weekend.
* The no. of discharges increased and the mortality rate decreased in March 2017 in line when the Trust had a discharge team over a weekend.
* During December 2016 and January 2017, the number of deaths reached 95 and 103. During the same months there weren't many more
additional patients discharged compared to other recent months.
Produced by Information Services
1.0%
1.1%
1.2%
1.3%
1.4%
1.5%
1.6%
1.7%
1.8%
Apr
May Jun
Jul
Aug Sep Oct
No
v
De
c
Jan
Feb
Ma
r
Apr
May Jun
Jul
Aug Sep Oct
No
v
Dec Ja
n
Feb
Ma
r
Apr
May Jun
Jul
Aug Sep Oct
No
v
Dec Ja
n
Feb
Ma
r
Apr
May Jun
Jul
Aug Sep Oct
No
v
2014 2015 2016 2017
Die
d in
Hos
pita
l Rat
e
NLAG Moving Annual Total (MAT) Crude Mortality vs Peer Group
NLAG Mean UCL LCL Peer Average National Average
4,2
75
4,3
98
4,4
10
4,7
11
4,2
49
4,3
66
4,7
48
4,3
08
4,2
22
4,3
26
4,0
66
4,5
65
4,3
94
4,3
00
4,4
98
4,6
03
4,3
43
4,3
15
4,5
95
4,1
35
4,3
13
4,1
67
4,2
30
4,2
45
4,1
78
4,3
36
4,4
09
4,2
92
4,2
93
4,1
29
4,1
13
4,0
93
4,1
01
4,3
32
4,2
27
4,9
19
4,0
25
4,3
11
4,5
25
4,3
34
4,3
17
4,4
22
4,7
02
4,6
47
63 55 7558
60 64
51
65 69 10470
7571 66
49 5376 67
67
6869
76 55 75 6853 52 59 63
69 64 60 95103 77
72
8477
6155 53 67
75 61
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
0
1,000
2,000
3,000
4,000
5,000
6,000
Apr
May Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Ap
r
May Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Aug Sep
Oct
Nov
2014 2015 2016 2017
Die
d In
Hos
pita
l Rat
e
No.
Dis
char
ges
DPOW Died in Hospital Rate
Other Discharges Died In Hospital Died In Hospital Rate
3,7
83
3,9
47
4,0
08
4,1
59
3,6
26
3,9
06
4,2
60
3,8
01
4,1
25
3,8
70
3,7
07
4,0
43
3,7
97
3,7
96
3,9
61
4,1
31
3,6
52
3,8
95
4,1
48
3,9
52
4,0
39
3,9
23
4,0
25
4,0
31
3,8
46
3,9
22
4,1
54
3,8
41
3,9
99
3,8
87
3,9
47
4,0
57
3,8
35
3,8
56
3,6
87
4,1
54
3,7
62
3,8
64
3,9
88
3,9
67
3,8
64
3,9
26
4,2
33
4,1
03
7251 54
45
7054
56
6582
8967
7557 79
5943
5156
5653 70 68 73 85
76 6063
5755 55 54 73
65 8072
66
69 64 65 64 40 5673 59
0.00%
0.50%
1.00%
1.50%
2.00%
2.50%
3.00%
0
1,000
2,000
3,000
4,000
5,000
6,000
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Ap
r
May Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Aug Sep
Oct
Nov
2014 2015 2016 2017
Die
d in
Hos
pita
l Rat
e
No.
Dis
char
ges
SGH Died In Hospital Rate
Other Discharges Died In Hospital Died In Hospital Rate
2.6%
2.8%
3.0%
3.2%
3.4%
3.6%
3.8%
4.0%
Apr
May Jun
Jul
Aug Sep Oct
No
v
De
c
Jan
Feb
Mar
Ap
r
May Jun
Jul
Aug Sep Oct
No
v
De
c
Jan
Feb
Ma
r
Apr
May Jun
Jul
Aug Sep Oct
No
v
Dec Ja
n
Feb
Ma
r
Apr
May Jun
Jul
Aug Sep Oct
No
v
2014 2015 2016 2017
Die
d in
Hos
pita
l Rat
e
Non Elective Mortality: NLAG Moving Annual Total (MAT) Crude Non Elective Mortality vs Peer Group
NLAG Mean UCL LCL Peer Average National Average
Directorate of Governance & Assurance, December 2017 Page 8 of 18
* When comparing the two sites admissions units, note the similar number of discharges, especially during late 16, early 17.
* Note the difference in LOS, with SGH averaging 0.5 days LOS compared to DPOW 1.5 days LOS, although DPOW LOS decreased in the last few months.
* Note the seeming impact of the Ambulatory Care / FEAST initiative at SGH.
* The numbers of zero LOS patients at DPoW is increasing since the introduction of Ambulatory Care in Sep-2017.
* DPoW Medicine non-elective stay was ranging from 8-10 days, but has decreased in the last few months. * SGH Medicine non-elective LOS, ranging from 4-6 days, is notably shorter than DPoW non-elective LOS.
* The died in hospital rate for DPoW non-elective medical patients ranges between 6-10% compared to SGH rate of around 4-6%.
* DPoW medical outliers are considerably more than those on the SGH site. * SGH medical outliers are notably lower than DPOW, peaking during January 2017 at approx 50 or 4%.
* At DPoW, during the pressures during late 2016/early 2017, the number of medical outliers peaked at nearly 250 per month or 15%.
* Note the apparent difference between the two sites in terms of the numbers of patients discharged on the same day as admission - or in other
words, those patients with a ZERO length of stay.
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
0.0
2.0
4.0
6.0
8.0
10.0
12.0
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
2014 2015 2016 2017
Die
d In
Hos
pita
l Rat
e
Av
era
ge
LO
S (d
ay
s)
DPOW Medicine Non Elective Average Length of Stay
Medicine Non Elective Average Spell LOS Medicine Non Elective Died In Hospital Rate
DPOW Ambulatory Care implemented
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
0.0
2.0
4.0
6.0
8.0
10.0
12.0
Apr
May Jun
Jul
Aug Sep
Oct
Nov
De
c
Jan
Feb
Mar
Apr
May Jun
Jul
Aug Sep Oct
Nov
De
c
Jan
Feb
Mar
Apr
May Jun
Jul
Aug Sep Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Aug Sep Oct
Nov
2014 2015 2016 2017
Die
d In
Hos
pita
l Rat
e
Ave
rage
LO
S (d
ays)
SGH Medicine Non Elective Average Length of Stay
Medicine Non Elective Average Spell LOS Medicine Non Elective Died In Hospital Rate
SGH Ambulatory Care and FEAST implemented
0.0%
4.0%
8.0%
12.0%
16.0%
20.0%
0
50
100
150
200
250
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Ma
r
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Ma
r
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Ma
r
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
2014 2015 2016 2017
Med
ical
Out
lier
Rat
e
No.
Med
ical
Out
liers
DPOW Medical Outliers
Medical Outliers Medical Outliers Rate
0.0%
4.0%
8.0%
12.0%
16.0%
20.0%
0
50
100
150
200
250
Ap
r
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Ap
r
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Ap
r
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Ap
r
May Jun
Jul
Au
g
Sep
Oct
Nov
2014 2015 2016 2017
Med
ical
Out
lier
Rat
e
Med
ical
Out
liers
SGH Medical Outliers
Medical Outliers Medical Outliers Rate
0
50
100
150
200
250
300
350
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Ap
r
May Jun
Jul
Au
g
Sep
Oct
Nov
2014 2015 2016 2017
No
. Ze
ro L
oS
NLAG Medicine Non Elective Zero Length of Stay (did not stay overnight)
DPOW SGH
SGH Ambulatory Care and FEAST implemented
DPOW Ambulatory Care implemented
0.0
1.0
2.0
3.0
0
100
200
300
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
2014 2015 2016 2017
Ave
rage
Spe
ll Lo
S (d
ays)
No
. D
isch
arg
es
DPOW AMU Discharges and Average Spell LoS
Discharges Average Spell LoS
0.0
1.0
2.0
3.0
0
100
200
300
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
Dec Ja
n
Feb
Mar
Apr
May Jun
Jul
Au
g
Sep
Oct
Nov
2014 2015 2016 2017
Ave
rage
Spe
ll Lo
S (d
ays)
No.
Dis
char
ges
SGH MAU Discharges and Average Spell LoS
Discharges Average Spell LoS
Directorate of Governance & Assurance, December 2017 Page 9 of 18
* A&E 12 hour waits increased considerably during the pressure seen during late 2016/early 2017. * Both sites, as nationally, have struggled to achieve the 4 hour target (95% national, 90% local target) but this was achieved in the latest month.
* Whilst the number of 12 hour waits reduced at SGH, these continued on the DPoW site. * Patient flow will have had an impact on performance.
* There is a known correlation between excessive waits in A&E and mortality indices.
0
20
40
60
80
100
120A
pr
Ma
y
Jun
Jul
Aug Sep Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep Oct
Nov
Dec Ja
n
Feb
Mar
Apr
Ma
y
Jun
Jul
Aug Sep Oct
Nov
2014 2015 2016 2017
No.
12
Hou
r W
aits
A&E Twelve Hour Waits (Arrival to Departure)
01 DPW 02 SGH
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
Apr
May Jun
Jul
Aug Sep Oct
No
v
De
c
Jan
Feb
Mar
Apr
May Jun
Jul
Aug Sep Oct
No
v
De
c
Jan
Feb
Mar
Apr
May Jun
Jul
Aug Sep Oct
No
v
De
c
Jan
Feb
Mar
Apr
May Jun
Jul
Aug Sep
Oct
No
v
2014 2015 2016 2017
4 H
our
Com
plia
nce
Rat
e
No.
4 H
our
Bre
ache
s
A&E Four Hour Breaches
01 DPW - 4 Hour Breaches 02 SGH - 4 Hour Breaches 01 DPW - 4 Hour Compliance Rate 02 SGH - 4 Hour Compliance Rate
Directorate of Governance & Assurance, December 2017 Page 10 of 18
1.0 Introduction
2.0 Board Action
3.0 Key Points
At a Glance – Mortality Dashboard
At a Glance – Flow & Staffing Dashboard
This section…
4.0 Mortality Indicators
4.1 Crude Mortality Overview
4.2 Crude Mortality by Diagnosis Groupings
4.3 Standardised Mortality Indicators Overview
4.4 Nationally Published SHMI
5.0 Clinical Coding Indicators
6.0 Glossary
Directorate of Governance & Assurance, December 2017 Page 11 of 18
4.0 MORTALITY INDICATORS
The following section of the Trust’s Mortality Report is compiled by Information Services. It contains high level analysis of NLAG’s crude mortality, Summary Hospital Level Mortality Indicator (SHMI), Hospital Standardised Mortality Ratio (HSMR) and Risk Adjusted Mortality Index (RAMI)
4.1 Crude Mortality Overview 4.1.1 In-Hospital Crude Mortality Indicators Dashboard
The following dashboard looks at our crude mortality indicators at Trust and site level.
Source: Information Services / CHKS
.
.
Dec-2016 to
Nov-2017Prev 12 mths
Annual
ChangePeer
Compared to
Peer
CRUDE MORTALITY
Trust 1.58% 1.50% 0.08% 0.12%
DPOW 1.64% 1.48% 0.16% 0.18%
SGH 1.59% 1.63% -0.04% 0.13%
GDH 0.81% 0.63% 0.18% -0.65%
Trust 3.72% 3.37% 0.35% 0.82%
DPOW 4.02% 3.41% 0.61% 1.12%
SGH 3.32% 3.23% 0.09% 0.42%
GDH 11.32% 10.08% 1.24% 8.42%
Trust 1,696 1,590 106
DPOW 880 763 117
SGH 773 789 -16
GDH 43 38 5
M2Non Elective Crude
Mortality Rate2.90%
Indicator
M1 Crude Mortality Rate 1.46%
M3 Number of Deaths n/a n/a
Directorate of Governance & Assurance, December 2017 Page 12 of 18
4.2 Crude Mortality by Diagnosis Groupings
4.2.1 Patients Died In hospital by Primary Diagnosis Summary Group
The following table shows the number of deaths in each diagnosis group (with the Respiratory group also shown split by site). The diagnosis group is based on the primary diagnosis at the time of death.
Source: Information Services
Comment: The latest month of November 2017 had 31 fewer deaths than the previous month and 19 fewer deaths than the previous twelve months’ average. Respiratory, Infection, Cardiology, General Surgery, Gastroenterology and Stroke are the six diagnosis areas with the higher number of deaths.
2016 2017Grand
Total
Primary Diag Summary Group Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov
Respiratory 59 74 55 66 61 52 34 44 32 32 42 29 580 -13 -19
DPOW 36 43 27 39 24 29 10 18 16 12 14 13 281 -1 -10
SGH 22 28 26 27 36 20 22 25 14 17 27 16 280 -11 -7
GDH 1 3 2 0 1 3 2 1 2 3 1 0 19 -1 -2
Infection 16 22 15 10 26 23 19 18 15 28 38 28 258 -10 7
Cardiology 22 22 20 13 14 21 13 12 12 13 20 13 195 -7 -3
General Surgery 13 12 7 12 9 8 13 9 10 10 10 9 122 -1 -1
Gastroenterology 14 12 17 8 6 6 10 10 6 9 11 9 118 -2 -1
Stroke 11 8 6 6 11 7 9 2 7 3 13 14 97 1 6
Renal 5 10 7 2 11 8 8 11 3 8 6 5 84 -1 -2
Trauma and Orthopaedics 7 9 7 9 9 3 6 5 4 7 2 6 74 4 0
Miscellaneous 3 5 5 4 5 3 3 0 1 2 3 4 38 1 1
Haematology 1 3 3 1 2 2 2 4 3 3 4 1 29 -3 -1
Vascular 2 2 1 1 1 4 4 1 2 4 1 0 23 -1 -2
Neurological 2 2 3 2 2 1 3 1 0 2 2 2 22 0 0
Urinary Tract 0 3 1 0 0 2 2 1 1 2 0 0 12 0 -1
Diabetes and Endocrine 1 0 4 0 0 1 2 1 0 1 0 1 11 1 0
Psychological 2 0 0 1 0 0 0 2 0 3 0 1 9 1 0
Gynaecology 1 0 0 0 0 3 0 0 1 0 1 0 6 -1 -1
Rheumatoid 1 0 1 1 0 1 1 1 0 0 0 0 6 0 -1
ENT 0 0 0 2 0 1 0 0 0 1 0 0 4 0 0
DVT/PE 2 1 0 0 1 0 0 0 0 0 0 0 4 0 0
Neonatal 0 1 0 2 0 0 0 0 0 1 0 0 4 0 0
Grand Total 162 186 152 140 158 146 129 122 97 129 153 122 1696 -31 -19
Latest Month
Vs. Previous
Month
Latest Month
Vs. 12
Months Avg
12 Months' Trend
Directorate of Governance & Assurance, December 2017 Page 13 of 18
4.3 Standardised Mortality Indicators
The following section provides high level analysis of NLAG’s performance in three of the key mortality indicators - Summary Hospital Level Mortality Indicator (SHMI), Hospital Standardised Mortality Ratio (HSMR), and Risk Adjusted Mortality Index (RAMI).
4.3.1 Summary Dashboard
Source: Information Services / NHS Digital / HED / CHKS
Jul16-Jun17
SHMI
Apr16-Mar17
SHMI
Quarterly
Change
National
PeerBanding
Nationally Published SHMI
Sep16 -
Aug17Prev 12 mths
Annual
Change
National
Peer
Compared to
Peer
HED SHMI
Trust 114 106 8 14
DPOW 123 113 10 23
SGH 107 98 9 7
GDH 121 12 109 21
Trust 106 99 7 6
DPOW 116 107 9 16
SGH 97 93 4 -3
GDH 149 55 94 49
Trust 135 140 -5 35
DPOW 139 151 -12 39
SGH 132 130 2 32
GDH 57 71 -14 -43
Oct16 -
Sep17Prev 12 mths
Annual
Change
National
Peer
Compared to
Peer
HSMR
Oct16 -
Sep17Prev 12 mths
Annual
Change
National
Peer
Compared to
Peer
RAMI
Indicator
100Higher than
Expected
Indicator
M6a Provisional SHMI 100
M5
Summary Hospital
Level Mortality
Indicator (SHMI)
Trust 119 114 5
Indicator
M7
Hospital
Standardised
Mortality Ratio
(HSMR)
Trust 115 105 10
Indicator
M8
Risk Adjusted
Mortality Index
(RAMI) - All
Conditions
Trust 99 98
100 15
1 90 9
M6bIn Hospital
Provisional SHMI100
M6cOut of Hospital
Provisional SHMI100
Directorate of Governance & Assurance, December 2017 Page 14 of 18
4.4 Nationally Published SHMI
The Summary Hospital-level Mortality Indicator (SHMI) is the nationally agreed mortality indicator which reports mortality at trust level across the NHS (acute care trusts only) in England using standard and transparent methodology. This indicator is produced and published officially by NHS Digital.
It is a ratio of the observed deaths in a trust over a period of time divided by the expected number given the characteristics of patients treated by that trust. It includes deaths occuring up to 30 days post discharge and excludes day cases. The national average SHMI is 100.
4.4.1 Latest Nationally Published SHMI Mortality Position
The most recent Summary Hospital Level Mortality Indicator (SHMI) was published in December 2017 and covers the July 2016 – June 2017 time period. The Trust’s SHMI score was 119 and this is in the “higher than expected” range. This is an increase from the Trust’s previous SHMI score of 114 which was also in the “higher than expected” range.
4.4.2 NLAG’s SHMI in National Context
The following chart illustrates the Trust’s most recent SHMI position in relation to all acute Trusts nationally.
Source: Information Services / NHS Digital
4.4.3 Deaths Occurring In and Out of Hospital
NLAG had 66.7% of SHMI deaths occurring in hospital – an increase from our rate of 65.5% in the previous SHMI release and lower than the national rate of 71.1%. The SHMI indicator is not solely a hospital-based mortality indicator, but is influenced by wider community-based healthcare also.
Directorate of Governance & Assurance, December 2017 Page 15 of 18
1.0 Introduction
2.0 Board Action
3.0 Key Points
At a Glance – Mortality Dashboard
At a Glance – Flow & Staffing Dashboard
4.0 Mortality Indicators
This section…
5.0 CLINICAL CODING INDICATORS
5.1 Clinical Coding Dashboard
6.0 Glossary
Directorate of Governance & Assurance, December 2017 Page 16 of 18
Clinical Coding Dashboard
(year to Nov-2017) (year to Nov-2017) (year to Nov-2017) (year to Nov-2017) (year to Nov-2017)
(year to Nov-2016) (year to Nov-2016) (year to Nov-2016) (year to Nov-2016) (year to Nov-2016)
Produced by Information Services, December 2017
Change Episode Same WardDepth of Coding No. Co-Morbidity Codes
106998
100163
5.2
Palliative Care Episodes
5.0
1.0%
1.1%
1.2%
1.2%
9.1%
9.0%
R Code Admissions
Definition: percentage of first episodes where there is a second
episode started on the same ward within the first 24 hours.
The closing and opening of episodes for administrative reasons may
affect the recording and coding of the fullest diagnoses in the first
episode - which the SHMI score which looks at for the primary
diagnosis and co-morbidities.
Definition: percentage of episodes with palliative care code.
The recording and coding of palliative care for appropriate patients
will exclude these patients from the RAMI indicator. The code is
also used to adjust the HSMR statistic, however the SHMI indicator
makes no adjustment for palliative care.
Definition: total number of co-morbidity codes.
The recording and coding of co-morbidities affects the risk given to
the patient in the SHMI model. If co-morbidities are not recorded
this could be reducing the “expected deaths” and potentially raising
our SHMI score (will also affect RAMI and HSMR).
Definition: average number of diagnosis codes per episode.
A high depth of coding reflects a wide source of clinical information
recorded in case-notes and may be an advantage in relation to
mortality indicators such as SHMI, RAMI and HSMR as it helps to
accurately reflect the total number of “expected deaths”.
Definition: percentage of admissions with R code as primary
diagnosis - primary diagnosis of R Code in the first two episodes for
multi episode spell or in the only episode for single episode spell.
Definition: percentage of all first episodes with primary diagnosis
of R code.
If a diagnosis is recorded as a query or is not specific, then this is coded as a signs and symptoms "R" code. These R codes hold a lower risk
in the SHMI model which reduces the "expected deaths" having the outcome of a higher SHMI score (will also affect RAMI and HSMR).
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Depth of Coding Trend - All Episodes
DPOW SGH GDH Peer
0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
Percentage of Episodes with a Z515 Palliative Care Code -Trust v Peer
Trust Peer
0
1000
2000
3000
4000
5000
6000
Number of Co-morbidities Trend - All Episodes
DPOW SGH GDH
0%
2%
4%
6%
8%
10%
12%
14%
Percentage of Admissions with R Code Diagnosis
DPOW SGH GDH
0%
2%
4%
6%
8%
10%
12%
14%
Percentage of First Episodes with Primary Diagnosis of R
Code - Trust v Peer
NLAG Peer
Directorate of Governance & Assurance, December 2017 Page 17 of 18
6.0 Glossary ‘As Expected’ mortality: This is usually expressed as a funnel chart, using confidence intervals. Using the ‘official’ SHMI definitions, ‘as expected’ mortality is explained within the 95% confidence intervals. Non-official uses of SHMI to calculate ‘outliers’ have been used by Dr Foster within recent months, they have used the 99% confidence limits. Outside of the ‘as expected’ grouping means an organisation is either an outlier in terms of mortality performance.
Benchmark Peer Group: Calderdale and Huddersfield NHS Foundation Trust, Chesterfield & North Derbyshire Royal Hospital NHS Trust, Countess of Chester NHS Foundation Trust, County Durham and Darlington NHS Foundation Trust, Doncaster and Bassetlaw Hospitals NHS Trust, North Cumbria University Hospitals NHS Trust, North Tees & Hartlepool NHS Trust, Rotherham NHS Foundation Trust, Royal Bolton Hospital NHS Foundation Trust, The Pennine Acute Hospitals NHS Trust, University Hospitals of Morecambe Bay NHS Trust
Common Cause Variation: an inherent part of the process, stable and “in control”. We can make predictions about the future behaviour of the process within limits. When a system is stable, displaying only common cause variation, only a change in the system will have an impact.
Control Limits: indicate the range of plausible variation within a process. They provide an additional tool for detecting special cause variation. A stable process will operate within the range set by the upper and lower control limits which are determined mathematically (3 standard deviations above and below the mean). The upper control limit is displayed in blue throughout this report. The lower control limit is displayed in teal throughout this report.
Crude Mortality Rate: The crude mortality rate is based on actual numbers. Unlike the SHMI which features adjustment based on population demographics and related mortality expectations.
H.E.D. – Health Evaluation Data: The official national data publications are released quarterly, six months after the event. The Trust therefore reports its performance to its Board every month using provisional data published by the University of Birmingham through its Hospital Evaluation Data system (HED). This is normally three months behind the current position, and has been validated as virtually identical to the official published data.
Hospital Standardised Mortality Rate (HSMR): The HSMR is a method of comparing mortality levels in different years, or between different hospitals. The ratio is of observed to expected deaths, multiplied conventionally by 100. Thus, if mortality levels are higher in the population being studied than would be expected, the HSMR will be greater than 100. This methodology allows comparison between outcomes achieved in different trusts, and facilitates benchmarking. See next page for a 1 page comparison between the different SMRs.
Moving Annual Total (MAT): The most recent months performance with the previous 11 months included thus providing an annual average. This is an effective way of presenting monthly performance data in a way that reduces some of the expected variation in the system i.e. seasonal factors providing a much smoother view of performance allowing trends to be more easily discerned.
National Peer Group: All other UK NHS Trusts, enabling the Trust to benchmark itself against all other UK hospitals.
R-Code: An R-Code is defined as an unspecific diagnosis when compared with the Connecting for Health NHS Coding Guidelines. Any diagnosis prefixed with “differential diagnosis”, “possible”, “likely”, “maybe”, “suspected” or “?” will be regarded, against the Coding Rules as non-specific, or in other words, an R-Code. R-codes, because they are regarded more as signs or symptoms and not a specific diagnosis, does not result in a patient’s full risk being recorded as part of the in-hospital coding. Therefore indicators such as SHMI, that rely on calculating actual mortality vs. expected mortality, calculated from the patient’s risk factors, will inflate the mortality position due to the patient’s actual risk, being under reported. In such cases, if a patient should die, their risk of mortality ‘on paper’ will be significantly lower than their actual mortality risk.
Risk Adjusted Mortality Indicator (RAMI): This is a risk adjusted standardised mortality ratio used by CHKS software which has been purchased by the Trust to monitor and analyse it’s data. See next page for a 1 page comparison between the different SMRs.
Summary Hospital-Level Mortality Indicator (SHMI): The most recently developed mortality ratio designed to be used to allow comparison between NHS organisations. This indicator also includes mortality within 30 days post discharge, so represents in hospital and out of hospital (within 30 days) mortality. The SHMI is the NHS ‘Official’ marker of mortality and is published on a quarterly basis. Because of its inclusion of mortality data within 30 days of hospital discharge, when published, the most recent information available is quite historic, at best 6 months behind the present day. See next page for a 1 page comparison between the different SMRs.
Sigma: A sigma value is a description of how far a sample or point of data is away from its mean, expressed in standard deviations usually with the Greek letter σ or lower case s. A data point with a higher sigma value will have a higher standard deviation, meaning it is further away from the mean.
Special Cause Variation: the pattern of variation is due to irregular or unnatural causes. Unexpected or unplanned events (such as extreme weather recently experienced) can result in special cause variation. Systems which display special cause variation are said to be unstable and unpredictable. When systems display special cause variation, the process needs sorting out to stabilise it. This report includes two types of special cause variation, trends and outliers. If a trend, the process has changed in some way and we need to understand and adopt if the change is beneficial or act if the change is a deterioration. The outlier is a one-off condition which should not result in a process change. These must be understood and dealt with on their own (i.e. response to a major incident).
Standard Deviation: Standard deviation is a widely used measurement of variability or diversity used in statistics and probability theory. It shows how much variation or "dispersion" there is from the "average" (mean, or expected/budgeted value). A low standard deviation indicates that the data points tend to be very close to the mean, whereas high standard deviation indicates that the data are spread out over a large range of values.
Glo
ss
ary
Directorate of Governance & Assurance, December 2017 Page 18 of 18
Understanding Standardised Mortality Ratios – SHMI, HSMR and RAMI
SHMI HSMR RAMI
Numerator Total number of observed deaths in hospital and within 30 days of discharge from the hospital
All spells culminating in death at the end of the patient pathway, defined by specific diagnosis codes for the primary diagnosis of the spell: uses 56 diagnosis groups which contribute to approx. 80% of in hospital deaths in England
Total number of observed in hospital deaths
Denominator Expected number of deaths Expected number of deaths Expected number of deaths
Adjustments Sex
Age group
Admission method
Co-morbidities based on Charlson score
Year index
Diagnosis group
No adjustment is made for palliative care.
Details of the categories above can be referenced from the methodology specification document at http://www.ic.nhs.uk/services/summary-hospital-level-mortality-indicator-shmi
Sex
Age in bands of five up to 90+
Admission method
Source of admission
History of previous emergency admissions in last 12 months
Month of admission
Socio economic deprivation quintile (using Carstairs)
Primary diagnosis based on the clinical classification system
Diagnosis sub-group
Co-morbidities based on Charlson score
Palliative care
Year of discharge
Sex
Age
Clinical grouping (HRG)
Primary and secondary diagnosis
Primary and secondary procedures
Hospital type
Admission method
No adjustment is made for palliative care.
Further detailed methodology information is included in CHKS products, or specific enquiries to CHKS www.chks.co.uk
Exclusions Excludes specialist, community, mental health and independent sector hospitals; Stillbirths, Day cases, regular day and night attenders. Palliative care patients not excluded.
Excludes regular attendees. Palliative care patients not excluded.
Excludes mental illness, obstetrics, babies born in or out of hospital, day cases, and patients admitted as emergencies with a zero length of stay discharged alive and spells coded as palliative care (Z515)
Whose data is being compared and how much data is used for comparison e.g. all trusts or certain proportion etc.
All England non-specialist acute trusts except mental health, community and independent sector hospitals via SUS/HES and linked to ONS data for out of hospital deaths. Deaths that occur within 30 days are allocated to the last hospital the patient was discharged from.
All England provider trusts via SUS/HES
UK database of Trust data and HES
Source: Information Services