The women who died 2009-12
Marian Knight
Maternal Lead – MBRRACE-UK
Maternal
Mortality in the UK
2009-12
2006-08
1952-54
90 per 100,000
maternities
11 per 100,000
maternities
10 per 100,000
maternities
New activities
• Confidential case reviews of selected maternal
morbidity cases as well as all deaths up to one
year after delivery
• Annual reports for both maternal and
stillborn/infant programmes
• Maternal Death Enquiry Ireland (MDE Ireland)
cases from 2009 onwards included with the UK
confidential enquiries
New work – maternal morbidity
• New morbidity topic selected annually
• Confidential enquiry of a sample of
approximately 30-40 cases nationally
• Cases can be identified through a variety of
sources depending on the topic
• 2014 morbidity enquiry – sepsis
Call for topic proposals open to 31 December
Maternal Morbidity and Mortality
Annual Report Topics
• Year 1 (This report): Sepsis, haemorrhage,
AFE, anaesthetic, neurological, respiratory,
endocrine and other indirect
• Year 2 (2015): Psychiatric, thrombosis,
malignancy, late and coincidental
• Year 3 (2016): Pre-eclampsia and eclampsia,
cardiac, early pregnancy
MBRRACE-UK: METHODS
Reporting of maternal deaths - UK
• Data on most 2009 cases collected through
CMACE
• Cases from 2010 and early 2011 reported to
CMACE
• Cases from 2011 and 2012 reported through
MPMN portal and national offices in Scotland
and Northern Ireland
• Cases from mid-2012 in England and Wales
reported to MBRRACE-UK
Additional identification
• Other reports e.g. coroners, procurators
fiscal, LSAMOs, media
• Women with pregnancy-related conditions
listed as the cause of death identified from
death registration data
• Birth registration data also linked with data
on deaths of all women of reproductive
age and any additional maternal deaths
identified
MBRRACE–UK
Methods
Confidential Enquiry Assessors
•
•
•
•
•
•
•
•
•
•
•
•
15 Obstetricians
16 Anaesthetists
3 Obstetric Physicians
4 Cardiologists
2 Neurologists
15 Midwives
8 GPs
7 Intensive care consultants
8 Pathologists
12 Psychiatrists
8 Infectious disease physicians
1 Emergency medicine consultant
Sepsis Confidential Enquiry
•
•
•
•
Topic Expert Group convened
Key standards identified
All maternal deaths included
34 morbidity cases selected
– Stratified random sample of women with septic shock,
identified from UKOSS severe sepsis study
• Medical records obtained and local clinician
reports requested
• Care assessed using standard MBRRACE-UK
methodology
Maternal Deaths - Definitions
• Direct: As a consequence of a disorder specific to
pregnancy
– E.g. Haemorrhage, pre-eclampsia, genital tract sepsis
• Indirect: Deaths resulting from previous existing
disease, or diseases that developed during pregnancy,
and which were not due to direct obstetric causes but
aggravated by pregnancy
– E.g. Cardiac disease, other infections (sepsis)
• Coincidental: Incidental/accidental deaths not due to
pregnancy or aggravated by pregnancy
– E.g. Road traffic accident
• Late: Deaths occurring more than 42 days but
less than one year after the end of pregnancy
Presentation of results
• Previous reports included data collected over
three years
• Surveillance data in this report covers four years
(2009-12)
• For comparison, most figures are presented
either for 2009-11 alone, or for both 2009-11
and 2010-12
• Note that figures for 2009-11 and 2010-12
therefore include some of the same women
(those who died in 2010 and 2011)
MBRRACE-UK:
SURVEILLANCE RESULTS
Notifications for 2009-12
• 265 deaths notified through CMACE (complete data on
only 91)
• 154 deaths notified through MPMN portal
• 50 deaths notified direct to MBRRACE-UK
• 44 additional deaths identified through case checking
directly with units
• 263 additional deaths identified through linkage with
national death/birth reports – mostly late deaths
• Total:
– 357 deaths in pregnancy or up to 42 days postpartum
– 419 late deaths (42-364 days after the end of pregnancy)
The women who died
• 357 women died during pregnancy or up
to 42 days postpartum in 2009-12
• 36 women’s deaths were classified as
coincidental
• Thus there were a total of 321 women who
died
– 253/2379014 in 2009-11 (10.63/100,000)
– 243/2401624 in 2010-12 (10.12/100,000)
Maternal death rate 2003-12
(Three year rolling averages)
16
14
Rate per 100 000 maternities
12
10
8
6
4
27% reduction in maternal death rate, p<0.001
2
0
2004
2005
2006
2007
2008
Mid-year of each three-year period
2009
2010
2011
Maternal death rate 2003-12
(Three year rolling averages)
16
14
Direct and Indirect maternal death rate
Rate per 100 000 maternities
12
10
Indirect maternal death rate
8
6
Direct maternal death rate
4
2
0
2004
2005
2006
2007
2008
Mid-year of each three-year period
2009
2010
2011
Maternal death rate 2003-12
(Three year rolling averages)
16
14
Direct and Indirect maternal death rate
Rate per 100 000 maternities
12
10
Indirect maternal death rate
8
6
Direct maternal death rate
4
2
48% reduction in direct maternal death rate, p<0.001
No significant decrease in indirect maternal deaths, p=0.73
0
2004
2005
2006
2007
2008
Mid-year of each three-year period
2009
2010
2011
Maternal death rate 1985-2011
(Three year periods)
16
Rate per 100 000 maternities
14
12
10
8
6
4
2
0
1985-1987 1988-1990 1991-1993 1994-1996 1997-1999 2000-2002 2003-2005 2006-2008 2009-2011
Triennium
Direct maternal deaths 1985-2011
(Three year periods)
10
9
Rate per 100 000 maternities
8
7
6
5
4
3
2
1
0
1985-1987 1988-1990 1991-1993 1994-1996 1997-1999 2000-2002 2003-2005 2006-2008 2009-2011
Triennium
Indirect maternal deaths 1985-2011
(Three year periods)
10
9
Rate per 100 000 maternities
8
7
6
5
4
3
2
1
0
1985-1987 1988-1990 1991-1993 1994-1996 1997-1999 2000-2002 2003-2005 2006-2008 2009-2011
Triennium
International comparisons
• MBRRACE-UK
– 10.1 per 100,000 maternities (95%CI 8.9-11.5) for
2010-12
• MDE Ireland
– 8.6 per 100,000 maternities (95%CI 5.2-16.6) for
2009-11
• France Confidential Enquiry
– 10.3 per 100,000 live births (95%CI 9.1–11.6) for
2007–09
• Netherlands Confidential Enquiry
– 5.0 per 100,000 live births (95%CI 3.5-6.9) for
2009-12
Maternal mortality in the
Netherlands
Rate per 100,000 live births
12
10
8
6
Direct
Indirect
4
2
0
93-'95
96-'98
99-'02
03-'05
Triennium
06-'08
Source: Dutch Maternal Mortality Committee 2014
09-'12
Maternal mortality in the UK and
the Netherlands
8
7
Rate per 100,000
6
5
Direct
Indirect
4
3
2
1
0
Netherlands 2009-12
UK 2010-12
Sources: Dutch Maternal Mortality Committee 2014, MBRRACE-UK 2014
Maternal mortality in the UK,
France and the Netherlands
8
7
Rate per 100,000
6
5
Direct
Indirect
4
3
2
1
0
Netherlands 2009-12
France 2007-09
UK 2010-12
Sources: Dutch Maternal Mortality Committee 2014, MBRRACE-UK 2014,
CNEMM 2013
International comparative
maternal mortality ratio (MMR)*
Triennium
1985-87
No. of deaths
identified
through death
certificates
174
1988-90
MMR
95% CI
Denominator number
of live births
7.67
6.61-8.90
2,268,766
171
7.24
6.24-8.42
2,360,309
1991-93
150
6.48
5.52-7.60
2,315,204
1994-96
158
7.19
6.15-8.40
2,197,640
1997-99
128
6.03
5.70-7.17
2,123,614
2000-02
136
6.81
5.76-8.05
1,997,472
2003-05
149
7.05
6.00-8.27
2,114,004
2006-08
155
6.76
5.78-7.92
2,291,493
2009-11
134
5.57
4.67-6.60
2,405,251
*Deaths identified solely from death certificates, live births used as denominator
Causes of maternal death
Rate per 100,000 maternities
2.50
2.00
1.50
1.00
0.50
0.00
Solid bars show indirect causes, hatched bars show direct causes
Causes of maternal death
Causes of direct maternal death
• Thrombosis and thromboembolism is now the
leading cause of direct maternal death
• Significant decrease in the maternal mortality
rate from pre-eclampsia and eclampsia – now
the lowest ever rate
– 0.38 per 100,000 maternities
• The mortality rate from genital tract sepsis has
more than halved between 2006-8 and 2010-12;
a statistically significant decrease
– RR 0.44; 95% CI 0.22-0.87, p=0.016
Maternal deaths from genital
tract sepsis
1.80
1.60
Rate per 100,000 maternities
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
1985-87
1988-90
1991-93
1994-96
1997-99
Triennium
2000-02
Rate in 2010-12: 0.50 per 100,000 (95%CI 0.26-0.87)
2003-05
2006-08
2009-11
Sepsis – all causes
Influenza
Causes of indirect maternal
death
• Major contribution from influenza and
other non-genital tract sepsis deaths
• Overall rate of maternal mortality from
infectious causes in 2009-12 was 2.0 per
100,000 maternities (95%CI 1.6-2.6)
• Cardiac disease remains the largest single
cause of indirect maternal deaths; rate
more than doubled since 1985-87
(RR 2.2, 95%CI 1.4-3.6)
Psychiatric deaths
• Rate of psychiatric deaths appears low
• However, this includes only maternal
deaths up to 42 days postpartum
• Psychiatric causes make a significant
contribution to late deaths:
– 95 of 419 late maternal deaths between 200912 were due to psychiatric causes
• Confidential Enquiry into psychiatric
deaths will be included in the 2015 report
Late deaths
• Note that the numbers of late deaths
reported by MBRRACE-UK are not
comparable with those reported by
CMACE due to different methodology
• Deaths counted in CMACE 2006-8 report
were only those which underwent
confidential enquiry; deaths counted in
MBRRACE-UK 2009-12 report are all late
deaths
Maternal mortality rate by age
Age
<20
20 – 24
25 – 29
30 – 34
35 – 39
≥ 40
Rate per
100,000
maternities
95% CI
Relative
risk
(RR)
95% CI
8.7
6.9
8.3
9.6
15.2
22.7
4.9-14.3
4.9-9.3
6.5-10.4
7.7-11.9
12.0-19.0
15.1-32.8
1.26
1
1.21
1.40
2.22*
3.30*
0.65-2.33
*Significantly raised compared to women aged 20-24
0. 81-1.81
0.96-2.09
1.50-3.32
1.96-5.47
Maternal mortality rate by area
of residence (IMD quintile)
IMD Quintiles (England only)
I (Least deprived/
highest 20%)
II
III
IV
V (Most deprived/
lowest 20%)
Rate per
100,000
maternities
95% CI
Relative
risk (RR)
95% CI
8.2
5.6 to 11.6
1
-
8.2
8.9
11.0
5.6 to 11.4
6.4 to 12.0
8.5 to 14.0
1.00
1.09
1.34
0.60 to 1.67
0.67 to 1.77
0.87 to 2.12
12.1
9.7 to 14.9
1.48*
1.00 to 2.29
*Significantly raised compared to women in least deprived areas
Maternal mortality rate by ethnic
group
Rate per
100,000
maternities
95% CI
Relative
risk (RR)
95% CI
White (inc. not known)
9.0
7.8 -10.4
1
Indian
20.5
11.9-32.8
2.27*
1.30-3.74
Pakistani
13.9
7.8-22.8
1.53
0.84-2.60
Bangladeshi
11.1
3.0-28.4
1.23
0.33-3.20
Other Asian
8.1
2.9-17.6
0.90
0.32-1.99
Caribbean
18.5
6.0-43.2
2.05
0.66-4.87
African
26.9
17.6-39.4
2.98*
1.90-4.51
Others/ mixed
10.2
5.6-17.1
1.13
0.61-1.94
Ethnicity (England only)
*Significantly raised compared to white women
Maternal mortality rate
according to country of birth
Woman's
country of
birth
UK
Outside UK
Rate per
100,000
maternities
8.6
95% CI
Relative risk
(RR)
95% CI
7.5 to 9.8
1
-
15.2
12.5 to 18.3
1.77*
1.39 to 2.24
Bangladesh
India
Pakistan
Sri Lanka
Ghana
Nigeria
Somalia
Poland
9.0
14.5
10.9
29.4
22.0
34.2
17.8
8.9
1.9 to 26.3
6.3 to 28.6
4.7 to 21.4
8.0 to 75.1
4.5 to 64.2
16.4 to 62.9
4.8 to 45.5
3.6 to 18.3
1.05
1.69
1.27
3.42
2.56
3.99*
2.07
1.03
0.21 to 3.11
0.72 to 3.39
0.54 to 2.54
0.92 to 8.89
0.52 to 7.59
1.88 to 7.48
0.56 to 5.38
0.41 to 2.17
*Significantly raised compared to women born in the UK
Other characteristics of women
who died
Medical condition/
characteristic
Body mass index
(BMI) kg/m2
<18
18 – 24
25 – 29
≥ 30
Missing
Mental health
problems
Yes
No
Any pre-existing
medical condition
(excluding obesity)
Yes
No
Missing
Direct (n=106)
Frequency (%)
Indirect (n=215)
Frequency (%)
Total (n=321)
Frequency (%)
1 (0.9)
35 (33.0)
28 (26.4)
31 (29.3)
11 (10.4)
5 (2.3)
89 (41.4)
44 (20.5)
56 (26.0)
21 (9.8)
6 (1.9)
124 (38.6)
72 (22.4)
87 (27.1)
31 (10.0)
12 (11.3)
87 (82.1)
42 (19.5)
165 (76.7)
54 (16.8)
252 (78.5)
74 (69.8)
25 (23.6)
7 (6.6)
163 (75.8)
44 (20.5)
8 (3.7)
237 (73.8)
69 (21.5)
15 (4.7)
Other characteristics of women
who died
Medical condition/
characteristic
Body mass index
(BMI) kg/m2
<18
18 – 24
25 – 29
≥ 30
Missing
Mental health
problems
Yes
No
Any pre-existing
medical condition
(excluding obesity)
Yes
No
Missing
Direct (n=106)
Frequency (%)
Indirect (n=215)
Frequency (%)
Total (n=321)
Frequency (%)
1 (0.9)
35 (33.0)
28 (26.4)
31 (29.3)
11 (10.4)
5 (2.3)
89 (41.4)
44 (20.5)
56 (26.0)
21 (9.8)
6 (1.9)
124 (38.6)
72 (22.4)
87 (27.1)
31 (10.0)
12 (11.3)
87 (82.1)
42 (19.5)
165 (76.7)
54 (16.8)
252 (78.5)
74 (69.8)
25 (23.6)
7 (6.6)
163 (75.8)
44 (20.5)
8 (3.7)
237 (73.8)
69 (21.5)
15 (4.7)
Antenatal care
• Only 29% of women who had antenatal care
received the recommended level of care
according to NICE antenatal care guidelines
(booking at 10 weeks or less and no routine
antenatal visits missed);
• Almost two thirds received a minimum level of
antenatal care (booking at less than 12 weeks
and three or fewer antenatal visits missed);
• 25% did not receive even this minimum level of
care.
The women who survived:
septic shock associated with
pregnancy
• Between June 2011 and May 2012, 71 women
were reported to UKOSS who had septic shock
in association with pregnancy
• 9.1 cases per 100,000 maternities (95%CI 7.111.5)
• 69 women survived; the care of 34 of these
women was examined in the confidential enquiry
Acosta, Kurinczuk et al. PLoS Med 2014
Selected characteristics of women
who survived septic shock
Characteristic
Age
<25
25 – 34
≥ 35
Ethnicity
White European
Other
Body mass index (BMI) kg/m2
<18
18 – 24
25 – 29
≥ 30
Missing
Any pre-existing medical condition (excl. obesity)
Yes
No
Total (n=69)
Frequency (%)
16 (23)
33 (48)
20 (29)
44 (64)
25 (36)
1 (1)
38 (55)
18 (26)
9 (13)
3 (4)
18 (26)
51 (74)
RESULTS: CONFIDENTIAL
ENQUIRIES
Case assessments completed
Assessor type
Number of reviews
completed as first
assessor
Number of reviews
completed as second
assessor
Obstetrician
237
52
Midwife
Anaesthetist
237
237
51
30
Physician
89
10
GP
Intensive care
24
110
8
-
Pathologist
203
28
Psychiatrist
1
1
Infectious Disease
64
2
Emergency Medicine
17
-
Local clinicians reports
Professional group
Obstetrics
Returned N (%)
(n=203)
33 (16)
Midwifery
33 (16)
Anaesthetics/critical care
27 (13)
GP
36 (18)
Emergency care
8 (*)
Medical specialties
15 (*)
*Not required for all deaths therefore percentage not calculated
Overall assessment of care
• Good care; no improvements identified
• Improvements in care* identified which would
have made no difference to the outcome
• Improvements in care* identified which may
have made a difference to the outcome
*Improvements in care are interpreted to include adherence
to guidelines, where these exist and have not been
followed, as well as other improvements which would
normally be considered part of good care, where no formal
guidelines exist.
Overall assessment of care
Classification of care received for women who died and are included in
the confidential enquiry chapters (n=203)
Classification of care received
Number of women
(%) (n=203)
Good care
58 (29)
Improvements to care which would have
made no difference to outcome
39 (19)
Improvements to care which may have made a
difference to outcome
106 (52)
Narrative versus EvidenceBased Medicine—And, Not Or
“Facts and figures are essential, but insufficient, to translate the data
and promote the acceptance of evidence-based practices and
policies…. narratives, when compared with reporting statistical
evidence alone, can have uniquely persuasive effects in overcoming
preconceived beliefs.
Stories help the public make sense of population-based evidence.
Guideline developers and regulatory scientists must recognize, adapt,
and deploy narrative to explain the science of guidelines to patients and
families, health care professionals, and policy makers to promote their
optimal understanding, uptake, and use.”
Meisel and Karlawish JAMA. 2011;306(18):2022-2023.
SUMMARY
Key messages - 1
• Overall there has been a statistically significant
decrease in the maternal death rate between
2006-8 and 2009-12 in the UK.
• This decrease is predominantly due to a decrease
in direct maternal deaths.
• There has been no significant change in the rate
of indirect maternal death over the last 10 years,
a time during which direct maternal deaths have
halved. This needs action across a wide range of
health services and not just maternity services.
Key messages - 2
• The number of women dying from genital tract
sepsis has significantly decreased since 2006-8.
• Influenza was an important cause of death
during this period; half of the associated deaths
occurred after a vaccine became available and
can therefore be considered preventable.
• The importance of seasonal influenza
immunisation for pregnant women cannot be
over-emphasised; increasing immunisation rates
in pregnancy against seasonal influenza must
remain a priority.