Content uploaded by Chukwuemeka Nwaneri
Author content
All content in this area was uploaded by Chukwuemeka Nwaneri
Content may be subject to copyright.
http://dvd.sagepub.com/
Disease
The British Journal of Diabetes & Vascular
http://dvd.sagepub.com/content/13/4/192
The online version of this article can be found at:
DOI: 10.1177/1474651413495703
2013 13: 192 originally published online 15 July 2013British Journal of Diabetes & Vascular Disease
Chukwuemeka Nwaneri, Helen Cooper and David Bowen-Jones
meta-analysis
Mortality in type 2 diabetes mellitus: magnitude of the evidence from a systematic review and
Published by:
http://www.sagepublications.com
can be found at:The British Journal of Diabetes & Vascular DiseaseAdditional services and information for
http://dvd.sagepub.com/cgi/alertsEmail Alerts:
http://dvd.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- Jul 15, 2013OnlineFirst Version of Record
- Aug 21, 2013Version of Record >>
by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from by guest on October 11, 2013dvd.sagepub.comDownloaded from
The British Journal of
Diabetes & Vascular Disease
13(4) 192 –207
© The Author(s) 2013
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1474651413495703
dvd.sagepub.com
Introduction
Type 2 diabetes remains one of the most challenging
global health problems of our time and with increasing
prevalence the burden will escalate.
1
The disease is asso-
ciated with an increase in mortality rate. In 1983, this
rate was quoted as four to five times that in the general
population;
2
more recent studies have suggested that the
rate is twice that seen in the population without diabe-
tes.
3
The degree of diabetes-related mortality is age-
related however, so tends to be less in patients diagnosed
in their late 70s (47–51% are diagnosed aged 65 years or
more),
4
when compared with those in their mid-40s.
5
Given these differential mortality rates associated with
differential diagnostic rates, and that 50% also have evi-
dence of complications at diagnosis,
6
the fact that type 2
diabetes reduces life expectancy does not translate
directly (linearly) to an ability to predict death rates in
these patients. However, what is known is that 89% of
people diagnosed with type 2 diabetes have one or more
modifiable risk factors
7
so that the proportion of pre-
ventable deaths should not be underestimated.
8
In type 2 diabetes, macrovascular disease is the pre-
dominant cause of mortality, with CVD accounting for
52–80% of deaths, followed by renal disease (heralded by
albuminuria) 10–20% of mortality,
9
and cerebrovascular
disease 15%, which is approximately twice that seen in
the population without diabetes in the first five years
following diagnosis.
10
Various researchers have also
Mortality in type 2 diabetes mellitus: magnitude
of the evidence from a systematic review and
meta-analysis
Chukwuemeka Nwaneri,
1,2
Helen Cooper
1,3
and David Bowen-Jones
2
Abstract
Background: Type 2 diabetes is associated with an increase in age-related mortality. A systematic review and meta-
analysis were performed to define the relative risks (RR) of all-cause or cause-specific mortality in type 2 diabetes and
to determine gaps in current research.
Methods: A comprehensive literature search was undertaken for studies (published 1990–2010) on mortality in type
2 diabetes. The study reports on the measure of mortality as defined by RR for all-cause and cause-specific mortality,
heterogeneity, sensitivity analyses and biases.
Results: In total 35 studies (220,689 patients; mean follow-up of 10.7 years) were eligible for inclusion: 33 studies
reported increased mortality risks; 24 had full data on 95% confidence intervals (CIs), one study reported no excess
mortality in men diagnosed after 65 years whereas three reported increased mortality in similar age groups in both
sexes. Meta-analysis showed RR = 1.85 (95% CI 1.79–1.92) for all-cause mortality [men RR=1.57 (95% CI 1.46–1.68);
women RR=2.0 (95% CI 1.89–2.12)], RR=1.76 (95% CI 1.66–1.88) for cardiovascular mortality and RR=2.26 (95% CI:
1.7-3.02) for stroke. There was no statistically significant evidence of publication bias.
Conclusion: Type 2 diabetes increases mortality approximately two-fold increase and macrovascular disease is the
principal cause of death.
Keywords
meta-analysis; mortality rate; odds ratio; relative risk; systematic review; type 2 diabetes
1
Department of Community Health & Wellbeing, Faculty of Health &
Social Care, University of Chester, UK
2
Department of Medicine, Centre for Diabetes & Endocrinology, Wirral
University Teaching Hospital NHS Foundation Trust, Arrowe Park
Hospital, Upton, Wirral, UK
3
Research and Development, Alder Hey Children’s NHS Foundation
Trust, Eaton Road, Liverpool, UK
Corresponding author:
Dr. Chukwuemeka Nwaneri, Department of Community Health &
Wellbeing, Faculty of Health & Social Care, University of Chester,
Riverside, Castle Drive, Chester, CH1 1SL, UK.
E-mail: c.nwaneri@chester.ac.uk
495703
DVD13410.1177/1474651413495703e British Journal of Diabetes & Vascular DiseaseNwaneri et al.
2013
Current Topics
Nwaneri et al. 193
documented increased relative mortality risk in type 2
diabetes associated with cancers, and more recently there
has been inconclusive and contradictory evidence of the
effects of various insulin therapies or secretagogues asso-
ciated with increased malignancy risks.
11,12
Tight glycaemic regulation has been shown to be diffi-
cult to maintain in type 2 diabetes. Evidence suggests that
in itself it has little effect on mortality from cardiovascular
disease. Efforts are therefore aimed at continual optimisa-
tion of treatments for the atherogenic risk factors in CVD
including hypertension, cholesterol/hyperlipidaemia, obe-
sity, smoking, and alcohol consumption; alongside satisfac-
tory glycaemic control. However, past studies have used the
hazard ratio (which measures the rate at which events have
occurred rather than how many events have occurred) to
estimate mortality risks in type 2 diabetes, as well as there
being issues around small sample size. Given the conflict-
ing evidence concerning all-cause or cause-specific mor-
tality in type 2 diabetes when compared with the population
without diabetes, the purpose of this systematic review and
meta-analysis was to explore, review, and summarise the
evidence to clarify current knowledge and research gaps,
and to identify possible future strategies for improvement.
Methods
Sources of data
A systematic review of published and unpublished litera-
ture was undertaken including grey literature. The review
followed the guidance of the CRD (March 2001) report
number 4.
13
An electronic database search was conducted
using MeSH and search terms as shown in Table 1.
References of all retrieved articles were checked for rele-
vant studies and if needed experts were contacted for
advice, which identified additional published and
unpublished references. The process of conducting the
search was documented as it developed to ensure trans-
parency.
Study selection
Inclusion and exclusion criteria
Studies were included if they fulfilled the following crite-
ria:
• Participants: people with type 2 diabetes
• Outcomes: all-cause or overall-cause or total mor-
tality expressed as RRs or risk ratios
• Evaluation of outcomes: mortality after diagnosis
of type 2 diabetes; coronary heart disease, MI,
nephropathy, cerebrovascular disease (stroke), or
ischaemic heart disease.
• Designs: existing systematic reviews, randomised
controlled trials, cohort studies and epidemiologi-
cal studies
• Reporting: English language studies only were
included as evidence shows that excluding non-
English language studies does not change outcome
Table 1. Data source and search terms.
Data source
Cochrane Library, Centre for Reviews and Dissemination (CRD; DARE-database of abstracts of reviews of effects, HTA-health
technology assessments, and NHS EED databases), Ovid Medline/PubMed, Medical Subject Headings (MeSH), CINAHL, Web of
Science (Web of Knowledge), World Health Organisation Library and Information Network for knowledge database (WHOLIS), The
Centre for Evidenced-Based Medicine, PsycInfo, Google Scholar, EMBASE, National Library for Health, Ongoing Reviews database,
British Nursing Index and SCOPUS. Others include UK National Research Register (NRR), ReFeR, Kings Fund and Conference
Papers Index. Also grey literature search was conducted for unpublished articles using FADE, Proquest Dissertation and Theses, and
other Indexed Citations up to 2010.
Search terms
Keywords and phrases, including Medical Subject Headings (MeSH) included; ‘Type 2 diabetes mellitus mortality’, ‘Type 2 diabetes
mortality’, ‘mortality and type 2 diabetes’, ‘excess mortality and type 2 diabetes’, ‘mortality rates and type 2 diabetes’, ‘mortality and
diabetes’, ‘type II diabetes and mortality’, ‘type 2 DM and mortality’, ‘type II DM and mortality’, ‘determinants of mortality and type 2
diabetes’, ‘mortality predictors and diabetes’, and an effective combination of search terms were conducted.
Abbreviations:
AMI acute myocardial infarction
CHD coronary heart disease
CI confidence interval
CVD cardiovascular disease
CRD Centre for Reviews and Dissemination
HbA
1C
glycated haemoglobin A
1C
HDL high-density lipoprotein
LDL low-density lipoprotein
MeSH Medical Subject Headings
MI myocardial infarction
PRISMA Statement Preferred Reporting Items for
Systematic Reviews and Meta-Analyses
PVD peripheral vascular disease
RR relative risk
UKPDS United Kingdom Prospective Diabetes
Study
194 The British Journal of Diabetes & Vascular Disease 13(4)
results.
14,15
Studies also had to provide sufficient
detail including: a review of mortality indices in
type 2 diabetes and partial or complete review of
mortality in type 2 diabetes.
Studies only available as an abstract or studies that
included both type 1 and type 2 diabetes were excluded
unless specific analysis for each type was published and
the sample size for each was >100. Studies that did not
specify the type of diabetes were excluded. Studies that
did not have a comparison population were excluded.
Poor quality studies as decided by quality appraisal
13
and
studies that did not have adequate data or lacked infor-
mation necessary for the synthesis of the data were
excluded.
The search was limited to studies published between
1990 and 2010 because two systematic reviews had
already been published, one in 1983 and the other in
1999; also diabetes was undifferentiated into type 1 and
type 2 diabetes in Medline-PubMed. In addition changes
in the clinical diagnostic criteria (as defined by the World
Health Organisation and the American Diabetes
Association) occurred in 1999 making standardisation of
methods and reporting comparable in different settings.
All these affected the reporting of diabetes mortality out-
come data before 1990. Where there was an overlap in
the study population the relevant paper was included
and the others were used to provide context. Titles and
abstracts of articles were checked by two reviewers (C.N.
and H.C.). Full texts of selected studies were assessed for
inclusion by one reviewer (C.N.) and triangulated by the
other reviewers (H.C. and D.B.J.). Any variations in com-
ments were resolved through discussion.
Quality appraisal
The internal validity of individual studies was assessed
independently (C.N. and H.C.), using CRD’s guidance
checklist for quantitative research,
13
and was evaluated in
accordance with the inclusion/exclusion criteria, meth-
odological quality (that met the critical appraisal frame-
work), and relevance to the research questions. Figure 1
shows the flow chart of the study selection based on the
inclusion and exclusion criteria. To assess the quality of
the included studies we used the scoring criteria outlined
by CRD and followed the PRISMA Statement, through
the course of the study.
13,16,17
Data extraction and synthesis of evidence
Data extraction was undertaken by one reviewer (C.N.),
triangulated by the second reviewer (H.C.), and where
there was uncertainty by the third reviewer (D.B.J.).
Data from individually selected studies were extracted as
follows: author, year of publication, study design, setting/
location, year of study, comparison population, study
size, number of deaths, patients’ characteristics, follow-up
years, outcome measure of mortality, and degree of mor-
tality in RRs or odds ratio. Adjustments for confounding
in the analyses were factored in the selected papers
including age, smoking, sex, hypertension, body mass
index, previous acute MI and hypercholesterolaemia.
Meta-analysis was carried out to show precise esti-
mate of mortality risk. Sub-analyses of the effects of sex,
drugs, CVD, smoking and alcohol, age at diagnosis and
cancer on mortality risks were also assessed. Evidence
from data on mortality in type 2 diabetes was synthesised
through a descriptive epidemiological review from
included studies.
Statistical analysis
Using the MIX 2.0 Pro software package, we analysed
individual effects using RRs, displayed in the forest plot
in Figure 2. The point estimate, lower and upper limits of
95% confidence intervals were log scaled and the effects
observed in the studies were pooled to produce a
weighted average effect of all the studies (meta-analysis).
Larger studies were weighted preferentially by fixed
effects statistical models; smaller studies were weighted
preferentially by random effects models thereby enhanc-
ing the robustness of the summary effects according to
variations in statistical methods. The test for heterogene-
ity of the various studies combined was computed using
the statistical heterogeneity funnel plots, as shown in
Figure 3, Cochran’s Q score, inconsistency index (I²),
Tau-square estimate (t²) and evidence of p-value <0.01.
I
2
is the measure of quantity of amount of heterogeneity
and is deemed more reliable in assessing inconsistency
between studies, and t
2
is the measure of study variance
in random effect which will be followed if the I
2
is greater
than 50%. Cochran’s Q score is the summation of all the
z-scores of the studies. A sensitivity analysis plot and
meta-regression were used to assess the subgroups on
the effects of RR of combined studies.
Results
Search Results
An initial search resulted in 13,217 abstracts with a total
of 196 relevant non-duplicate abstracts. Papers (n=69)
were excluded if they failed to meet the inclusion criteria.
There was retrieval of 127 papers for full text searching,
out of which 83 papers were excluded due to RRs or risk
ratios not being used to measure degree of mortality. The
bibliographies of the selected papers were searched and
35 papers were finally selected that met the inclusion cri-
teria (Figure 1), adapted from the PRISMA statement.
17
Nwaneri et al. 195
A total of 35 studies satisfied the inclusion criteria
(Table 2), representing a total of 220,689 patients with a
mean follow-up of 10.7 years. Twenty of the studies were
prospective,
3
18-36, 12 studies were retrospective
37–48
,
and three were systematic reviews.
49–51
Each systematic
review had examined different parameters in different
locations. One study in the Netherlands
51
described
the relationship between blood glucose and mortality,
another from Canada
50
examined the relationship between
microalbuminuria and mortality, and the third study in
the UK
49
evaluated the mortality risks and type 2 diabe-
tes in the elderly.
Of the studies 22 included populations in Europe: United
Kingdom,
3,22,28,38,48,49
Finland,
24
Denmark,
19,23
France,
20
Switzerland,
39
Sweden,
30,31,40
Netherlands,
43,47,51
Italy,
26,45
Germany,
32
Spain
34
and Iceland.
36
Other populations
included United States of America,
18,25,29,33,35,44,46
Israel,
21
Canada,
50
New Zealand,
41
Fuji,
37
Argentina,
42
and Japan.
27
Figure 1. Flow chart of study selection for inclusion in the systematic review.
196 The British Journal of Diabetes & Vascular Disease 13(4)
Figure 3. Heterogeneity funnel plot for the included studies to detect publication bias in the meta-analysis. The individual studies
are symmetrically aligned demonstrating limited publication bias.
Figure 2. Relative risk for overall mortality in type 2 diabetes population selected. The size of the box represents the weight assigned
to individual effects (point estimate of effect). The horizontal lines indicate the lower and upper limits of 95% confidence intervals.
Nwaneri et al. 197
Table 2. Summary of 35 studies included in the systematic review.
1
st
author
(year of
publication)
Design of study Setting location
[year of study]
Comparison
population
Study size
[Number
deceased]
Patient’s
characteristics
(age range,
male–female
distribution)
Follow up
(years)
Outcome
measure of
mortality
Degree of mortality,
i.e. RR or OR (95%
CI)
Comments
Al-Delaimy
(2001)
18
Prospective Boston, USA
[1996]
With
diabetes
7401
[724]
F
30–55 years
20 years All-cause
(Ex-smoker)
All-cause
(Current
smoker)
RR: 1.31 (1.11–1.55)
RR: 1.43 (0.96–2.14)
Smoking: ↑ mortality in a dose
response manner
Almdal
(2004)
19
Prospective Copenhagen,
Denmark
[1997]
Without
diabetes
13105
[NA]
M=5907
F=7198
20 years All-cause
Acute MI (M)
Acute MI (F)
RR: 2.0 (1.7–2.7)
RR: 2.0
RR: 4.5
↑ mortality between 1.5 and
2.0, although questionnaire is
non-validated
Balkau
(1993)
20
Prospective Paris, France
[1972]
Without
diabetes
7166
[975]
M
44–55 years
15.6 years All-cause
Coronary Heart
Disease
RR: 2.31 (1.81–2.93)
RR: 2.26 (1.25–3.79)
Hazard ratio was inferred to
be the relative risk
Barnett
(2006)
49
Systematic
review
United
Kingdom
[2004]
General
population
>7268
[NA]
M & F
>60 years
7.3 years All-cause (M)
60–70 years
>70 years
All-cause (F)
60–70 years
>70 years
RR: 1.38 (1.08–1.76)
RR: 1.13 (0.88–1.45)
RR: 1.40 (1.10–1.79)
RR: 1.19 (0.93–1.52)
↑ mortality in all patients
diagnosed in elderly in 12 out
of 14 eligible studies
Behar
(1997)
21
Prospective Israel
[1983]
Without
diabetes
463
[NA]
53–74 years
M=302
F=161
10 years AMI (M; insulin)
AMI (M; non-
insulin)
AMI (F; insulin)
AMI (F; non-
insulin)
RR: 1.75 (1.26–2.45)
RR: 1.32 (1.10–1.58)
RR: 2.59 (1.89–3.56)
RR: 1.41 (1.10–1.82)
Women worse than men after
AMI, and those treated with
insulin had worst short and
long term prognosis after AMI
in both sexes
Chaturvedi
(1996)
22
Prospective London, UK
[1995]
Afro-
Caribbean &
Europeans
227
[75]
35–55 years
M=122
F=105
20 years All cause
cardiovascular
Ischaemic heart
disease
RR: 0.59 (0.32–1.10)
RR: 0.33 (0.15–0.70)
RR: 0.37 (0.16–1.85)
Afro-Caribbean with type 2
diabetes maintain a lower risk
of heart diseases. Small group
of study
Collins
(1996)
37
Retrospective Suva, Fuji
[1991]
Without
diabetes
231
[86]
M=102
F=129
11 years All-cause
(Melanesians) M
F
All-cause
(Indians) M
F
RR: 1.7 (0.9–3.1)
RR: 2.0 (1.1–3.7)
RR: 4.2 (2.7–6.5)
RR: 3.2 (1.9–5.7)
↑ mortality due to CVD and
CHD more in Indians than in
Melanesians
Croxson
(1994)
38
Retrospective Melton, United
Kingdom
[1992]
Without
diabetes
861
[95]
M & F
>65 years
4.5 years All-cause
Known + newly
diagnosed
Known diabetes
Newly diagnosed
RR: 4.5 (2.9–7.0)
RR: 5.2 (3.2–8.5)
RR: 3.0 (1.3–6.6)
Elderly diabetes show
↑ mortality risks when
compared with non-diabetes
(Continued)
198 The British Journal of Diabetes & Vascular Disease 13(4)
1
st
author
(year of
publication)
Design of study Setting location
[year of study]
Comparison
population
Study size
[Number
deceased]
Patient’s
characteristics
(age range,
male–female
distribution)
Follow up
(years)
Outcome
measure of
mortality
Degree of mortality,
i.e. RR or OR (95%
CI)
Comments
Diem
(2003)
39
Retrospective Bern,
Switzerland
[NA]
Subgroups of
cohorts
287
[70]
40–53 years
M=162
F=125
12.6 years All-cause
1–15g Alcohol/
day
16–30g Alcohol/
day
30 g Alcohol/day
RR: 1.27 (0.68–2.28)
RR:0.36 (0.09–0.99)
RR:1.66 (0.76–3.33)
Moderate alcohol
consumption in T2DM;16-30 g/
day decreased mortality from
CHD and from all cause.
Dinneen
(1997)
50
Systematic
review
Ontario,
Canada [1985]
Without
diabetes
2138
[NA]
M & F
≥52 years
6.4 years Total
Cardiovascular
OR: 2.4 (1.6–3.1)
OR: 2.0 (1.4–2.7)
Micro-albuminuria is a
predictor of total and CVD
mortality
Eliasson
(2008)
40
Retrospective Umea,
Sweden[2004]
Without
diabetes
3589
[NA]
M & F
40–84 years
5 years All-cause M
F
RR: 3.5 (2.8–4.4)
RR: 4.2 (3.3–5.4)
↑ mortality in females than in
males
Florkowski
(2001)
41
Retrospective Canterbury,
New Zealand
[1999]
Without
diabetes
447
[187]
30–82 years
M=208
F=239
10 years Cause-specific
Age
Pre-existing
CAD
Albuminuria
Hypertension
PVD
Smoking
HbA
1C
RR: 2.0 (1.6–2.5)
RR: 1.7 (1.2–2.4)
RR: 1.58 (1.1–2.3)
RR: 1.9 (1.0–3.7)
RR: 2.4 (1.3–4.5)
RR: 2.6 (1.2–5.8)
RR: 1.6 (1.1–2.3)
↑ mortality attributable to
multiple risk factors
Gagliardino
(1997)
42
Retrospective La Plata,
Argentina [NA]
Without
diabetes
1040
[NA]
≥48 years
M=856, F=184
NA AMI RR: 1.73 (1.00–2.98) ↑ mortality in type 2 diabetes
following AMI
Gall
(1995)
23
Prospective Hvidore,
Denmark
[1993]
Without
diabetes
328
[NA]
M=109
F=82
5 years All-cause
Pre-existing
CHD
Albuminuria
HbA
1C
Age
RR 2.7 (0.93–7.69)
RR: 2.9 (1.6–5.1)
RR: 1.9 (1.4–2.6)
RR: 1.2 (1.0–1.4)
RR: 1.08 (1.03–1.13)
↑ albumin excretion rate
(AER) was associated with ↑
all-cause mortality
Grauw
(1995)
43
Retrospective Netherlands
[1989]
T2DM
cohorts
265
[71]
>75 years
M=112
F=153
6.8 years All-cause
mortality
CVD Mortality
RR: 1.54 (1.07–2.23)
RR: 2.05 (1.24–3.37)
↑ mortality due to
cardiovascular death
Groeneveld
(1999)
51
Systematic
review
Netherlands
[1998]
Type 2
diabetes
cohorts
13,869
[3212]
M & F NA All-cause
mortality
RR: 1.31 (0.92–1.87) ↑ mortality in 13 out of the 23
studies
Gu
(1998)
44
Retrospective
(NHANES-1
study)
USA Without
diabetes
710
[486]
25–74 years
M=293
F=417
20 years All-cause
25-44 years
45-64 years
65-74 years
RR: 3.6 (p<0.001)
RR: 2.2 (P<0.001)
RR: 1.5 (p<0.001)
↑mortality except in those
who died from malignant
neoplasms.
Guzder
(2007)
3
Prospective Southampton,
United
Kingdom
[2003]
Without
diabetes
736
[147]
M=403
F=333
5.25 years All-cause
<60 years
60–74 years
≥75 years
OR: 2.47 (1.74–3.49)
OR: 3.20 (1.17–8.73)
OR: 3.00 (1.67–5.38)
OR: 2.00 (1.24–3.23)
For females between 60–74
years OR: 7.00 (2.09-23.47).
Too large CI. Type 2 diabetes
shows 2.5 fold ↑ in risk of
death in both sexes
(Continued)
Table 2. (Continued)
Nwaneri et al. 199
1
st
author
(year of
publication)
Design of study Setting location
[year of study]
Comparison
population
Study size
[Number
deceased]
Patient’s
characteristics
(age range,
male–female
distribution)
Follow up
(years)
Outcome
measure of
mortality
Degree of mortality,
i.e. RR or OR (95%
CI)
Comments
Hanninen
(1999)
24
Prospective Mikkeli, Finland
[1992]
Without
diabetes
252
[NA]
36–64 years
M=134
F=118
5 years CHD +
Albuminuria
RR: 3.43 (1.63–7.19) ↑ mortality worse when CHD
co-existed with albuminuria.
No association between HbA
1c
and mortality in both the
surviving and deceased cohort
Hu
(2001)
25
Prospective Maryland, USA
[1996]
Without
diabetes
121,046
[8464]
F
30–55 years
20 years All-cause
Hx type 2
diabetes, no
CHD
Hx of CHD, no
type 2 diabetes
Both type 2
diabetes & CHD
RR: 3.12 (2.83–3.44)
RR: 2.55 (2.12–3.07)
RR: 5.08 (3.47–7.43)
In the multivariate analysis;↑
mortality in women with type
2 diabetes as well as in CHD,
worse when both co-existed
Mannucci
(2004)
45
Retrospective Florence, Italy
[NA]
Type 2
diabetes
cohort
927 M=493
F=434
4.59 years All-cause
(SU+BG) F
M
RR: 2.08 (1.18–3.67)
RR: 1.68 (1.91–2.79)
Combination of SU and
BG had ↑ mortality when
compared with SU or BG
alone
Mazza
(2007)
26
Prospective
(CASTEL
study)
Rovigo, Italy
[NA]
T2DM
Cohorts
591
[369]
≥65 years
M=202, F=389
12 years CHD mortality RR: 1.76 (1.18–2.27) ↑ serum uric acid and CHD
mortality in a J-shaped pattern
Katakura
(2007)
27
Prospective Nagano, Japan
[2004]
T2DM
cohort
388
[77]
≥65 years
M=176
F=212
6 years Cause-specific
mortality
Prior stroke
Age
Serum creatinine
Current smoking
RR: 2.24 (1.33–3.77)
RR: 1.12 (1.08–1.18)
RR: 2.95 (2.15–4.03)
RR: 1.63 (1.00–2.66)
↑ mortality associated with
renal dysfunction, prior stroke,
high LDL-Cholesterol, prior
obesity and current smoking
Neil
(1993)
28
Prospective Oxford, UK
[1982]
Type 2
diabetes
cohort
246
[93]
M=125
F=121
6.1 years Cause specific
UAC (40–200
μg/L)
Severe
retinopathy
RR: 2.2 (1.3–3.7)
RR: 3.4 (1.9–6.0)
Micro-albuminuria is predictive
factor
O’Connor
(1997)
29
Prospective Navajos, USA
[NA]
Without
diabetes
144
[43]
M=77
F=77
18 years All-cause
Bivariate
Multivariate
RR: 3.12 (p<0.01)
RR: 3.02 (1.12–7.53)
↑ mortality in T2DM
Olsson
(2000)
30
Prospective Malmo,
Sweden
[1994]
Type 2
diabetes
cohorts
910
[NA]
M & F 6.1 years All-cause (Both
SU + MF)
IHD mortality
Stroke mortality
OR: 1.63 (1.27–2.09)
OR: 1.73 (1.17–2.55)
OR: 2.33 (1.17–4.63)
↑ CVS mortality in type 2
diabetes taking SU and MF in
combination than those taking
only SU
Ostgren
(2002)
31
Prospective Skara, Sweden
[2000]
Type 2
diabetes
cohorts
400
[131]
M=202
F=198
5.9 years All-cause
mortality
Age
RR: 0.80 (0.56–1.13)
RR: 1.57 (1.41–1.75)
↑ mortality:HbA
1C
, lipid,
hypertension, micro-
albuminuria and previous CVD.
Table 2. (Continued)
(Continued)
200 The British Journal of Diabetes & Vascular Disease 13(4)
1
st
author
(year of
publication)
Design of study Setting location
[year of study]
Comparison
population
Study size
[Number
deceased]
Patient’s
characteristics
(age range,
male–female
distribution)
Follow up
(years)
Outcome
measure of
mortality
Degree of mortality,
i.e. RR or OR (95%
CI)
Comments
Sawicki
(1998)
32
Prospective Dusseldorf,
Germany
[NA]
Type 2
diabetes
cohorts
216
[158]
M & F
>40 years
15-16 years All-cause
mortality
RR: 3.3 (p< 0.001) In an unselected type 2
diabetes population shows
independent mortality risk
factors
Sievers
(1992)
46
Retrospective
(NHI studies)
Pima, USA
[1984]
Without
diabetes
1266
[512]
≥35 years
M=536, F=730
10 years All-cause
mortality
RR: 1.7 (1.4–2.2) ↑ mortality with increasing
diabetic duration
Siscovick
(2010)
33
Prospective Washington,
USA
[1994]
Without
diabetes
342
[NA]
50–79 years
M=214
F=128
14 years AMI: No
diabetes
type 2 diabetes,
no microvascular
type 2 diabetes
plus micro-
vascular
OR: 1.24 (0.98–1.57)
OR: 1.73 (1.28–
2.234)
OR: 2.66 (1.84–3.85)
Mortality ↑with onset of Type
2 diabetes and worsens with
presence of microvascular
disease
Spijkerman
(2002)
47
Retrospective
(Hoorn study)
Hoorn,
Netherland
[2001]
Without
diabetes
174
[NA]
M & F
50–75 years
M=76, F=98
10 years All-cause
mortality
(<6.2 years)
SDD
(≥6.2 years)LDD
RR: 2.06 (1.04–4.10)
RR: 3.19 (1.64–6.20)
Mortality risk significantly
↑ with increasing diabetes
duration
Ta n
(2004)
48
Retrospective Scotland, UK
[2002]
Without
diabetes
3594
[909]
≥65 years
M=1725
F=1869
4.6 years All-cause
mortality
M
F
RR: 1.36 (1.06–1.73)
RR: 1.06 (0.94–1.19)
RR: 1.29 (1.15–1.45)
Men diagnosed after age
65 years had no ↑ risk of
mortality.
Valdes
(2009)
34
Prospective
(Asturias
study)
Asturias, Spain
[NA]
Without
diabetes
1015
[42]
M & F
30–75 years
6 years All-cause
Diagnosed type
2 diabetes
Undiagnosed
type 2 diabetes
Prediabetes
RR: 2.5 (1.0–6.3)
RR: 2.7 (1.1–6.7)
RR: 1.6 (0.7–4.0)
Diagnosed and undiagnosed
type 2 diabetes had a risk of
2.5 to 3.0 ↑ than control. Pre-
diabetes had ↑ mortality
Valmadrid
(2000)
35
Prospective Wisconsin,
USA
Without
diabetes
840
[364]
M=378
F=462
12 years All-cause
Micro-
albuminuria
Gross
proteinuria
RR: 1.68 (1.35–2.09)
RR: 2.47 (1.97–3.10)
Both micro- and gross
proteinuria ↑ mortality from
all cause and for CVD
Vilbergsson
(1998)
36
Prospective Reykjavik,
Iceland
Without
diabetes
18912
[4380]
34–79 years
M=9139
F=9773
17 years All-cause M
F
CHD specific M
F
RR: 1.9 (1.6–2.3)
RR: 1.7 (1.3–2.1)
RR: 2.0 (1.5–2.6)
RR: 2.4 (1.6–3.6)
Type 2 diabetes carries twice
the risk of CHD death in both
sexes
Key: AMI: acute myocardial infarction; BG: biguanide; CAD: coronary artery disease; CHD: coronary heart disease; CVS: cardiovascular disease; F: females; HbA
1C
: glycated haemoglobin A
1C
;
Hx: history; LDD: long duration of diabetes; M: males; NHANES: National Health and Nutrition Examination Survey; MF: metformin; OR: odds ratio; PVD: peripheral vascular disease; RR: relative risk; SDD:
short duration of diabetes; SU: sulfonylurea; UAC: uric acid concentration; ↑: increased.w
Table 2. (Continued)
Nwaneri et al. 201
Overall mortality rates
Thirty-three studies reported increased mortality in type 2
diabetes but only 26 had full data on 95% confidence inter-
vals on all-cause mortality (Figure 2 and Table 2), of which
24 demonstrated increased mortality. One study reported
no excess mortality in men diagnosed with type 2 diabetes
after age 65 years,
48
whereas three reported an increased
mortality in a similar age group in both sexes.
3,38,49
In Israel
one researcher reported that relative mortality was worse
in women when compared to men after AMI.
21
Two studies reported decreased relative mortality in type
2 diabetes
22,31
although the association was statistically non-
significant. However, they found that increased mortality
was associated with poor HbA
1C
, lipid profile, hypertension,
microalbuminuria and history of previous CVD.
Gender and mortality risk
There was conflicting evidence concerning gender. For
two studies, women showed increased relative mortality
compared to men,
40
worse still after AMI,
21
whereas two
other studies reported opposite effects in that men had
increased mortality compared to women.
24,
32
Surprisingly,
one researcher reported that men diagnosed after age 65
years had no increased mortality.
48
Figures 4 and 5 illus-
trate the forest plots for males and females respectively.
Drugs
Certain drugs used in the treatment of type 2 diabetes were
reported in three studies. In those treated with insulin there
were worse short- and long-term prognoses after AMI in
both sexes.
21
Those treated with a combination of sulfonyl-
ureas and metformin had increased relative mortality when
compared with sulfonylureas or metformin alone,
30,45
and
this increased mortality was worse in women.
Cardiovascular diseases
Ten studies reported increased relative mortality risk
associated with CVD, CHD or AMI in those with type 2
Figure 4. Relative risks of mortality in type 2 diabetes males. The size of the box represents the weight assigned to individual
effects (point estimate of effect). The horizontal lines indicate the lower and upper limits of 95% confidence intervals.
202 The British Journal of Diabetes & Vascular Disease 13(4)
diabetes.
23–25,31,35,37,41–43
A history of type 2 diabetes alone
had worse relative mortality risks and when combined
with CHD, the mortality risks doubled.
25
One of the
studies reported that type 2 diabetes carries twice the risk
of CHD death in both sexes.
36
Another study reported an
increased relative mortality risk in those with PVD.
41
Hypertension was also reported to be associated with
worsening mortality risks.
23,31,32,41
Retinopathy was
reported to be associated with increased risk of mortal-
ity,
32,33
as well as stroke.
27
Figures 6 and 7 demonstrate the forest plots for car-
diovascular and stroke risks respectively.
Chronic renal diseases
Microalbuminuria or increased albumin excretion ratio
were reported to be associated with increased relative
mortality risk in people with type 2 diabetes in 8
studies.
23,24,27,28,31,33,35,41
Increased serum uric acid had an
association with CHD mortality risk in a J-shaped
fashion.
26
Glycaemic control (HbA
1C
). Poor glycaemic control was
associated with increased relative mortality in type 2 diabe-
tes populations in two studies,
21
,
39
whereas in two other
studies there were no associations with mortality risk.
24
,
31
Three studies revealed that long-term diabetes duration
increased relative mortality risk.
32
,
47
,
50
Lipid control
Poor lipid profile: low HDL-cholesterol,
24,31,32
total cho-
lesterol,
32
and high LDL-cholesterol
27
were associated
with increased relative mortality risk in type 2 diabetes.
Obesity was associated with increased relative mortality
in one study,
27
whilst in another it was not associated
with increased mortality, similar to the UKPDS find-
ings.
32
Figure 5. Relative risks of mortality in type 2 diabetes females. The size of the box represents the weight assigned to individual
effects (point estimate of effect). The horizontal lines indicate the lower and upper limits of 95% confidence intervals.
Nwaneri et al. 203
Figure 6. Relative risks of cardiovascular mortality in type 2 diabetes population. The size of the box represents the weight assigned
to individual effects (point estimate of effect). The horizontal lines indicate the lower and upper limits of 95% confidence intervals.
Smoking and alcohol
One study found both ex-smokers and current smokers
had increased relative mortality in a dose dependent man-
ner,
18
while two other studies reported that being a current
smoker is associated with increased mortality.
27,41
One study reported moderate alcohol consumption in
type 2 diabetes (in the range of 1.6–3 units per day) was
associated with decreased relative mortality, whereas less
than < 1.5units per day or > 3 units per day had adverse
effects on mortality,
39
but the association was not statisti-
cally significant in effect.
Age at diagnosis
Age at diagnosis was associated with increased absolute
and relative mortality,
38,41,49
in all ages but worse for
those aged less than 44 years, followed by those between
45 and 64 years and then aged 65 years or more.
3,32,44
Relative mortality was worse for women aged 65–74
years at diagnosis, with men having no excess mortality
when compared to the population without diabetes.
48
Cancer
No meta-analysis for cancer studies was included because
the studies had used hazard ratio or standardised mor-
tality ratio as proxy for RRs, highlighting a gap in this
area.
In summary, Figures 2–7 show the forest plots of all-
cause mortality RR of 1.85 (95% CI 1.79–1.92), CVD
mortality of 1.76 (95% CI 1. 66–1.88), stroke mortality of
2.26 (95% CI 1.7–3.02) and sex-specific mortality of 1.57
(95% CI 1.46–1.68) for men and 2.0 (95% CI 1.89–2.12)
for women. Details of RRs, point estimates and 95% con-
fidence intervals and weighted averages are demonstrated
204 The British Journal of Diabetes & Vascular Disease 13(4)
Figure 7. Relative risks of stroke mortality in type 2 diabetes population. The size of the box represents the weight assigned to
individual effects (point estimate of effect). The horizontal lines indicate the lower and upper limits of 95% confidence intervals.
in Figure 3, 4, 5 and 6. However, the studies demonstrated
unavoidable statistical heterogeneity (I²=91.2%, Cochran’s
Q=284, t²=0.11). Two studies
25,42
were identified as major
contributors to the heterogeneity after exclusion sensitiv-
ity analysis plot was computed as a result of the diverse
nature of patients’ characteristics, design, follow-up and
methodologies (p=0.79); I
2
=85.7%, t
2
=0.13 and Q
score=160 in the all-cause mortality. The overall RR
changes each time a study which contributed to the appar-
ent heterogeneity is excluded, with only a little variation in
I
2
. However, the ‘desired heterogeneity threshold’ in terms
of I
2
statistics was exclusion of six studies
22,23,25,31,39,42
, with
I
2
=85.3%, RR=1.74 (95% CI 1.64–1.84). This level of het-
erogeneity is acceptable since there were predefined eligi-
bility criteria, application of random- and fixed effect
models in the meta-analysis. The heterogeneity funnel
plot was symmetrical, and the dissemination sensitivity
test using Begg’s test and Egger’s regression test were 0.55
and 0.71 respectively, indicating a low probability of pub-
lication bias.
Discussion
In the past various researchers have analysed the mortal-
ity risks in type 2 diabetes showing decreases over the
years. However, there has been an absence of a compre-
hensive, rigorous review of current evidence (post 1999)
which analyses actual RRs of mortality in type 2 diabetes.
This review has found that in past studies there were sev-
eral flaws in the measure of binary dichotomous out-
comes as researchers had used hazard ratios (measures of
the rate at which events happen), standardised mortality
ratios, and mortality rates as approximates of RR (mea-
sures of how many events have occurred). These approx-
imates are known to yield measurement errors.
52
Nwaneri et al. 205
Following wide variations, inconsistencies and con-
flicting results in the actual RRs of mortality, this study
was undertaken to summarise the evidence to answer
research questions relating to current knowledge on RRs
in all-cause or cause-specific mortality in type 2 diabetes
when compared to the population without diabetes, and
to identify research gaps and possible strategies for
improvement. This systematic review of 35 studies shows
that type 2 diabetes is associated with an unequivocal
increase in mortality risk regardless of the age at diagno-
sis when compared to the general population. It shows
varying independent predictors (age, micro and/or gross
albuminuria, gender, smoking, hypertension, PVD and
duration of diabetes), as well as cause-specific mortalities
(cardiovascular, cerebrovascular, stroke, retinopathy and
nephropathy). It has found a RR of 1.85 (95% CI 1.79–
1.92) for all-cause mortality, with men and women hav-
ing RRs of 1.57 (95% CI 1.46–1.68) and 2.0 (95% CI
1.89–2.12) respectively. This shows a high degree of
association between diagnosis of T2D and mortality
although lower than those reported in previous studies.
2
The evidence is consistent as almost all of the studies
reported an increased RR of mortality in type 2 diabetes
regardless of age, with the middle aged group having
worse mortality predictions than the elderly.
43,44,49
In
contrast with the results of our review, Barnett et al.
reported a RR of mortality in the elderly with type 2 dia-
betes of 1.38 (95% CI 1.08–1.76) in males between 60
and 70 years and 1.13 (95% CI 0.881.45) for those aged ≥
70 years or more, while females had a RR of 1.40 (95% CI
1.10–1.79) and 1.19 (95% CI 0.93–1.52) respectively.
49
Their findings signify that even though there is an
increased relative mortality in the elderly diagnosed with
type 2 diabetes, it is lower than the general older diabetic
population. Inarguably, comparison of diabetes-related
mortality between patients with diabetes taking different
classes of medications is hard to interpret in an observa-
tional setting given that the prescription of medication
could be related to disease severity and co-morbidities.
Although two key studies from the UK and Sweden
reported a lower relative mortality than in the general
population, both had small population sizes,
22,31
and the
age range was between 35 and 55 years in a selective pop-
ulation.
22
The associations of these two studies were not
statistically significant. However another study
44
reported worse mortality in a similar age group. These
conflicting findings need further research with larger
population sizes.
Strong and positive associations were made between
CVD mortality and type 2 diabetes in both men and
women, although of varying degrees and age groups in all
the studies that reported on this relationship. The RRs of
1.76 (95% CI 1.66–1.88) for CVD mortality and 2.26 (95%
CI 1.7–3.02) for stroke are similar to those reported in the
literature.
36,50
Other consistent associations with increased
mortality were smoking, retinopathy, PVD, hypertension,
low or high alcohol intake, poor lipid profile, age and sex,
combination of drugs such as sulfonylureas and met-
formin and in those on insulin. Nephropathy (microalbu-
minuria or gross-albuminuria) showed increased relative
mortality risk, whereas HbA
1C
and duration of diabetes
had inconsistent outcomes with mortality.
In all the studies the duration of follow-up was satisfac-
tory, although not analysed to ascertain its RR. However,
despite this, evidence abounds that both absolute and rela-
tive mortality risks rise directly with increasing diabetes
duration even after multivariate adjustments.
46,47
The distribution of the study populations across
Europe, USA, Canada and New Zealand, as well as Japan,
Fuji, Argentina and Israel makes the population heterog-
enous and invariably minimises bias. The pathophysio-
logical mechanism of the two-fold increase in relative
mortality risk in type 2 diabetes has been shown to be due
to macrovascular and microvascular complications
involving the cardiovascular, cerebrovascular and renal
vessels. This study reinforces the complexity surrounding
mortality in type 2 diabetes and the interplay of various
independent mortality predictors. Unarguably, patients
with type 2 diabetes should be at the centre of preventive
management strategies making education key to improv-
ing outcomes. However, given the increasing prevalence
of type 2 diabetes, the fact that 50% of people develop
complications before their diagnosis, then the need for
public health interventions aimed at whole populations is
obvious. This systematic review and meta-analysis has
answered the question of the degree of RR of mortality in
type 2 diabetes. It has also highlighted gaps including
relationships with malignancies, ethnicity, obesity,
HbA
1C
, social deprivation and various medications used
in treatment of type 2 diabetes. Regardless of these gaps,
evidence using survival measures have shown decreased
survival and life expectancies in type 2 diabetes.
The selected studies reported on both significantly
positive and negative impact results. In addition, the
impact of publication bias was reduced because of the
wide geographic and population size studied, and the
weighted fixed and random effects. Although the selec-
tion of articles in English language is accepted as a limi-
tation, statistics were not limited to western countries as
it included research from non-English speaking coun-
tries like Argentina, Japan, Fuji and Israel which reduced
bias. Many of the studies had their estimates of RR s
adjusted for confounders for age, smoking and sex.
The strength of this study lies in the fact that an appro-
priate measure of mortality in type 2 diabetes has been
used. Given the global epidemic of diabetes, and the asso-
ciated rises in complications, the need to have an accurate
measure of mortality risk is paramount as countries rely
on this as one of the key indicators for resource alloca-
tion, health planning and health system prioritization.
206 The British Journal of Diabetes & Vascular Disease 13(4)
In summary, this paper has presented the outcomes of
a systematic review and meta-analysis showing approxi-
mately a two-fold increase in the mortality risks in type 2
diabetes when compared with the general population.
Our evidence illustrates a strong association of increased
relative mortality with type 2 diabetes with a RR of 1.85
(95% CI 1.79–1.92) for all-cause mortality and 1.76 (95%
CI 1.66–1.88) for CVD related mortality, and 2.26 (95%
CI 1.7–3.02) for stroke. Evidence has shown that the con-
trol of cardiovascular, cerebrovascular and renal compli-
cations with appropriate interventions can reduce
mortality in type 2 diabetes.
53
However, further primary
research is needed to ascertain the clinical benefits of
pro-active management of these micro- and macro-vas-
cular complications. To guide this process, further
research is now required to explore the associations
between morbidity, mortality and malignancies, socio-
economic status, HbA
1C
, obesity, ethnicity and various
medications used in the treatment of type 2 diabetes.
Acknowledgements
We are indebted to the services of the Librarian at the Riverside
Campus of the University of Chester, and the McArdle Library
at the Arrowe Park Hospital, Upton, and also wish to thank the
researchers of the selected studies.
Declaration of conflicting interests
The authors have no conflicts of interest
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
References
1. Whiting DR, Guariguata L, Weil C, Shaw J. IDF Diabetes
Atlas: Global estimates of the prevalence of diabetes for
2011 and 2030. Diab Research Clin Pract 2011; 94: 311-21.
2. Fuller JH, Elford J, Goldblatt P, Adelstein AM. Diabetes
Mortality: New light on an underestimated public health
problem. Diabetologia 1983; 24: 336-41.
3. Guzder RN, Gatling W, Mullee MA, Byrne CD. Early
Mortality from the time of diagnosis of Type 2 Diabetes: A
5-Year Prospective Cohort Study With A Local Age- And
Sex-Matched Comparison Cohort. Diabet Med 2007; 24:
1164-167.
4. Gatling W, Guzder RN, Turnbull JC et al. The Poole
Diabetes Study: how many cases of type 2 diabetes are
diagnosed each year during normal health care in a defined
community? Diabetes Res Clin Pract 2001; 53: 107-12.
5. Roglic G, Unwin N, Bennett PH etal. The burden of mor-
tality attributable to diabetes: realistic estimates for the
year 2000. Diabetes Care 2005; 28: 2130-135.
6. Diabetes UK. Diabetes in the UK. A report from Diabetes
UK October 2004. http://www.library.nhs.uk/diabetes/
ViewResource.aspx?resID=82430 Accessed November 4,
2010.
7. National Academy On An Aging Society. At Risk:
Developing chronic conditions later in life. Challenges for
the 21st century. Chronic and Disabling Conditions Series
2000; 4.
8. Matsushima M, LaPorte RE, Maruyama M, Shimizu K,
Nishimura R, Tajima N. Geographic variation in mor-
tality among individuals with youth-onset diabetes
across the world. DERI mortality Study Group. Diabetes
Epidemiology Research International. Diabetologia 1997;
40: 212-16.
9. Morrish NJ, Wang SL, Stevens LK, Fuller JH, Keen H.
Mortality and causes of death in the WHO multinational
study of vascular disease in diabetes. Diabetologia 2001;
44(suppl 2): S14-S21.
10. Jeerakathil T, Johnson JA, Simpson SH, Majumdar SR.
Short-term risk for stroke is doubled in persons with
newly treated type 2 diabetes compared with persons
without diabetes: a population based cohort study. Stroke
2007; 38: 1739-43.
11. Currie CJ, Poole CD, Gale EAM. The influence of glu-
cose-lowering therapies on cancer risk in type 2 diabetes.
Diabetologia 2009; 52: 1766-77.
12. Pollack M, Russell-Jones D. Insulin analogues and cancer
risk: cause for concern or cause celebre? Int J Clin Pract
2010; 64: 628-36.
13. Centre for Research and Development Report. Under-
taking Systematic Reviews of Research on Effectiveness
CRD’s Guidance for those Carrying Out or Commissioning
Reviews. CRD Report 2001; 4 (2nd Edition). http://www.
york.ac.uk/search/?q=Undertaking+systematic+reviews+
of+research+on+effectiveness&client=yorkweb&output=
xml_no_dtd&proxystylesheet=yorkweb&site=crd-main.
Accessed October 20, 2010.
14. Moher D, Pham B, Klassen TP et al.What contributions
do languages other than English make on the results of
meta-analyses? J Clin Epidemiol 2000; 53: 964-72.
15. Juni P, Holenstein F, Sterne J etal. Direction and impact
of language bias in meta-analyses of controlled trials:
Empirical study. Int J Epidemiol 2002; 31: 115-23.
16. Loveman E, Cave C, Green C etal. The clinical and cost-
effectiveness of patient education models for diabetes:
a systematic review and economic evaluation. Health
Technol Assess 2003; 7: 1-211.
17. Moher D, Liberati A, Tetzlaff J, Altman DG. The PRISMA
Group. Preferred Reporting Items for Systematic Reviews
and Meta-Analyses: The PRISMA Statement. Ann Intern
Med 2009; 151: 264-9.
18. Al-Delaimy WK, Willett WC, Manson JE etal. Smoking and
mortality among women with type 2 diabetes: The Nurses’
Health Study Cohort. Diabetes Care 2001; 24: 2043-8.
19. Almdal T, Scharling H, Jensen JS, Vestergaard H. The
independent effect of type 2 diabetes mellitus on ischemic
heart disease, stroke, and death: A population-based study
of 13 000 men and women with 20 years of follow-up.
Arch Intern Med 2004; 164: 1422-6.
20. Balkau B, Eschwège E, Papoz L etal. Risk factors for early
death in non-insulin dependent diabetes and men with
known glucose intolerance status. BMJ 1993; 307: 295-9.
21. Behar S, Boyko V, Reicher-Reiss H, Goldbourt U. Ten-Year
Survival After Acute Myocardial Infarction: Comparison
Nwaneri et al. 207
Of Patients With And Without Diabetes. Am Heart J 1997;
133: 290-6.
22. Chaturvedi N, Jarrett J, Morrish N et al. Differences In
Mortality And Morbidity In African Caribbean And
European People With Non-Insulin Dependent Diabetes
Mellitus: Results Of 20 Year Follow Up Of A London
Cohort Of A Multinational Study. BMJ 1996; 313: 848-2.
23. Gall M, Borch-Johnsen K, Hougaard P etal. Albuminuria
And Poor Glycemic Control Predict Mortality In NIDDM.
Diabetes 1995; 44: 1303-9.
24. Hänninen J, Takala J, Keinänen-Kiukaanniemi S.
Albuminuria and other risk factors for mortality in
patients with non-insulin-dependent diabetes mellitus
aged under 65 years: a population-based prospective
5-year study. Diabetes Res Clin Pract 1999; 43: 121-6.
25. Hu FB, Stampfer MJ, Solomon CG et al. The impact of
diabetes mellitus on mortality from all causes and coro-
nary heart disease in women: 20 years of follow-up. Arch
Intern Med 2001; 161: 1717-23.
26. Mazza A, Zamboni S, Rizzato E etal . Serum uric acid shows
a J-shaped trend with coronary mortality in non-insulin-
dependent diabetic elderly people. The CArdiovascular
STudy in the ELderly (CASTEL). Acta Diabetol 2007; 44:
99-105.
27. Katakura M, Naka M, Kondo T etal. Normal mortality in
the elderly with diabetes under strict glycemic and blood
pressure control: outcome of 6-year prospective study.
Diabetes Res Clin Pract 2007; 78: 108-14.
28. Neil A, Hawkins M, Potok M, Thorogood M, Cohen D,
Mann J. A prospective population-based study of microal-
buminuria as a predictor of mortality in NIDDM. Diabetes
Care 1993; 16: 996-1003.
29. O’Connor PJ, Crabtree BF, Nakamura RM. Mortality
experience of Navajos with type 2 diabetes mellitus. Ethn
Health 1997; 2: 155-62.
30. Olsson J, Lindberg G, Gottsäter M etal. Increased mortal-
ity in Type II diabetic patients using sulphonylurea and
metformin in combination: a population-based observa-
tional study. Diabetologia 2000; 43: 558-60.
31. Ostgren CJ, Lindblad U, Melander A, Rastam L. Survival In
Patients With Type 2 Diabetes In A Swedish Community:
Skaraborg Hypertension And Diabetes Project. Diabetes
Care 2002; 25: 1297-302.
32. Sawicki PT, Kiwitt S, Bender R, Berger M. The Value
of QT-Interval Dispersion For Identification Of Total
Mortality Risk In Non-Insulin-Dependent Diabetes
Mellitus. J Intern Med 1998; 243: 49-56.
33. Siscovick DS, Sotoodehnia N, Rea TD etal. Type 2 dia-
betes mellitus and the risk of sudden cardiac arrest in the
community. Rev Endocr Metab Disord 2010; 11: 53-9.
34. Valdés S, Botas P, Delgado E, Díaz Cadórniga F. Mortality
risk in Spanish adults with diagnosed diabetes, undiag-
nosed diabetes or pre-diabetes. The Asturias study 1998–
2004. Rev Esp Cardiol 2009; 62: 528-34.
35. Valmadrid CT, Klein R, Moss SE, Klein BEK. The Risk
Of Cardiovascular Disease Mortality Associated With
Microalbuminuria And Gross Proteinuria In Persons
With Older-Onset Diabetes Mellitus. Arch Intern Med
2000; 160: 1093-100.
36. Vilbergsson S, Sigurdsson G, Sigvaldason H, Sigfusson
N. Coronary heart disease mortality amongst
non-insulin-dependent diabetic subjects in Iceland: the
independent effect of diabetes. The Reykjavik Study
17-year follow-up. J Intern Med 1998; 244: 309-16.
37. Collins VR, Dowse GK, Ram P et al. Non-insulin-
dependent diabetes and 11-year mortality in Asian Indian
and Melanesian Fijians. Diabet Med 1996; 13: 125-32.
38. Croxson SCM, Price DE, Burden M, Jagger C, Burden
AC. Mortality of elderly people with diabetes. Diabet Med
1994; 11 :250-2.
39. Diem P, Deplazes M, Fajfr R et al. Effects Of Alcohol
Consumption On Mortality In Patients With Type 2
Diabetes Mellitus. Diabetologia 2003; 46: 1581-5.
40. Eliasson M, Talbäck M, Rosén M. Improved survival in both
men and women with diabetes between 1980 and 2004-a
cohort study in Sweden. Cardiovasc Diabetol 2008; 7: 32.
41. Florkowski CM, Scott RS, Coope PA, Moir CL. Predictors
of mortality from type 2 diabetes mellitus in Canterbury,
New Zealand; a ten-year cohort study. Diabetes Res Clin
Pract 2001; 53: 113-20.
42. Gagliardino JJ, Werneke U, Olivera EM et al.
Characteristics, clinical course, and in-hospital mortal-
ity of non-insulin-dependent diabetic and non-diabetic
patients with acute myocardial infarction in Argentina. J
Diabetes Complications 1997; 11: 163-71.
43. Grauw WJC, van de Lisdonk EH, van den Hoogen HJM,
van Weel C. Cardiovascular morbidity and mortality in
type 2 diabetic patients: a 22-year historical cohort study
in Dutch general practice. Diabet Med 1995; 12: 117-22.
44. Gu K, Cowie C, Harris M. Mortality in adults with and
without diabetes in a national cohort of the U.S. popula-
tion, 1971–1993. Diabetes Care 1998; 21: 1138-45.
45. Mannucci E, Monami M, Masotti G, Marchionni N. All-
cause mortality in diabetic patients treated with combina-
tions of sulfonylureas and biguanides. Diabetes Metab Res
Rev 2004; 20: 44-7.
46. Sievers ML, Nelson RG, Knowler WC, Bennett PH.
Impact of NIDDM on mortality and causes of death in
Pima Indians. Diabetes Care 1992; 15: 1541-9.
47. Spijkerman AMW, Dekker JM, Nijpels G etal. Impact Of
Diabetes Duration And Cardiovascular Risk Factors On
Mortality In Type 2 Diabetes: The Hoorn Study. Eur J Clin
Invest 2002; 32: 924-30.
48. Tan HH, McAlpine RR, James P etal. Diagnosis of type 2 dia-
betes at an older age: effect on mortality in men and women.
DARTS/MEMO collaboration. Diabetes Care 2004; 27: 2797-9.
49. Barnett KN, McMurdo MET, Ogston SA, Morris AD.
Mortality in people diagnosed with type 2 diabetes at an
older age: a systematic review. Age and Aging 2006; 5: 463–8.
50. Dinneen SF, Gerstein HC. The Association of
Microalbuminuria and Mortality in Non-Insulin-
Dependent Diabetes Mellitus: A Systematic Overview of
the Literature. Arch Intern Med 1997; 157: 1413-8.
51. Groeneveld Y, Petri H, Hermans J, Springer MP. Relationship
between blood glucose level and mortality in type 2 diabe-
tes mellitus: a systematic review. Diabet Med 1999; 16: 2-13.
52. Duerden M. What are Hazard ratios? London: Hayward
Medical Communications, 2009. Available at www.whats-
series.co.uk Accessed October 20, 2011.
53. Gæde P, Lund-Andersen H, Parving H, Pedersen O. Effect
of a multifactorial intervention on mortality in type 2 dia-
betes. N Engl J Med 2008; 358: 580-91.
