RESEARCH

Global burden of type 1 diabetes in adults aged 65 years and

BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
older, 1990-2019: population based study
Kaijie Yang,1 Xue Yang,1 Chenye Jin,2 Shuangning Ding,1 Tingting Liu,1 Bing Ma,3 Hao Sun,3
Jing Zhang,4 Yongze Li1
1
Department of Endocrinology ABSTRACT The most significant decrease in DALYs was observed
and Metabolism and the Institute OBJECTIVES among those aged <79 years: 65-69 (−0.44% per year
of Endocrinology, NHC Key To estimate the burden, trends, and inequalities of (95% CI −0.53% to −0.34%)), 70-74 (−0.34% per
Laboratory of Diagnosis and
Treatment of Thyroid Diseases, type 1 diabetes mellitus (T1DM) among older adults year (−0.41% to −0.27%)), and 75-79 years (−0.42%
First Hospital of China Medical at global, regional, and national level from 1990 to per year (−0.58% to −0.26%)). Mortality fell 13 times
University, Shenyang, China 2019. faster in countries with a high sociodemographic index
2
Department of Rheumatology versus countries with a low-middle sociodemographic
and Immunology, First Hospital DESIGN
of China Medical University, Population based study. index (−2.17% per year (95% CI −2.31% to −2.02%)
Shenyang, China v −0.16% per year (−0.45% to 0.12%)). While the
POPULATION
3
Department of Clinical highest prevalence remained in high income North
Epidemiology and Evidence- Adults aged ≥65 years from 21 regions and 204
America, Australasia, and western Europe, the highest
based Medicine, First Hospital countries and territories (Global Burden of Disease
DALY rates were found in southern sub-Saharan Africa,
of China Medical University, and Risk Factors Study 2019)from 1990 to 2019.
Shenyang, China Oceania, and the Caribbean. A high fasting plasma
4
School of Public Health,
MAIN OUTCOME MEASURES glucose level remained the highest risk factor for
Shenzhen University Medical Primary outcomes were T1DM related age DALYs among older adults during 1990-2019.
School, Shenzhen, China standardised prevalence, mortality, disability adjusted
CONCLUSIONS
Correspondence to: Y Li life years (DALYs), and average annual percentage
yzli87@cmu.edu.cn The life expectancy of older people with T1DM has
change.
(ORCID 0000-0001-8782-3314) increased since the 1990s along with a considerable
Additional material is published RESULTS decrease in associated mortality and DALYs. T1DM
online only. To view please visit The global age standardised prevalence of T1DM related mortality and DALYs were lower in women
the journal online. among adults aged ≥65 years increased from 400 aged ≥65 years, those living in regions with a high
Cite this as: BMJ 2024;385:e078432 (95% uncertainty interval (UI) 332 to 476) per 100 000
http://dx.doi.org/10.1136/ sociodemographic index, and those aged <79 years.
bmj‑2023‑078432 population in 1990 to 514 (417 to 624) per 100 000 Management of high fasting plasma glucose remains
population in 2019, with an average annual trend a major challenge for older people with T1DM, and
Accepted: 15 April 2024
of 0.86% (95% confidence interval (CI) 0.79% to targeted clinical guidelines are needed.
0.93%); while mortality decreased from 4.74 (95%
UI 3.44 to 5.9) per 100 000 population to 3.54 (2.91 Introduction
to 4.59) per 100 000 population, with an average Type 1 diabetes mellitus (T1DM) was traditionally
annual trend of −1.00% (95% CI −1.09% to −0.91%), considered a disease that manifested during childhood
and age standardised DALYs decreased from 113 and adolescence and could have a profound effect
(95% UI 89 to 137) per 100 000 population to 103
on life span.1 2 Variable evidence from recent studies
(85 to 127) per 100 000 population, with an average
suggests that the life expectancy of an increasing
annual trend of −0.33% (95% CI −0.41% to −0.25%).
number of older and elderly people with T1DM has
improved as has diabetes care and the management
WHAT IS ALREADY KNOWN ON THIS TOPIC of complications.3 4 Current and comprehensive data
Type 1 diabetes mellitus (T1DM) has been considered a disease to have a on the burden of T1DM is, however, lacking in most
deleterious effect on life expectancy countries and territories worldwide.5 6 Nearly all of the
Diabetes care has improved substantially since the 1990s, with an increasing existing clinical practices and guidelines lack targeted
number of older people with T1DM reported in studies content for the management of T1DM among older
In most countries and regions worldwide, comprehensive and current data on the people.7
burden of T1DM are lacking
As the global population ages, substantial gaps in
data related to T1DM need to be bridged urgently. A
WHAT THIS STUDY ADDS modelling study estimated that a diagnosis of T1DM
Using the Global Burden of Disease Study model, this study found that the age was missed in 3.7 million people in 2021, and it
standardised prevalence of T1DM among people aged ≥65 years worldwide predicted a 60-107% increase in the number of people
increased during 1990-2019 concomitant with a substantial decrease in with a diagnosis between 2021 and 2040 globally,
associated mortality highlighting the opportunity to raise the standard
of care and save millions of lives in the coming
A high fasting plasma glucose level was the major contributor to disability
decade.8 This increasing prevalence could represent
adjusted life years, indicating that hyperglycaemia management remained
an increasing number of people with T1DM living to
challenging for the older people with T1DM
elderly age. Importantly, based on current evidence,
Mortality fell 13 times faster in countries with a high sociodemographic index
the number of older people with T1DM has already
versus countries with a low-middle sociodemographic index (−2.17% per year v
likely increased owing to an overall increase in life
−0.16% per year) expectancy.4 Continuous T1DM care of people in older

the bmj | BMJ 2024;385:e078432 | doi: 10.1136/bmj-2023-078432 1

 RESEARCH
age, diabetes related complications, and impaired
cognitive and physiological function from ageing are
challenging for both people with T1DM and healthcare
resources.9 Understanding the changes in mortality
and DALYs among older people (≥65 years) with T1DM
is critical for their care.
We investigated T1DM associated prevalence,
mortality, and disability adjusted life years (DALYs) in
adults aged ≥65 years at global, regional, and national
level during 1990-2019 by social developmental level
and age and sex of the older population. We also
assessed factors that might have an influence on DALYs
among older people with T1DM.
Methods
Study population and data collection
In our analysis of the Global Burden of Disease Study
2019, we accessed repeated cross sectional data from
the Global Health Data Exchange, encompassing the
global burden of 369 diseases and injuries and 87 risk
factors, including T1DM, across 21 regions and 204
countries and territories from 1990 to 2019. The study
population comprised people with T1DM aged 65 years
and older, including those with diabetes diagnosed
before age 65 years.10-13 From the Global Burden of
Disease Study 2019 we extracted information on
T1DM in people aged ≥65 years; location, age, and
sex specific prevalence; mortality; numbers and rates
for DALYs; and DALYs attributable to each risk factor
(with corresponding 95% uncertainty intervals (UIs)).
Attributable DALYs are a metric for quantifying the
contribution of specific risk factors to the burden of
disease. DALYs signify the changes and reduction
in current burden of disease if a change occurs to
population level exposure to a particular risk factor.
After adjusting for comorbidity, we used micro-
simulation to obtain the final estimate of years lived
with disability. Years of life lost were calculated by
multiplying the estimated number of deaths from
T1DM by the standard life expectancy at the age of
death. DALYs were equal to the sum of years lived
with disability and years of life lost. The methodology
employed in the Global Burden of Disease Study 2019
is described elsewhere (also see supplementary file,
methods section).14 15
In the Global Burden of Disease Study, T1DM is
defined as doctor diagnosed disease identified through
a diabetic registry or hospital records. To estimate non-
fatal burdens of T1DM and complications from T1DM,
a bayesian meta-regression modelling tool, DisMod-
MR 2.1, was used to analyse 1527 location years of
data from the scientific literature, survey microdata,
and insurance claims.14 Estimates of both non-fatal
and fatal outcomes in people with T1DM include
those attributed to complications from the disease.
The Global Burden of Disease Study extrapolates and
models data based on the reported T1DM, juvenile
onset diabetes, and insulin dependent diabetes to
develop a database applicable to most countries, with
no age restrictions on the input raw data. To allow
for smoothing over age, time, and location in areas
2
without complete datasets, spatiotemporal Gaussian
process regression was used to model the Global
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
Burden of Disease Study 2019 input data.14 The Global
Burden of Disease Study analysed the spatiotemporal
Gaussian process regression.16 The methods section in
the supplementary file provides detailed information
on the input data and methodology for T1DM.
In this study, we collected data on T1DM from
21 regions of countries that are geographically
proximate and have similar epidemiological profiles,
encompassing seven age groups (65-69 years, 70-
74 years, 75-79 years, 80-84 years, 85-89 years, 90-
94 years, ≥95 years) for both men and women. We
also calculated the sociodemographic index for each
country, which is a composite indicator of the social
and economic conditions influencing health outcomes
in each locality. The sociodemographic index ranges
from 0.005 to 1. In this study 1 represented the highest
education level, highest per capita income, and lowest
fertility rate. The sociodemographic index is divided
into five categories: low, low-middle, middle, high-
middle, and high.
Statistical analysis
A descriptive analysis was performed to characterise
the burden of T1DM among adults aged ≥65 years
on a global scale. We compared the age standardised
prevalence (per 100 000 population), age
standardised mortality (per 100 000 population), and
age standardised DALYs (per 100 000 population) of
T1DM across different age groups, sexes, regions, and
countries. Based on the data on T1DM and associated
risk factors obtained from the Global Burden of Disease
Study, we further calculated the age standardised rates
and corresponding 95% confidence intervals (CIs)
based on the world standard population reported in the
Global Burden of Disease Study 2017 for comparison
between regions, and further estimated average annual
percentage changes (AAPCs) by joinpoint regression
to measure the temporal trend.14 Our estimates are
shown per 100 000 population using the top equation
in figure 1.
AAPCs are used to represent the average increase
or rate of change of a specific variable over a specified
period. In this study, it is the annual change percentage
transformed from the weighted average of the slope
coefficients of the underlying joinpoint regression
model from 1990 to 2019.17 The AAPC value denotes
the percentage annual change (increase, decrease, or
no change). If annual percentage change estimates and
95% CIs were both >0 (or both <0), we considered the
corresponding rate to be in an upward (or downward)
trend. AAPC was calculated using the lower equation
in figure 1.
The measures we chose are correlated
epidemiologically, including prevalence, DALYs (one
DALY represents the loss of one year of full health
owing to premature death or disability), and mortality.
For example, the reduction in mortality from a chronic
disease could lead to an increase in both prevalence
and DALYs, in terms of more patients living longer
doi: 10.1136/bmj-2023-078432 | BMJ 2024;385:e078432 | the bmj
 RESEARCH

the secondary data from this collaborative work, and

Σ Ai=1 a w we did not have direct access to the participants. No

BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
i i
Age standardised rate = patients were involved in setting the research question
Σ Ai=1 w i or the outcome measures, nor were they involved in
the design and implementation of the study. The BMJ
where ai is the age specific rate and wi is the weight in the same age
subgroup of the chosen reference standard population (in which i denotes invited a patient reviewer as a member of the public to
the ith age class) and A is the upper age limit read our manuscript after submission.

( )
AAPC = exp
Σ wb
Σw
i i

i
–1 x 100

bi is the slope coefficient for the i segment with i indexing the segments
th
in the desired range of years, and wi is the length of each segment in the
range of years
Fig 1 | Equations used to calculate age standardised rate and average annual
percentage change (AAPC)
with that disease. An increased number of people with
incident disease or a longer disease duration can both
contribute to increased prevalence. The increasing
prevalence of T1DM among people aged ≥65 years
could be explained by longer life with disease because
of improved medical care.
All statistical analyses were conducted using
GraphPad Prism (version 8.0), Joinpoint Regression
program (version 5.0.2), and R (version 4.2.3).
Patient and public involvement
The Global Burden of Disease Study is a collaborative
scientific effort allowing the effects of different health
conditions to be compared and replicated between
age, sex, and geographical locations for specific points
in time. The study has generated substantial scientific,
policy, and public interest worldwide. Our study used
Results
Global trends
Globally, the prevalence of T1DM among older people
aged ≥65 years increased by 180% between 1990
and 2019, from 1.3 million to 3.7 million. The age
standardised prevalence rate of T1DM among this
age group increased by 28%, from 400 per 100 000
population in 1990 to 514 per 100 000 population in
2019, with an average annual trend of 0.86% (table
1). Furthermore, the proportion of older people with
T1DM has shown a consistent upward trend relative
to the overall number of people with T1DM, increasing
from 12% in 1990 to 17% in 2019 (see supplementary
figure 1). Compared with the overall number of people
with T1DM, the increasing trend in prevalence of
T1DM among those aged ≥65 years was noticeable
from 1990 to 2019 (see supplementary figure 2). The
disease burden (all cause DALYs) from T1DM has been
increasing over the past 30 years in the population
aged ≥65 years (see supplementary figure 3). The
age standardised mortality from T1DM among this
age group significantly decreased by 25%, from 4.7
per 100 000 population in 1990 to 3.5 per 100 000
population in 2019, with an average annual trend of
−1.00% (see supplementary table 1).
Compared with prevalence and mortality, the trend
of DALYs from T1DM among people aged ≥65 years
was less noticeable during the same period. Age

Table 1 | Age standardised prevalence and AAPC of T1DM in people aged ≥65 years at global and regional level,
1990-2019
Prevalence (95% UI)
No of people with Age standardised rate No of people with Age standardised rate
T1DM in 1990 (000s) in 1990 (per 100 000) T1DM in 2019 (000s) in 2019 (per 100 000) AAPC (95% CI))
Global 1284 (1064 to 1527) 400 (332 to 476) 3688 (2992 to 4478) 514 (417 to 624) 0.86 (0.79 to 0.93)
Sex:
Female 734 (608 to 871) 399 (331 to 474) 1955 (1587 to 2370) 495 (402 to 600) 0.74 (0.65 to 0.83)
Male 550 (457 to 657) 401 (333 to 479) 1733 (1400 to 2111) 536 (433 to 653) 1.00 (0.95 to 1.06)
Age group (years):
65-69 456 (377 to 543) 369 (305 to 440) 1218 (986 to 1490) 471 (381 to 575) 0.83 (0.72 to 0.93)
70-74 325 (269 to 386) 384 (318 to 457) 941 (765 to 1140) 503 (409 to 611) 0.94 (0.85 to 1.04)
75-79 254 (212 to 302) 414 (346 to 492) 668 (543 to 805) 526 (428 to 633) 0.82 (0.76 to 0.89)
80-84 153 (127 to 181) 434 (362 to 514) 464 (378 to 562) 550 (448 to 665) 0.82 (0.75 to 0.89)
85-89 69.1 (57.1 to 82.2) 459 (379 to 546) 257 (207 to 311) 590 (475 to 715) 0.86 (0.72 to 1.01)
90-94 21.8 (17.9 to 26.0) 495 (405 to 590) 108 (86.9 to 131) 642 (516 to 778) 0.89 (0.79 to 1.00)
≥95 5.32 (4.29 to 6.44) 517 (417 to 626) 31.9 (25.6 to 38.7) 668 (535 to 810) 0.88 (0.82 to 0.94)
SDI level:
High 678 (570 to 794) 688 (578 to 805) 1666 (1361 to 2012) 913 (746 to 1103) 0.98 (0.93 to 1.03)
High-middle 327 (269 to 392) 369 (303 to 442) 956 (770 to 1167) 519 (418 to 634) 1.19 (1.09 to 1.28)
Middle 129 (103 to 158) 165 (132 to 203) 561 (450 to 692) 272 (218 to 335) 1.73 (1.66 to 1.80)
Low-middle 104 (82.5 to 128) 236 (186 to 293) 374 (297 to 461) 336 (266 to 415) 1.21 (1.12 to 1.29)
Low 45.9 (36.2 to 57.2) 263 (206 to 329) 130 (103 to 161) 338 (267 to 420) 0.86 (0.85 to 0.88)
AAPC=average annual percentage change; CI=confidence interval; SDI=sociodemographic index; T1DM=type 1 diabetes mellitus; UI=uncertainty interval.

the bmj | BMJ 2024;385:e078432 | doi: 10.1136/bmj-2023-078432 3

RESEARCH
standardised DALYs decreased by 8.9%, from 113
per 100 000 population in 1990 to 103 per 100 000
population in 2019, with an average annual trend of
−0.33% (see supplementary table 1).
Global trends by sex
From 1990 to 2019, the age standardised prevalence of
T1DM among people aged ≥65 years increased for both
men and women worldwide (men: from 0.6 million to
1.7 million; women: from 0.7 million to 2 million). The
increase in T1DM prevalence was more rapid among
men than among women (AAPC 1.00% v 0.74%) (table
1). During the same period, the age standardised
mortality from T1DM decreased for both men and
women aged ≥65 years, although the reduction
was smaller in men (AAPC −0.58% v −1.29%) (see
supplementary table 1).
From 1990 to 2019, the reduction in age
standardised DALYs from T1DM among older people
was less pronounced in men than in women. The
decrease in DALYs was more substantial in women aged
≥65 years, with an average annual trend of −0.58%.
In 1990, among people aged ≥65 years, women had
higher DALYs from T1DM per 100 000 population
than men (118 v 106). In 2019, however, women aged
≥65 years had fewer DALYs from T1DM per 100 000
population than men (100 v 106) (see supplementary
table 1). This sex difference remained consistent
regardless of changes in sociodemographic index,
with the burden of diseases higher in men compared
with women, especially in countries with a low-middle
sociodemographic index (see supplementary figures 4
and 5).
Global trends by age subgroup
Globally, during 1990-2019 the prevalence of T1DM at
least tripled in every age subgroup of people aged ≥65
years (65-69 years: from 0.46 million to 1.2 million;
70-74 years: from 0.32 million to 0.94 million; 75-79
years: from 0.25 million to 0.67 million; 80-84 years:
from 0.15 million to 0.46 million; 85-89 years: from
0.07 million to 0.26 million) and even increased 5-6
times for those >90 years (90-94 years: from 0.02
million to 0.11 million; ≥95 years: from 0.005 million
to 0.03 million). Notably, among all age subgroups,
the upward trend for men consistently surpassed that
for women, especially among men aged 90-94 years
(AAPC 1.18%) (see supplementary figure 6). The
age standardised prevalence of T1DM across all age
subgroups increased at a noticeable rate during 1990-
2019, especially among those aged 70-74 years (AAPC
0.94%) (table 1). The age standardised mortality of
T1DM decreased across all age subgroups among older
people worldwide, especially among those younger
than 79 years. In 2019, the mortality rates for T1DM
per 100 000 population among people aged ≥65
years increased with age, from 2.8 among those aged
65-69 years to 10 among those aged ≥95 years (see
supplementary table 1).
The age standardised DALYs associated with T1DM
among older people decreased in all age subgroups
4
at various rates. The most significant decrease was
observed among those aged <79 years, including 65-69
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
years (AAPC −0.44%), 70-74 years (−0.34%), and 75-
79 years (−0.42%). In 2019, the DALYs attributable to
T1DM were highest in those aged 75-79 years (111 per
100 000 population) and ≥95 years (111 per 100 000
population) (see supplementary table 1).
Global trends by sociodemographic index
The age standardised prevalence of T1DM among
people aged ≥65 years increased across all subgroups of
sociodemographic index during 1990-2019, especially
in countries with a middle sociodemographic index
(AAPC 1.73%). Regardless of sociodemographic level,
the increase in prevalence among those aged ≥65
years has consistently been higher than in the general
population (see supplementary figures 7 and 8). In
2019, among older people the highest prevalence of
T1DM was in countries with a high sociodemographic
index (913 per 100 000 population) (table 1). While
T1DM related mortality among older people decreased
across all sociodemographic subgroups during the
same period, the largest decrease was attributed to
countries with a high sociodemographic index (AAPC
−2.17%), which was more than 13 times faster than
that in countries with a low-middle sociodemographic
index (AAPC −0.16%). In 2019, the highest mortality
rate was in countries with a low sociodemographic
index (6.4 per 100 000 population), which was more
than two times the mortality in countries with a high
sociodemographic index (2.2 per 100 000 population)
(see supplementary table 1 and supplementary figures
4 and 5).
The age standardised DALYs from T1DM among
older people significantly decreased across all
sociodemographic subgroups, except in countries
with a low-middle sociodemographic index
(AAPC 0.01%) (see supplementary table 1). The
downward trend in DALYs for T1DM became more
pronounced after surpassing a sociodemographic
index threshold of 0.7 (see supplementary figure 9).
In 2019, the DALYs were highest in countries with
a low sociodemographic index (141 per 100 000
population) and lowest in countries with a high-
middle sociodemographic index (86 per 100 000
population) (see supplementary table 1).
Regional trends
From 1990 to 2019, none of the 21 regions showed a
decrease in prevalence of T1DM among people aged
≥65 years. The most rapid increase in age standardised
prevalence for older with T1DM was observed in
North Africa and the Middle East (AAPC 2.40%), East
Asia (2.17%), and western Europe (1.95%), and the
slowest increase in prevalence was in high income
North America (0.60%) during the same period (see
supplementary figure 10). In 2019, among the 21
regions, the highest age standardised prevalence
for T1DM among people aged ≥65 years was in high
income North America (1248 per 100 000 population),
Australasia (1080 per 100 000 population), and
doi: 10.1136/bmj-2023-078432 | BMJ 2024;385:e078432 | the bmj

western Europe (1077 per 100 000 population) (see
supplementary table 2). No difference was observed
after sex stratification (see supplementary table 3).
Most regions experienced a reduction in DALYs from
T1DM among older people at various rates during
1990-2019. However, a large increase in DALYs was
found in central Asia (AAPC 1.79%). The largest
reduction in DALYs from T1DM among older people
was in Central Latin America (AAPC −1.30%) (see
supplementary figure 10). In 2019, the highest DALYs
from T1DM among older people were in southern sub-
Saharan Africa (178 per 100 000 population), Oceania
(178 per 100 000 population), and the Caribbean (177
per 100 000 population). The lowest DALYs were in
East Asia (32 per 100 000 population), the high income
Asia Pacific region (51 per 100 000 population), and
eastern Europe (77 per 100 000 population) (see
supplementary table 4). After stratifying by sex, no
significant differences were found (see supplementary
figure 11).
National trends
At the national level, from 1990 to 2019, United Arab
Emirates had the highest increase in age standardised
prevalence of T1DM among people aged ≥65 years, with
0.28 to <0.57 0.57 to <0.76
0.76 to <0.89 0.89 to <1.04
1.04 to <1.18 1.18 to <1.36
1.36 to <1.57 1.57 to <1.95
1.95 to <2.45 2.45 to <3.61
Caribbean and Central America Persian Gulf Balkan Peninsular
Fig 2 | Map showing average annual percentage change in global prevalence of type 1 diabetes mellitus among people aged ≥65 years, 1990-2019
the bmj | BMJ 2024;385:e078432 | doi: 10.1136/bmj-2023-078432
RESEARCH
an average annual trend of 3.61%, followed by Libya
(AAPC 3.28%) and Greece (3.16%). Over the same
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
period, Cuba showed the most substantial decrease in
age standardised DALYs for older people with T1DM
(AAPC −2.40%), followed by the Republic of Korea
(−2.29%). The country with the greatest increase in age
standardised DALYs for older people with T1DM was
Uzbekistan (AAPC 3.60%). In 2019, Finland had the
highest age standardised prevalence of T1DM among
older people (1693 per 100 000 population), while
Oman had the highest age standardised DALYs for T1DM
among older people (418 per 100 000 population) (fig
2, fig 3, fig 4, and supplementary table 5).
Risk factors
A detailed analysis of global data from 1990 to 2019
revealed three primary risk factors associated with
DALYs for T1DM among people aged ≥65 years,
including high fasting plasma glucose levels, low
temperature, and high temperature. In 2019, these
factors accounted for 103 per 100 000 people, 3 per
100 000 people, and 1 DALY per 100 000 people,
respectively. From 1990 to 2019, the corresponding
AAPCs for these factors were −0.33%, −2.27%, and
1.92%. In countries with a high sociodemographic
South East Asia West Africa Eastern
Mediterranean
Northern Europe
5
RESEARCH
-7.26 to <-2.82 -2.82 to <-2.00
-2.00 to <-1.70 -1.70 to <-1.12
-1.12 to <-0.90 -0.90 to <-0.68
-0.68 to <-0.41 -0.41 to <-0.14
-0.14 to <0.44 0.44 to <4.71
Caribbean and Central America Persian Gulf Balkan Peninsular
Fig 3 | Map showing average annual percentage change in global mortality among people with type 1 diabetes mellitus aged ≥65 years, 1990-2019
index, the most substantial reduced burden from
1990 to 2019 was associated with a high fasting
plasma glucose level (AAPC −0.37%) and low
temperature (−2.65%). In contrast, countries with a
low sociodemographic index contributed the most to
the burden attributed to the three risk factors (table 2).
Discussion
Worldwide during 1990-2019, the age standardised
prevalence of T1DM increased among older people
aged ≥65 years, concomitant with a substantial
decrease in associated mortality. The results suggest
that T1DM is no longer a contributory factor in
decreased life expectancy owing to improvements in
medical care over the three decades. DALYs from T1DM
among older people decreased at a much slower rate,
with inequality identified between countries with
different sociodemographic levels. Optimal blood
glucose control remains challenging among older
people. Our study extended the current understanding
of the increasing global burden of T1DM by focusing
on trends in older people (≥65 years) with T1DM. Our
study also advocates for urgent attention to coping
strategies for ageing populations and older people
6
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
South East Asia West Africa Eastern
Mediterranean
Northern Europe
with T1DM, rational allocation of health resources, and
the provision of targeted guidelines. The findings are
important for health practice and future research, and
provide optimistic evidence for all people with T1DM,
especially those with a diagnosis at a young age.
Age differences in burden of T1DM among older
people
Our study found that in 2019, possibly owing to
notable advances in medicine, mortality from T1DM
significantly decreased and the life expectancy of
affected people improved. These findings may be
related to recent achievements in development goals
aimed at improving accessibility and coverage of
healthcare services, as well as progress in economic
growth, reduced poverty, and social protection efforts.18
We found that the prevalence of T1DM substantially
increased in every age subgroup among those aged ≥65
years. Populations worldwide are ageing, and caring
for older people with T1DM involves both healthy
ageing and management of T1DM. In recent years, with
the increasing prevalence and accessibility of scientific
technologies, an increasing number of older people are
turning to technology to improve the management of
doi: 10.1136/bmj-2023-078432 | BMJ 2024;385:e078432 | the bmj

-2.40 to <-1.29 -1.29 to <-0.91
-0.91 to <-0.69 -0.69 to <-0.51
-0.51 to <-0.33 -0.33 to <-0.15
-0.15 to <-0.10 -0.10 to <0.34
0.34 to <0.67 0.67 to <3.60
Caribbean and Central America Persian Gulf Balkan Peninsular
Fig 4 | Map showing average annual percentage change in global DALYs among people with type 1 diabetes mellitus aged ≥65 years, 1990-2019
their diabetes.19 The widespread adoption of insulin
analogues and the improved utilisation of insulin
pumps, along with improved education about diabetes
and its management, in countries such as France and
Sweden, have contributed to the improvement of the
disease burden associated with T1DM.20-22 Although
no cure exists for T1DM, the disease is manageable.23
Sociodemographic differences in burden of older
people with T1DM
The DALYs from T1DM among people aged ≥65 years
decreased slower than mortality and unequally
between countries. Firstly, against a background of
increasing T1DM prevalence, a small decrease in
DALYs is not necessarily harmful because of the huge
improvement in survival for this population, and after
accounting for more morbidity. Secondly, most of the
prevalent cases of T1DM occurred among older people
in high income regions and countries and the highest
DALYs occurred in the least developed regions and
countries. As the sociodemographic index decreased,
the burden of T1DM among older people appeared to
gradually increase. This aligns with previous studies,
which identified limited access to insulin and testing
the bmj | BMJ 2024;385:e078432 | doi: 10.1136/bmj-2023-078432
RESEARCH
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
South East Asia West Africa Eastern
Mediterranean
Northern Europe
equipment as crucial reasons why people with T1DM
in countries with a low sociodemographic index
do not receive timely management, which leads to
low awareness, poor treatment, low control rates,
and silent progression and deterioration among
patients.24 Although this unequal distribution of T1DM
associated DALYs between countries with different
sociodemographic index reduced in the past three
decades, it was still substantial in 2019.
Sex difference in burden of T1DM among older
people
In people aged ≥65 years, we observed a significantly
increased overall disease burden from T1DM among
men compared with women. T1DM is an autoimmune
disease that results from activation of immune cells
that destroy insulin producing pancreatic beta cells,
and sex hormones are major contributors to this sex
difference in the onset and progression of T1DM.
Evidence suggests that the prevalence of T1DM is
higher in girls during prepuberty, but high in men
after puberty, findings that have also been observed
in animal models.25 This sex inequality, however,
is attributed not only to physiological differences
7
RESEARCH

Table 2 | Main risk factors for age standardised T1DM related DALYs, and AAPC, among people aged ≥65 years, 1990-2019

BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
Age standardised DALYs (per 100 000) (95% UI)
Risk factors by SDI 1990 2019 AAPC (95% CI))
High fasting plasma glucose
Global 113 (89 to 137) 103 (85 to 127) −0.33 (−0.41 to −0.25)
High SDI 121 (93 to 151) 109 (82 to 143) −0.37 (−0.48 to −0.26)
High-middle SDI 95 (78 to 117) 86 (68 to 109) −0.37 (−0.47 to −0.28)
Middle SDI 105 (82 to 132) 91 (77 to 119) −0.48 (−0.55 to −0.41)
Low-middle SDI 127 (73 to 176) 128 (95 to 166) 0.01 (−0.17 to 0.20)
Low SDI 153 (84 to 230) 141 (95 to 184) −0.29 (−0.40 to −0.18)
Low temperature
Global 5 (3 to 8) 3 (1 to 5) −2.27 (−2.57 to −1.97)
High SDI 8 (4 to 12) 4 (2 to 6) −2.65 (−3.18 to −2.10)
High-middle SDI 6 (3 to 10) 3 (2 to 5) −2.37 (−2.70 to −2.04)
Middle SDI 4 (2 to 5) 2 (1 to 3) −1.84 (−2.11 to −1.57)
Low-middle SDI 3 (0.2 to 6) 2 (−0.2 to 5) −0.89 (−1.09 to −0.70)
Low SDI 3 (0.2 to 8) 3 (−0.4 to 7) −0.70 (−2.14 to 0.76)
High temperature
Global 0.7 (−2 to 3) 1 (0.2 to 3) 1.92 (1.54 to 2.29)
High SDI 0.1 (−0.1 to 0.3) 0.1 (0.004 to 0.2) 0.28 (−0.79 to 1.35)
High-middle SDI 0.2 (−0.5 to 0.9) 0.3 (0.03 to 0.9) 1.92 (1.43 to 2.42)
Middle SDI 0.9 (−3 to 5) 1 (0.2 to 4) 1.09 (0.62 to 1.55)
Low-middle SDI 2 (−7 to 10) 4 (0.6 to 9) 1.62 (1.26 to 1.97)
Low SDI 2 (−11 to 13) 4 (−0.5 to 10) 1.91 (1.58 to 2.24)
AAPC=average annual percentage change; CI=confidence interval; DALYs=disability adjusted life years; SDI=sociodemographic index; T1DM=type 1
diabetes mellitus; UI=uncertainty interval.

between the sexes in terms of anatomy and metabolic
processes but also to lifestyle, educational attainment,
socioeconomic factors, and cultural factors.26
Studies have suggested that women exhibit a more
proactive approach to healthcare utilisation and
disease management, showing a higher awareness of
their condition and greater adherence to treatment.
These factors could mitigate the risk of diabetes in
women and reduce the incidence of complications.27
Additionally, high risk factors such as smoking
and lack of physical activity may, to some extent,
account for this trend shift. These risk factors are also
considered important prognostic factors for T1DM.
Population based interventions aimed at reducing
the presence of these risk factors are therefore crucial
for managing and mitigating the disease burden of
T1DM.28 Consequently, consideration of further sex
specific treatment strategies is necessary to derive
enhanced treatment benefits, and treatment plans can
be more targeted.
Risk factors in burden of T1DM among older people
A high fasting plasma glucose level was identified as
the major contributor to DALYs from T1DM among
people aged ≥65 years, indicating that blood glucose
control is still suboptimal and a challenge among this
population. On the one hand, most people with T1DM
are unable to remain within normal glucose range
for a large part of the day.29 The difficulty of optimal
blood glucose control for people lies in the dynamic
adjustment of insulin dose shots, which depends
on factors such as daily nutrient intake, dietary
patterns, insulin sensitivity factors, and daily activity.
Difficulties, such as adjustment and calculation of
insulin sensitivity factors according to dietary patterns,
become even more challenging for older people owing
to changes in metabolism and other factors associated
with ageing. On the other hand, doctors might prefer to
deal with hyperglycaemia rather than hypoglycaemia
when caring for older people with diabetes, because
complications from hyperglycaemia take longer to
manifest than hypoglycaemia, the latter leading to
immediate adverse outcomes such as unconsciousness,
falls, brain damage, and cardiovascular events. Tighter
control measures also increase stress for older patients
at home and their caregivers. The objective should be
to aim for normoglycaemia without putting people at
risk of hypoglycaemia.30 We therefore suggest active,
but not tight, control of blood glucose, especially in
people aged ≥65 years with T1DM. The development
of self-management skills, education about nutrition,
and training are crucial for people with T1DM and
their caregivers, all of which should be highlighted in
guidelines for managing T1DM in older people. The
latest advances in a new hybrid close loop insulin
delivery system have been found to be safe and
efficient for glucose instability among young people
with T1DM,31 but evidence for its use in older people
with T1DM is lacking.
Additionally, we observed that extreme temperatures
were major contributors to the burden of T1DM among
older people. A previous study suggested a positive
correlation between high and low temperatures
and diabetes incidence and mortality, particularly
among older people.32 Temperature is one of the main
environmental factors, and changes in temperature can
influence insulin sensitivity.33 34 Another explanation
might be that temperature affects fasting plasma

8 doi: 10.1136/bmj-2023-078432 | BMJ 2024;385:e078432 | the bmj


glucose levels through alterations in dietary and
exercise patterns, heightened susceptibility to infections,
activation of the sympathetic nervous system, and the
triggering of catecholamine secretion, leading to faster
onset of diabetes.35 Also, not only does high or low
temperature result in physiological complications but
it also affects people’s daily routines, activities, and
dietary habits—all important factors in the wellbeing of
people with T1DM. Notably, we found harm to be more
noticeable during extreme temperatures in countries
with a low sociodemographic index. This finding could
be attributed to the management of T1DM, as extreme
temperatures can impair insulin storage and reduce
insulin efficacy.36 The capacity to adapt to variations
in temperature is closely associated with the economic
status of individual countries.37 Despite forecasts
indicating that the global economic burden of diabetes
in the adult population will escalate to $20tn (£16tn;
€18tn) by 2030, the unmet demands for diabetes care
in middle and low income countries persist.38 39
Strengths and limitations of this study
For older people with T1DM and their families
worldwide, the decreasing mortality and DALYs
associated with this disease is encouraging. For policy
makers, health resource preparedness is needed for the
growing number of people with both T1DM and ageing
related problems, especially in countries with a middle
and low sociodemographic index. In clinical practice,
older people with T1DM are encouraged to take
active control of their blood glucose levels, with the
development of self-management skills and education
about nutrition, and this should be reinforced in
guidelines. Suitable training is also required for doctors
and caregivers. For future research, intervention
for glycaemic control that proved effective among
adolescents and young adults with T1DM needs to be
examined for older people too, with evidence based
strategies. More innovative studies are needed in the
area of healthy ageing in people with T1DM, and the
cost effective delivery of treatment and care.
This study has several limitations. Firstly, the
data were extrapolated from countries with existing
epidemiological data. For our T1DM model, we used
DisMod-MR 2.1 to generate estimates for older age
groups based on age pattern. Interpreting the results in
the real world requires caution. Secondly, despite using
rigorous statistical methods in our study, variations in
health information systems and reporting mechanisms
across countries and regions, particularly in low and
middle income countries and in areas experiencing
conflict, could lead to incomplete data and bias,
potentially affecting the accuracy of the results.
Thirdly, the data on disease burden includes a time lag.
Fourthly, we primarily relied on modelling processes
for the estimates in this study, and the choice of models
and parameter settings could have influenced the
results. Finally, the diagnosis of T1DM in older people
presents challenges,40 41 and variations in healthcare
systems, healthcare policies, and medical practices
across countries may have an affect on disease burden.
the bmj | BMJ 2024;385:e078432 | doi: 10.1136/bmj-2023-078432
RESEARCH
As a result, further high quality real world research is
needed to validate the findings of this study.
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
Conclusions
Mortality and DALYs among older people (≥65 years)
with T1DM decreased considerably from 1990 to
2019. Both were lower in women, those living in
countries or regions with a high sociodemographic
index, and those younger than 79 years. Management
of high fasting plasma glucose levels remains a major
challenge for older people with T1DM, and targeted
clinical guidelines are needed.
Contributors: KJY, XY, and CYJ contributed equally to this work. YZL
and JZ contributed equally to this work and are joint senior authors.
YZL and JZ conceived and designed the study. SND, HS, BM, and
TTL supervised the study. KJY, XY, and CYJ performed the statistical
analysis. All authors contributed to the acquisition, analysis, or
interpretation of data. KJY, YZL, and JZ drafted the manuscript. All
authors revised the report and approved the final version before
submission. YZL is the guarantor and attests that all listed authors
meet authorship criteria and that no others meeting the criteria have
been omitted.
Funding: This work was supported by the National Natural Science
Foundation of China (grant Nos 82000753 and 82304201) and the
China Postdoctoral Science Foundation (grant Nos 2021MD703910
and 2023T160724). The funder of the study had no role in the study
design, data collection, data analysis, data interpretation, or the
writing of the report. The corresponding author had full access to all
the data in the study and had final responsibility for the decision to
submit for publication.
Competing interests: All authors have completed the ICMJE uniform
disclosure form at https://www.icmje.org/disclosure-of-interest/
and declare: support from the National Natural Science Foundation
of China and the China Postdoctoral Science Foundation for the
submitted work; no financial relationships with any organisations that
might have an interest in the submitted work in the previous three
years; no other relationships or activities that could appear to have
influenced the submitted work.
Ethical approval: Not required as this study used secondary data
aggregated at both country and global level.
Data sharing: The data used for analyses are publicly available at
https://ghdx.healthdata.org/gbd-results-tool. All data will be made
available on request to the corresponding author. Proposals will be
reviewed and approved by the sponsor, investigator, and collaborators
based on scientific merit. After approval of a proposal, data will be
shared through a secure online platform after the signing of a data
access agreement.
Transparency: The lead authors (YZL and JZ) affirm that the
manuscript is an honest, accurate, and transparent account of the
study being reported; that no important aspects of the study have
been omitted; and that any discrepancies from the study as planned
(and, if relevant, registered) have been explained.
Dissemination to participants and related patient and public
communities: Our work will be distributed to clinicians and the
wider community by a plain language summary through social
media platforms. We will also disseminate our findings through
presentations at international forums and academic societies.
Provenance and peer review: Not commissioned; externally peer
reviewed.
Publisher’s note: Published maps are provided without any warranty
of any kind, either express or implied. BMJ remains neutral with regard
to jurisdictional claims in published maps.
This is an Open Access article distributed in accordance with the
terms of the Creative Commons Attribution (CC BY 4.0) license, which
permits others to distribute, remix, adapt and build upon this work,
for commercial use, provided the original work is properly cited. See:
http://creativecommons.org/licenses/by/4.0/.
1 Livingstone SJ, Levin D, Looker HC, et al, Scottish Diabetes Research
Network epidemiology groupScottish Renal Registry. Estimated life
expectancy in a Scottish cohort with type 1 diabetes, 2008-2010.
JAMA 2015;313:37-44. doi:10.1001/jama.2014.16425
2 World Health Organization (WHO) Diabetes. 2023. https://www.who.
int/news-room/fact-sheets/detail/diabetes.
9
RESEARCH
3 Miller RG, Secrest AM, Sharma RK, Songer TJ, Orchard TJ.
Improvements in the life expectancy of type 1 diabetes: the
Pittsburgh Epidemiology of Diabetes Complications study cohort.
Diabetes 2012;61:2987-92. doi:10.2337/db11-1625
4 Toschi E. Type 1 Diabetes and Aging. Endocrinol Metab Clin North
Am 2023;52:389-403. doi:10.1016/j.ecl.2022.10.006
5 Hayes L, Cheetham T, Muirhead C, et al. Type 1 diabetes in North East
England and North Cumbria: patterns and time trends in 0-14-year-
olds from 2012 to 2020. Front Public Health 2023;11:1193403.
doi:10.3389/fpubh.2023.1193403
6 Green A, Hede SM, Patterson CC, et al. Type 1 diabetes in 2017:
global estimates of incident and prevalent cases in children and
adults. Diabetologia 2021;64:2741-50. doi:10.1007/s00125-021-
05571-8
7 Meneilly GS, Knip A, Miller DB, Sherifali D, Tessier D, Zahedi A,
Diabetes Canada Clinical Practice Guidelines Expert Committee.
Diabetes in Older People. Can J Diabetes 2018;42(Suppl 1):S283-
95. doi:10.1016/j.jcjd.2017.10.021
8 Gregory GA, Robinson TIG, Linklater SE, et al, International Diabetes
Federation Diabetes Atlas Type 1 Diabetes in Adults Special
Interest Group. Global incidence, prevalence, and mortality of type
1 diabetes in 2021 with projection to 2040: a modelling study.
Lancet Diabetes Endocrinol 2022;10:741-60. doi:10.1016/S2213-
8587(22)00218-2
9 American Diabetes Association. 16. Diabetes Advocacy: Standards
of Medical Care in Diabetes-2021. Diabetes Care 2021;44(Suppl
1):S221-2. doi:10.2337/dc21-S016
10 LeRoith D, Biessels GJ, Braithwaite SS, et al. Treatment of Diabetes
in Older Adults: An Endocrine Society* Clinical Practice Guideline. J
Clin Endocrinol Metab 2019;104:1520-74. doi:10.1210/jc.2019-
00198
11 LeRoith D, Biessels GJ, Braithwaite SS, et al. Treatment of Diabetes
in Older Adults: An Endocrine Society* Clinical Practice Guideline. J
Clin Endocrinol Metab 2019;104:1520-74. doi:10.1210/jc.2019-
00198
12 ElSayed NA, Aleppo G, Aroda VR, et al, on behalf of the American
Diabetes Association. 13. Older Adults: Standards of Care in
Diabetes-2023. Diabetes Care 2023;46(Suppl 1):S216-29.
doi:10.2337/dc23-S013
13 Diabetes Atlas IDF. International Diabetes Federation. https://
diabetesatlas.org/.
14 GBD 2019 Diseases and Injuries Collaborators. Global burden of 369
diseases and injuries in 204 countries and territories, 1990-2019:
a systematic analysis for the Global Burden of Disease Study 2019.
Lancet 2020;396:1204-22. doi:10.1016/S0140-6736(20)30925-9.
15 Zhang J, Ma B, Han X, Ding S, Li Y. Global, regional, and national
burdens of HIV and other sexually transmitted infections in
adolescents and young adults aged 10-24 years from 1990 to 2019:
a trend analysis based on the Global Burden of Disease Study 2019.
Lancet Child Adolesc Health 2022;6:763-76. doi:10.1016/S2352-
4642(22)00219-X
16 Foreman KJ, Lozano R, Lopez AD, Murray CJ. Modeling causes
of death: an integrated approach using CODEm. Popul Health
Metr 2012;10:1. doi:10.1186/1478-7954-10-1
17 Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint
regression with applications to cancer rates. Stat Med 2000;19:335-
51. doi:10.1002/(SICI)1097-0258(20000215)19:3<335::AID-
SIM336>3.0.CO;2-Z
18 Marmot M. The health gap: the challenge of an unequal world.
Lancet 2015;386:2442-4. doi:10.1016/S0140-6736(15)00150-6
19 Toschi E, Munshi MN. Benefits and Challenges of Diabetes
Technology Use in Older Adults. Endocrinol Metab Clin North
Am 2020;49:57-67. doi:10.1016/j.ecl.2019.10.001
20 Roze S, Smith-Palmer J, Valentine W, et al. Cost-Effectiveness of
Sensor-Augmented Pump Therapy with Low Glucose Suspend Versus
Standard Insulin Pump Therapy in Two Different Patient Populations
with Type 1 Diabetes in France. Diabetes Technol Ther 2016;18:75-
84. doi:10.1089/dia.2015.0224
21 Jendle J, Smith-Palmer J, Delbaere A, et al. Cost-Effectiveness Analysis
of Sensor-Augmented Insulin Pump Therapy with Automated Insulin
Suspension Versus Standard Insulin Pump Therapy in Patients
with Type 1 Diabetes in Sweden. Diabetes Ther 2017;8:1015-30.
doi:10.1007/s13300-017-0294-z
22 Iqbal A, Heller SR. The role of structured education in the
management of hypoglycaemia. Diabetologia 2018;61:751-60.
doi:10.1007/s00125-017-4334-z
23 Stephens E. Insulin therapy in type 1 diabetes. Med Clin North
Am 2015;99:145-56. doi:10.1016/j.mcna.2014.08.016
10
24 Godman B, Basu D, Pillay Y, et al. Ongoing and planned activities to
improve the management of patients with Type 1 diabetes across
BMJ: first published as 10.1136/bmj-2023-078432 on 12 June 2024. Downloaded from http://www.bmj.com/ on 17 June 2024 by guest. Protected by copyright.
Africa; implications for the future. Hosp Pract (1995) 2020;48:
51-67.
25 Chapter 15 Sex Differences in Autoimmune Type-1 Diabetes, Sex
Differences in Cardiovascular Physiology and Pathophysiology. 2019,
Pages 239-249. https://www.sciencedirect.com/science/article/abs/
pii/B9780128131978000154?via%3Dihub.
26 Nordström A, Hadrévi J, Olsson T, Franks PW, Nordström P.
Higher Prevalence of Type 2 Diabetes in Men Than in Women Is
Associated With Differences in Visceral Fat Mass. J Clin Endocrinol
Metab 2016;101:3740-6. doi:10.1210/jc.2016-1915
27 Krag MØ, Hasselbalch L, Siersma V, et al. The impact of gender on the
long-term morbidity and mortality of patients with type 2 diabetes
receiving structured personal care: a 13 year follow-up study.
Diabetologia 2016;59:275-85. doi:10.1007/s00125-015-3804-4
28 GBD 2019 Diabetes in the Americas Collaborators. Burden of
diabetes and hyperglycaemia in adults in the Americas, 1990-2019:
a systematic analysis for the Global Burden of Disease Study 2019.
Lancet Diabetes Endocrinol 2022;10:655-67. doi:10.1016/S2213-
8587(22)00186-3.
29 Agiostratidou G, Anhalt H, Ball D, et al. Standardizing Clinically
Meaningful Outcome Measures Beyond HbA1c for Type 1 Diabetes:
A Consensus Report of the American Association of Clinical
Endocrinologists, the American Association of Diabetes Educators,
the American Diabetes Association, the Endocrine Society, JDRF
International, The Leona M. and Harry B. Helmsley Charitable Trust,
the Pediatric Endocrine Society, and the T1D Exchange. Diabetes
Care 2017;40:1622-30. doi:10.2337/dc17-1624
30 Ceriello A, Prattichizzo F, Phillip M, Hirsch IB, Mathieu C, Battelino T.
Glycaemic management in diabetes: old and new approaches.
Lancet Diabetes Endocrinol 2022;10:75-84. doi:10.1016/S2213-
8587(21)00245-X
31 Bergenstal RM, Nimri R, Beck RW, et al, FLAIR Study Group. A
comparison of two hybrid closed-loop systems in adolescents and
young adults with type 1 diabetes (FLAIR): a multicentre, randomised,
crossover trial. Lancet 2021;397:208-19. doi:10.1016/S0140-
6736(20)32514-9
32 Song X, Jiang L, Zhang D, et al. Impact of short-term exposure to
extreme temperatures on diabetes mellitus morbidity and mortality?
A systematic review and meta-analysis. Environ Sci Pollut Res
Int 2021;28:58035-49. doi:10.1007/s11356-021-14568-0
33 Chondronikola M, Volpi E, Børsheim E, et al. Brown adipose tissue
improves whole-body glucose homeostasis and insulin sensitivity in
humans. Diabetes 2014;63:4089-99. doi:10.2337/db14-0746
34 Moses RG, Patterson MJ, Regan JM, Chaunchaiyakul R, Taylor NA,
Jenkins AB. A non-linear effect of ambient temperature on apparent
glucose tolerance. Diabetes Res Clin Pract 1997;36:35-40.
doi:10.1016/S0168-8227(97)01391-0
35 Li S, Zhou Y, Williams G, et al. Seasonality and temperature effects
on fasting plasma glucose: A population-based longitudinal
study in China. Diabetes Metab 2016;42:267-75. doi:10.1016/j.
diabet.2016.01.002
36 Minuto N, Tambroni B, Vannati M, et al. Diabetic ketoacidosis
caused by exposure of insulin to low temperature. Diabetes Technol
Ther 2010;12:745-6. doi:10.1089/dia.2010.0055
37 GBD 2019 Risk Factors Collaborators. Global burden of 87 risk factors
in 204 countries and territories, 1990-2019: a systematic analysis for
the Global Burden of Disease Study 2019. Lancet 2020;396:1223-49.
doi:10.1016/S0140-6736(20)30752-2.
38 Bommer C, Sagalova V, Heesemann E, et al. Global Economic Burden
of Diabetes in Adults: Projections From 2015 to 2030. Diabetes
Care 2018;41:963-70. doi:10.2337/dc17-1962
39 Manne-Goehler J, Geldsetzer P, Agoudavi K, et al. Health system
performance for people with diabetes in 28 low- and middle-income
countries: A cross-sectional study of nationally representative
surveys. PLoS Med 2019;16:e1002751. doi:10.1371/journal.
pmed.1002751
40 Thomas NJ, Jones AG. The challenges of identifying and studying type
1 diabetes in adults. Diabetologia 2023;66:2200-12. doi:10.1007/
s00125-023-06004-4
41 Harding JL, Wander PL, Zhang X, et al. The Incidence of Adult-Onset
Type 1 Diabetes: A Systematic Review From 32 Countries and Regions.
Diabetes Care 2022;45:994-1006. doi:10.2337/dc21-1752
Supplementary information: Additional methodology,
tables, and figures, and the STROBE statement
doi: 10.1136/bmj-2023-078432 | BMJ 2024;385:e078432 | the bmj