Type 1 diabetes — epidemiology and burden¶
TL;DR — A modeling study estimated 8.4 million people living with type 1 diabetes in 2021 and projected 13.5–17.4 million by 2040; most prevalent cases were adults and the modeled median onset age was 39 years (Gregory 2022, PMID 36113507). These estimates reconcile incomplete registries and should not be mistaken for direct counts. Outcome inequality is larger than the uncertainty interval: modeled remaining life expectancy for a 10-year-old diagnosed in 2021 ranged from 13 years in a low-income country to 65 years in a high-income country (Gregory 2022, PMID 36113507). In children, 41.9% worldwide present in diabetic ketoacidosis, with country estimates from 15.6% to 78.5% (Zhang 2026, PMID 42303108).
Global count and projection¶
| Measure | Estimate | Year/method | Source |
|---|---|---|---|
| People living with T1D | 8.4 million (95% UI 8.1–8.8) | 2021 microsimulation/model | Gregory 2022, PMID 36113507 |
| New diagnoses | About 0.5 million | 2021 model | Gregory 2022, PMID 36113507 |
| Deaths within 12 months of symptom onset without diagnosis | About 35,000 | 2021 model | Gregory 2022, PMID 36113507 |
| Projected prevalence | 13.5–17.4 million | 2040 scenarios | Gregory 2022, PMID 36113507 |
| Relative growth | 60–107% | 2021–2040 scenarios | Gregory 2022, PMID 36113507 |
| Adults aged 20–59 among prevalent cases | 64% | 2021 model | Gregory 2022, PMID 36113507 |
| Median age at onset | 39 years | Modeled global distribution | Gregory 2022, PMID 36113507 |
The global T1D model combines incidence, prevalence, mortality, demographic, and health-system data where direct surveillance is sparse (Gregory 2022, PMID 36113507). Its strength is comparability; its weakness is dependence on assumptions in precisely the countries where diagnosis and vital registration are least complete.
GBD estimates usually aggregate diabetes types and therefore answer a different question. The GBD 2021 diabetes analysis estimated total diabetes burden and projected more than 1.3 billion people with diabetes by 2050, but that number must not be presented as a T1D count (GBD 2021 Diabetes Collaborators 2023, PMID 37356446).
Incidence is geographically heterogeneous¶
Childhood incidence varies greatly between countries and over time. Registry comparisons are affected by age limits, case ascertainment, migration, completeness, and whether adult-onset autoimmune diabetes is recognized. The global model's adult-onset result directly challenges the common shortcut that T1D epidemiology can be measured using pediatric registries alone (Gregory 2022, PMID 36113507; Evans-Molina 2025, PMID 40230204).
Adult-onset disease is especially vulnerable to classification error. Adults may retain C-peptide, lack ketoacidosis, and initially appear to have type 2 diabetes. Genetic risk scores, autoantibodies, clinical course, and C-peptide can improve classification, but performance varies by ancestry and time from diagnosis (Sharp 2019, PMID 30655379; Evans-Molina 2025, PMID 40230204).
Diabetic ketoacidosis at diagnosis¶
The 2026 systematic review included 233 studies, 380,191 children, and 58 countries. Pooled DKA prevalence at diagnosis was 41.9% (95% CI 39.7–44.0), with reported country-level estimates ranging from 15.6% in Sweden to 78.5% in Thailand (Zhang 2026, PMID 42303108).
| Interpretation | What the number can indicate | What it cannot establish alone |
|---|---|---|
| High DKA proportion | Delayed recognition, access barriers, rapid onset, lower awareness | Which single barrier caused each case |
| Low DKA proportion | Earlier recognition, awareness, screening, health-system access | That all cases were diagnosed presymptomatically |
| Change over time | Possible effect of campaigns, access, diagnostic practices | Causality without a comparative design |
DKA is both a clinical emergency and a health-system endpoint. Population antibody screening can reduce symptomatic discovery among screened children, but campaigns and primary-care recognition may also matter; comparisons need to distinguish these mechanisms (Ziegler 2020, PMID 31990315; Zhang 2026, PMID 42303108).
Glycemic outcomes in routine care¶
In the US T1D Exchange registry during 2016–2018, only 17% of youth and 21% of adults met contemporaneous ADA HbA1c targets. Mean HbA1c peaked at 9.3% in ages 15–18, and pump/CGM uptake differed across racial and socioeconomic groups (Foster 2019, PMID 30657336).
The registry is not population-representative of all US T1D. It nevertheless shows that access to specialist centers and technology does not guarantee target attainment, and it exposes adolescence as a high-burden transition period.
Mortality and life expectancy¶
| Population | Outcome | Estimate | Source |
|---|---|---|---|
| Scotland, age 20 in 2008–2010 | Life-years lost, men | 11.1 years | Livingstone 2015, PMID 25562264 |
| Scotland, age 20 in 2008–2010 | Life-years lost, women | 12.9 years | Livingstone 2015, PMID 25562264 |
| Sweden, onset before age 10 | All-cause mortality hazard vs controls | 4.1 | Rawshani 2018, PMID 30129464 |
| Sweden, onset before age 10 | Coronary heart disease hazard vs controls | About 30 | Rawshani 2018, PMID 30129464 |
| Modeled low-income setting, diagnosis age 10 | Remaining life expectancy | 13 years | Gregory 2022, PMID 36113507 |
| Modeled high-income setting, diagnosis age 10 | Remaining life expectancy | 65 years | Gregory 2022, PMID 36113507 |
The Scottish analysis quantifies the residual gap in a high-income universal system (Livingstone 2015, PMID 25562264). Swedish registry data show that age at onset modifies lifetime cardiovascular exposure: onset before 10 years was associated with particularly high all-cause and coronary risks (Rawshani 2018, PMID 30129464). A separate Swedish analysis found residual excess mortality and cardiovascular risk across glycemic-control strata, indicating that HbA1c is central but not the sole mediator (Rawshani 2017, PMID 28402770).
Burden is multidimensional¶
Daily treatment replaces a continuously regulated hormone with repeated decisions about insulin, food, activity, illness, and sleep. Severe hypoglycemia and DKA remain acute burdens; microvascular and cardiovascular disease accumulate over decades; distress, fear, stigma, school/work disruption, and cost are not captured by HbA1c alone (Zhang 2020, PMID 33091198; Embick 2024, PMID 38361327).
Automated insulin delivery improves time in range in trials and real-world studies, but the benefits depend on device availability, training, consumable supply, and sustained use (Brown 2019, PMID 31618560; Yang 2024, PMID 38888056). Technology can therefore lower physiological burden while adding device and access burdens.
Measurement cautions¶
- Modeled is not counted. Global totals fill surveillance gaps with assumptions (Gregory 2022, PMID 36113507).
- Pediatric is not all-age. Adult onset is common and often misclassified (Evans-Molina 2025, PMID 40230204).
- Clinic registries are selected. T1D Exchange describes participating centers and users, not every person with T1D (Foster 2019, PMID 30657336).
- Diabetes-type aggregation obscures T1D. GBD all-diabetes numbers should not be relabeled (GBD 2021 Diabetes Collaborators 2023, PMID 37356446).
- Life expectancy is era-specific. Historical cohorts mix prior insulin, monitoring, and cardiovascular-treatment eras.
Incidence trajectories are regional, not universal¶
A synthesis of 51 registry studies (433,727 participants, 38 countries) estimated childhood incidence at 20.16 per 100,000 person-years (95% CI 16.49–23.84), but heterogeneity was I²=89%. European estimates averaged 22.2 per 100,000 versus 8.90 in Asia; registries using ISPAD criteria yielded 26.45 versus 19.15 with EURODIAB criteria, although the criteria subgroup difference was not statistically significant (Chauhan 2026, PMID 42551241).
| Registry/time | Quantified trajectory | Interpretation |
|---|---|---|
| Saxony, 1999–2019 | 17.1 to 24.7/100,000; 98% recent ascertainment | Rise with a recent point-prevalence plateau |
| Sweden, 1978–2007 | 21.6 to 43.9/100,000 | Long rise; youngest-cohort reversal after 2000 |
| Italy, 1990–2003 | 12.26/100,000; +2.94%/year (95% CI 2.22–3.67) | Strong regional and temporal variation |
| Austria, 1989–2021 | +4.6%/year to 2011, then 0.78%/year through 2020 | Plateau followed by a 2021 peak |
Saxony projected 34.8 per 100,000 by 2030 if its trend continued, but the 95% CI was 24.4–49.6 and its recent segment suggested slowing (Manuwald 2021, PMID 34972197). Swedish data likewise showed that a long rise can contain a younger-age reversal (Berhan 2011, PMID 21270269). Italy documented a 40% period increase and large Sardinian excess within one health system (Bruno 2010, PMID 20566665).
Adult and older-onset ascertainment gap¶
A systematic review identified 46 population-based studies from 32 countries/regions, with sparse evidence from low- and middle-income countries. Adult incidence was lowest in Asian and highest in Nordic settings, generally higher in men, and too inconsistent to establish whether it declines with age or changes over calendar time (Harding 2022, PMID 35349653). Adult autoimmune diabetes can initially resemble type 2 diabetes and need not require insulin immediately (Leslie 2021, PMID 34670785).
Reviews of onset after 60 report incidence peaks in some populations comparable to or exceeding childhood peaks, but increasing misclassification and the absence of a gold-standard discriminator destabilize comparisons (Tomic 2025, PMID 39448829). Systems defining T1D only by age at insulin initiation can undercount slowly progressive autoimmune disease and miscount insulin-treated type 2 diabetes.
Improving outcomes and residual risk¶
In 11,766 Finnish people diagnosed before age 15 and followed for a median 29.6 years, 1,761 developed cardiovascular disease. Risk fell 3.8% per later calendar year of diagnosis (HR 0.96, 95% CI 0.96–0.97), yet people diagnosed in the 1990s retained standardized incidence ratios of 8.9 (3.9–17.5) for coronary disease and 2.9 (1.3–5.7) for stroke (Harjutsalo 2021, PMID 34303414). Improvement and substantial excess are simultaneously true.
Inequity at multiple levels¶
| Level | Estimate | Boundary |
|---|---|---|
| Pump access | 61% White, 26% Black, 39% Hispanic among 10,704 US children | Registry clinics, not a probability sample |
| Glycemia | Adjusted HbA1c 9.6% Black, 8.4% White, 8.7% Hispanic | SES adjustment did not remove Black–White gap |
| Income/race | Black participants at ≥$100,000 income had higher HbA1c than White participants below $25,000 | Income and race are not interchangeable exposures |
| Pediatric utility | 0.91 pooled (95% CI 0.89–0.93); adolescents 0.85 vs children 0.90 | Six-study evidence base |
The T1D Exchange disparities also included more DKA and severe hypoglycemia among Black participants (Willi 2015, PMID 25687140). A later age-diverse analysis found a steeper inverse income–HbA1c gradient in Black than White participants, while absolute racial differences persisted at high income (Bell 2024, PMID 38965795). Health-utility synthesis included 1,276 participants and is useful for models but geographically fragile (Xie 2024, PMID 37915225).
Open questions¶
- What fraction of adult-onset T1D is currently misclassified, and how does that bias national incidence estimates? (Evans-Molina 2025, PMID 40230204)
- Which intervention explains cross-country reductions in DKA at diagnosis: antibody screening, public awareness, primary-care access, or background system strength? (Zhang 2026, PMID 42303108; Ziegler 2020, PMID 31990315)
- Will AID-era cohorts narrow the 11–13-year high-income life-expectancy gap, or will cardiovascular risk persist despite better glucose profiles? (Livingstone 2015, PMID 25562264; Rawshani 2017, PMID 28402770)
- How should global models quantify people dying before diagnosis when those deaths are least likely to be coded as T1D? (Gregory 2022, PMID 36113507)
Related pages¶
- staging and natural history — the preclinical population missing from conventional incidence counts.
- complications — drivers of long-term morbidity and mortality.
- patient experience and advocacy — burden beyond biomedical outcomes.
- screening and early detection — approaches to reducing DKA at diagnosis.
References¶
- Gregory GA, et al. Global incidence, prevalence, and mortality of type 1 diabetes in 2021 with projection to 2040. Lancet Diabetes Endocrinol. 2022;10:741-760. PMID 36113507
- Zhang T, et al. Worldwide prevalence of diabetic ketoacidosis at diagnosis of type 1 diabetes: A systematic review and meta-analysis. Prev Med. 2026;210:108625. PMID 42303108
- Livingstone SJ, et al. Estimated life expectancy in a Scottish cohort with type 1 diabetes, 2008-2010. JAMA. 2015;313:37-44. PMID 25562264
- Rawshani A, et al. Excess mortality and cardiovascular disease in young adults with type 1 diabetes in relation to age at onset. Lancet. 2018;392:477-486. PMID 30129464
- Rawshani A, et al. Mortality and Cardiovascular Disease in Type 1 and Type 2 Diabetes. N Engl J Med. 2017;376:1407-1418. PMID 28402770
- Foster NC, et al. State of Type 1 Diabetes Management and Outcomes from the T1D Exchange in 2016-2018. Diabetes Technol Ther. 2019;21:66-72. PMID 30657336
- GBD 2021 Diabetes Collaborators. Global, regional, and national burden of diabetes from 1990 to 2021, with projections to 2050. Lancet. 2023;402:203-234. PMID 37356446
- Evans-Molina C, et al. Type 1 diabetes presenting in adults: Trends, diagnostic challenges and unique features. Diabetes Obes Metab. 2025;27 Suppl 6:57-68. PMID 40230204
- Sharp SA, et al. Development and Standardization of an Improved Type 1 Diabetes Genetic Risk Score. Diabetes Care. 2019;42:200-207. PMID 30655379
- Ziegler AG, et al. Yield of a Public Health Screening of Children for Islet Autoantibodies in Bavaria, Germany. JAMA. 2020;323:339-351. PMID 31990315
- Zhang Y, et al. Fear of hypoglycemia in patients with type 1 and 2 diabetes: a systematic review. J Clin Nurs. 2020. PMID 33091198
- Embick R, et al. The impact of stigma on the management of type 1 diabetes: A systematic review. Diabet Med. 2024;41:e15299. PMID 38361327
- Brown SA, et al. Six-Month Randomized, Multicenter Trial of Closed-Loop Control in Type 1 Diabetes. N Engl J Med. 2019;381:1707-1717. PMID 31618560
- Yang Q, et al. Real-world glycaemic outcomes of automated insulin delivery in type 1 diabetes: A meta-analysis. Diabetes Obes Metab. 2024;26:3753-3763. PMID 38888056
- Chauhan R, et al. Epidemiological and clinical insights from Type 1 Diabetes Global Registries. Diabetes Metab Syndr. 2026;20:103455. PMID 42551241
- Manuwald U, et al. Trends in incidence and prevalence of type 1 diabetes in Saxony. PLoS One. 2021;16:e0262171. PMID 34972197
- Berhan Y, et al. Thirty years of nationwide childhood type 1 diabetes incidence in Sweden. Diabetes. 2011;60:577-581. PMID 21270269
- Bruno G, et al. Age-period-cohort analysis of childhood diabetes in Italy. Diabetes. 2010;59:2281-2287. PMID 20566665
- Harding JL, et al. The Incidence of Adult-Onset Type 1 Diabetes: A Systematic Review From 32 Countries and Regions. Diabetes Care. 2022;45:994-1006. PMID 35349653
- Leslie RD, et al. Adult-Onset Type 1 Diabetes: Current Understanding and Challenges. Diabetes Care. 2021;44:2449-2456. PMID 34670785
- Tomic D, et al. The epidemiology of type 1 diabetes mellitus in older adults. Nat Rev Endocrinol. 2025;21:92-104. PMID 39448829
- Harjutsalo V, et al. Long-term population-based trends in cardiovascular disease in type 1 diabetes from Finland. Lancet Diabetes Endocrinol. 2021;9:575-585. PMID 34303414
- Willi SM, et al. Racial-ethnic disparities in management and outcomes among children with type 1 diabetes. Pediatrics. 2015;135:424-434. PMID 25687140
- Bell MJ, Neff OT. Exploring the intersection of income and race in people with Type 1 diabetes. Sociol Health Illn. 2024;46:1792-1807. PMID 38965795
- Xie S, et al. Health state utility values for children and adolescents with type 1 diabetes. Diabet Med. 2024;41:e15251. PMID 37915225