Not yet independently audited
This page was written from verified sources but has not yet passed the independent citation audit. Treat its claims as provisional.
Statistics — type 1 diabetes
Last curated: 2026-10-15
Global burden
| Measure |
Estimate |
Year/population/method |
Source |
| People living with T1D |
8.4M (95% UI 8.1–8.8) |
2021 global model |
PMID 36113507 |
| New diagnoses |
~0.5M |
2021 global model |
PMID 36113507 |
| Deaths undiagnosed within 12 months of symptoms |
~35,000 |
2021 model |
PMID 36113507 |
| Projected prevalence |
13.5–17.4M |
2040 scenarios |
PMID 36113507 |
| Projected increase |
60–107% |
2021–2040 |
PMID 36113507 |
| Prevalent cases aged 20–59 |
64% |
2021 model |
PMID 36113507 |
| Median onset age |
39 years |
modeled global distribution |
PMID 36113507 |
T1D Index version 3.0 portal estimates (checked 2026-10-15)
| Measure |
Estimate |
Year/population/method |
Source |
| People living with T1D |
9,500,194 |
2025; all ages; global simulation model, version 3.0 |
T1D Index — "T1D Numbers", https://numbers.t1dindex.org/, accessed 2026-10-15 |
| New diagnoses |
513,125 |
2025; same model |
Same portal, accessed 2026-10-15 |
| Deaths |
173,516 |
2025; same model |
Same portal, accessed 2026-10-15 |
These are modeled estimates, not an enumeration or an observed trend from the 2021 estimate above. The portal identifies version 3.0 as the current release; publication/update timing within this sweep window was not established (T1D Index — "Data & Methods", https://numbers.t1dindex.org/data-methods, accessed 2026-10-15).
Diagnosis and natural history
| Measure |
Estimate |
Population/method |
Source |
| Multiple-antibody progression at 10 years |
69.7% (95% CI 65.1–74.3) |
pooled prospective children |
PMID 23780460 |
| Multiple-antibody progression at 15 years |
84% |
pooled prospective children |
PMID 23780460 |
| Single-antibody progression at 10 years |
14.5% |
pooled prospective children |
PMID 23780460 |
| No-antibody diabetes by age 15 |
0.4% |
pooled prospective children |
PMID 23780460 |
| Fr1da presymptomatic yield |
0.31% (95% CI 0.27–0.35) |
90,632 Bavarian children |
PMID 31990315 |
| Pediatric DKA at diagnosis |
41.9% (95% CI 39.7–44.0) |
233 studies; 380,191 children; 58 countries |
PMID 42303108 |
| Country DKA range |
15.6–78.5% |
meta-analysis country estimates |
PMID 42303108 |
Risk refinement and monitored diagnosis
| Measure |
Estimate |
Population/method |
Source |
| IA-2A and progression, single-antibody state |
5.3-fold higher 5-year risk |
4,577 antibody-positive TrialNet relatives |
PMID 40016443 |
| IA-2A and progression, stage 1 |
2.2-fold higher 5-year risk |
TrialNet relatives |
PMID 40016443 |
| IA-2A and progression, stage 2 |
1.3-fold higher 5-year risk |
TrialNet relatives |
PMID 40016443 |
| Five-year risk across antibody-titer strata |
6–75% |
1,604 confirmed-positive children |
PMID 34758977 |
| Screen at age 10 predicting diabetes by 18 |
sensitivity 90% (95% CI 86–95); PPV 66% |
genetically/familially enriched adolescents |
PMID 36681087 |
| Screens at ages 10 and 14 |
sensitivity 93%; PPV 55% |
same harmonized cohorts |
PMID 36681087 |
| HbA1c ≥5.7%, one-year progression with one antibody |
children 38%; adults 13% |
5,024 antibody-positive TrialNet relatives |
PMID 42090204 |
| TEDDY DKA under surveillance |
6.1% (23/379) |
prospectively monitored children |
PMID 35043162 |
| TrialNet single-center DKA among progressors |
0/51 |
4,046 relatives; median 9.9-year follow-up |
PMID 40439773 |
| Three-month secretion-loss marker |
specificity 95% (95% CI 86–100); sensitivity 19% (8–30) |
TN-10 metabolic reanalysis |
PMID 39560746 |
Screening economics, assays, and burden
| Measure |
Estimate |
Context |
Source |
| ASK research-program cost per case detected |
$4,700 |
Colorado model |
PMID 32327420 |
| ASK routine-care scenario cost per case detected |
$14,000 |
Colorado model |
PMID 32327420 |
| Scenario needed for conventional cost-effectiveness |
20% DKA reduction plus sustained 0.1-point HbA1c benefit |
model assumption, not observed effect |
PMID 32327420 |
| Initial parental anxiety after positive screen |
46.1±11.2 |
280 ASK families; clinical cutoff 40 |
PMID 37673098 |
| Initial accurate understanding of increased risk |
48.9% |
ASK follow-up |
PMID 37673098 |
| Harmonized IA-2A performance |
>99% specificity; 64% sensitivity |
assay-harmonization program |
PMID 20444913 |
| GADA retest discordance |
15.4% to 2.7% |
TEDDY samples before vs after harmonization |
PMID 20444913 |
Treatment effects
| Intervention/outcome |
Estimate |
Design |
Source |
| DCCT retinopathy onset |
76% reduction |
RCT |
PMID 8366922 |
| DCCT microalbuminuria |
39% reduction |
RCT |
PMID 8366922 |
| DCCT clinical neuropathy |
60% reduction |
RCT |
PMID 8366922 |
| CGM HbA1c difference, injection users |
−0.6% (95% CI −0.8 to −0.3) |
24-week RCT |
PMID 28118453 |
| iDCL TIR difference |
+11 points (95% CI 9–14) |
6-month RCT |
PMID 31618560 |
| Very-young-child closed-loop TIR |
+8.7 points (95% CI 7.4–9.9) |
crossover RCT |
PMID 35045227 |
| Bionic pancreas HbA1c |
−0.5% (95% CI −0.6 to −0.3) |
13-week adult RCT |
PMID 36173236 |
| Teplizumab time to stage 3 |
48.4 vs 24.4 months |
76-person RCT |
PMID 31180194 |
| Teplizumab stage-3 hazard |
HR 0.41 (95% CI 0.22–0.78) |
RCT |
PMID 31180194 |
| Verapamil C-peptide |
30% higher at 52 weeks |
pediatric RCT |
PMID 36826844 |
| Zimislecel insulin independence |
10/12 full-dose recipients |
phase 1–2 uncontrolled |
PMID 40544428 |
| HCL meta-analysis TIR |
+10.87 points (95% CI 9.38–12.37) |
22 trials lasting 12–96 weeks |
PMID 37759290 |
| Youth AID meta-analysis TIR |
+11.5 points (95% CI 9.3–13.7) |
randomized trials |
PMID 40920375 |
| Youth AID nighttime TIR |
+19.7 points (95% CI 17.0–22.4) |
randomized trials |
PMID 40920375 |
| Real-world AID TIR |
+11.61 points (95% CI 10.47–12.76) |
before–after meta-analysis; 101,704 users |
PMID 38888056 |
| HARPdoc vs BGAT severe-event rate |
IRR 1.25 (95% CI 0.51–3.09) at 12 months |
active-comparator RCT |
PMID 35484106 |
| HypoCOMPaSS severe events |
8.9 to 0.4/person-year at 24 months |
intervention cohort follow-up |
PMID 29661916 |
| Immunotherapy pooled C-peptide |
SMD 0.221 (95% CI 0.069–0.373) |
19 trials; 1,852 participants |
PMID 41267047 |
| Immunotherapy pooled HbA1c |
SMD 0.033 (95% CI −0.070 to 0.136) |
same meta-analysis |
PMID 41267047 |
| Stage-1 abatacept, high-secretor subgroup |
HR 0.46 (95% CI 0.25–0.84) |
post hoc TrialNet reanalysis |
PMID 41237315 |
Cell-replacement outcomes and harms
| Outcome |
Estimate |
Population/design |
Source |
| Islet-transplant composite at 1 year |
87.5% |
48-person phase 3; HbA1c <7% without severe hypoglycemia |
PMID 27208344 |
| Islet-transplant composite at 2 years |
71% |
same cohort |
PMID 27208344 |
| Bleeding requiring transfusion |
5/48 (10.4%) |
75 portal-infusion procedures |
PMID 27208344 |
| Ten-year insulin independence |
28% (95% CI 13–45) |
28-person French cohort |
PMID 31615852 |
| Ten-year graft function |
78% (95% CI 57–89) |
same cohort |
PMID 31615852 |
| TRIMECO modified β-score ≥6 |
64% (95% CI 43–82) vs 0% (0–15) |
randomized islets vs intensive insulin at 6 months |
PMID 29776895 |
| TRIMECO portal-infusion bleeding |
4/55 infusions (7%) |
randomized trial/transplant procedures |
PMID 29776895 |
| CITR 3-year insulin independence |
27% to 44% |
1999–2002 vs 2007–2010 eras |
PMID 22723582 |
Outcomes and inequity
| Measure |
Estimate |
Population/year |
Source |
| Life-years lost at age 20, men |
11.1 years |
Scotland 2008–2010 |
PMID 25562264 |
| Life-years lost at age 20, women |
12.9 years |
Scotland 2008–2010 |
PMID 25562264 |
| Remaining life expectancy after diagnosis age 10 |
13 years |
modeled low-income setting, 2021 |
PMID 36113507 |
| Remaining life expectancy after diagnosis age 10 |
65 years |
modeled high-income setting, 2021 |
PMID 36113507 |
| ADA HbA1c goal attainment, youth |
17% |
T1D Exchange 2016–2018 |
PMID 30657336 |
| ADA HbA1c goal attainment, adults |
21% |
T1D Exchange 2016–2018 |
PMID 30657336 |
| Mean HbA1c ages 15–18 |
9.3% |
T1D Exchange 2016–2018 |
PMID 30657336 |
| Diabetes-distress prevalence |
38.9% |
global systematic review/meta-analysis; heterogeneous definitions |
PMID 42486409 |
| DEPS-R positive screen |
13.2% (95% CI 8.8–17.5) |
228 people with T1D in Hong Kong |
PMID 37259043 |
| Interview-defined eating disorder |
4.4% (95% CI 2.1–7.9) |
same study |
PMID 37259043 |
| Impaired awareness in UK adult cohort |
21% |
782 adults; 89% using CGM |
PMID 40386839 |
| Recent severe hypoglycemia in UK adult cohort |
5.3% |
same cross-sectional cohort |
PMID 40386839 |
| Severe-event odds per 1-point glucose CV |
OR 1.14 |
same cohort |
PMID 40386839 |
PROTECT per-protocol update
| Measure |
Estimate |
Population/year/method |
Source |
| Insulin dose, teplizumab minus placebo at week 78 |
−0.17 U/kg/day (95% CI −0.26 to −0.08); nominal P<0.001 |
2026 PP report: 180 vs 95 participants; observed insulin-dose data at week 78 in 86 vs 45; missing values multiply imputed |
PMID 42720752, Table 2 |
| HbA1c change difference at week 78 |
−0.13 percentage points (95% CI −0.46 to 0.20); nominal P=0.454 |
Same PP report; not a new randomized trial |
PMID 42720752, Table 2 |
These selected-population estimates supplement the original intention-to-treat results and do not resolve long-term clinical effectiveness.
Known conflicts and caveats
- Global counts are modeled, not complete enumeration (PMID 36113507).
- GBD all-diabetes totals must not be relabeled as T1D-specific (PMID 37356446).
- Birth-cohort progression estimates are strongest for genetically enriched children, not adult-onset disease.
- Trial effects are comparator-, device-, and support-dependent.
- Life expectancy is era- and health-system-specific; conflicting estimates should remain side by side.
- Registry goal attainment describes participating clinics, not a national probability sample.
- Screening cost-effectiveness rows are model outputs that depend on assumed DKA and long-term HbA1c effects.
- Subgroup hazard ratios, particularly stage-1 abatacept high secretors, are hypothesis-generating until prospectively replicated.
- Transplant success estimates come from highly selected people with severe hypoglycemia and must be paired with procedural, renal, infection, and immunosuppression outcomes.
- Psychosocial prevalence depends strongly on instrument, threshold, language, and sampling frame.