T1D Index data downloads, modelling methodology and publications
Data Downloads
Datasets are distributed in Feather format. Read them with the Arrow package in R (arrow::read_feather()) or Python (pyarrow.feather.read_feather()).
T1D Index Version 3.0
Current releaseThe recommended dataset for new work. Country-level prevalence, incidence, mortality and missing-persons estimates underpinning the IDF Diabetes Atlas, 11th Edition, plus expanded subnational coverage. Use this version unless you are reproducing an older publication.
Subnational estimates
State- and region-level incidence and prevalence for the US, Australia, the UK, New Zealand, Canada, India and Brazil. The US state-level estimates for 2024 are those reported in Wang et al., JAMA Network Open (2026). Explore them in the Data Explorer, where any view can be exported.
Subnational modelling is still experimental â results and geographic boundaries may change as methods are refined.
Previous releases
Kept available so published analyses stay reproducible. Estimates differ from Version 3.0 because the underlying methods and input data changed between releases.
Methodology
The T1D Index is a Markov model that follows simulated cohorts through onset, diagnosis, treatment and death, using every relevant incidence, prevalence and mortality input available for a country. Full derivations are in the appendices linked below.
Markov cohort model
Country populations move year by year through diagnosed and undiagnosed states, with modelled changes in incidence and mortality over time. Prevalence, missing persons and life-years lost all fall out of the same simulation, so the published measures are internally consistent.
Version 3.0 appendix âIncidence over time
Registry series are pooled to estimate how incidence has trended within and across countries, and to project forward where observations are sparse. Countries without registry data are informed by comparable settings rather than assumed static.
Version 1.0 appendix âAdult incidence
Because most registries only cover children, the Index builds an incidence-by-age curve from published studies and registry data covering adults, allowing onset after age 20 to be estimated rather than truncated â a substantial share of the global total.
Version 3.0 appendix âStandardised mortality modelling
Standardised mortality ratios are estimated from health-system and country characteristics using machine learning, so that countries without T1D mortality studies still receive an evidence-based estimate of excess mortality and of deaths before diagnosis.
Limitations and development âSubnational modelling
For the US, the Index model is applied to observed incidence from the SEARCH for Diabetes in Youth study together with adult incidence estimates and state-level demographic data. Prevalent cases are derived from state populations and the national standardised mortality ratio, which is assumed constant across states. Confidence intervals come from 1000 model runs with incidence drawn from a log-normal distribution parameterised by the SEARCH intervals.
JAMA Network Open supplement âPublications
Type 1 Diabetes Incidence and Prevalence in the US
Authors: Fei Wang; Aveni Haynes; Dana Dabelea; David M. Maahs; Catherine Pihoker; Dianna J. Magliano; Stephanie A. Pearson; Aaron Turner-Phifer; Graham D. Ogle
Journal: JAMA Network Open
Year: 2026
Global type 1 diabetes prevalence, incidence, and mortality estimates 2025: Results from the International diabetes Federation Atlas, 11th Edition, and the T1D Index Version 3.0
Authors: Graham D. Ogle; Fei Wang; Aveni Haynes; Gabriel A. Gregory; Thomas W. King; Kylie Deng; Dana Dabelea; Steven James; Alicia J. Jenkins; Xia Li; Ronald C.W. Ma; David M. Maahs; Richard A. Oram; Catherine Pihoker; Jannet Svensson; Zhiguang Zhou; Dianna J. Magliano; Jayanthi Maniam
Journal: Diabetes Research and Clinical Practice
Year: 2025
The T1D Index: Implications of Initial Results, Data Limitations, and Future Development
Authors: Graham D Ogle, Gabriel A Gregory, Fei Wang, Thomas IG Robinson, Jayanthi Maniam, Dianna J Magliano, Trevor John Orchard
Journal: Current Diabetes Reports
Year: 2023
Global incidence, prevalence, and mortality of type 1 diabetes in 2021 with projection to 2040: a modelling study
Authors: Gabriel A Gregory, MD; Thomas I G Robinson, LLB; Sarah E Linklater, PhD; Fei Wang, PhD; Prof Stephen Colagiuri, MD; Prof Carine de Beaufort, PhD; Prof Kim C Donaghue, PhD; Prof Dianna J Magliano, PhD; Jayanthi Maniam, PhD; Prof Trevor J Orchard, MD; Priyanka Rai, MSc; Prof Graham D Ogle
Journal: The Lancet Diabetes & Endocrinology
Year: 2022