International Journal of Women's Health · Nov 11, 2026

Changing Global Modelled Excess Incidence of Type 2 Diabetes Among Women with Polyendocrine Metabolic Ovarian Syndrome, 1990-2023: A Risk-Translation Study

By Miao P, Jin Y, Lu J

PURPOSE: To estimate modelled excess type 2 diabetes (T2D) incidence associated with polyendocrine metabolic ovarian syndrome (PMOS), previously termed polycystic ovary syndrome (PCOS), examine benchmark-year changes and geographic heterogeneity, and assess dependence on transported cohort associations. METHODS: Global Burden of Disease (GBD) 2023 inputs for women aged 20-49 years in 204 locations were combined with a US BMI-matched/adjusted hazard ratio (HR) of 1.47 and a baseline prevalence-allocation odds ratio of 2.41. Fixed risk parameters were applied to six benchmark years during 1990-2023, with five-factor Shapley decomposition. Conditional model uncertainty intervals (MUIs) used 10,000 Monte Carlo draws; alternative cohort effects, input dependence and exclusion of ages 20-24 were examined separately. RESULTS: The 2023 country-age model estimated 78,075 annual excess cases (95% conditional MUI 67,645-88,983), or 15.3 per 10,000 T2D-free women with PMOS. This interval is conditional on the selected US source-effect distribution and simulation assumptions; it excludes between-cohort structural uncertainty. Excluding ages 20-24 yielded 69,070 cases (59,711-78,948). Benchmark estimates increased from 27,950 cases in 1990 to 77,874 in 2023, a 178.6% increase. Population growth and background T2D incidence accounted for 41.3% and 34.9% of the modelled net increase. Alternative cohort HRs generated 78,075-345,335 cases, while globally shared input ranks widened the primary-source MUI to 46,444-122,357. CONCLUSION: These conditional risk-translation scenarios describe how population inputs and selected BMI-matched/adjusted associations generate excess T2D incidence. They do not identify the total PMOS-associated burden across adiposity pathways or causally attributable events. Geographical variation and effect uncertainty require local calibration before surveillance or prevention applications.

doi.org/10.2147/ijwh.s658542Open accessPubMed 43036709Populations

AgingResearch.News