Evaluation of Diabetes Risk Models Applied to the Prediction of Incident Dysglycaemia in African and European Populations
Authors' conclusion
Existing diabetes risk prediction models can be applied to predict incident dysglycaemia in both Sub-Saharan African and European populations, but only after recalibration to local risk levels.
Study context
- Population
- Prospective cohorts of African (Ghanaian migrants and non-migrants; urban Black South Africans) and European (Swedish) populations; dysglycaemia by American Diabetes Association criteriathree cohorts
- Participants
- 1,607 / 515 / 3,044
- Comparison
- African Diabetes Risk, Finnish Diabetes Risk, DESIR and personal and clinical Framingham models
- RODAM-Prosn = 1,607
- Middle-aged Soweto cohort (MASC)n = 515
- Västerbotten Intervention Programme (VIP)n = 3,044
- Outcome
- Discrimination (c-statistic) and calibration of risk models for incident dysglycaemia, before and after recalibration
Methodological detailsthree prospective cohorts, recalibration
- Three prospective cohorts
- Recalibration
Performance
As reported in the abstract
Model performance for incident dysglycaemia
All models demonstrated poor-to-modest discrimination
c-statistic < 0.7
Calibration error after recalibration, RODAM-Pros and VIP
≤ 2.3%
Calibration error after recalibration, MASC
≤ 3.4%
Further findings
Incident dysglycaemia · median follow-up 6.7, 6.7 and 9.9 years, respectively
Original publication
Evaluation of Diabetes Risk Models Applied to the Prediction of Incident Dysglycaemia in African and European Populations. Diabetes Obes Metab. Published 1 September 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →