Risk prediction models for type 2 diabetes using either fasting plasma glucose or HbA1c in Chinese, Malay, and Indians: Results from three multi-ethnic Singapore cohorts
Authors' conclusion
Simplified risk models including HbA1c or FPG had good discrimination in predicting incidence of T2D in three major Asian ethnic groups. Risk functions with HbA1c performed as well as those with FPG.
Study context
- Population
- Chinese, Malay and Indian adults without T2D at baseline, multiethnic Singapore cohorts; mean follow-up 8.3 yearsvalidation cohorts N = 12,720
- Participants
- 6,217 participants
- Comparison
- Established fasting plasma glucose (FPG)-based T2D risk models vs simplified models with HbA1c or FPG
- Outcome
- Discrimination (AUC) and fit of models predicting incident T2D
- Follow-up
- Average 8.3 years
Methodological detailsprospective cohort, multiethnic, external validation in two independent cohorts
- Prospective cohort
- Multiethnic
- External validation in two independent cohorts
Performance
As reported in the abstract
| Test or model | Value | AUC | Fit |
|---|---|---|---|
| Established models | 0.762 - 0.828 | 0.762 - 0.828 | P-values < 0.05 |
| Simplified models | ≥ 0.810 | ≥ 0.810 |
Value
Established models0.762 - 0.828
Simplified models≥ 0.810
AUC
Established models0.762 - 0.828
Simplified models≥ 0.810
Fit
Established modelsP-values < 0.05
Simplified models— not stated in the abstract
- Established models: moderate-to-good discrimination but lack of fit
- P < 0.05
- Simplified models: good discrimination in all cohorts, sufficiently captured differences between ethnic groups
- AUCs ≥ 0.810
Further findings
Calibration of simplified models after recalibration in validation cohorts
- Recalibration improved fit; miscalibration remained in Chinese
- P ≤ 0.012
Original publication
Risk prediction models for type 2 diabetes using either fasting plasma glucose or HbA1c in Chinese, Malay, and Indians: Results from three multi-ethnic Singapore cohorts. Diabetes Res Clin Pract. Published 15 August 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →