Key result
Standard CVD risk models systematically overestimate fatal CVD risk in Mexican adults despite good discrimination.
Why the study?
Most CVD risk equations were developed in high-income countries with limited validation in Latin America, where CVD is the leading cause of mortality.
Cohort (n=112,262)
No
Standard cardiovascular risk prediction models systematically overestimate 10-year fatal CVD risk in Mexican adults, highlighting the critical need for population-specific recalibration to accurately guide preventive therapies.
Answers are grounded in this paper and cite their sources.
May warrant caution with uncalibrated models in Mexican adults; leaves open prospective validation of recalibrated Globorisk-fatal.
Background: Cardiovascular disease (CVD) is the leading cause of mortality in Latin America, yet most CVD risk equations were developed in high-income countries with limited validations in these settings. Here, we externally validated CVD risk prediction models for fatal CVD, and recalibrated the Globorisk-fatal model in Mexican population. Methods: We analyzed 112,262 adults ≥40 years, 67% (75,320) female and 33% (36,942) males, mean age 54.6 ± 10.7 years without prior CVD from the Mexico City Prospective Study. The primary outcome was 10-year fatal CVD, including myocardial infarction (MI) and stroke. CVD risk was estimated using the laboratory- and office-based Framingham, Globorisk, Globorisk-LAC, WHO, and SCORE2 equations. Discrimination was assessed with Harrell's c-statistic and AUROCs, calibration with mean estimates, slopes, and calibration curves, and overall performance with Brier scores. Sex-specific recalibration of the Globorisk-fatal model was performed using observed risks of fatal CVD. Findings: During 10 years of follow-up, 2429 fatal CVD events were recorded (1667 MI, 762 stroke). All models showed good discrimination, with c-statistics ranging from 0.761 to 0.805 in males and 0.797-0.831 in females. The Globorisk-fatal model had the highest c-statistic in females (0.831, 95% CI 0.821-0.841), and the laboratory-based WHO-MI model in males (0.805, 95% CI 0.783-0.827). Despite this, all equations consistently overestimated CVD risk, particularly in females. Calibration analyses revealed systematic overprediction at higher risk levels. Recalibration of the Globorisk-fatal model improved agreement between predicted and observed risks. Interpretation: CVD risk models showed good discrimination but consistently overestimated risk in this Mexican cohort. The recalibrated Globorisk-fatal model improves risk estimation of fatal CVD in Mexican adults. Funding: This research was supported by Instituto Nacional de Geriatría in Mexico.
No takes yet. Share an insight, caveat, or question.
Perezalonso-Espinosa et al. (2026) conducted a cohort in Without prior cardiovascular disease (n=112,262). Cardiovascular risk prediction models (Globorisk, Framingham, SCORE2, WHO) vs. Observed outcomes was evaluated on 10-year fatal cardiovascular disease (c-statistic 0.831, 95% CI 0.821-0.841). Major cardiovascular risk models showed good discrimination but systematically overestimated 10-year fatal CVD risk in Mexican adults, with recalibration of the Globorisk-fatal model improving estimation.
Synapse has enriched 10 related papers. Consider them for comparative context: