Cardiovascular risk models for South Asian populations: a systematic review.
OBJECTIVES: To review existing cardiovascular risk models applicable to South Asian populations. METHODS: A systematic review of the literature using a combination of search terms for "South Asian", "cardiovascular", "risk"/"score" and existing risk models for inclusion. South Asian was defined as those residing in or with ancestry belonging to the Indian subcontinent. RESULTS: The literature search including MEDLINE and EMBASE identified 7560 papers. After full-text review, 4 papers met the inclusion criteria. Only 1 reported formal measures of model performance. In that study, both a modified Framingham model and QRISK2 showed similar good discrimination with AUROCs of 0.73-0.77 with calibration also reasonable in men (0.71-0.93) but poor in women (0.43-0.52). CONCLUSIONS: Considering the number of South Asians and prevalence of cardiovascular disease, very few studies have reported performance of risk scores in South Asian populations. Furthermore, it was difficult to make comparisons, as many did not provide measures of discrimination, accuracy and calibration. There is a need for further research to evaluate risk models in South Asians, and ideally derive and validate cardiovascular risk models within South Asian populations.