Abstract
OBJECTIVE: To assess the relationship between risk factor clusters and cardiovascular disease (CVD) incidence in Asian and Caucasian populations and to estimate the burden of CVD attributable to each cluster. SETTING: Asia Pacific Cohort Studies Collaboration. PARTICIPANTS: Individual participant data from 34 population-based cohorts, involving 314 024 participants without a history of CVD at baseline. OUTCOME MEASURES: Clusters were 11 possible combinations of four individual risk factors (current smoking, overweight, blood pressure (BP) and total cholesterol). Cox regression models were used to obtain adjusted HRs and 95% CIs for CVD associated with individual risk factors and risk factor clusters. Population-attributable fractions (PAFs) were calculated. RESULTS: During a mean follow-up of 7 years, 6203 CVD events were recorded. The ranking of HRs and PAFs was similar for Australia and New Zealand (ANZ) and Asia; clusters including BP consistently showed the highest HRs and PAFs. The BP-smoking cluster had the highest HR for people with two risk factors: 4.13 (3.56 to 4.80) for Asia and 3.07 (2.23 to 4.23) for ANZ. Corresponding PAFs were 24% and 11%, respectively. For individuals with three risk factors, the BP-smoking-cholesterol cluster had the highest HR (4.67 (3.92 to 5.57) for Asia and 3.49 (2.69 to 4.53) for ANZ). Corresponding PAFs were 13% and 10%. CONCLUSIONS: Risk factor clusters act similarly on CVD risk in Asian and Caucasian populations. Clusters including elevated BP were associated with the highest excess risk of CVD.