Utility of China-PAR Stroke Equations for Predicting 10-Year Stroke Risk in the Rural Inner Mongolian Population in China

Background The aim of this study is to investigate the capability using the China-PAR stroke equations for predicting 10-year risk of stroke among the Inner Mongolian population in China. Methods A prospective cohort study was conducted among 2535 rural Inner Mongolian residents from June 2002 to July 2012. Participants were categorized into four subgroups according to their 10-year predicted stroke risks calculated using the China-PAR stroke equations: <5%, 5-9.9%, 10-19.9%, and >= 20%. Results The C-statistic of the China-PAR stroke equations for 10-year stroke was 0.58, and the result from Hosmer-Lemeshow 'goodness-of-fit' test showed that the China-PAR stroke equations fitted the Inner Mongolian women well (chi(2) = 11.18, P = 0.192). The adjusted hazard ratios of stroke were 3.86 (95% CI: 1.12-13.29) for 5-9.9% category, 10.37 (95% CI: 2.70-39.84) for 10-19.9% category, and 17.00 (95% CI: 3.54-81.63) for >= 20% category among Inner Mongolian women using the <5% category as reference (P for trend <0.001). However, the China-PAR stroke equations underestimated the 10-year stroke risk in Inner Mongolian men, and the calibration was unsatisfactory (chi(2) = 15.82, P = 0.045). Conclusion The China-PAR stroke equations have potential predictive ability for 10-year stroke risk in the rural Inner Mongolian women, while it might not suit the rural Inner Mongolian men well. The performance of China-PAR stroke equations in other ethnic groups in China will need to be further evaluated.