Prediction of cardiovascular disease in China

Abstract

Atherosclerotic cardiovascular disease (ASCVD) is a leading cause of death globally and now also in China. Many clinical practice guidelines have been developed to help identify people at high risk of cardiovascular disease for preventive interventions using validated risk assessment tools. The recent 2013 American College of Cardiology/the American Heart Association (ACC/AHA) cholesterol management guidelines recommended statin therapy initiation for people at a 10-year ASCVD risk of ≥7.5% based on the pooled cohort equation (PCE). However, as the PCE was derived from non-Hispanic Caucasian and African Americans, the potential implications of the new guidelines in other settings, such as Mainland Chinese men and women, remains unverified.

This thesis investigated the treatment implications of the new 2013 ACC/AHA guidelines using two population-based Southern Chinese cohorts, the Guangzhou Biobank Cohort Study (GBCS) and the Hong Kong Cardiovascular Risk Factor Prevalence Study (CRISPS).

Using GBCS, I first assessed eligibility for statin treatment at baseline based on the ACC/AHA guidelines. The numbers recommended for statin use were then compared with the actual numbers being treated at follow-up. Weighting was used to extrapolate the results to the Guangzhou population aged 50 to 75 years. I, then, evaluated the validity of the PCE in GBCS by assessing discrimination and calibration using the Harrell’s concordance statistic and Hosmer-Lemeshow chisquared statistic respectively. When an overestimation or underestimation of actual risk occurred, recalibration was performed replacing the average 5-year survival rate at baseline and the mean of risk factors in GBCS while remaining all the original beta-coefficients unchanged. Finally I investigated the difference between the risk predictions in GBCS and CRISPS.

More than 50% of participants in GBCS were recommended for statin therapy based on the 2013 ACC/AHA guidelines. Only 5.1% of these people started lipid-modulating medications during an average of 7.1 years of follow-up. Among 40.6% of individuals with estimated 10-year ASCVD risk ≥7.5%, only 4.1% were started on treatment. The new guidelines had greater impact on older people aged above 60 years. Direct application of the PCE to GBCS led to systematic overestimation of the 5-year ASCVD risk. However, after recalibration, prediction of the PCE was substantially improved in both sexes. In GBCS there was a lower ASCVD death rate than in Hong Kong.

In a developing Southern Chinese setting, the prevalence of cholesterol-lowering drug use was much lower than the recommendations from the 2013 ACC/AHA guidelines. Given the escalating prevalence of dyslipidemia and increasing burden of cardiovascular disease in China, such marked shortfall in lipid drug use may suggest an opportunity for improvement of cardiovascular disease prevention in China. However, given the lower rate of ASCVD in China the lack of treatment may reflect a different level of risk. Further validation of the treatment guidelines in a Chinese context is urgently needed to facilitate ASCVD primary prevention.