Dyslipidemia prevalence, treatment, and control in the Multi-Ethnic Study of Atherosclerosis (MESA): gender, ethnicity, and coronary artery calcium

Abstract

Background: To assess the implementation challenge facing the Third Report of the Adult Treatment Panel (ATP III) of the National Cholesterol Education Program, we determined the prevalence, treatment, and control of dyslipidemia, including ethnic and gender differences, in persons free of known clinical cardiovascular disease (CVD). In addition, this report provides information about the presence of coronary artery calcium (CAC) across groups defined by risk and recommendations for the use of lipid-lowering drugs.

Methods and results: The Multi-Ethnic Study of Atherosclerosis (MESA) is a multicenter cohort study of 6814 persons aged 45 to 84 years who were free of clinical CVD at baseline (2000-2002). Participants with complete fasting lipid profiles (n=6704) were evaluated for CVD risk and self-reported use of lipid-lowering therapy. CAC was assessed by CT. Drug treatment thresholds and goals were defined according to ATP III. Models were constructed to adjust for age, clinic site, risk factors, socioeconomic characteristics, and healthcare access variables with the use of Poisson regression. Overall, 29.3% (1964/6704) had dyslipidemia, among whom lipid-lowering drug therapy was reported by 54.0% (1060/1964). Control to ATP III goal was observed in 75.2% (797/1060) of participants with treated dyslipidemia and 40.6% (797/1964) of participants with dyslipidemia. Men were more likely than women to qualify for drug therapy and less likely to be treated and controlled. Relative to non-Hispanic whites, Chinese Americans were less likely to qualify for drug treatment, but no differences in treatment and control rates were observed. Black and Hispanic Americans had prevalence of dyslipidemia that was comparable to that of non-Hispanic whites but were less likely to be treated and controlled. Ethnic disparities were attenuated substantially by adjustment for healthcare access variables; however, the gender disparities persisted despite adjustment for risk factors, socioeconomic characteristics, and healthcare access variables. Control of dyslipidemia was achieved less commonly in the CVD high- and intermediate-risk groups than in the low-risk group. Among high-risk individuals, 19.7% of those who did not qualify for lipid-lowering drug treatment had CAC >400. The proportion of drug treatment-qualifying persons who were not treated differed by presence and severity of CAC, with 48.0%, 46.8%, and 39.6% of eligible persons with no CAC, with CAC >0 and <400, and with CAC >400 not receiving treatment, respectively (P for difference=0.04).

Conclusions: Dyslipidemia is common among persons without CVD. The quality of care for dyslipidemia is suboptimal in general and variable by CVD risk group, ethnicity, and gender. The utility of incorporating CAC screening into the risk stratification and treatment process should be investigated in light of the substantial proportions of persons with CAC who are currently classified as not requiring treatment. Research and quality improvement programs are needed to optimize management of dyslipidemia.