| Abstrakt: |
Introduction: The 2013 pooled cohort equations (PCE) may misestimate cardiovascular event (CVE) risk, particularly for black patients. Alternatives to the original PCE (O-PCE) to assess potential statin benefit for primary prevention-a revised PCE (R-PCE) and US Preventive Services Task Force (USPSTF) algorithms-have not been compared in contemporary US patients in routine office-based practice.Methods: We performed retrospective, cross-sectional analysis of a nationally representative, US sample of office visits made from 2011 to 2014. Sampling criteria matched those used for PCE development: aged 40 to 79 years, black or white race, no cardiovascular disease. Original PCE, R-PCE, and USPSTF algorithms were applied to biometric and demographic data. Outcomes included estimated 10-year CVE risk, percentage exceeding each algorithm's statin-treatment threshold (>7.5% risk for O-PCE and R-PCE, and >10% O-PCE plus >1 risk factor for USPSTF), and percentage prescribed statin therapy.Results: In 12 556 visits (representing 285 330 123 nationwide), 10.8% of patients were black, 27.1% had diabetes, and 15.7% were current smokers. Replacing O-PCE with R-PCE decreased mean (95% confidence interval [CI]) estimated CVE risk from 12.4% (12.0%-12.7%) to 8.5% (8.2%-8.8%). Significant (P < 0.05) racial disparity in the rate of CVE risk >7.5% was identified using O-PCE (black and white patients [95% CI], respectively: 58.8% [54.6%-62.9%] vs 52.8% [51.1%-54.4%], P = .006) but not R-PCE (41.6% [37.6%-45.7%] vs 39.9% [38.3%-41.5%], P = .448). Revised PCE and USPSTF recommendations were concordant for 90% of patients. Significant racial disparity in guideline-concordant statin prescribing was found using O-PCE (black and white patients, respectively, 35.0% [30.5%-39.9%] vs 41.8% [39.9%-44.4%], P = .013), but not R-PCE (40.6% [35.0%-46.6%] vs 43.0% [40.0%-45.9%], P = .482) or USPSTF recommendations (39.0% [33.8%-44.5%] vs 44.4% [41.5%-47.5%], P = .073).Conclusions: Use of an alternative to O-PCE may reduce racial disparity in estimated CVE risk and may facilitate shared decision-making about primary prevention. [ABSTRACT FROM AUTHOR] |
