Abstract TP163: The Relationship Between PAP Adherence and Optimal Cardiovascular Disease Protective Regimen Prescription Depends on Race Among Veterans With Obstructive Sleep Apnea

Introduction: Obstructive sleep apnea (OSA) severity is an independent predictor of cardiovascular disease (CVD) including stroke. Black race has been associated with higher likelihood of CVD morbidity and lower PAP adherence, an unhealthy patient behavior. Data are equivocal for “healthy user” effect: whether a patient’s regular PAP use is associated with physician behavior of optimal CVD prescription regimen. It is unknown if this relationship varies by patient race. We aim to describe whether the association of PAP use with optimal CVD treatment regimen prescription depended on race in veterans with OSA. Methods: We conducted a cross-sectional study of existing PAP users at a VA Medical Center. Data on demographics, medications, comorbidities, and OSA were collected. Optimal CVD medication regimen required prescription of antiplatelet, antihypertensive, and statin. Objective: PAP use was measured by % of days with use ≥ 4 hours. Using logistic regression, we examined whether the association of patient PAP use with physician optimal medication prescription varied by race. The model included race, PAP use, PAP use by race interaction term, and pertinent covariates. Results: A total of 499 veterans (95% male, 38% Black) were included. Two hundred seventy-three veterans (55%) were prescribed an optimal CVD prevention regimen, while 226 (45%) were prescribed a suboptimal regimen. In logistic regression analyses, older age (OR 1.06, 95% CI 1.0-1.1), Black race (OR 1.8, 95% CI 1.1-2.9), higher BMI (OR 1.1, CI 1.1-1.2), and higher Charlson Index (OR 1.3, 95% CI 1.2-1.5), and PAP use by race interaction term were significantly associated with optimal medication regimen prescription. Among white veterans, level of PAP use was not associated to likelihood of optimal CVD prescriptions regimen; However, among Black veterans, those with high levels of PAP use were significantly more likely to be prescribed optimal medication regimen than individuals with low levels of PAP use. Conclusions: A “healthy user” effect was detected for the relationship between patient PAP adherence and physician optimal CVD regimen prescription in Black, but not white, veterans. Longitudinal studies are needed to determine the temporal order of these health behavior associations.