Impact of Non-Adherence on Renal and Cardiovascular Outcomes in US Veterans.

BACKGROUND: Adherence is paramount in treating hypertension; however, no gold standard method is available for non-adherence screening, delineating high-risk patients. An International Classification of Diseases 9th Edition non-adherence diagnostic code (V15.81) has been available for decades; but, its utility is poorly studied. We examined the association between the V15.81 code assigned prior to the initiation of anti-hypertensive drugs (AHDs) and renal and cardiovascular outcomes. METHODS: This was a historical prospective cohort study involving 312,489 newly treated hypertensive individuals (mean age 53.8 years, 90.9% males, 20.3% black, median follow-up 8.0 years). We used crude and Cox models adjusted for baseline socio-demographic characteristics, estimated glomerular filtration rate (eGFR), body mass index, blood pressure, comorbidities, and prospective AHD adherence (measured as proportion of days covered, PDC). RESULTS: In the unadjusted analysis, the V15.81 code was associated with higher risks for faster eGFR decline (hazard ratio, HR 1.22, 95% CI 1.11-1.33), incident CKD (HR 1.17, 95% CI 1.09-1.27), end-stage renal disease (ESRD) (HR 2.53, 95% CI 1.72-3.72), incident coronary artery disease (CAD) (HR 1.26, 95% CI 1.15-1.38), and stroke (HR 1.55, 95% CI 1.38-1.73). In the adjusted model, the V15.81 code remained predictive of increased risk of CKD (HR 1.33, 95% CI 1.22-1.45), ESRD (HR 1.81, 95% CI 1.18-2.78), incident CAD (HR 1.26, 95% CI 1.14-1.40), and stroke (HR 1.46, 95% CI 1.29-1.65). Additional adjustment for PDC did not alter adverse associations between V15.81 code and studied outcomes. CONCLUSIONS: Assignment of V15.81 code prior to AHD therapy was associated with higher risks of renal and cardiovascular outcomes in incident hypertensive US veterans. Previous history of non-adherence is a poor prognostic marker in hypertensive individuals; therefore, patients with V15.81 code may require close monitoring. The observational nature of this study limits our ability to make firm recommendations for clinical practice.