To investigate multiple medication adherence (MMA) and its impact on microvascular and macrovascular complications using instrumental variables (IVs).
A retrospective observational study was conducted using administrative claims and electronic medical records from a large physician group in Southern California (N=2334).
We identified individuals between January 2006 and June 2009 newly starting oral diabetes (DM) or hypertension (HTN) medications with preexisting comorbid HTN or DM prescription history.
MMA was defined as a proportion of days covered where both DM and HTN medications were simultaneously available over a 33-month follow-up period. Microvascular or macrovascular complications included myocardial infarction, stroke, renal failure, and diabetic retinopathy. Multivariable logistic regressions and an IV estimation using physician-related variables were implemented.
MMA was supoptimal as the mean (SD) proportion of days covered was 0.53 (0.32). Patients were more adherent to medications for a preexisting condition in comparison with those for the newer disease. Older age, number of index medications [OR (95% CI)=1.36 (1.22–1.52)], receiving care from a physician who prescribed statin more frequently [OR (95% CI)=2.63 (1.67–4.14)], and receiving care from the same physician for both DM and HTN [OR (95% CI)=1.57 (1.08–2.27)] were significant factors of being adherent. Using physician-related IVs, MMA reduced microvascular and macrovascular complications. The increase in MMA from 50% to 80% reduced the average predicted probability of microvascular or macrovascular complication rate by 29.5%.
Adherence to medications for DM and HTN were differed and higher MMA reduced microvascular or macrovascular complications when controlling for endogeneity bias.