Abstract MP62: The Role of Primary Care Physician Visits in Promoting Medication Adherence as Part of a Randomized Controlled Diabetes Management Intervention in American Samoa

Abstract

Background: The Native Hawaiian/Pacific Islander group is one of the fastest growing US ethnic groups, and Samoans make up the second largest Pacific Islander group in the US. Samoans are at a disproportionately high risk of type 2 diabetes mellitus (T2DM) and other cardiovascular disease risk factors, with prevalence rates of T2DM in adults in the US Territory of American Samoa around 20%, nearly double that of US Hispanics. A randomized, controlled trial of a community health worker (CHW)-facilitated diabetes management intervention in American Samoans with T2DM showed improved HbA1c levels in part through promoting health behaviors associated with diabetes management, including diet, physical activity, and most strongly, medication adherence. The current study sought to examine a potential intermediate variable through which the CHW intervention influenced health behaviors, in particular, whether the CHW intervention resulted in more frequent primary care physician (PCP) visits, which in turn, were associated with increased engagement in health behaviors. We also examined if relationships differed by baseline PCP utilization. Methods: Participants were 266 Samoan adults diagnosed with T2DM, randomized to the CHW intervention or wait-list control condition. Participants were additionally classified as meeting American Diabetes Association guidelines for PCP utilization at baseline (i.e., ≥4 PCP visits in the year prior to the intervention, “high utilizers”) or not (“low utilizers”). Regression models were used to examine the association between treatment assigned and frequency of PCP utilization, and whether PCP utilization was associated with the probability of engaging in diabetes management health behaviors (at least moderate intensity physical activity, healthy diet, medication adherence). Results: After adjustment for covariates including age, gender, and comorbidities, results indicate that CHW participants had greater rates of PCP visits over the intervention period, but only amongst low utilizers (RR = 1.94, 95% CI= 1.27, 2.97). A greater number of PCP visits, in turn, was associated with a higher odds of medication adherence (but not diet or physical activity), only amongst low utilizers (OR = 1.40, 95% CI= 1.12, 1.74). Conclusions: Results suggest that a CHW-facilitated diabetes intervention in the Samoan population may have promoted diabetes medication adherence (which has previously been associated with lower HbA1c in this cohort) by increasing the frequency with which participants encountered and interacted with their PCPs, specifically those participants in greatest need (i.e., those who had fewer PCP visits than recommended at baseline). Future research should further examine how increasing engagement with PCPs can serve as a mechanism through which to elicit behavior change in diabetic minority populations at high risk for cardiovascular disease.