Abstract P3044: Co-occurring immigrant-related sanctuary policies in U.S. and receipt of antidiabetic prescriptions

Abstract

The sociopolitical environment can affect healthcare seeking patterns for immigrants. Local-level sanctuary policies, which limit local law enforcement’s involvement with federal immigration law, may facilitate care by reducing exposure to policing, but their effects on immigrants’ healthcare seeking are understudied. For people with diabetes, receiving glucose-lowering medications is a necessary part of care. In prior work, we have not observed associations between antidiabetic prescription rates and a continuous index or high/low classification of county-level sanctuary policy environments. Questions about individual policy effects remain, but sanctuary policies rarely occur in isolation. Thus, we identified plausible groupings of sanctuary policies based on their typology or co-occurrence. We advance this work by examining associations between three policy groupings and receipt of antidiabetic prescriptions. We used the Immigrant Legal Resource Center’s data on seven county-level sanctuary policies in existence in 2017 and linked this with electronic health record data (using patient addresses) for non-pregnant immigrant and non-immigrant adults (18+ years old), diagnosed with diabetes and seen at OCHIN community health centers across the U.S. Exposures were three groups of sanctuary policies: Group (G)1) declines 287(g) and detention contracts and holds other sanctuary policies vs. only declines 287(g)/contracts; G2) 2-policy group of highly correlated policies vs. neither policy; and G3) 4-policy group of moderately correlated co-occuring policies vs. none of those policies. Using multilevel logistic regression, adjusted for individual- and county-level confounders, we quantified associations between policy groups and patients’ receipt of at least one antidiabetic prescription during the study period (2017-2019). Interaction with nativity status was examined. In a sample of 21,151 patients, we observed a moderate effect of the 4-policy group on receiving antidiabetic prescriptions (OR G3 : 1.32 (95%CI: 1.00-1.74)), but did not observe effects of the other groups (OR G1 : 1.02 (95%CI: 0.80-1.31); OR G2 : 1.13 (0.90-1.42)). No interaction between policies and nativity status was observed. Our findings suggest that some combinations of sanctuary policies are more impactful on receipt of antidiabetic prescriptions than others. Future work will examine the effect of these policies on other CVD-related outcomes.