Background: Two classes of newer glucose-lowering drugs (GLDs), sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists, improve cardiovascular and renal outcomes among patients with type 2 diabetes (T2D). However, racial and ethnic minority groups carry higher cardiovascular risks but have lower access to newer GLDs. Contextual-level social determinants of health (SDOH) may be the underlying factor associated with newer GLD adoption.
Objective: To identify the association between contextual-level SDOH and real-world adoption of newer GLDs among Medicare beneficiaries and to examine the nonstationarity in the associations.
Methods: Data were from 15% random samples of January 2017 to December 2018 nationwide Medicare beneficiaries. We identified patients with T2D who did not use newer GLDs in the year before the index date-January 1, 2018-and followed the cohort for 1 year to record their status of initiating a newer GLD. We used a geographically weighted multivariable Poisson regression model to determine to what extent the SDOH-newer GLD initiation association (β coefficient) varied geographically.
Results: We identified 795,469 eligible Medicare beneficiaries with T2D during the study period from our dataset. Of the study cohort, mean age was 73.1 (SD = 10.5) years, 424,312 (53.3%) were female, 562,994 (70.8%) were non-Hispanic White, 96,891 (12.2%) were non-Hispanic Black, 84,744 (10.6%) were Hispanic, and 29,645 (3.7%) were Asian/Pacific Islander. Newer GLD initiation was negatively associated with the percentage of the population reporting non-Hispanic Black race, Hispanic ethnicity, and unemployment, as revealed by nonspatial regression analyses. The county-level median household income was also associated with higher newer GLD initiation. The spatial analysis presented distinct distributions of local parameter estimates for each contextual-level SDOH.
Conclusions: We identified key contextual-level SDOH associated with real-world adoption of newer GLDs and explored their geographic variation through spatially explicit, data-driven analytical approaches. Identifying areas of strong association between SDOH and newer GLD initiation is crucial for policymakers to allocate resources and develop interventions that address structural inequities.