Effects of Multilevel and Multidomain Interventions on Glycemic Control in U.S. Hispanic Populations.

Hispanic populations in the U.S. have a high prevalence of type 2 diabetes and its complications. It has been proposed that interventions targeting multiple levels and domains of influence are needed to address health disparities, but more evidence is needed regarding the most effective approaches. We aimed to review the effects of non-pharmacological interventions on glycemic control among Hispanic persons with diabetes, overall and by level and domain of intervention. A systematic review (PubMed, Scopus, PsycInfo, CINAHL; 1985-2019) identified randomized trials reporting HbA1c outcomes for Hispanic populations. Article review, data extraction, and risk of bias assessment were completed by independent reviewers. Level and domain of intervention were assigned based on the National Institute on Minority Health and Health Disparities Research Framework. Random-effects meta-analyses estimated pooled effect sizes. Quality of evidence was rated based on the GRADE framework. Forty-eight trials met inclusion criteria, representing various Hispanic populations (n = 18 Mexican, n = 5 Puerto Rican, n = 1 Dominican, n = 4 multiple, n = 20 unspecified) and enrolling 9185 total participants. Overall, interventions decreased HbA1c by -0.32% (95% CI: -0.44% to -0.20%, I2 = 68%, strength of evidence: moderate). Multi-level, multi-domain interventions decreased HbA1c by -0.41% (-0.61% to -0.21%, I2 = 74%, strength of evidence: moderate). Few interventions addressed community (n = 3), society (n = 0), or physical/built environment (n = 1). Non-pharmacological interventions have modestly decreased HbA1c among Hispanic persons with diabetes. Multi-level, multi-domain interventions are promising, but more research is needed on interventions that target social and environmental structures.