Abstract P081: Psychosocial Resources Are Protective of Diabetes Among African Americans in the Jackson Heart Study

Abstract

Introduction: African Americans (AAs) have among the highest prevalence of type 2 diabetes in the U.S. Research has shown that positive affect and supportive networks are associated with better health outcomes and may improve regulation of physiological processes. We examined the extent to which psychosocial resources were protective of diabetes outcomes among a sample of 5,306 AAs. Hypothesis: Psychosocial resource measures are inversely associated with prediabetes and diabetes [defined by hemoglobin (Hb)A1c categories] and prevalent diabetes (defined by self-report diabetes status and medication use). Methods: Using data from the Jackson Heart Study (JHS), we evaluated the cross-sectional associations of four psychosocial-resource indicators (social support, optimism, religiosity, social networks) with two diabetes outcomes [1) HbA1c categories: normal (HbA1c ≤ 5.7%), at risk/pre-diabetic (5.7% < HbA1c < 6.5%), diabetic (HbA1c ≥ 6.5%) and 2) prevalent diabetes (vs. no diabetes)]. For each psychosocial-resource measure, we created high vs. low categories (median split) and continuous standard deviation (SD) units. Associations with HbA1c categories were examined using multinomial logistic regression to estimate odds ratios (OR 95% confidence interval-CI) of pre-diabetes (vs. normal) and diabetes (vs. normal). Associations with prevalent diabetes were examined using Poisson regression to estimate prevalence ratios (PR 95% CI) of diabetes (vs. no diabetes). Models adjusted for demographics, SES, waist circumference, health behaviors, and depression. Results: Participants with diabetes reported fewer psychosocial resources than those with pre-diabetes and normal HbA1c ( p <0.01). After full adjustment, 1-SD unit increase in social support was associated with an 11% lower odds of pre-diabetes (vs. normal HbA1c) (OR 0.89, 95% CI 0.81-0.99). High (vs. low) religiosity was associated with an increased odds of diabetes (vs. normal Hba1c) (OR 1.29, 95% CI 1.01-1.64) after full adjustment. Optimism and social networks were only associated with lower diabetes prevalence after adjustment for demographics and education, respectively. Conclusion: With the exception of religiosity, psychosocial-resource measures were inversely associated with diabetes. Social support and social networks, especially, should be considered when addressing the reduction of diabetes burden among AAs.