Longitudinal Changes in Allostatic Load during a Randomized Church-based, Lifestyle Intervention in African American Women

Abstract

Introduction: African American (AA) women have disproportionately higher risk of cardiovascular disease than White women, which may be explained by the uniquely higher allostatic load (AL) found in AA women. No studies have tested the effect of lifestyle interventions on AL in AA women. Our objectives were to assess the change in allostatic load following a lifestyle intervention and explore the roles of lifestyle behaviors and socioeconomic factors on allostatic load change.

Methods: Participants were non-diabetic (mean age and SD: 48.8±11.2 y) AA women (n=221) enrolled in a church-based, cluster randomized trial testing a standard diabetes prevention program (DPP) and a faith-enhanced DPP with 4-months of follow-up. We assessed the relationships of changes in diet, physical activity, neighborhood disadvantage, individual socioeconomic factors, and other lifestyle variables to changes in AL at 4-months using a multilevel multinomial logistic regression model.

Results: Average AL decreased (-.13±.99, P=.02) from baseline to 4-months. After adjusting for other variables, a high school education or less (OR:.1, CI:.02-.49) and alcohol use (OR: .31, CI: .09-.99) contributed to increased AL. Living in a disadvantaged neighborhood was responsible for increased AL, though it was not statistically significant. There were no statistically significant associations between AL and other health behavior changes.

Conclusions: Lower education levels may dampen the benefits of lifestyle interventions in reducing AL. Although a significant reduction in AL was found after participation in a lifestyle intervention, more research is needed to determine how lifestyle behaviors and socioeconomic factors influence AL in AA women.

Keywords: Allostatic Load; Church-based Program; Health Behaviors; Lifestyle Intervention.

Conflict of interest statement

Competing Interests: None declared.

Figures

Figure 1.. Two-level hierarchical multinomial logistic regression model showing cluster randomization and church-level correlation