Impact of social integration on metabolic functions: evidence from a nationally representative longitudinal study of US older adults

Abstract

Background: Metabolic functions may operate as important biophysiological mechanisms through which social relationships affect health. It is unclear how social embeddedness or the lack thereof is related to risk of metabolic dysregulation. To fill this gap we tested the effects of social integration on metabolic functions over time in a nationally representative sample of older adults in the United States and examined population heterogeneity in the effects.

Methods: Using longitudinal data from 4,323 adults aged over 50 years in the Health and Retirement Study and latent growth curve models, we estimated the trajectories of social integration spanning five waves, 1998-2006, in relation to biomarkers of energy metabolism in 2006. We assessed social integration using a summary index of the number of social ties across five domains. We examined six biomarkers, including total cholesterol, high-density lipoprotein cholesterol, glycosylated hemoglobin, waist circumference, and systolic and diastolic blood pressure, and the summary index of the overall burden of metabolic dysregulation.

Results: High social integration predicted significantly lower risks of both individual and overall metabolic dysregulation. Specifically, adjusting for age, sex, race, and body mass index, having four to five social ties reduced the risks of abdominal obesity by 61% (odds ratio [OR] [95% confidence interval {CI}] = 0.39 [0.23, 0.67], p = .007), hypertension by 41% (OR [95% CI] = 0.59 [0.42, 0.84], p = .021), and the overall metabolic dysregulation by 46% (OR [95% CI] = 0.54 [0.40, 0.72], p < .001). The OR for the overall burden remained significant when adjusting for social, behavioral, and illness factors. In addition, stably high social integration had more potent metabolic impacts over time than changes therein. Such effects were consistent across subpopulations and more salient for the younger old (those under age 65), males, whites, and the socioeconomically disadvantaged.

Conclusions: This study addressed important challenges in previous research linking social integration to metabolic health by clarifying the nature and direction of the relationship as it applies to different objectively measured markers and population subgroups. It suggests additional psychosocial and biological pathways to consider in future research on the contributions of social deficits to disease etiology and old-age mortality.

Figure 1

Representation of the latent growth curve model and hypothesized relationships between social integration and metabolic dysregulation.

Figure 2

Estimated mean effects of social integration (modeled as latent growth curves) on individual metabolic risk factors.