Abstract P227: Association Between Depressive Symptoms and Insulin Resistance in Midlife Women: The Study of Women’s Health Across the Nation (SWAN)

Abstract

Introduction: Recent meta-analyses and cross-sectional studies propose that insulin resistance (IR) is a transient and not a lasting feature of depression. However, longitudinal evidence is lacking, particularly in midlife women across the menopause and ethnically diverse cohorts. The goals of this study were to 1) assess the longitudinal association between IR and concurrent and prior depressive symptoms among midlife women and 2) determine if these associations vary by menopausal status and ethnicity. Methods: We included 2694 women (48.4% White, 27.3% Black, 6.5% Hispanic, 8.1% Chinese, and 9.7% Japanese) from The Study of Women’s Health Across the Nation (SWAN) followed for approximately 20 years. The Center for Epidemiologic Studies Depression Scale (CES-D) score was used to assess depressive symptoms (CES-D ≥16 indicating high depressive symptoms) and IR was obtained from the Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) and log-transformed for analysis. Multivariable linear mixed regression was used to estimate IR as a function of concurrent and prior depressive symptoms adjusting for baseline and time-varying covariates. We tested interactions between depressive symptoms and 1) time-varying menopausal status and 2) race/ethnicity in relation to IR. Results: At baseline women (mean age 45.86 ± 2.70 years, 48.4% white) were premenopausal (54.7%) or early perimenopausal (45.3%). In separate adjusted models, neither concurrent high depressive symptoms nor high depressive symptoms from the prior visit were statistically associated with log-HOMA-IR. When both concurrent and prior visit depressive symptoms were included in the model, the estimated effect of concurrent high depressive symptoms was comparable and statistically significant. High depressive symptoms were associated with a 0.03 increase in log-HOMA-IR (95% CI: 0.01, 0.06), equivalent to a 3% change in HOMA-IR compared to women with low depressive symptoms. Interaction effects were detected between high depressive symptoms and race/ethnicity (P=0.005) but not with menopausal status in relation to HOMA. The relationship between concurrent high depressive symptoms and IR was driven by the association observed among Hispanic women [β=0.18, 95% CI: 0.18 (0.08, 0.27), a 20% change in HOMA-IR]. Conclusion: Current reporting of high depressive symptoms, but not prior reporting, was associated with elevated future IR over time among midlife women. This study provides longitudinal evidence in support of IR as a transient and not a lasting feature of depression. Moreover, the results suggest that Hispanic women with high depressive symptoms may be particularly susceptible to IR. Supported by NIH/DHHS {Grants U01NR004061; U01AG012505, U01AG012535, U01AG012531, U01AG012539, U01AG012546, U01AG012553, U01AG012554, U01AG012495, and U19AG063720}