CES-D SUBCOMPONENTS AND CLINICALLY MEANINGFUL WEIGHT LOSS: LONGITUDINAL ASSOCIATIONS IN OLDER WOMEN. (11th November 2018)

• Record Type: Journal Article
• Authors: Ranker, L; Breaud, A; Heeren, T; Smith, M; Ensrud, K; Fredman, L
• Is Part Of: Innovation in aging. Volume 2(2018)Supplement 1
• Page Start: 480
• Page End: 480
• Publication Date: 2018-11-11
• Subjects: Aging -- Periodicals; Gerontology -- Periodicals; 612.67
• DOI: 10.1093/geroni/igy023.1792
• Languages: English
• ISSNs: 2399-5300

Abstract

Abstract: Depression often leads to weight loss in older adults and weight loss of 5% or greater increases the risk of mortality. Studies suggest weight loss may be driven by absence of positive affect (PA), not presence of negative affect (NA). Using data on 1027 women from the Caregiver-Study of Osteoporotic Fractures, we tested the hypothesis that women with high PA would have reduced risk of weight loss compared to women with low PA, whereas other depression components (NA, somatic symptoms, interpersonal difficulties) would be unassociated with weight loss. The 20-item Center for Epidemiologic Studies Depression Scale (CES-D) and measured weight were collected at three interviews (1999–2003; 2.01 mean years follow-up). We dichotomized each CES-D subscale score into High (top quartile) versus Low (all others), calculated annualized weight loss of 5%+ from baseline to follow-up, and imputed missing values. Logistic regression included all four CES-D subcomponents and baseline covariates. At baseline, mean age was 81.7 (+/-3.6); 35% were caregivers; 20% lost 5%+ weight over follow-up. Women with high PA were less likely to lose weight than women with low PA: adjusted odds ratio (aOR)=0.71, 95%CI 0.51–0.99. Weight loss odds was lower in women with high versus low NA (aOR=0.64, 95%CI 0.41–0.98), and non-statistically significantly elevated in those with high somatic symptoms (aOR=1.21, 95%CI 0.82–1.78) and interpersonal difficulties (aOR=1.14, 95%CI 0.71–1.82). In conclusion, both high PA and high NA were associated with reduced weight loss odds in older women. Results may reflect different underlying mechanisms or analytic choices, and warrant further exploration.