Sara Carmel1,2, Hava Tovel1,2, Victoria H Raveis3, Norm O'Rourke1,2. 1. Center for Multidisciplinary Research in Aging, Ben-Gurion University of the Negev, Beer-Sheva, Israel. 2. Department of Public Health, Ben-Gurion University of the Negev, Beer-Sheva, Israel. 3. Psychosocial Research Unit on Health, Aging and the Community, College of Dentistry and Nursing, New York University, New York, New York.
Abstract
OBJECTIVES: To assess the causal directions of interaction between depression and decline in will to live (WTL) over 2 years in community-dwelling older adults. DESIGN: Community-based longitudinal study. SETTING: Three large cities in Israel. PARTICIPANTS: Individuals aged 75 and older (N=870). MEASUREMENTS: We tested a cross-lagged structural equation model in which WTL and depressive symptoms (DS) were tested as latent variables at each of three annual points of measurement. RESULTS: Of the health and sociodemographic covariates examined, only self-rated health predicted WTL and DS, and age predicted WTL. WTL predicted DS at each point of measurement. In addition to these cross-sectional effects, WTL also predicted DS 12 months later at the second and third year. In other words, the direction of prediction was exclusively from WTL to DS, concomitantly and over time. CONCLUSION: Our results indicate that WTL predicts DS rather than vice versa, suggesting that decline in WTL contributes and leads to depression at present and in future. Health strategies designed to promote WTL in older adults may help forestall depression. Moreover, psychotherapeutic interventions targeting WTL might be effective in the treatment of depression.
OBJECTIVES: To assess the causal directions of interaction between depression and decline in will to live (WTL) over 2 years in community-dwelling older adults. DESIGN: Community-based longitudinal study. SETTING: Three large cities in Israel. PARTICIPANTS: Individuals aged 75 and older (N=870). MEASUREMENTS: We tested a cross-lagged structural equation model in which WTL and depressive symptoms (DS) were tested as latent variables at each of three annual points of measurement. RESULTS: Of the health and sociodemographic covariates examined, only self-rated health predicted WTL and DS, and age predicted WTL. WTL predicted DS at each point of measurement. In addition to these cross-sectional effects, WTL also predicted DS 12 months later at the second and third year. In other words, the direction of prediction was exclusively from WTL to DS, concomitantly and over time. CONCLUSION: Our results indicate that WTL predicts DS rather than vice versa, suggesting that decline in WTL contributes and leads to depression at present and in future. Health strategies designed to promote WTL in older adults may help forestall depression. Moreover, psychotherapeutic interventions targeting WTL might be effective in the treatment of depression.