André Hajek1, Hans-Helmut König2. 1. Department of Health Economics and Health Services Research, Hamburg Center for Health Economics, University Medical Center Hamburg-Eppendorf, Martinistr. 52, 20246 Hamburg, Germany. Electronic address: a.hajek@uke.de. 2. Department of Health Economics and Health Services Research, Hamburg Center for Health Economics, University Medical Center Hamburg-Eppendorf, Martinistr. 52, 20246 Hamburg, Germany.
Abstract
BACKGROUND: While previous studies have examined the determinants of loneliness (i) using a longitudinal approach and (ii) using data from nationally representative samples, only few studies have done both at once. Hence, the purpose of our study was to clarify which factors are associated with loneliness longitudinally based on nationally representative data. METHODS: Data were taken from wave 5 to 7 of the "Survey of Health Ageing, and Retirement in Europe "(SHARE; covering 27 European countries and Israel in total) (in our analytical sample, n = 101,909 observations). Loneliness was assessed using the three-item loneliness scale. As explanatory variables, we included age, marital status, income, self-rated health, depressive symptoms, functional decline, cognitive functioning and chronic diseases. Exploiting the features of panel data and mitigating the problem of unobserved heterogeneity, linear FE regressions were used. RESULTS: FE regressions showed that loneliness increased with increasing age (β = .02, p < .001), changes from married and living together with spouse/registered partnership to another marital status (β=-.71, p < .001), decreases in log income (β=-.01, p < .05), worsening self-rated health (β = .04, p < .001), functional decline (β = .09, p < .001), increases in depressive symptoms (β = .13, p < .001) and decreases in cognitive functioning (β=-.01, p < .001), whereas it was not associated with changes in chronic diseases. CONCLUSION: Our longitudinal study based on nationally representative SHARE data contributed to identify the determinants of loneliness among older Europeans using panel data methods. Tackling the identified risk factors may assist in avoiding loneliness in older adults living in Europe.
BACKGROUND: While previous studies have examined the determinants of loneliness (i) using a longitudinal approach and (ii) using data from nationally representative samples, only few studies have done both at once. Hence, the purpose of our study was to clarify which factors are associated with loneliness longitudinally based on nationally representative data. METHODS: Data were taken from wave 5 to 7 of the "Survey of Health Ageing, and Retirement in Europe "(SHARE; covering 27 European countries and Israel in total) (in our analytical sample, n = 101,909 observations). Loneliness was assessed using the three-item loneliness scale. As explanatory variables, we included age, marital status, income, self-rated health, depressive symptoms, functional decline, cognitive functioning and chronic diseases. Exploiting the features of panel data and mitigating the problem of unobserved heterogeneity, linear FE regressions were used. RESULTS:FE regressions showed that loneliness increased with increasing age (β = .02, p < .001), changes from married and living together with spouse/registered partnership to another marital status (β=-.71, p < .001), decreases in log income (β=-.01, p < .05), worsening self-rated health (β = .04, p < .001), functional decline (β = .09, p < .001), increases in depressive symptoms (β = .13, p < .001) and decreases in cognitive functioning (β=-.01, p < .001), whereas it was not associated with changes in chronic diseases. CONCLUSION: Our longitudinal study based on nationally representative SHARE data contributed to identify the determinants of loneliness among older Europeans using panel data methods. Tackling the identified risk factors may assist in avoiding loneliness in older adults living in Europe.
Authors: Anna-Maria Lahti; Tuija M Mikkola; Minna Salonen; Niko Wasenius; Anneli Sarvimäki; Johan G Eriksson; Mikaela B von Bonsdorff Journal: Int J Environ Res Public Health Date: 2021-11-23 Impact factor: 3.390