Michikawa T, Nishiwaki Y, Nakano M, Iwasawa S, Yamada M, Asakura K, Yoshioka N, Kuwahara E, Takebayashi T. Am. J. Geriatr. Psychiatry 2013; 21(11): 1154-1163.
Affiliation
Department of Preventive Medicine and Public Health, School of Medicine, Keio University, Tokyo, Japan; Environmental Epidemiology Section, Center for Environmental Health Sciences, National Institute for Environmental Studies, Tsukuba, Japan. Electronic address: tmichikawa@nies.go.jp.
OBJECTIVE: Dehydroepiandrosterone sulfate (DHEAS) appears to have a protective effect against depression, but evidence from prospective cohort studies is sparse. Therefore, we examined the association between serum DHEAS levels and depressive symptoms in older community-dwelling Japanese.
DESIGN: A community-based cohort study. SETTING: Kurabuchi Town, Gunma Prefecture, Japan. PARTICIPANTS: A total of 554 residents (248 men and 306 women) age 65 years or older without depressive symptoms at baseline. MEASUREMENTS: We performed a baseline examination of the subjects between 2005 and 2006 to determine serum DHEAS levels. The subjects were categorized into three groups based on age strata- and sex-specific tertiles of DHEAS. Depressive symptoms were assessed with the Geriatric Depression Scale 15-item version (GDS-15) in face-to-face home visit interviews carried out once in 2007 and once in 2008. The association of DHEAS with depressive symptoms (GDS-15 ≥ 6) was analyzed with the use of logistic regression models.
RESULTS: The incidence of depressive symptoms was 12.1% in men and 19.6% in women. In men, the multiadjusted odds ratio of depressive symptoms was 0.24 (95% confidence interval: 0.06-0.94, Wald χ2 = 4.20, degrees of freedom = 1, p = 0.04) for the highest tertile compared with the lowest. The association observed for the highest versus the lowest remained significant even after adjustment for physical performance and cognitive function. In women, DHEAS was not associated with depressive symptoms.
CONCLUSIONS: In this study, higher serum DHEAS levels were found to be protectively and independently associated with the risk of developing depressive symptoms in men, but not in women.