Abstract
BACKGROUND: Existing satisfaction measures cover general feelings of well-being among older adults, but it is not clear whether life satisfaction is associated with depressive symptoms that decrease psychological well-being. We developed a new life satisfaction scale to assess the associations of health, social factors, and interpersonal relationships with overall life satisfaction. The structural and predictive validity of the scale regarding the onset of depressive symptoms was examined. METHODS: A 13-item questionnaire was developed based on a literature review. The response options for all of the questions were: 1 = poor, 2 = not very good, 3 = good, and 4 = excellent. For the analysis, a total satisfaction score was calculated by summing the individual scores (range = 13-52), and higher scores indicated higher overall satisfaction. Baseline data were obtained using the Geriatric Depression Scale-Short Form with a 30-month follow-up. Older Japanese adults (n = 1,792, mean age 70.1 ± 6.3 years, 46.4% male) participated in both surveys. An exploratory factor analysis and a logistic regression analysis were used to verify the construct and predictive validity. RESULTS: In the exploratory factor analysis, a four-factor structure consisting of Personal Satisfaction, Societal Satisfaction, Community Satisfaction, and Health Satisfaction explained 63% of the total variance, with factor weights between 0.42 and 0.90, and internal reliability was acceptable (Cronbach's alpha = 0.86). About 13.8% of the sample was identified as having depressive symptoms during the follow-up period. The Youden index determined the cutpoints regarding the development of depressive symptoms identified by the life satisfaction scale as 39 points. The participants were classified as high life satisfaction (60%, n = 1,068) or low life satisfaction (40%, n = 724). The logistic regression analysis revealed that the scale (cutpoint at 39 points) predicted depressive symptoms while controlling for the effects of other relevant variables (OR = 5.14, 95% Confidence Interval 3.76-7.04, p < 0.001). The relative risk of the low life satisfaction group for developing depressive symptoms compared to the high life satisfaction group was 2.39. CONCLUSIONS: The 13-item questionnaire is a valid instrument for measuring the risk of depressive symptoms in older adults.