Rumination and Loneliness Independently Predict Six-Month Later Depression Symptoms among Chinese Elderly in Nursing Homes

Abstract

Background: Previous studies conducted in Western countries independently demonstrated that loneliness and rumination are remarkable risk factors of depression among the elderly in both community and nursing homes. However, knowledge on the relationship between these three constructs among the elderly in Eastern countries is scarce. The current study aims to determine the relationship between loneliness, rumination, and depression among Chinese elderly in nursing homes.

Methods: A total of 71 elderly participants with an average age of 82.49 years completed this six-month longitudinal study. Physical reports indicated that none of the participants were clinically depressed before the study. At Time 1, their loneliness and rumination were measured using UCLA-8 Loneliness Scale and Ruminative Responses Scale. Six months later, the participants completed the Center for Epidemiologic Studies Depression Scale to assess depressive symptoms (Time 2).

Results: Multiple regression analysis revealed that both loneliness and rumination at Time 1 were the predictors of depression symptoms at Time 2 among the Chinese elderly in nursing homes. However, in the mediation analysis using PROCESS, the indirect effect between loneliness at Time 1 and depression symptoms at Time 2 was insignificant.

Conclusions: Results suggest that previous loneliness and rumination thinking are predictors of future depression symptoms among the Chinese elderly in nursing homes. However, the insignificant mediation further suggests that the differences between loneliness and rumination should be explored in future studies. Findings have important implications for mental health professionals in nursing homes in China.

Fig 1. The Estimation of the Direct and Indirect Effect of Loneliness in Affecting Rumination and Depression among the Elderly in Nursing Homes of China. (N = 71).

Note. Brackets indicated the bias-corrected and accelerated 90% confidence interval. Bolded lines indicate the significant effects. Illness history is controlled as covariates.