Latent profile analysis of depression among older adults living alone in China

Bailing Hou,Huijun Zhang
DOI: https://doi.org/10.1016/j.jad.2022.12.154
2023-03-15
Abstract:Background: The number of older adults living alone has increased significantly. Depression is one of the significant mental health problems they face; classifying depressive conditions into homogeneous subgroups can help discover hidden information. Methods: The data comes from the Chinese Longitudinal Healthy Longevity Survey (CLHLS). Latent profile analysis (LPA) was used to identify depression subgroups among elderly living alone, Chi-square tests and Kruskal-Wallis tests were used to univariate analysis, multinomial logistic regression was used to analyze the related factors. Results: 1831 older adults living alone were identified and classified as low-level (30.4 %), moderate-level (55.3 %) and high-level (14.4 %). All variables, except age, were significant in the univariate analysis. Multinomial logistic regression showed that not participating in exercise, sometimes interacting with friends, anxiety symptoms, and impaired IADL were associated with the moderate- and high-level of depression in older adults living alone; good or fair self-rated health and life satisfaction were associated with the low-level of depression in older adults living alone. Anxiety symptoms were associated with high-level of depression in older adults living alone compared to moderate-level; good or fair self-rated health and life satisfaction were associated with moderate-level of depression in older adults living alone. Limitations: The CES-D-10 cannot fully determine the presence of depression in elderly people living alone at high-level. Conclusions: In future primary health care, it would be more meaningful to provide targeted interventions for different subgroups of depression in older adults living alone.
What problem does this paper attempt to address?