Latent profile analysis of cognitive function and depressive symptoms in Chinese elderly: results from the China Health and Retirement Longitudinal Study

Abstract

Background: Geriatric cognitive impairment often occurs in conjunction with depressive symptoms. This study focuses on categorising the Chinese elderly with such co-occurring symptoms into homogeneous groups using latent profile analysis (LPA), a person-centred statistical approach. Methods: Information on cognitive function and depressive states of the elderly was extracted from the China Health and Retirement Longitudinal Study. The underlying characteristics were identified by LPA, and based on those findings, differences in demographic characteristics of different subgroups were explored by chi-squared test and analysis of variance. Results: A total of 6710 Chinese elderly who met the inclusion criteria were selected from the dataset. Four subgroups were identified among this sample by LPA, based on cognitive function associated with depressive symptoms, and named in this study as follows: mild cognitive impairment (n = 3747, 55.84%), moderate cognitive impairment (n = 1306, 19.46%), mild cognitive impairment combined with depressive symptoms (n = 1114, 16.6%), and moderate cognitive impairment combined with depression (n = 543, 8.09%). Age, gender, marital status, and educational level were all significantly different across subgroups (P < 0.001); religious belief and pension mode, however, were not (P > 0.05). Conclusions: In the present study, four subgroups of cognitive function combined with depressive symptoms were found in Chinese elderly individuals, and differences in demographic factors were noted between the subgroups. In clinical practise, these findings could help clinical workers identify patients accurately and consider the demographic features of each subgroup when designing medical treatment, care, and rehabilitation programmes for those with cognitive impairment and concomitant depressive symptoms.

Paranormal belief, psychopathological symptoms, and well-being: Latent profile analysis and longitudinal assessment of relationships

“…This article was produced as part of a large, multiple time-point project investigating relationships between cognitive-perceptual factors, paranormal belief, and wellbeing. This study was unique because it combined latent profile analysis with a multiple time point design to assess whether changes in relationships between sub-groups were mediated by conceptually important factors over time [ 76 ]. This was not possible within the general analytical framework.…”