CHARLS comparison finds persistent physician-recognition gap for probable depression in China

A two-wave national comparison found that 95.61% of observations classified as having probable depression among Chinese adults aged 45 and older still lacked self-reported physician recognition in 2020. The proportion was 98.23% in 2011, an absolute decrease of 2.62 percentage points. Even after adjustment for measured differences between the survey populations and their respondents, the odds of lacking recognition were lower in 2020: the reported odds ratio was 0.45, with a 95% confidence interval of 0.32–0.64 and P < 0.001.

The findings show improvement in the modeled comparison while documenting a large remaining recognition gap. The analysis drew on data from the China Health and Retirement Longitudinal Study, or CHARLS. It included 10,546 wave-specific observations from 8,754 unique respondents aged 45 years or older in the 2011 and 2020 survey waves.

Probable depression was defined through the study’s operational threshold on the ten-item Center for Epidemiologic Studies Depression Scale, or CES-D-10: a score of 10 or higher was classified as positive. An observation was considered to have undiagnosed probable depression when it met that threshold but included no self-reported physician recognition of an emotional, nervous or psychiatric problem. To compare the two survey waves, the researchers used survey-weighted logistic regression and accounted for clustering at the village or community level.

The models adjusted for predisposing, enabling and need-related characteristics, allowing the year comparison to account for differences in the characteristics of respondents included in the analyses. The adjusted odds ratio of 0.45 means that, conditional on the variables included in the model, observations in 2020 had 55% lower odds of lacking physician recognition than observations in 2011. This is an adjusted association across two survey waves, not evidence that a particular policy, service or other factor caused the difference.

The percentages describe the extent of physician recognition among observations that met the study’s probable-depression threshold. They should not be read as the prevalence of probable depression among all CHARLS respondents: 95.61% of a selected group had no reported recognition, rather than 95.61% of the entire study population having probable depression without recognition. The distinction matters because the headline figures are conditional on meeting the CES-D-10 threshold.

Within that group, however, the scale of the remaining gap was clear. In both waves, recognition was absent for the overwhelming majority of observations, although the 2020 estimate was modestly lower than the 2011 estimate. Several features limit how the comparison can be interpreted.

The CES-D-10 score and cutoff were used to define probable depression for the analysis; the study did not equate that scale-based classification with a clinical diagnosis. Physician recognition was based on participant self-report rather than verification against medical records, so the analysis measured reported recognition rather than every aspect of formal or informal mental-health assessment. The surveys also captured two specific years rather than continuous change throughout the interval, and the observational model did not establish causation.

Published as an open-access research article in BMC Psychology on September 2, 2026, the study concluded that the lower proportion of undiagnosed probable depression in 2020 coexisted with a persistent recognition gap that warrants continued attention in research and public health practice.

Subject of Research: Two-wave CHARLS analysis of physician recognition of probable depression among Chinese adults aged 45 and older

Article Title: Undiagnosed probable depression among middle-aged and older adults in China: a two-wave national comparison using CHARLS, 2011 and 2020

Article References: Wu, Y. (2026). Undiagnosed probable depression among middle-aged and older adults in China: a two-wave national comparison using CHARLS, 2011 and 2020. BMC Psychology. https://doi.org/10.1186/s40359-026-05351-2

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05351-2

Keywords: CHARLS, probable depression, CES-D-10, physician recognition, China, older adults, national survey, mental health, Undiagnosed, probable, depression, middle-aged