Commuting time and depressive symptoms among employed adults in China: pooled and fixed-effects estimates from the China Family Panel Studies
Abstract
Background: Daily commuting is a routine exposure experienced by large segments of the working population, but its association with depressive symptoms in rapidly urbanising settings remains insufficiently understood. We examined this association among employed adults in China, and explored dose–response form, job satisfaction and effect modification as secondary questions.
Methods: We analysed 33,322 person-year observations from 18,389 employed respondents aged 16–65 in four biennial waves (2016, 2018, 2020 and 2022) of the China Family Panel Studies (CFPS). Depressive symptoms were assessed using the six-item negative-affect component of the Center for Epidemiologic Studies Depression Scale (CES-D-6). We estimated pooled ordinary least-squares (OLS) models with individual-clustered standard errors and two-way individual-and-year fixed-effects (FE) models. Sensitivity analyses addressed missing working-hours data, selection into the analytic sample, clustering level and survey weighting.
Results: Longer one-way commuting time was positively associated with CES-D-6 scores in the fully adjusted pooled model ( β = 0.00387 per minute, 95% confidence interval [CI] 0.00225 to 0.00548; P < 0.001), or 0.039 points per 10 min. The two-way fixed-effects estimate was positive but not statistically significant ( β = 0.00154, 95% CI − 0.00078 to 0.00386; P = 0.192). Only 9,674 of 18,389 individuals (52.6%) contributed at least two waves and only 7,650 (41.6%) showed any within-person variation in commuting time, so the within-person analysis was imprecise. Removing weekly working hours from both the sample restriction and the adjustment set increased the sample to 39,821 person-years, with similar estimates (pooled β = 0.00347, P < 0.001; fixed-effects β = 0.00143, P = 0.155). Weighted and inverse-probability-weighted pooled analyses were consistent with the main pooled result. Analyses of secondary outcomes, dose–response form, job satisfaction and effect modification are reported as exploratory.
Conclusions: Among employed adults in China, longer commuting time was associated with more depressive symptoms in between-person comparisons. The main within-person estimate pointed in the same direction but was statistically inconclusive, partly because the biennial panel provided limited within-person variation and was underpowered for associations of the magnitude implied by the comparable pooled model. The present design does not establish a causal effect.
Related
- Prevalence of depressive symptoms among intensive care unit nurses: a systematic review and meta-analysis of observational studies
- Association Between Perceived Stress and Andropause Symptoms Among Employed Men: A Cross-sectional Study
- Sleep disturbance and its associations with night shift work, anxiety symptoms, and depressive symptoms among urban operation-supporting workers: a cross-sectional study
- Self-reported pain, depressive symptoms and risk for cardiovascular diseases
- Association of Milk and Dairy Intake with Anxiety and Depressive Symptoms in Children and Adolescents: A Systematic Review and Meta-Analysis
BibTeX
@article{zhang2026commuting,
title = {{Commuting time and depressive symptoms among employed adults in China: pooled and fixed-effects estimates from the China Family Panel Studies}},
author = {Ying Zhang and Chuhan Wang and Chanyu Guo},
journal = {BMC Public Health},
year = {2026},
month = oct,
doi = {10.1186/s12889-026-29768-x},
url = {https://link.springer.com/10.1186/s12889-026-29768-x},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Commuting time and depressive symptoms among employed adults in China: pooled and fixed-effects estimates from the China Family Panel Studies
AU - Ying Zhang
AU - Chuhan Wang
AU - Chanyu Guo
PY - 2026
DA - 2026/10/05
JO - BMC Public Health
DO - 10.1186/s12889-026-29768-x
UR - https://link.springer.com/10.1186/s12889-026-29768-x
AB - Background: Daily commuting is a routine exposure experienced by large segments of the working population, but its association with depressive symptoms in rapidly urbanising settings remains insufficiently understood. We examined this association among employed adults in China, and explored dose–response form, job satisfaction and effect modification as secondary questions. Methods: We analysed 33,322 person-year observations from 18,389 employed respondents aged 16–65 in four biennial waves (2016, 2018, 2020 and 2022) of the China Family Panel Studies (CFPS). Depressive symptoms were assessed using the six-item negative-affect component of the Center for Epidemiologic Studies Depression Scale (CES-D-6). We estimated pooled ordinary least-squares (OLS) models with individual-clustered standard errors and two-way individual-and-year fixed-effects (FE) models. Sensitivity analyses addressed missing working-hours data, selection into the analytic sample, clustering level and survey weighting. Results: Longer one-way commuting time was positively associated with CES-D-6 scores in the fully adjusted pooled model ( β = 0.00387 per minute, 95% confidence interval [CI] 0.00225 to 0.00548; P < 0.001), or 0.039 points per 10 min. The two-way fixed-effects estimate was positive but not statistically significant ( β = 0.00154, 95% CI − 0.00078 to 0.00386; P = 0.192). Only 9,674 of 18,389 individuals (52.6%) contributed at least two waves and only 7,650 (41.6%) showed any within-person variation in commuting time, so the within-person analysis was imprecise. Removing weekly working hours from both the sample restriction and the adjustment set increased the sample to 39,821 person-years, with similar estimates (pooled β = 0.00347, P < 0.001; fixed-effects β = 0.00143, P = 0.155). Weighted and inverse-probability-weighted pooled analyses were consistent with the main pooled result. Analyses of secondary outcomes, dose–response form, job satisfaction and effect modification are reported as exploratory. Conclusions: Among employed adults in China, longer commuting time was associated with more depressive symptoms in between-person comparisons. The main within-person estimate pointed in the same direction but was statistically inconclusive, partly because the biennial panel provided limited within-person variation and was underpowered for associations of the magnitude implied by the comparable pooled model. The present design does not establish a causal effect.
SN - 1471-2458
ER -