Associations between ozone exposure and anxiety and depressive symptoms were found across outdoor, indoor, and personal exposure metrics among young adults, with sleep quality mediating 10-30% and 16-49% of the ozone-anxiety and ozone-depression associations, respectively.
Key Findings
Results
The prevalence of anxiety symptoms, depressive symptoms, and sleep disturbance among young adult participants was 14%, 19%, and 39%, respectively.
Study included 13,028 participants from a nationwide cross-sectional survey conducted in 2023-2024.
Survey spanned 261 Chinese municipalities.
Anxiety symptoms were assessed using GAD-7, depressive symptoms using PHQ-9, and sleep quality using PSQI.
Participants were young adults.
Results
Higher outdoor ozone exposure was associated with increased odds of anxiety and depressive symptoms.
Per interquartile range increase in outdoor ozone concentration, the odds ratio for anxiety symptoms was 1.17.
Per interquartile range increase in outdoor ozone concentration, the odds ratio for depressive symptoms was 1.14.
Annual maximum 8-hour ozone concentrations over the preceding year were used as the exposure metric.
Estimates were adjusted for sociodemographic, behavioral, and health condition covariates using multivariable logistic regression.
Results
Higher indoor ozone exposure was associated with increased odds of anxiety and depressive symptoms.
Per interquartile range increase in indoor ozone concentration, the odds ratio for anxiety symptoms was 1.19.
Per interquartile range increase in indoor ozone concentration, the odds ratio for depressive symptoms was 1.13.
Associations with indoor exposure were modestly different from those with outdoor exposure.
Indoor ozone concentrations were estimated at the indoor exposure level based on annual maximum 8-hour concentrations.
Results
Higher personal ozone exposure was associated with increased odds of anxiety and depressive symptoms.
Per interquartile range increase in personal ozone concentration, the odds ratio for anxiety symptoms was 1.14.
Per interquartile range increase in personal ozone concentration, the odds ratio for depressive symptoms was 1.10.
Personal exposure estimates were distinct from ambient measurements and intended to better represent actual individual exposure.
Associations with personal exposure varied modestly compared to outdoor and indoor metrics.
Results
Sleep quality mediated a significant portion of the associations between ozone exposure and both anxiety and depressive symptoms.
Sleep quality mediated 10-30% of the association between ozone exposure and anxiety symptoms across outdoor, indoor, and personal exposure metrics.
Sleep quality mediated 16-49% of the association between ozone exposure and depressive symptoms across outdoor, indoor, and personal exposure metrics.
Mediation analysis was conducted within a logistic framework.
Sleep disturbance was assessed using the Pittsburgh Sleep Quality Index (PSQI).
The mediation proportion was larger for the ozone-depression association than for the ozone-anxiety association.
Results
Associations between ozone and mental health outcomes varied only modestly across outdoor, indoor, and personal exposure metrics.
Odds ratios for anxiety were 1.17, 1.19, and 1.14 for outdoor, indoor, and personal exposure, respectively.
Odds ratios for depression were 1.14, 1.13, and 1.10 for outdoor, indoor, and personal exposure, respectively.
The authors noted that 'associations with anxiety and depressive symptoms varied modestly across outdoor, indoor, and personal ozone exposure.'
Incorporating indoor and personal exposure metrics was suggested to 'provide additional insight into ozone-related mental health risks.'
Methods
The study estimated ozone exposure at three distinct levels—outdoor, indoor, and personal—using annual maximum 8-hour concentrations over the preceding year.
This approach was used to address the concern that 'reliance on ambient measurements may misrepresent personal exposure.'
The study was designed as a nationwide cross-sectional survey across 261 Chinese municipalities.
Survey was conducted during 2023-2024.
Multivariable logistic regression was used for primary analyses, adjusting for sociodemographic, behavioral, and health condition covariates.
Wang C, Zhou R, Yang C, Zhou Y, Zhang R, Hu L, et al.. (2026). Ozone Exposure (Outdoor-Indoor-Personal) and Mental Health in Young Adults.. Environmental science & technology. https://doi.org/10.1021/acs.est.5c14296