Long-term ambient benzene exposure is associated with higher risks of psychiatric disorders, partly through accelerated biological aging, with nonlinear dose-response patterns showing disproportionately steeper risks at lower concentrations.
Key Findings
Results
Each interquartile range increase in ambient benzene exposure was significantly associated with elevated risk of any psychiatric disorder.
Hazard ratio (HR) for any psychiatric disorder: 1.040 (95% CI 1.028–1.053)
Hazard ratio for multiple psychiatric disorders (MPD): 1.062 (95% CI 1.039–1.086)
Over 14.80 years of median follow-up, 64,270 of 410,207 participants developed psychiatric disorders
Annual ambient benzene concentrations were estimated at residential addresses using 1×1 km spatial grids
Cox proportional hazards models were used adjusting for confounders
Results
Ambient benzene exposure was significantly associated with elevated risks of mood disorders, major depressive disorder, anxiety disorders, and sleep disorders, with most showing significant linear trends across exposure quartiles.
Nine psychiatric outcomes were identified using ICD-10 codes from linked health records
Significant associations were found for mood disorders, major depressive disorder, anxiety disorders, and sleep disorders
Most of these disorder categories showed significant linear trends across exposure quartiles
Associations were evaluated in a historical cohort of 410,207 UK Biobank participants
Results
Nonlinear dose-response patterns indicated disproportionately steeper risks of psychiatric disorders at lower benzene concentrations.
Dose-response analyses revealed nonlinear relationships between benzene exposure and psychiatric outcomes
Risk increases were disproportionately steeper at lower concentrations compared to higher concentrations
This pattern suggests there may not be a safe lower threshold for benzene's psychiatric effects
These findings support refinement of benzene emission standards and monitoring policies
Results
Age and urbanicity were significant effect modifiers, with stronger associations between benzene exposure and psychiatric disorders observed among younger and rural participants.
Younger participants showed stronger benzene-psychiatric associations than older participants
Rural participants showed stronger associations than urban participants
Urbanicity as an effect modifier is notable given that benzene is typically considered an urban pollutant
These subgroup differences were identified through effect modification analyses within the Cox model framework
Results
Accelerated biological aging, measured by PhenoAge acceleration, significantly mediated the association between benzene exposure and psychiatric disorders.
Mediation proportion for any psychiatric disorder: 7.85% (95% CI 5.57%–11.12%)
Mediation proportion for multiple psychiatric disorders (MPD): 17.26% (95% CI 9.01%–80.54%)
Similar mediation was observed for anxiety disorders, substance use disorders, and sleep disorders
Mediation analysis used a bootstrap-based method
Biological aging was assessed using PhenoAge acceleration, a composite biomarker-based measure
Methods
The study population comprised 410,207 UK Biobank participants with a median follow-up of 14.80 years.
This was a historical cohort study design
Annual ambient benzene concentrations were spatially estimated at residential addresses on 1×1 km grids
Psychiatric outcomes were ascertained through ICD-10 codes from linked health records
64,270 participants developed at least one psychiatric disorder during follow-up
Accelerated PhenoAge was assessed using available biomarker data within the UK Biobank
What This Means
This research suggests that people who live in areas with higher long-term levels of benzene in the outdoor air face a meaningfully greater risk of developing psychiatric disorders, including depression, anxiety, and sleep disorders. Analyzing data from over 410,000 UK adults followed for nearly 15 years, researchers found that for every standard increase in benzene exposure, the risk of developing any psychiatric condition rose by about 4%, and the risk of developing multiple psychiatric conditions rose by about 6%. Notably, the relationship was not straightforwardly linear — the risk increased more steeply at lower benzene concentrations, suggesting that even relatively modest exposures may carry significant mental health consequences.
The study also found that a biological process called accelerated aging — where a person's body appears physiologically older than their calendar age, as measured by a blood-based marker called PhenoAge — partly explains how benzene affects mental health. This 'biological aging' pathway accounted for roughly 8% of the link to any psychiatric disorder and 17% of the link to having multiple psychiatric disorders. Additionally, younger individuals and people living in rural areas appeared to be more susceptible to these effects, which was an unexpected finding given that benzene is more commonly associated with urban pollution sources.
This research suggests that ambient benzene — a pollutant released from vehicle exhaust, industrial emissions, and other combustion sources — may have broader public health consequences for mental health than previously recognized. The findings point to the potential value of tightening benzene emission standards and improving air quality monitoring, particularly in areas and for populations currently considered lower-risk. The biological aging mechanism identified also opens possible avenues for understanding how air pollution more broadly accelerates health decline.
Wu Y, Gasparrini A, Su B, Talifu Z, Mukadam N, Tian Y, et al.. (2026). Long-term ambient benzene exposure, acceleration of biological aging, and risk of incident psychiatric disorders: A cohort study in the UK Biobank.. Ecotoxicology and environmental safety. https://doi.org/10.1016/j.ecoenv.2026.120733