Skip to main content
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Environ Int. 2022 Jan 1;160:107072. doi: 10.1016/j.envint.2021.107072

Fig. 5: Associations between long-term ozone exposure and risk of cognitive impairment by subpopulations.

Fig. 5:

Hazard ratio (HR) equal to 1 represents no association (as indicated by the solid grey vertical line). The dashed orange vertical line indicates the fully-adjusted HR for all participants (HR=1.104). Error bars represent 95% confidence intervals (CIs). HR in each subpopulation was estimated using the Cox proportional hazards model with an interaction term between the potential effect modifier and ozone exposure. The significance of effect modification was assessed by comparing the reduction in log partial likelihood of models with and without an interaction term. Covariates include age, sex, education, geographic region, residence, marital status, financial support, economic status, smoking status, drinking status, regular exercise, social and leisure activity, personality, chronic disease, disability, PM2.5, and air temperature. Abbreviation: HR, hazard ratio; CI, confidence interval.