Risk/benefit trade-off of habitual physical activity and air pollution on mortality: A large-scale prospective analysis in the UK Biobank.

ABSTRACT

BACKGROUND: Previous observational studies have indicated associations of physical activity (PA) and air pollution with mortality. A few studies have evaluated air pollution and PA interactions for health. Still, the trade-off between the harmful effects of air pollution exposure and the protective effects of PA remains controversial and unclear. OBJECTIVE: This study aimed to investigate the joint association of air pollution and PA with mortality risks. METHODS: This prospective cohort study included 322,092 participants from 2006 to 2010 and followed up to 2021 in the UK Biobank study. The concentrations of air pollutants (2006-2010), including particulate matter (PM) with diameters <=2.5 mm (PM2.5), <=10 mm (PM10), and between 2.5 and 10 mm (PM2.5-10), and nitrogen oxides (NO2 and NOx) were obtained. Information on PA measured by the International Physical Activity Questionnaire short form (2006-2010) and wrist-worn accelerometer (2013-2015) were collected. All-cause and cause-specific mortalities were recorded. Cox proportional hazard models were used to investigate the associations of air pollution exposure and PA with mortality risks. The additive and multiplicative interactions were also examined. RESULTS: During a mean follow-up of 11.83 years, 16629 deaths were recorded. Compared with participants reporting low PA, higher PA was negatively associated with all-cause [hazard ratio (HR), 0.74; 95% CI, 0.71-0.78], cancer (HR, 0.85; 95% CI, 0.80-0.90), CVD (HR, 0.79; 95% CI, 0.71-0.87), and respiratory disease-specific mortality (HR, 0.51; 95% CI, 0.44-0.60). Exposure to PM2.5 (HR, 1.05; 95% CI, 1.00-1.09) and NOx (HR, 1.06; 95% CI, 1.02-1.10) was connected with increased all-cause mortality risk, and significant PM2.5-associated elevated risks for CVD mortality and NOx-associated elevated risks for respiratory disease mortality were observed. No obvious interaction between PA and PM2.5 or NOx exposure was detected. CONCLUSIONS: Our study provides additional evidence that higher PA and lower air pollutant levels are independently connected with reduced mortality risk. The benefits of PA are not significantly affected by long-term air pollution exposure, indicating PA can be recommended to prevent mortality regardless of air pollution levels. Our findings highlight the importance of public health policies and interventions facilitating PA and reducing air pollution in reducing mortality risks and maximizing health benefits. Future investigation is urgently needed to identify these findings in areas with severe air pollution conditions.