Long-term air pollution exposure is associated with sick leave 20 years later

Abstract

Background: Little is known on outdoor air pollution in a long-term perspective and societal costs such as sick leave. In the Nordic countries, recent pollution health impact assessments have had to rely on outdated studies.

Aims: To investigate if air pollution exposure is associated with sick leave 20 years later.

Methods: We analysed self-reported sick leave (all-cause and respiratory) in 7 466 subjects from Bergen, Gothenburg, Umea, Uppsala in the RHINE3 study in 2010-12. Home addresses were geocoded and linked to annual average concentrations of PM_2.5, PM₁₀ and NO₂ at RHINE3, 10 years earlier and 20 years earlier, using existing land-use regression (LUR) models. We performed multilevel logistic regression clustered by centre, and adjusted for sex, smoking, education and previous health-related workplace change.

Results: Age range in RHINE3 was 40-66 yrs, 34% and 4% reported all-cause and respiratory sick leave during the last year. In the adjusted analyses all-cause sick leave was associated with PM_2.5 20 years earlier (OR per interquartile range (IQR) difference (2.6 µg/m³) 1.12 (95%CI 1.01, 1.24)), and borderline with NO₂ (OR per IQR diff (8.1 µg/m³) 1.09 (95%CI 0.99, 1.19)). Respiratory sick leave was associated with PM₁₀ 20 years earlier (OR per IQR diff (3.92 µg/m³) 1.54 (95%CI 1.06, 2.25)), and borderline with PM_2.5 (OR per IQR diff 1.31 (95%CI 0.97, 1.76)). Pollution exposures at present as well as 10 years earlier were not significantly associated with sick leave.

Conclusions: Air pollution exposure in a general population is associated with sick leave in a 20-year perspective. Our findings suggest that even low air pollution levels such as in Northern Europe have societal costs over time.