Long-Term Exposure to Wind Turbine Noise and Incidence of Myocardial Infarction: The Danish Nurse Cohort

Abstract

Introduction: Growing evidence supports the idea that traffic noise (road, rail and air) leads to long-term health complications other than annoyance, including myocardial infarction (MI), and similar effects may be expected from wind turbine noise. We examined the association between long-term exposure to wind turbine noise and incidence of MI.Methods: We used the Danish Nurse Cohort with 28,731 female nurses (>44 years) recruited in 1993 or 1999 from entire Denmark, and obtained data on incidence of MI from the National Patient and Causes of Death Registries until 2013. The noise from all wind turbines within a 6000 meters radius within nurses residence between 1982 and 2014 was estimated by Nord2000 method, as the annual mean of a weighted 24-hour average (Lden) at the most exposed façade. We used time-varying Cox regression to link the baseline year mean, 11-, 5- and 1-year moving averages of noise levels to MI incidence.Results: Of 23,994 nurses in the analyses 727 developed MI until 2013. There were 5,601 wind turbines in operation in Denmark between 1982 and 2013, and number of nurses exposed (≥ 1 wind mill within 6000m radius from home) to wind turbine noise was 10% at the cohort baseline in 1993 or 1999, and increased to almost 25% in 2013. Compared to non-exposed nurses at the cohort baseline, we found positive associations with annual mean residential level of wind turbine noise and MI incidence, with hazard ratio (95% confidence intervals) for noise <21.5 dB: 1.29 (0.81-2.03); 21.5-25.4 dB: 1.79 (1.21-2.62); 25.4-29.9 dB: 1.12 (0.70-1.79), and > 29.9 dB: 1.02 (0.62-1.67). Similar pattern and weaker associations were observed with 11-, 5-, and 1-year moving averages, with strongest associations in second quartile of exposure. Conclusion: We found a suggestive evidence of an association between wind turbine noise and MI, but relatively small number of exposed subjects and MI cases, and lack of dose-response relationship demand more studies.