Abstract

Objectives: To assess the association between hypertension and traffic noise.

Methods: The prevalence and incidence of hypertension were examined in a Swedish municipality partly affected by noise from a highway (20 000 vehicles/24 h) and a railway (200 trains/24 h). A-weighed 24 h average sound levels (L_Aeq,24h) from road and railway traffic were calculated at each residential building using a geographical information system and a validated model. Physician-diagnosed hypertension, antihypertensive medication and background factors were evaluated in 1953 individuals using postal questionnaires (71% response rate). Prevalence ratios and odds ratios (ORs) were calculated for different noise categories. Based on year of moving into the residence and year of diagnosis, person-years and incidence rates of hypertension were estimated, as well as relative risks including covariates, using Poisson and Cox regression.

Results: When road traffic noise, age, sex, heredity and body mass index were included in logistic regression models, and allowing for >10 years of latency, the OR for hypertension was 1.9 (95% CI 1.1 to 3.5) in the highest noise category (56−70 dBA) and 3.8 (95% CI 1.6 to 9.0) in men. The incidence rate ratio was increased in this group of men, and the relative risk of hypertension in a Poisson regression model was 2.9 (95% CI 1.4 to 6.2). There were no clear associations in women or for railway noise.

Conclusions: The study shows a positive association between residential road traffic noise and hypertension among men, and an exposure–response relationship. While prevalence ratios were increased, findings were more pronounced when incidence was assessed.