Long-term effects on cardiovascular and non-accidental mortality by exposure to NOx in the primary prevention study cohort - preliminary findings

摘要

Background and aims: Exposure to air pollution increases cardiopulmonary mortality. Based on the American Cancer Society study an estimated 5000 premature deaths per year in Sweden are caused by long-term exposure to fine particles. The composition of air pollution, as well as other factors, varies over the world though. To provide a risk assessment for Sweden we studied the long term health effects of air pollution in a well-characterized cohort. Methods: Out of a random sample of 10,000 Gothenburg men aged 47- 55 years at baseline, 7495 men were examined in 1970-73 including questionnaires, physical examination and blood samples, to determine predictors of cardiovascular disease. NOx-levels for 1975 were modeled and individual exposures at residential addresses for those living in Gothenburg set (≈90% success rate). Using Cox regression we calculated the risk of non-accidental death and cardiovascular death by exposure to NOx. Results: Between 1975 and 2007 over 4000 non-accidental deaths and over 2000 cardiovascular deaths occurred in the cohort, identified through the Swedish hospital discharge and cause specific death-registries. The risks of both were significantly associated with exposure to NOx (hazard ratios 1.12 and 1.15 in those with higher than median exposure compared to those with lower). Smoking and age were correlated with NOx-levels and thus true confounders reducing the hazard ratios. In a model including smoking, age, social class, body mass index, self-perceived stress and cholesterol levels the effect of NOx was only borderline significant. Discussion and conclusions: This indicates that exposure to NOx at residential address is associated with increased risk of cardiovascular and non-accidental mortality in this cohort of men in Gothenburg, but the increased risk is largely explained by age and smoking. This first preliminary analysis is based on initial exposure 1975. Yearly individual exposures 1975-2006 are being assigned and will be included in the final model, as well as further information on risk factors.