Long-term exposure to black carbon and mortality: A 28-year follow-up of the GAZEL cohort

International audienceBACKGROUND: The current evidence on health effects of long-term exposure to outdoor airborne black carbon (BC) exposure remains scarce. OBJECTIVES: To examine the association between long-term exposure to BC and mortality in a large population-based French cohort, with 28 years of follow-up. METHODS: Data from the GAZEL cohort were collected between 1989 and 2017. Land use regression model with temporal extrapolation wa used to estimate yearly BC and PM(2.5) exposure at the residential addresses from 1989 until censoring for 19,906 participants. Time-varying Cox models with attained age as time-scale was used to estimate the associations between BC and all-cause and cardiovascular mortality, after adjusting for individual and area-level covariates. To handle confounding by PM(2.5), we used the residual of BC regressed on PM(2.5) as an alternate exposure variable. For all-cause mortality, we also examined effect modification by sex, smoking status, BMI and fruit/vegetable intake. RESULTS: The median of 20-year moving average of BC exposure was 2.02 10(-5)/m in study population. We found significant associations between BC exposure and all-cause mortality (n = 2357) using both 20-year moving average of BC and residual of BC, with corresponding hazard ratios (HR) of 1.14 (95 %CI: 1.07-1.22) and 1.17 (95 %CI: 1.10-1.24) for an inter-quartile range (IQR) increase (0.86 10(-5)/m for BC and 0.57 10(-5)/m for residual of BC). We found a similar association between BC and cardiovascular mortality (n = 277) with a HR of 1.15 (95 %CI: 0.95-1.38). The dose-response relationship between BC and all-cause mortality was monotonic but nonlinear with a steeper slope at high BC levels. In addition, the effect of BC was higher among never-smokers and among those having fruit/vegetables less than twice a week. CONCLUSIONS: There was a positive association between long-term exposure to BC and increased mortality risk, reinforcing the emerging evidence that BC is a harmful component of PM(2.5)