The Camp fire and perinatal health: an example of the generalized synthetic control method to identify susceptible windows of exposure

Abstract

The November 2018 Camp fire was the most destructive wildfire in California history, but its effects on reproductive health are not known. We linked California birth records from 2017-2019 to daily smoke levels using US EPA Air Quality System (AQS) PM2.5 data and NOAA Hazard Mapping System smoke plume polygons during the Camp fire. In the main analysis, pregnancies were considered exposed if they had median AQS PM2.5 levels above 50 μg/m3 for at least 7 days during November 8-22, 2018. We calculated rates of preterm birth and the infant sex ratio based on week of conception and used the generalized synthetic control method to estimate the average treatment effect on the treated and to propose a novel approach to identify potential critical weeks of exposure during pregnancy. We found associations between Camp fire-related smoke exposure and rates of preterm birth, with a risk difference (RD) of 0.005 and a 95% confidence interval (CI) of 0.001-0.010. Exposure during week 10 of pregnancy was consistently associated with increased preterm birth (RD, 0.030; 95% CI, 0.004-0.056). We did not observe differences in the infant sex ratio. Camp fire smoke exposure was associated with increased rates of preterm birth, with sensitive windows in the first trimester. This article is part of a Special Collection on Environmental Epidemiology.

Keywords: natural experiment; pregnancy; preterm birth; sex ratio; susceptible windows; wildfire.

Figure 1

Map of camp fire burn area perimeter and Census places exposed to wildfire-related PM_2.5 levels at least 50 μg/m3 for at least 7 days based on AQS ground monitor data during the November 8-22, 2018 period.

Figure 2

Estimated average treatment effect on the treated (ATT) for risk of preterm birth by week of exposure during pregnancy using an exposure threshold of 50 μg/m³ of PM_2.5 for at least 7 days based on Air Quality System (AQS) ground monitor data during the November 8-22, 2018 period.

Figure 3

Estimated average treatment effect on the sex ratio by week of exposure during pregnancy using an exposure threshold of 50 μg/m³ of PM_2.5 for at least 7 days during the November 8-22, 2018, period.