County-level hurricane exposure and birth rates: application of difference-in-differences analysis for confounding control

Abstract

BackgroundEpidemiological analyses of aggregated data are often used to evaluate theoretical health effects of natural disasters. Such analyses are susceptible to confounding by unmeasured differences between the exposed and unexposed populations. To demonstrate the difference-in-difference method our population included all recorded Florida live births that reached 20 weeks gestation and conceived after the first hurricane of 2004 or in 2003 (when no hurricanes made landfall). Hurricane exposure was categorized using ≥74 mile per hour hurricane wind speed as well as a 60 km spatial buffer based on weather data from the National Oceanic and Atmospheric Administration. The effect of exposure was quantified as live birth rate differences and 95 % confidence intervals [RD (95 % CI)]. To illustrate sensitivity of the results, the difference-in-differences estimates were compared to general linear models adjusted for census-level covariates. This analysis demonstrates difference-in-differences as a method to control for time-invariant confounders investigating hurricane exposure on live birth rates.ResultsDifference-in-differences analysis yielded consistently null associations across exposure metrics and hurricanes for the post hurricane rate difference between exposed and unexposed areas (e.g., Hurricane Ivan for 60 km spatial buffer [−0.02 births/1000 individuals (−0.51, 0.47)]. In contrast, general linear models suggested a positive association between hurricane exposure and birth rate [Hurricane Ivan for 60 km spatial buffer (2.80 births/1000 individuals (1.94, 3.67)] but not all models.ConclusionsEcological studies of associations between environmental exposures and health are susceptible to confounding due to unmeasured population attributes. Here we demonstrate an accessible method of control for time-invariant confounders for future research.Electronic supplementary materialThe online version of this article (doi:10.1186/s12982-015-0042-7) contains supplementary material, which is available to authorized users.

Impact of Hurricane Exposure on Reproductive Health Outcomes, Florida, 2004

“…We previously published an article comparing spatial generalized linear models with the difference-in-difference model and demonstrated possible uncontrolled confounding in the relationship between hurricanes and birth rates. 11 We showed in this article that traditional spatial methods (ie, comparing surrounding counties to exposed counties) indicated associations between hurricane exposure and birth rates but difference-in-difference methods showed little to no associations.…”

“…Results investigating Hurricanes Charley and Ivan were previously published as an example in our difference-in-difference methods article. 11 There was some possible evidence of negative associations between Hurricane Ivan >39 mph wind speed and live birth rate (rate difference: −0.40; 95% CI: −0.64, −0.16); however, there was no consistency across hurricanes or wind speed cutoffs.…”

Caesarean section rate and cost control effectiveness of case payment reform in the new cooperative medical scheme for delivery: evidence from Xi County, China

“…Additionally, because there was no control group to compare with Xi County, the effectiveness of the CPR in Xi County was tested using only a before-and-after contrast analysis. It was not confirmed with a difference-in-difference estimation, which is an effective method for reducing bias when comparing groups and can produce considerably accurate and effective estimates [ 33 ].…”

Implications of Prenatal Exposure to the Spring 2011 Alabama and Missouri Tornadoes on Birth Outcomes

“…30 These conflicting findings have been attributed to potential confounding that has not been accounted for in the analyses, variations in how exposure to natural disasters were defined, and study populations that may differ. 30,31 Our findings show there was no statistical association between tornado exposure and birth weight outcomes, and infant mortality rates. These findings are consistent with other studies, which did not find a significant effect of exposure to natural disasters on preterm birth and low birth weight.…”