Seven county study of air quality and birth defects in Texas, 1997--2000

Background and methods. The Seven County Study of Air Quality and Birth Defects is a population-based case-control study that investigated the association between maternal exposure to five criteria air pollutants (carbon monoxide, nitrogen dioxide, ozone, sulfur dioxide and particulate matter less than ten micrometers in aerodynamic diameter) during weeks three through eight of pregnancy, and the risk of selected cardiac birth defects and oral clefts among live births and fetal deaths between 1997 and 2000 in seven Texas counties. The study consisted of a record linkage analysis and a computer-assisted telephone interview sub-study. Non-malformed controls were frequency matched to cases on year, vital status, and maternal county of residence at delivery. Case data came from the Texas Birth Defects Registry and control data from the live birth or fetal death certificate. Stationary monitoring data were used to estimate air pollution exposure using geocoded maternal residence at delivery. Automatic and manual geocoding methods were employed. Multiple tracing methods were used to locate potential participants for the computer-assisted telephone interview. Results. The record linkage analysis consisted of 3667 controls and 4570 cases. Logistic regression analyses showed positive associations between carbon monoxide and tetralogy of Fallot, particulate matter and isolated atrial septal defects, and sulfur dioxide and isolated ventricular septal defects and inverse associations between nitrogen dioxide and ventricular septal defects and sulfur dioxide and isolated atrial septal defects. In the sub-study, 38% of mothers were located, and among located mothers, 38% completed the computer-assisted telephone interview. Case mothers were more likely than control mothers to be located (44% vs. 30%), and if located, to be interviewed (43% vs. 31%). Among both cases and controls, older mothers (>=30 years) were more likely than younger mothers to be located and interviewed. Conclusions. Results of the record linkage analysis supported a previously reported finding of an association between ozone exposure and pulmonary artery and valve defects. More associations were seen with particulate matter than with any other pollutant. However, the under-representation of Latina women in the geocoded population remained a source of potential selection bias. Poor participation and response rates for the computer-assisted telephone interview indicated the need for caution in follow-back studies using data sources such as vital records and state-based registries.