Exploring Associations between Environmental Risk Factors and Low Birth Weight Using Geographic Big Data
dc.contributor.advisor | Zhan, F. Benjamin | |
dc.contributor.author | Gong, Xi | |
dc.contributor.committeeMember | Hagelman, Ronald R. | |
dc.contributor.committeeMember | Chow, T. Edwin | |
dc.contributor.committeeMember | Bell, Michelle L. | |
dc.date.accessioned | 2019-07-11T19:17:24Z | |
dc.date.available | 2019-07-11T19:17:24Z | |
dc.date.issued | 2016-05 | |
dc.description.abstract | Low birth weight (LBW) is defined as newborns who are born weighing less than 2,500 grams. LBW is not only one of the adverse pregnancy outcomes, but also an important predictor of infants’ health. LBW is associated with many risk factors, among which environmental risk factors account for an important portion. This dissertation examines the association between maternal residential exposure to environmental risk factors and LBW in offspring, focusing on ambient air pollution and ionizing radiation near nuclear facilities. Although a growing body of literature has studied environmental risk factors and their relationships with LBW, especially during the last decade, there are some limitations in these reported studies. First, the impact of a significant number of air pollutants on LBW has not been investigated. Second, most studies have used predefined exposure windows (e.g. entire pregnancy, trimesters, and months), which restricted the discovery of more critical exposure windows with more flexible time durations and starting times. Third, few studies have taken into account exposure windows before conception. Fourth, there has been little research about the influence of ionizing radiation near nuclear facilities on LBW. This dissertation fills these gaps in the literature through examinations of (1) association between maternal residential exposure to chemicals released into the air from Toxic Release Inventory (TRI) facilities and LBW in offspring in Texas, (2) association between maternal residential exposure to chemicals monitored by the Texas Commission on Environmental Quality (TCEQ) air quality monitors and LBW in offspring in Texas, and (3) association between maternal residential proximity to nuclear facilities and LBW in offspring in Texas. I call them the TRI Chemicals-LBW association study, the TCEQ Chemicals-LBW association study, and the Nuclear Facilities-LBW association study, respectively. All three studies used case-control study designs. The birth data used in the three studies were birth certificates for all registered births in Texas from 1996 to 2008 obtained from the Center for Health Statistics in the Texas Department of State Health Services (TX DSHS). The TRI Chemicals-LBW association study collected air emission data in Texas during 1996-2008 from the United States Environmental Protection Agency TRI program and air quality monitoring data in Texas during 1996-2008 from the TCEQ Texas Air Monitoring Information System (TAMIS) database. This study estimated maternal residential exposure to TRI chemicals using a modified version of Emission Weighted Proximity Model (EWPM). The model parameters for different TRI chemicals were calibrated through a geocomputational method. Binary logistic regression was used to generate odds ratios for the TRI Chemical-LBW associations. The odds ratios were adjusted for birth year, public health region of maternal residence, child’s sex, gestational weeks, maternal age, education, and race/ethnicity. Based on these adjusted odds ratios, this study identified ten chemicals that were most likely to be associated with LBW from the 449 TRI chemicals. These ten chemicals are styrene, n-hexane, benzene, cumene, methyl isobutyl ketone, cyclohexane, zinc (fume or dust), o-xylene, propylene, and ethylene. In this case-control study, case-mothers were more likely to have a higher level of exposure to these ten chemicals than control-mothers. For four of the ten chemicals (styrene, o-xylene, n-hexane, and benzene), LBW risks increased monotonically when the estimated exposure intensities increased. The TCEQ Chemicals-LBW association study collected air quality monitoring data in Texas during 1996-2008 from the TCEQ TAMIS database. This study estimated maternal residential exposure to TCEQ chemicals in different exposure windows using a spatio-temporal method that took into account residential distance to air quality monitors and ambient concentrations of chemicals within exposure windows. For each combination of the 367 TCEQ chemicals and various exposure windows, this study utilized binary logistic regression to generate odds ratios for the TCEQ Chemical-LBW associations. Based on these odds ratios, this study identified the top ten chemicals (benzaldehyde, 4-methyl-1-pentene, hexanaldehyde, sum of PAMS target compound, m-tolualdehyde, n-undecane, p-tolualdehyde, ethylene dibromide, n-butane, and trans-crotonaldehyde) and corresponding critical exposure windows that showed strongest impact on LBW in offspring. Findings from the study suggested that case-mothers were more likely to be exposed to higher intensities of these ten chemicals within the critical exposure windows than control-mothers. The critical exposure windows identified in the study had flexible time durations (e.g. 30 days, 90 days) and starting time (e.g. before conception and after conception). Most of the critical exposure windows after conception found in this study were located within the second or third trimester of pregnancy. Critical exposure windows before conception were also identified in eight of the ten TCEQ chemicals, which indicated that mothers who were prepared for pregnancy should pay close attention to the air quality in their living environment before conception. Methodologically, the study proposed a standardized protocol for interactively exploring critical exposure windows of air pollution-LBW associations based on the analysis of massive georeferenced air quality monitoring data. The Nuclear Facilities-LBW association study obtained data from United States Nuclear Regulatory Commission for nuclear facilities in operation during 1996-2008 in Texas. This study categorized the LBW case/control births into multiple proximity groups based on distances between their maternal residence and nuclear facilities. Then, this study used a binary logistic regression model to examine the association between maternal residential proximity to nuclear facilities and low birth weight in offspring. The odds ratios were adjusted for birth year, public health region of residence, child’s sex, gestational weeks, maternal age, education, and race/ethnicity. In addition, this study conducted sensitivity analyses using different distance thresholds. Compared with the reference group (>50 km), the exposed groups did not show a statistically significant increase in LBW risk (adjusted odds ratio (aOR) 0.91, (95% confidence interval (CI) 0.81, 1.03) for group 40-50 km; aOR 0.98 (CI 0.84, 1.13) for group 30-40 km; aOR 0.95 (CI 0.79, 1.15) for group 20-30 km; aOR 0.86 (CI 0.70, 1.04) for group 10-20 km; and aOR 0.98 (CI 0.59, 1.61) for group 0-10 km). These results were also confirmed by results of the sensitivity analyses. The results suggest that maternal residential proximity to nuclear facilities is not a significant factor for LBW in offspring. | |
dc.description.department | Geography and Environmental Studies | |
dc.format | Text | |
dc.format.extent | 162 pages | |
dc.format.medium | 1 file (.pdf) | |
dc.identifier.citation | Gong, X. (2016). Exploring associations between environmental risk factors and low birth weight using geographic big data (Unpublished dissertation). Texas State University, San Marcos, Texas. | |
dc.identifier.uri | https://hdl.handle.net/10877/8289 | |
dc.language.iso | en | |
dc.subject | air pollution | |
dc.subject | nuclear facilities | |
dc.subject | low birth weight (LBW) | |
dc.subject | exposure assessment | |
dc.subject | GIS | |
dc.subject | health | |
dc.subject | spatial modeling | |
dc.title | Exploring Associations between Environmental Risk Factors and Low Birth Weight Using Geographic Big Data | |
dc.type | Dissertation | |
thesis.degree.department | Geography | |
thesis.degree.discipline | Geographic Information Science | |
thesis.degree.grantor | Texas State University | |
thesis.degree.level | Doctoral | |
thesis.degree.name | Doctor of Philosophy |
Files
Original bundle
1 - 1 of 1