Investigating the association of preterm birth and residential stability in Georgia
By Xiang Ren
Background: Due to the past 30-40 years improvements in prenatal care and medical technology, outcomes for infants born after short gestations have declined both in mortality and morbidity. However, preterm birth still causes considerable morbidity and mortality in the United States. The mechanism for the environmental risk factors association with preterm birth is not fully explained so far. There is evidence that environmental level socioeconomic status influences maternal related health such as preterm birth. We selected residential stability that living in a same household for more than 5 years as exposure, in order to test its association with outcome preterm birth.
Methods: Simple logistic regression and hierarchical multilevel logistic regression were performed to investigate the association between preterm birth and residential stability while controlling for potential infants, maternal related risk factors and environmental level confounders. Our study sample was limited to the singleton births within Georgia from 1998 to 2002.
Results: Among the 338,435 observations, the overall risk of preterm birth is 10.38%, and the risk of preterm birth is increasing from 1998 to 2002. Previous preterm birth history, maternal use of tobacco and alcohol during pregnancy and prenatal care are strongly associated with the preterm birth. Residential stability is strongly associated with the preterm birth even considering location as a random effect. The residential stability showed statistically significant interaction with household tenure in the final logistic model without random effect. After considering the random effect of location, house hold tenure is not significant associated with preterm birth in the model.
Discussion: The result showed that residential stability is statistically significant risk factor in this study, even including location as random effect. The potential model of the mechanism indicated that residential stability would contribute to an increased risk of preterm birth.
Table of Contents
Table of Contents
Outcome -- Preterm birth. 4
Exposure -- Residential stability. 6
Other risk factors. 7
Descriptive analysis. 16
Table 1 Risk of preterm birth based on the characteristics of Georgia mothers of live singleton birth and the infant demographic information between 1998 and 2002. 25
Table 2 Risk of preterm birth based on the characteristics of Georgia mothers of live singleton birth and the maternal health related information between 1998 and 2002. 26
Table3 t-test of environmental related risk factors among different preterm birth outcomes. 27
Table 4 ANOVA test of association of residential stability (proportion calculated at the census tract level) and individual and environmental level risk factors. 28
Table 5 Correlation of the environmental risk factors and the residential stability. 29
Table 6 The estimates of the final simple logistic model 30
Table7 Logistic model fitness evaluation evidence. 31
Table8 Logistic model fitness evaluation evidence. 32
Table9 Solutions for Fixed Effects of GLIMMIX final model including location as random effect. 33
Table10 Solutions for Fixed Effects of GLIMMIX final model including location as random effect after dropping the interaction term (Residential stability*houseowned). 34
Table11 Solutions for Fixed Effects of GLIMMIX final model including location as random effect after dropping the interaction term (Residential stability*houseowned) and house tenure. 35
About this Master's Thesis
|Committee Chair / Thesis Advisor|
|Investigating the association of preterm birth and residential stability in Georgia ()||2018-08-28 12:11:55 -0400||