Date of Award


Degree Name

Doctor of Philosophy


Interdisciplinary Health Sciences


Christina Sobin


Introduction. Infant mortality (IM) is a critical indicator of the health of communities because societal factors that affect population health directly impact the mortality rate of infants. There is limited infant mortality data for the U.S. side of the U.S.-Mexico Border Region, defined as the areas within 100 km north of the border. Data from 2005-2007 show that the average IM rate for the border states combined was 5.8 infant deaths per 1,000 live births; for the border counties, the rate was 5.4 per 1,000. Although IM rates were lower in the border counties than in the border states, risk factors that predict IM were significantly higher in the border region. For example, from 2006-2008 preterm births were higher among Hispanics than non-Hispanic whites in the border counties (12.9% vs. 10.4%); in the border states overall (12.0% vs. 11.2%); and in the U.S. (12.2% vs. 11.4). Also, rates of low birthweight infants were higher among Hispanics than non-Hispanic whites in the border counties (7.3% vs. 6.6%). Models that simultaneously characterize IM while suggesting prevention approaches can be valuable. For example, the Perinatal Periods of Risk (PPOR) model provides an analytic framework for calculating excess fetal and infant mortalities while identifying the highest risk periods of infant care for specified subgroups. The model divides fetal and infant deaths into four chronological Perinatal Periods of Risk according to infant birth weight and age at death. Using this approach in the border region could help to identify periods of elevated risk which in turn would indicate specific types of needed interventions. The goals of this study were to test differences in: 1.) overall feto-infant mortality rate for the U.S. border counties in the four border states and a.) non-border counties in the four border states; b.) all counties in the border states; and c.) the U.S. national rates; 2.) excess feto-infant deaths in border counties and non-border counties in the four border states; 3.) specific periods of risk in which excess feto-infant deaths occur in border counties and non-border counties in the U.S. border states; 4.) specific risk factors associated with low birthweight and prematurity in the U.S. Mexico border region; 5.) maternal characteristics associated with an infant dying in the U.S. border states; and 6.) factors associated with the leading cause of death in the PPOR with the highest excess deaths in the U.S Mexico border region. Methods. This was a retrospective cohort study of 25,000 infant deaths occurring before the age of one year in the four border states of Arizona, California, New Mexico, and Texas from 2009-2013. Linked birth-death and fetal death datasets from the National Center for Health Statistics (NCHS) were used. The PPOR approach for fetal and infant mortality was used to first characterize all deaths by age at death (fetal, neonatal, post-neonatal) and birth weight (500-1499, 1500g or more) and then use this information to categorize deaths into one of four maternal/infant care time periods (as defined by the PPOR model) including Maternal Health/Prematurity (MHP), Maternal Care (MC), Newborn Care (NC), and Infant Health (IH). Mortality rates were calculated for each period to estimate the overall IM rate and then compared to a standard reference group, consisting of deliveries to White-non-Hispanic women, 20 years of age and older, residing in California to determine the preventable (excess) deaths. National infant death rates were calculated from the NCHS files and compared with NCHS yearly published rates. Chi-square tests were used to test whether specific factors increased the likelihood ratios of low birthweight, prematurity, and infant death. Also, multivariate logistic regression models were used to test which risk factors were associated with the leading cause of death in the PPOR with the highest excess deaths in the border counties overall. Results. The overall fetal-infant death rate in the border counties was 6.3 fetal-infant deaths per 1,000 live births and fetal deaths, from 2009 to 2013. This rate was lower than the national average (7.8 per 1,000), border states average (6.9 per 1,000), and the non-border counties average (7.0 per 1,000). The excess fetal-infant mortality rate of the border counties in the four border states was 0.8 fetal-infant deaths per 1,000 live births and fetal deaths, which equates to 460 preventable deaths. In the border counties, the highest excess death rate was found in the Maternal Health/Prematurity (MHP) perinatal period (0.4 per 1,000), similar to the non-border counties (MHP= 0.7 per 1,000). However, the MHP in the border counties contributed 52% to the overall excess death and the pathway associated with feto-infant death was birthweight distribution (babies born with very low birthweight (VLBW)); while VLBW babies contributed the most to the feto-infant death rate for the border counties. In the border counties smoking (AOR= 1.4; 95% CL: 1.2-1.7), and previous pre-term birth (AOR= 3.0; 95% CL: 2.5-3.5) were associated with having a VLBW baby. Border mothers that experienced infant death were, in fact, less likely to have a medical risk and to smoke during pregnancy, and more likely to be married, have no or late prenatal care and have >13 years of education as compared to mothers that experienced an infant death in a non-border state. Impact. The traditional IM indicator cannot provide an indication of causal factors. PPOR analyses, however, go beyond traditional IM indicators and provide a well-tested and systematic approach for determining in which chronological periods of risk excess infant deaths are occurring for a particular region. This, in turn, can guide community-based inventions with the highest likelihood of reducing infant deaths. Using this approach, these findings provide strong evidence that interventions to reduce infant deaths in border counties should focus on the MHP (maternal health/prematurity) period of risk. Moreover, this study substantially adds to the current body of knowledge regarding excess deaths and infant mortality in the U.S.-Mexico border region.




Received from ProQuest

File Size

113 pages

File Format


Rights Holder

Dyanne G Herrera