Date of Award

Fall 8-2020

Degree Name

Doctor of Philosophy (PhD)



Second Advisor


Third Advisor



Background: Recent trends show that the utilization of neonatal intensive care unit (NICU) has extended beyond severely ill infants and increased substantially across all birth weights. However, little is known about what drives the growth of NICU admission rates and whether these trends differ by race/ethnicity. Methods: The study used 2008-2018 Natality Files with restricted use of state and county-level information. Crude and risk-adjusted NICU admission rates, overall and stratified by birth weight group, were compared between black and white infants and between Hispanic and white infants. Kitagawa decomposition and Oaxaca-Blinder decomposition analyses were conducted for the temporal increase in NICU admission rates by race/ethnicity. Results: Overall NICU admission rates increased by 37% from 2008 to 2018, and the increasing trends were observed among all racial and ethnic groups. The absolute and percent increases were the smallest among white infants. NICU admission rates remained highest among black infants. Hispanic infants had the lowest NICU admission rates in early study years but reached rates similar to those of white infants in later years. Most differences in overall NICU admission rates by race/ethnicity disappeared after the risk adjustment but birth weight stratified analyses showed different patterns.

Racial/ethnic differences diminished in the very low birth weight and moderately low birth weight groups while risk-adjusted NICU admission rates remained higher among black and Hispanic infants in the normal to high birth weight group. Kitagawa decomposition found that the overall increase in NICU admission rates was decomposed into 3.4% attributed to changes in the birth weight distribution and 96.6% attributed to changes in the birth weightspecific NICU admission rate. Oaxaca-Blinder decomposition analysis showed that changes in infant health risk contributed 0.87 and 0.47 of NICU admission rate increase per 100 infants among black and Hispanic infants respectively, while it mitigated the increase by 0.14 among white infants. Increased NICU bed supply contributed 0.48, 0.04, and 0.28 per 100 infants among white, black, and Hispanic infants, respectively. Maternal socioeconomic characteristics did not change but changes in their association with NICU admission contributed most to the NICU admission increase among all race/ethnic groups. Conclusions: Racial/ethnic differences in risk-adjusted NICU admission rates diminished among high-risk infants while black and Hispanic infants maintained higher risk-adjusted NICU admission rates among low-risk infants. The contributions of the factors affecting NICU admission growth substantially differed by race/ethnicity