Objectives To describe the demographics, clinical characteristics and referral patterns of premature infants to a regional level IV neonatal intensive care unit (NICU); to determine the prevalence and predictors of back-transport of infants ≤ 32 weeks gestational age in a level IV NICU; for infants not back-transported closer to maternal residence, determine the length of stay beyond attainment of clinical stability. Methods: Data (2010-2014) from the Children's Hospital Neonatal Database and individual chart review for infants ≤ 32 weeks admitted to a level IV NICU whose maternal residence was outside the metro area were included. Bivariate associations of maternal and infant characteristics with back-transport were estimated using two-sample t tests and Fisher's exact test. Multivariable logistic regression was used to measure independent predictors of back-transport. Clinical stability was defined as the attainment of full volume enteral feedings and low flow nasal cannula. Results: A total of 223 infants were eligible for analysis; of whom 26% were back-transported after acute care. In the adjusted analysis, insurance status, distance from maternal residence and gestational age were significantly associated with back-transport. For infants not back-transported closer to maternal residence, median length of stay in the level IV NICU beyond attainment of clinical stability was 28.5 days. Conclusion: For Practice Predictors of back-transport include private insurance, greater distance of maternal residence from NICU and younger gestational age. Many preterm infants admitted to a regional NICU for acute care remained hospitalized in a level IV NICU after achieving clinical stability, for which care in a NICU closer to maternal residence may be appropriate.