BACKGROUND: Early-term infants (37-38 weeks) are at increased risk of short- and long-term morbidities compared with full term infants (39-40 weeks). In 2009, the American College of Obstetricians and Gynecologists (ACOG) issued guidelines to discourage early elective deliveries prior to 39 weeks of gestation, and Beth Israel Deaconess Medical Center (BIDMC) in Boston, MA, adopted a policy to implement these guidelines. The impact of this policy on Neonatal Intensive Care Unit (NICU) utilization at BIDMC is unknown. Objectives: The objectives of this study were to (1) examine the gestational age distribution of infants at BIDMC from 2004 to 2015 and confirm a reduction in proportion of early-term births (37-38 weeks) after policy implementation in 2009, (2) compare the incidence of NICU admissions among infants ≥37 weeks of gestation before and after policy implementation, and (3) compare the length of NICU stays among infants ≥37 weeks of gestation before and after policy implementation. METHODS: We conducted a medical record review of infants ≥37 weeks of gestation born from January 1, 2004, through November 10, 2015. We used chi-square tests to compare the incidence of early-term deliveries and NICU admissions in two time periods: 2004-2008 (pre-period) and 2010-2015 (post-period). We excluded infants born in 2009 from the analysis. We used logistic regression to calculate the odds ratio of both short (>4 to <24 hours) and long (≥24 hours) NICU admissions in the two time periods. NICU stays ≤4 hours were excluded as they most often occur among asymptomatic infants for evaluation of sepsis in the setting of maternal fever during labor. We also excluded infants transferred to other hospitals. We considered potential confounding variables such as multiple births, maternal age, race and ethnicity, parity, insurance, and marital status. We compared median lengths of stay using a Wilcoxon test. RESULTS: A total of 50,373 infants were born ≥37 weeks of gestation during the study period, 46,254 of whom were included in the analysis excluding 4,119 infants born in 2009, the washout period. The incidence of early-term delivery was lower in the post-period (27.1%) versus the pre-period (34.2%) (P <0.0001). We detected a slight but statistically insignificant decrease in the incidence of overall NICU admissions from 9.1% in the pre-period to 8.9% in the post-period (P = 0.3). The incidence of short NICU admissions also decreased from 5.3% in the pre-period to 4.6% in the post-period (P < 0.0001). Interestingly, there was an increase in the incidence of long NICU stays from 3.8% in the pre-period to 4.3% in the post-period (P = 0.006). Term infants born after 2009 had lower odds of short NICU stays in adjusted models (adj. OR 0.84; 95% CI 0.77, 0.91). Among NICU admissions >4 hours, the median length of stay (LOS) increased from 21 hours (pre-period) to 39 hours (post-period) (P <0.0001). CONCLUSION: A local policy aligned with ACOG national guidelines to reduce early elective deliveries was associated with a reduction in early-term births. We observed a concurrent reduction of short but not long NICU stays. Our findings suggest that a reduction in early elective deliveries before 39 weeks of gestation may lead to more opportunities for infants to stay with their families in the first 24 hours but may not affect the incidence of significant morbidities requiring longer NICU stays. / 2017-06-16T00:00:00Z