© 2017 Elsevier Inc. Objective To evaluate the impact of the Neonatal Resuscitation Program (NRP)-recommended low oxygen strategy (LOX) on neonatal morbidities, mortality, and neurodevelopmental outcomes in neonates born preterm. Study design In March 2011, Parkland Hospital changed from a high oxygen strategy (HOX) of resuscitation with initial 100% oxygen and targeting 85%-94% oxygen saturation for delivery room resuscitation to a LOX with initial 21% oxygen and titrating oxygen to meet NRP-recommended transitional target saturations. Neonates ≤28 weeks’ gestational age born between August 2009 and April 2012 were identified. In this retrospective, observational study, neonates exposed to LOX vs HOX were compared for short-term morbidity, mortality, and long-term neurodevelopmental outcomes. Regression analysis was performed to control for confounding variables. Results Of 199 neonates, 110 were resuscitated with HOX and 89 with LOX. Compared with HOX, neonates exposed to LOX had lower oxygen exposure in the delivery room (5.2 ± 1.5 vs 7.8 ± 2.8 [∑FiO2 × time min], P <.01), spent fewer days on oxygen (30 [5, 54] vs 46 [11, 82], P =.01), and had lower odds of developing bronchopulmonary dysplasia (aOR 0.4 [0.2, 0.9]). There was no difference in mortality (17 [20%] vs 20 [18%]), but neonates exposed to LOX had greater motor composite scores on Bayley Scales of Infant and Toddler Development–Third edition assessment (91 [85, 97] vs 88 [76, 94], P <.01). Conclusion The NRP-recommended LOX strategy was associated with improved respiratory morbidities and neurodevelopmental outcomes with no increase in mortality. Prospective trials to confirm the optimal oxygen strategy for the resuscitation of neonates born preterm are needed.