© 2017 Elsevier Inc. Objective To evaluate the temperature distribution among moderately preterm (MPT, 29-33 weeks) and extremely preterm (EPT, <29 weeks) infants upon neonatal intensive care unit (NICU) admission in 2012-2013, the change in admission temperature distribution for EPT infants between 2002-2003 and 2012-2013, and associations between admission temperature and mortality and morbidity for both MPT and EPT infants. Study design Prospectively collected data from 18 centers in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network were used to examine NICU admission temperature of inborn MPT and EPT infants. Associations between admission temperature and mortality and morbidity were determined by multivariable logistic regression. EPT infants from 2002-2003 and 2012-2013 were compared. Results MPT and EPT cohorts consisted of 5818 and 3213 infants, respectively. The distribution of admission temperatures differed between the MPT vs EPT (P <.01), including the percentage <36.5°C (38.6% vs 40.9%), 36.5°C-37.5°C (57.3% vs 52.9%), and >37.5°C (4.2% vs 6.2%). For EPT infants in 2012-2013 compared with 2002-2003, the percentage of temperatures between 36.5°C and 37.5°C more than doubled and the percentage of temperatures >37.5°C more than tripled. Admission temperature was inversely associated with in-hospital mortality. Conclusions Low and high admission temperatures are more frequent among EPT than MPT infants. Compared with a decade earlier, fewer EPT infants experience low admission temperatures but more have elevated temperatures. In spite of a change in distribution of NICU admission temperature, an inverse association between temperature and mortality risk persists.