Context: Delayed thyrotropin (TSH) elevation (dTSH) is defined as elevated TSH at the second neonatal screening (after normal TSH levels at the initial screening) in premature, low-birth-weight, and ill newborns, mostly in the neonatal intensive care unit (NICU) setting. The pathogenesis of dTSH is elusive. Objective: To identify the risk factors for dTSH development among newborns in the NICU. Design, Setting, and Patients: A retrospective medical record review of neonates with dTSH was conducted in eight university-affiliated NICUs. Two controls were selected for each patient, matched for sex and birth weight. The risk factors for dTSH were identified by univariate analysis, followed by multivariate analysis. Main Outcome Measures: Maternal variables, types of NICU treatments and procedures, syndromes, and various medical conditions were compared between dTSH patients and their matched controls. Results: We enrolled 100 dTSH patients and 200 matched controls and 46 variables were compared between the two groups. Twelve risk factors for dTSH were identified on univariate analysis: cesarean section, mechanical ventilation, patent ductus arteriosus (PDA), pneumothorax, and administration of cefotaxime, vancomycin, fluconazole, dopamine, ibuprofen, furosemide, insulin, and packed red blood cells. On multivariate analysis, four risk factors were identified: PDA and vancomycin, insulin, and furosemide administration. In 26 twin pairs, in which one twin had dTSH, all variables presented similarly in both twins. Conclusions: Although some variables had direct effects on pituitary-thyroid axis dysfunction, these variables, altogether, reflect the severity of the clinical conditions in the NICU, which is the common basis for dTSH.