Predictive Performance of Screening Methods for Neonatal Opioid Withdrawal Syndrome

BACKGROUND: Routine urine substance screening (USS) in pregnant women for perinatal substance exposure is associated with bias and has limited evidence supporting its use. We evaluated agreement between maternal disclosure, maternal and neonatal USS, and their predictive abilities for neonatal opioid withdrawal syndrome (NOWS) admission. METHODS: In this single-center retrospective study, we included pregnant women with gestational substance use and their neonates (February 2018 to December 2022). Agreement between screening tests was assessed using κ. Predictive abilities of the screening were evaluated using logistic regression and area under the receiver operating characteristic curve (AUC). RESULTS: We analyzed 1161 paired maternal USS and maternal disclosure of substance use and 907 paired maternal and neonatal USS results. Self-disclosure and maternal USS showed moderate agreement for opioids (κ = 0.54) and substantial agreement for methadone (κ = 0.80). Maternal and neonatal USS demonstrated moderate agreement for opioids (κ = 0.67) and almost perfect agreement for methadone (κ = 0.88). Among 952 infants born during the study period, 277 (29%) were admitted for NOWS and 176 (18.4%) required pharmacotherapy. Odds ratio (OR) for NOWS admission for self-disclosure, maternal USS, and neonatal USS were 9.9, 8.5, and 6.2, and the AUC for the same were 0.75, 0.74, and 0.71, respectively. Combined maternal history and maternal USS provided the strongest prediction (OR, 12.0; AUC, 0.78), whereas combined maternal disclosure, maternal USS, and neonatal USS had an OR of 10.8 and AUC of 0.75. CONCLUSIONS: Maternal self-disclosure demonstrated moderate agreement with maternal USS for perinatal opioid exposure. Although combining maternal self-disclosure with maternal USS provided the highest predictive ability for neonatal admission with NOWS, maternal self-disclosure alone remained a reliable predictor.