Adapting a Risk Prediction Tool for Neonatal Opioid Withdrawal Syndrome

Abstract

Background: The American Academy of Pediatrics recommends up to 7 days of observation for neonatal opioid withdrawal syndrome (NOWS) in infants with chronic opioid exposure. However, many of these infants will not develop NOWS, and infants with seemingly less exposure to opioids may develop severe NOWS that requires in-hospital pharmacotherapy. We adapted and validated a prediction model to help clinicians identify infants at birth who will develop severe NOWS.

Methods: This prognostic study included 33 991 births. Severe NOWS was defined as administration of oral morphine. We applied logistic regression with a least absolute shrinkage selection operator approach to develop a severe NOWS prediction model using 37 predictors. To contrast the model with guideline screening criteria, we conducted a decision curve analysis with chronic opioid exposure defined as the mother receiving a diagnosis for opioid use disorder (OUD) or a prescription for long-acting opioids before delivery.

Results: A total of 108 infants were treated with oral morphine for NOWS, and 1243 infants had chronic opioid exposure. The model was highly discriminative, with an area under the receiver operating curve of 0.959 (95% CI, 0.940-0.976). The strongest predictor was mothers' diagnoses of OUD (adjusted odds ratio, 47.0; 95% CI, 26.7-82.7). The decision curve analysis shows a higher benefit with the model across all levels of risk, compared with using the guideline criteria.

Conclusion: Risk prediction for severe NOWS at birth may better support clinicians in tailoring nonpharmacologic measures and deciding whether to extend birth hospitalization than screening for chronic opioid exposure alone.