Prediction of site-specific pharmacologic therapy among newborns with neonatal opioid withdrawal syndrome.

Abstract

Objective: To predict pharmacotherapy for NOWS based on factors available shortly after birth.

Study design: A multi-center, retrospective study of 1377 opioid exposed newborns between 2016 and 2017 dichotomized based on pharmacologic treatment (N = 665 treated, N = 712 not treated) was conducted. A multilevel mixed-effect logistic regression model that considered cluster effect from sites determined significant maternal and newborn factors associated with pharmacotherapy, which were combined in a nomogram to predict probability of treatment for infants at each participating site.

Results: Factors predictive of treatment were: no breastmilk (1.76, 1.30-2.39; p < 0.001), male sex (1.39, 1.05-1.83; p = 0.021), polysubstance exposure (1.53, 1.15-2.06; p = 0.004), inadequate prenatal care (1.51, 1.07-2.13; p < 0.018) and Methadone (6.08, 4.03-9.27; p < 0.001) or Buprenorphine (1.86, 1.29-2.69; p < 0.001). A site-specific nomogram provided the probability of treatment for individual newborns.

Conclusion: Six factors available shortly after birth can be combined to predict site-specific use of medication for NOWS. If confirmed, the nomogram may identify at-risk newborns.