Risk-Based Newborn Drug Testing in a Setting With a Low Prevalence of Maternal Drug Use

Our objective in this study was to determine the predictive value of an institutional risk-based newborn drug-testing tool for detecting maternal drug use during pregnancy.

For 5.5 months, the umbilical cords of all newborns born at the study institution were collected and analyzed at a national reference laboratory. In the context of usual clinical care, the decision to perform newborn drug testing is based on an institutional risk assessment tool. For the cohort without clinical indication for testing, cords were deidentified during the study period. Chart review was not performed. Study data were compared with a national data set during the same time period and to previous institutional data.

We tested 857 newborns, 257 of which had 1 or more identified risk factors. There were no drugs or drug metabolites that were significantly more common in the cohort without risk factors than in the clinical cohort. Alprazolam, methamphetamine, hydrocodone, and oxycodone were all significantly more commonly found in the risk-identified cohort. Amphetamine, methamphetamine, and cocaine were not detected in umbilical cords from any of the 600 newborns that would not have been identified for testing. Tetrahydrocannabinol (1.0%; n = 6) was the only illegal substance in the institution's state that would not have been detected.

Performing universal newborn drug testing in the study population would have identified an additional 6 newborns who were exposed prenatally to tetrahydrocannabinol out of 600 who were additionally tested. In areas with a low prevalence of maternal drug use, universal testing may not be cost-effective.