Perinatal screening for drugs of abuse: reassessment of current practice in a high-risk area.

Anonymous urine toxicology screening among parturient women during 1 month in 1990 and selective newborn testing during this and the subsequent 4-month period was done to assess prevalence of drug use among parturients in a municipal hospital in the Bronx and to assess impact of infant urine toxicology screening on discharge placement. Infant testing was performed for maternal history of drug use, poor prenatal care (5 or fewer visits), or infant symptoms. Urine was screened for cocaine, opiates, methadone, barbiturates, amphetamines, and benzodiazepines. Of 204 women screened, 9.3% were positive. Of these, 74% were positive for cocaine and 21% revealed polysubstance use. Only 28.6% of cocaine-positive mothers gave a history of use. Selective testing of 1196 newborns during this 5-month period revealed an apparent prevalence of cocaine exposure of 4.9%. Selective infant testing failed to identify 42.1% of newborns of cocaine-positive women. Social work evaluation was performed on all families and was the basis for reporting to state agencies for protective services. Only 6 of 83 drug-positive infants entered foster care, none because of positive toxicology per se. Selective infant toxicology studies miss many cocaine-exposed infants and has little impact on placement. Universal social work evaluation of families may be as effective and freer of bias than selective urine screening.