Sir.—I agree with Dr Fulroth and coauthors that we need to know more about the consequences of maternal abuse of cocaine and heroin in combination.1 However, a number of problems weaken the conclusions that can be drawn from their study.
First, the authors have chosen an insensitive method for the detection of maternal drug abuse. Study subjects were identified only by a history or suspicion of maternal drug abuse at the time of delivery. As the authors acknowledged, maternal history regarding drug abuse is often inaccurate2; in other studies, up to 24% of mothers with positive drug screen results have denied drug use.3 The authors asserted that "the problem of inaccurate histories was decreased by having the maternal drug history obtained by three separate individuals," but they offered no evidence to support such a conclusion. The study population thus represents a group detected by maternal report