The objective of this longitudinal study is to determine the long-term neurocognitive effects of in utero cocaine exposure in preschool children. Study participants were 476 full-term, inner-city, African-American infants, enrolled prospectively at birth and categorized into a cocaine-exposed (COC), polydrug/non-cocaine exposed (POLY), and drug-free control (CONT) group. Cocaine exposure was defined by a combination of maternal report and/or positive urine or meconium toxicology assays. Both the COC and POLY groups included infants also exposed to varying combinations of alcohol, nicotine, and marijuana. The Clinical Evaluation of Language Fundamentals-Preschool (CELF-P) was administered to participants as part of a detailed cognitive test battery at approximately 3 years, 3 months of age. The groups did not differ in average test age. A total of 434 children were seen(91% of the total sample). Of these, nine children were untestable due to extremely severe cognitive or behavioral deficits. The groups did not differ, however, in the proportions of children who were untestable. Maternal education, an important correlate of child cognitive/language functioning, was not significantly different among the groups. On the CELF-P (total n=425), Total Language Scores were significantly different between the COC and CONT groups (p<.006*), with similar trends for the Expressive and Receptive Language Scores. Analyses of cognitive functioning as measured by the McCarthy Scales of Children's Abilities revealed no significant differences among the groups. Cocaine remained a significant predictor of Total Language Scores after controlling for the effects due to alcohol, marijuana, and tobacco. Means and standard deviations are reported below.Table

Table 1
Full size table