A wide assortment of outcome measures were collected on a regular basis for 2 years, including mother and infant baseline hair samples, maternal hair samples every 4 months, urine toxicology, self-report, and measures sensitive to psychological changes and parenting. The mothers all had recently delivered babies, they were known to be drug users owing to hospital screens, and they gave their consent for participation and the random assignment process. Through random assignment, two thirds of the women received drug treatment services directly from the project. Repeated observations taken over a 2 year time period were examined in detail, both within repeated hair measurements of cocaine over time and between hair and urine measurements over time. These findings support the contention that levels of cocaine found within subjects follow predictable patterns of change and these values map onto external measures of cocaine exposure. The relationship between changes of cocaine-use in hair cocaine and changes in measures of psychological health and other evidence of positive treatment outcomes were reviewed. Findings endorse the utility of the hair cocaine exposure measure as a useful metric for gauging the impact of treatment and other efforts toward self-change. Measures of psychological health suggest that correlations between improved mental health and reductions in hair cocaine are reliable and valid. The objectivity of hair measurements is far more consistent with good science than the use of self-report to estimate past gestational exposure levels. Cocaine levels found in hair can be linked both to other exposure measures and to measures of functional change. It is a tool that can be put to work to help resolve important questions about child development. 3 figures, 6 tables, 58 references