Predicting post-treatment narcotic use among patients terminating from methadone maintenance

Post-treatment use of narcotics was examined among a group of 687 patients who left methadone-maintenance treatment in New York City in the middle 1970s. Multiple regression analysis was used to identify variables that predicted a return to daily use of narcotics. Variables were entered in the equation starting with those that describe the patient at the time of leaving treatment and proceeding backwards through time. The multiple R² was.44. The type of discharge, employment at discharge, and length of heroin use prior to treatment were the strongest predictors. Results from this study were also compared with those from three other large scale prediction studies; a high degree of consistency was noted across the four studies.