Participation and retention in drug abuse treatment services research.

We examined potential bias in treatment effectiveness research by studying client characteristics associated with initial participation and subsequent retention in research assessments among clients in substance abuse treatment receiving case management. Six hundred thirty-four residential and 429 outpatient clients were invited to participate. Participants were randomly assigned to one of four conditions. Logistic regression examined the relation between client characteristics and initial participation and retention in follow-up assessments. Females and clients with significant others were more likely to participate and remain in the study. Older persons and criminal justice-referred clients were less likely to participate. Older clients and urban outpatient clients were more likely, and criminal justice-referred clients were less likely, to be retained. Two case management conditions were associated with decreased retention. Client characteristics are important factors in participation and retention in treatment services research. Further, client characteristics may bias estimates of effectiveness generally, or for specific populations.