Reliability and validity of not-in-treatment drug users' follow-up self-reports

Based on a sample of 259 drug users not in treatment from five different sites, we examined the reliability of self-reported demographic and behavioral information and the validity of self-reported drug use. Data were collected twice with a 48-hr interval, using the Risk Behavior Follow-Up Assessment (RBFA; National Institute for Drug Abuse, 1992), a structured interview. We examined internal consistency and test-retest reliability and found that, overall, drug users were generally reliable reporters of information regarding their demographics, drug use, sexual behavior, work and income, and criminal behavior. Exceptions to these findings of reliability were noted for items that were phrased too broadly. Comparison of self-report data and urinalysis results indicates that drug users accurately report cocaine and opiate use, although reports of cocaine use were somewhat more valid. These findings indicate that self-report data from drug users, when collected by trained interviewers, can be considered reliable and valid.