Rita A. Stapulonis, MA, Survey and Information Systems, Mathematica Policy Research, Inc., P.O. Box 2393, Princeton, NJ 08543, 609-275-2363, rstapulonis@mathematica-mpr.com

Self-reported illicit drug use in surveys is often under-reported, which can bias study results and affect policy and program planning. Increasingly survey researchers are looking to other methods to verify and quantify drug use among survey respondents. Two such methods are urinalysis and hair analysis, both of which can identify and quantify illicit drug usage. Hair analysis has the added benefit of quantifying drug usage over a longer period (up to 90 days). Mathematica Policy Research, Inc. has recently completed a program evaluation that utilized both self-reported drug use and a hair sample for drug testing during a survey interview. The study consisted of a baseline interview conducted in jail, random assignment into a treatment or control group, tracking of sample members upon release from jail, and a follow-up interview and voluntary hair sample extraction 12 months after release from jail. We will present a data analysis on approximately 1,050 sample members who completed a 12-month interview, of whom 566 also provided a hair sample. The analysis will consist of 1) a comparison of self-reported drug use and hair test results among hair sample responders; 2) a comparison of self-reported drug use among hair sample responders and non-responders, and 3) a descriptive analysis of hair sample responders and non-responders. The analysis will examine the following questions: Do survey respondents under-report drug use? Are there differences in self-reported drug use among hair sample responders and non-responders? And what are the characteristics of hair sample responders and non-responders?

Learning Objectives: