Incentive effects on self-report of drug use and other measures of response quality in the alcohol and drug services study

Abstract

The Alcohol and Drug Services Study (ADSS) was conducted for the Substance Abuse and Mental Health Services Administration between 1996 and 1999 to assess the nation's substance abuse treatment system. The sample for ADSS was selected using a multi-stage stratified design. Clients were sampled from selected substance abuse treatment facilities as part of the Phase II sample. Client follow-up comprised interviews and urine specimen collection in Phase III of the survey. One component of ADSS examined the impact of different incentive payments on measures of response rate and response quality using ADSS Phase III client interview data. To test for the effects of four payment levels, several measures of response quality were used. The analysis of consistency involving respondent self-reports, abstracted record data, and urine test results showed no conclusive evidence that incentive payments have any positive or negative effect on data consistency. The analysis of item non-response rates showed that an increase in incentive payments was associated with a subtle decrease in item type non-response rates. Furthermore, mixed results were observed when analyzing the relationship between incentive payment and clients reporting more (or less) of certain types of behavior.