Experiment to Enhance the Reporting of Drug Use by Arrestees in Cleveland, Detroit, and Houston, 1997 (ICPSR 2890)

Time Period(s)

1997-07 -- 1997-11

Date of Collection

1997-07-08 -- 1997-11-01

Data Collection Notes

(1) Users are encouraged to review the documentation for the 1997 DUF data collection for the methodology and sampling used for the standard DUF interview protocol. (2) The user guide and the codebook and data collection instruments are provided by ICPSR as Portable Document Format (PDF) files. The PDF file format was developed by Adobe Systems Incorporated and can be accessed using PDF reader software, such as the Adobe Acrobat Reader. Information on how to obtain a copy of the Acrobat Reader is provided on the ICPSR Web site.

Study Purpose

Findings from the Drug Use Forecasting (DUF) Program [DRUG USE FORECASTING IN 24 CITIES IN THE UNITED STATES, 1987-1997 (ICPSR 9477)] indicate that many arrestees underreport their recent use of illicit drugs in spite of the many attempts to convince arrestees that their responses are part of a confidential and anonymous research study and cannot affect their cases. If arrestees fail to report drug use accurately in research interviews, and if the level of underreporting varies from one research site to another, then the accuracy and comparability of self-reported drug use data are suspect. This project involved an experiment conducted in three DUF sites to determine whether using a more detailed informed consent procedure and/or altering the sequence of the interview and urine specimen collection could enhance the validity of arrestees' self-reports of drug use, without adversely affecting study response rates. Further, results from the experiment had potential implications for procedures used in the DUF program's successor, the Arrestee Drug Abuse Monitoring (ADAM) Program.

Study Design

A 2x2 factorial design was used to assess the effects of the two manipulations. The first two experimental conditions involved administering either the standard DUF informed consent or an enhanced consent that told the arrestees more about the confidential nature of the research and the capabilities of urinalysis. The second two conditions involved collecting the urine specimen either before or after the interview was administered. The experiment included 2,015 adult arrestees from Cleveland, Detroit, and Houston who were randomly assigned to one of the four experimental cells. Cell A represented the typical DUF protocol in which interviewers administered the standard consent and requested the urine specimen after the interview had been completed. In Cell B, researchers also administered the standard consent, but asked for the specimen prior to conducting the interview. Cells C and D used the enhanced informed consent and asked for the specimen after or before the interview, respectively. Data collection was divided into two phases. During the first phase, arrestees were randomly assigned to Cells A or C, both of which required the collection of the specimen after the interview. After Phase I was completed, collection began for Cells B and D, where the urine specimen was collected before the interview. Four scripts were prepared representing the four experimental conditions. A single script was used for each arrestee approached, which automatically assigned the arrestee to one of the four cells. To randomize the two scripts for each phase, they were first photocopied so that the scripts alternated. The sheets were then broken into smaller stacks and shuffled by picking up varying numbers of sheets from each stack and merging them together. The shuffling step was then repeated several times. The scripts were constructed so that each guided the interviewer through the sequence of steps to be taken for the experimental condition assigned to the arrestee being interviewed. The experiment was designed so that the only variability across the interviews was the manipulation of informed consent and sequence of requesting the urine specimen. All other procedures of a standard DUF collection were followed. The DUF data processing contractor merged the interview data and urine test results and forwarded the merged data file to Center for Substance Abuse Research (CESAR) staff. CESAR staff constructed the final analysis file by merging the data file from the DUF contractor with the CESAR data containing variables denoting the cells to which the arrestee had been assigned and responses to additional questions about the experiment recorded on the script sheet. Data were collected in Cleveland between July 8 and August 22, 1997, in Detroit from August 4 to September 27, 1997, and in Houston from October 17 to November 1, 1997.

Sample

Sites were chosen to represent different interviewing conditions and to fit with the experiment's data collection schedule.

Universe

Booked adult arrestees in Cleveland, Detroit, and Houston in 1997.

Unit(s) of Observation

Individuals.

Data Source

personal interviews, urine specimens, and arrest records

survey data, clinical data, and administrative records data

experimental data

Description of Variables

Variables specific to this project include the experimental condition to which the arrestee was assigned, follow-up questions asking whether the arrestee would have responded differently if assigned to the other conditions, and several dummy variables on length and type of drug use. Data from the DUF interview provided detailed information about each arrestee's self-reported use of 15 drugs. For each drug type, arrestees were asked whether they had ever used the drug, the age at which they first used the drug, whether they had used the drug within the past three days, how many days they had used the drug within the past month, whether they had ever needed or felt dependent on the drug, and whether they were dependent on the drug at the time of the interview. Data from the DUF interview instrument also included alcohol/drug treatment history, information about whether arrestees had ever injected drugs, and whether they were influenced by drugs when the crime that they were charged with was committed. The data also include information about whether the arrestee had been to an emergency room for drug-related incidents and whether he or she had had prior arrests in the past 12 months. Urine tests screened for the presence of ten drugs, including marijuana, opiates, cocaine, PCP, methadone, benzodiazepines (Valium), methaqualone, propoxyphene (Darvon), barbiturates, and amphetamines (positive test results for amphetamines were confirmed by gas chromatography). Demographic data include the age, race, sex, educational attainment, marital status, employment status, and living circumstances of each respondent.

Response Rates

There were no differences in the interview or urine specimen response rates across the four conditions. Between 80 and 85 percent of all arrestees approached agreed to be interviewed and between 88 and 92 percent of the participants in the four conditions provided a urine specimen.

Presence of Common Scales

None.

Original Release Date

2001-04-12

Version Date

2001-04-12

Version History

2001-04-12 ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:

• Standardized missing values.
• Checked for undocumented or out-of-range codes.