Appendix A: The NTIES data
The National Opinion Research Center (NORC) at the University of Chicago and its subcontractor, the Research Triangle Institute (RTI), were responsible for the design, the collection of data, and the analysis of data from the National Treatment Improvement Evaluation Study (NTIES).
Four main client-based instruments used in data collection were: an interview at intake to determine recent and lifetime drug use and impairments or problems across a panel of domains including mental health, crime, employment, family situations, housing, etc.; an interview at discharge to record all aspects of the treatment experience and behavior during treatment; an interview (and in parallel, a urine specimen and collection of arrest record information) approximately one year after discharge, covering much the same ground in the post-treatment period as the intake interview did for the pre-treatment period; and an abstract summarizing information in the patient's treatment record, including details about receipt of services and source of funding for the treatment. The administration of patient intake questionnaires began in July, 1993 and continued through November, 1994. Discharge questionnaires were administered between July, 1993 and April, 1995. Follow-up questionnaires were fielded between August, 1994 and November, 1995.
In addition to this extensive patient-level data, two rounds of administrative data (baseline and follow-up) on staffing, clinical organization, patient flows, costs, and revenues were collected from each program's chief financial officer and chief executive officer, and questionnaires were also collected from all clinicians in each program. Administrative data collection began in July, 1993 and continued through November, 1994. A total of 519 service delivery units completed the baseline questionnaire.
The administrative data are described in some depth in this paper and to a greater extent in the sizable NTIES final report by Gerstein et al. (1997). The patient-level data are summarized in Table A-1 on the following page and are also fully described in the NTIES final report.
The next phase of multivariate, multilevel analysis will focus on 62 treatment programs in 16 cities, counties, or institutions with statewide catchment areas, in which more than 4,000 patients were included in the final analytical sample used to study patient outcomes. All four of the main patient-based instruments were successfully collected for 3,738 of those patients. Table A-2 compares the subsample of 62 treatment programs with the full sample of NTIES SDUs by CSAT program type and treatment modality. This table shows some differences in these two samples, such as the overrepresentation of adult correctional facilities and short-term residential programs, although other proportions are fairly comparable.