|
Activity Number:
|
252
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2003 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #301287 |
|
Title:
|
Comparisons of Natural History Reports of Substance Use, Treatment and Incarceration to Longitudinal Self-Reports and Administrative Data
|
|
Author(s):
|
Garrett Moran*+ and Cynthia Helba and Jennifer O'Brien and Jennifer Kawata and Bill Luckey
|
|
Companies:
|
Westat and Westat and Westat and Westat and Westat
|
|
Address:
|
1650 Research Blvd., Rockville, MD, 20850,
|
|
Keywords:
|
natural history interview ; substance use ; longitudinal data ; retrospective self-reports ; administrative data
|
|
Abstract:
|
Prior research suggests that natural history interviews can be used to acquire substance use and treatment information for a client's drug use career. This paper addresses the quality of data collected using natural history interviews by examining the congruence of natural history reports with longitudinal self-reports and administrative records from treatment providers and a state prison system. The data concern 230 participants in a three-year longitudinal study who also participated in a natural history interview for that same period. Comparing natural history reports of treatment to administrative records, we find a 61% congruence with chance congruence removed (kappa=.61). Agreement between natural history data and administrative records is higher for reports of incarceration, kappa=.79. Congruence between reports of substance use is not as high (kappa of .51 for any substance use and between .45 and .49 for specific substances). The paper explores the impact of temporality on reliability by examining congruence between reports at approximately six-month intervals and issues of saliency and reliability by examining agreement between reports of types of treatment.
|
