Research Papers

Remedial Programs for Convicted Drinking Drivers: Use of Regression-Discontinuity Analyses to Demonstrate Beneficial Effects of a Criterion-Based Increase in Program Duration

Filename Mann.pdf
Filesize 133 KB
Version 1
Date added June 6, 2010
Downloaded 2 times/fois
Category 2010 CMRSC XX Niagara
Tags Session 7B
Author/Auteur Robert E. Mann, Rosely Flam Zalcman, Gina Stoduto, Thomas Nochajski, Brian R. Rush, Marilyn Herie, Lyn Watkin-Merek, Anja Koski-Jännes, Rita K. Thomas, Christine M. Wickens, Jürgen Rehm

Abstract

Purpose: An assumption underlying most remedial programs for convicted drinking drivers is that individuals with more severe alcohol problems benefit from being assigned to longer or more intense remedial interventions, but this has never been tested directly. Using regression- discontinuity analyses (RD), we examined if assignment to a longer program at the point of discontinuity (the cutoff point for assignment based on scores on the Research Institute on Addictions Self-Inventory (RIASI -Total and -Recidivism) is associated with a significant reduction in number of days using alcohol and certain drugs.
Methods: Convicted drinking drivers in Ontario seeking relicensing are required to complete the Back on Track (BOT) remedial program. They are assigned to complete either an 8-hour education or a 16-hour treatment program based on alcohol problem severity as determined by an assessment interview including the RIASI. Between 2000 and 2005, 22,277 convicted drinking drivers registered for BOT and completed the follow-up interview six months after completion of the education or treatment programs. Clients were assigned to the longer treatment program based on problem severity as measured by a RIASI-T score of 9+ and a RIASI-R score of 6+.
Results: The main outcome measures were number of days of alcohol and drug use (e.g., cannabis, prescription opioids, tobacco) during the 90 days preceding the six-month follow-up interview. A significant reduction in the number of drinking days from the point of assignment to the treatment program was found. Specifically, there was a reduction of one drinking day (regression coeff. = -1.1) for the RIASI-T assignment strategy and two drinking days (regression coeff. = -1.96) for the RIASI-R assignment strategy. A significant reduction of less than one day of cannabis use (regression coeff. = -0.16) from the point of assignment to the 16-hour program based on the RIASI-T was also found.
Conclusions: These results provide support for longer or more intensive treatment interventions for convicted drinking drivers when severity-based criteria are met. They also illustrate the value of RD analysis when randomized designs are not possible.

Robert E. Mann, Rosely Flam Zalcman, Gina Stoduto, Thomas Nochajski, Brian R. Rush, Marilyn Herie, Lyn Watkin-Merek, Anja Koski-Jännes, Rita K. Thomas, Christine M. Wickens and Jürgen Rehm