Technology transfer through performance management: the effects of graphical feedback and positive reinforcement on drug treatment counselors’ behavior

https://doi.org/10.1016/S0376-8716(00)00207-6Get rights and content

Abstract

After drug treatment counselors at a community-based methadone treatment clinic were trained in implementing a contingency management (CM) intervention, baseline measures of performance revealed that, on average, counselors were meeting the performance criteria specified by the treatment protocol about 42% of the time. Counselors were exposed to graphical feedback and a drawing for cash prizes in an additive within-subjects design to assess the effectiveness of these interventions in improving protocol adherence. Counselor performance measures increased to 71% during the graphical feedback condition, and to 81% during the drawing. Each counselor's performance improved during the intervention conditions. Additional analyses suggested that counselors did not have skill deficits that hindered implementation. Rather, protocol implementation occurred more frequently when consequences were added, thereby increasing the overall proportion of criteria met. Generalizations, however, may be limited due to a small sample size and possible confounding of time and intervention effects. Nonetheless, present results show promise that feedback and positive reinforcement could be used to improve technology transfer of behavioral interventions into community clinic settings.

Introduction

Numerous studies have demonstrated that contingency-management (CM) interventions can improve the effectiveness of drug abuse treatment. In CM, clients earn vouchers redeemable for goods and services or other privileges (e.g. take-home bottles of methadone, changes in medication doses) for providing drug-free urine samples or for compliance with other treatment plan goals. When awarded contingently, vouchers and other privileges have reduced illicit drug use when compared with standard drug counseling (Milby et al., 1978, Hall et al., 1979, Havassy and Hargreaves, 1979, Stitzer et al., 1982, Stitzer et al., 1992, Higgins et al., 1986, Higgins et al., 1994, Iguchi et al., 1988, Silverman et al., 1996). However, Leshner (2000) noted that the transfer of these highly effective CM techniques from clinical research centers into community-based drug abuse treatment clinics is among the most significant challenges currently confronting drug abuse treatment. Kirby et al. (1999) identified specific challenges to broader clinical application, including philosophical and/or political objections and lack of appropriate training for designing and implementing CM.

One method of negotiating philosophical and/or political barriers is to use CM procedures that are more likely to be perceived as socially acceptable to clinic staff. Giving drug abusers special rewards for not using drugs is frequently seen as inappropriate, yet reinforcing cooperation with their counselors and treatment plan compliance is often more acceptable. Some studies have shown that contingent rewards for treatment compliance can reduce urinalysis verified drug use (Iguchi et al., 1997), however, the results of other studies have been equivocal (Morral et al., 1996, Iguchi et al., 2000).

While the reason for these discrepancies has not been determined, there are several possible explanations. One is that different researchers have implemented the CM procedures differently. Detailed guidelines for selecting treatment goals and verifying their completion are still under development. Further, some studies have shown that minor changes in the schedule of reinforcement delivery can alter the effectiveness of the CM program (Silverman et al., 1996; Kirby et al., 1998). However, Iguchi and his colleagues have themselves noted reductions in efficacy as they have moved toward implementing the intervention in more community-based settings (Morral et al., 1996, Iguchi et al., 2000). This suggests that methods for training and monitoring counselors in implementing CM interventions are necessary to ensure that procedures are implemented appropriately. Some researchers have noted that the success of behavioral interventions may ultimately rely on the effective training and management of the persons charged with their implementation, and not on their utility, per se (Kazdin, 1982, Fredericksen, 1984). The effectiveness of CM interventions in community-based substance abuse treatment, therefore, may rely heavily on the effective training and maintenance of counselor behavior.

Efforts to implement CM interventions may benefit from using performance feedback and positive reinforcement. Performance feedback provides employees information about the quality or quantity of their performance. Its ease, low cost, and flexibility make it one of the most popular and frequently used interventions in organizational settings. A review of 126 studies by Balcazar et al. (1986) found that a combination of performance feedback and positive reinforcement produced better performance than feedback alone. In a recent meta-analysis of feedback and other procedures, Stajkovic and Luthans (1997) found that the effectiveness of a procedure was moderated by the type of work being performed (i.e. manufacture vs. service). A combination of feedback and social rewards produced the strongest effects in manufacturing organizations, whereas financial rewards produced the strongest effects in service organizations. The authors cited the ambiguity of task definitions as a possible obstacle to feedback effectiveness in service organizations — it is easier to define desirable manufacturing behaviors (e.g. quality, quotas, safe behaviors) than desirable customer service behaviors (e.g. friendliness, expeditious help, convenience).

The present study was conducted as part of a larger investigation evaluating CM. Drug treatment counselors at a community-based methadone clinic were trained to implement a CM research protocol under the guidance of a research team. After admission to the clinic, clients were randomly assigned to a counselor and to one of four conditions; two conditions involved contingencies for treatment plan goals (TP conditions), and two conditions did not (non-TP conditions). The degree to which the counselors implemented the CM research protocol was assessed across three conditions. Specifically, performance feedback with and without positive reinforcement was implemented in an additive within-subjects experimental design.

Section snippets

Participants

Following approval from the appropriate institutional review committees, ten drug treatment counselors provided informed consent and participated in this study. Their level of education and experience as drug counselors varied. All had at least a bachelor's degree, several had master's degrees, but none had specific training in CM or behavioral psychology. The content of individual therapy sessions was not altered, except that the main study required counselors to initiate brief negotiations

Average proportion of criteria met

Fig. 2 presents the mean proportion of criteria met during each condition. The first bar represents baseline performance of the four counselors who were exposed only to baseline. The next three bars represent the mean proportion of criteria met during each condition for the remaining six counselors. The mean proportion of criteria met for the baseline only group was 0.43, while the mean during the baseline condition for the intervention-exposed group was comparable at 0.42. During the feedback

Discussion

The current results indicate that providing graphical feedback and performance-contingent incentives can improve implementation of a CM protocol in a community-based substance abuse treatment clinic. The greater overall change appeared during the graphical feedback condition, however the difference between the feedback and drawing approached significance (P=0.06). Given that the overall improvement between feedback and drawing was only about 10%, and given that the drawing cost $25 per week to

Acknowledgements

We thank Rebecca Sheppard, Karen Mosher Waite, Monica Cain, Heidi Kooiman, and Dana Becker for their assistance in executing the study. Thanks also to Stephen Weinstein and the counselors, staff, and patients of the Thomas Jefferson University Narcotics Addiction Research Project. We appreciate the feedback and comments from Dr Philip N. Hineline, Dr Saul Axelrod, and Dr Donald Hantula on earlier drafts of this manuscript. This research was supported by NIDA grant R01-DA10778.

References (22)

  • Iguchi, M.Y., Morral, A.R., Riehman, K.S., Zeller, M., Bullock, D., 2000. Applying contingency management in community...
  • Cited by (0)

    View full text