Schedule of voucher delivery influences initiation of cocaine abstinence

Temple University Douglas B. Marlowe, David S. Festinger, R. J. Lamb, and Jerome J. Platt Allegheny University of the Health Sciences This study examined whether voucher delivery arrangements affect treatment outcome. First, 90 cocaine-dependent adults were randomly assigned to behavioral counseling or counseling plus vouch- ers for cocaine-free urine samples. The value of each voucher was low at the beginning but increased as the patient progressed (Voucher Schedule 1 ). Voucher Schedule 1 produced no improvements relative to counseling only. Next, 23 patients received vouchers on either Voucher Schedule 1 or Voucher Schedule 2. Voucher Schedule 2 began with high voucher values, but requirements for earning vouchers increased as the patient progressed. Average durations of cocaine abstinence were 6.9 weeks on Voucher Schedule 2 versus 2.0 weeks on Voucher Schedule 1 (p = .02). This confirms that vouchers can assist in initiating abstinence and that voucher delivery arrangements are critical. Cocaine dependence is a serious national health problem con- tributing to the spread of crime, violence, infectious diseases, and neonatal drug exposure (Bux, Lamb, & Iguchi, 1995; Chais- son et al., 1989; De La Rosa, Lambert, & Gropper, 1990; Kain, Kain, & Scarpelli, 1992). Although recreational cocaine use recently has decreased, regular use by some subgroups (e.g., inner-city minorities) has remained stable or increased (Har- rison, 1992; National Institute on Drug Abuse, 1991). There is no widely accepted standard treatment for cocaine dependence, but behavioral treatments for cocaine and stimulant abuse are among the first interventions that have reported good treatment outcomes (Higgins, Budney, & Bickel, 1994; Higgins et al., 1991, 1993, 1994; Milby et al., 1996; Silverman et al., 1996; Tusel et al., 1996). These treatments assume that cocaine use is an operant behavior maintained by the reinforcing effects of the drug. This assumption is supported by an extensive exper- imental literature that demonstrates that drug self-administration is orderly and can be predicted and controlled by operant princi- ples (Henningfield, Lukas, & Bigelow, 1986; Stolerman & Gold- berg, 1986). Following this conceptual framework, these behav- ioral treatments attempt to rearrange the patient's environment Kimberly C. Kirby, Department of Counseling Psychology, Temple University; Douglas B. Marlowe, David S. Festinger, R.J. Lamb, and Jerome J. Platt, Division of Addiction Research and Treatment, Allegheny University of the Health Services. This research was conducted with support by Grant DA06986 from the National Institute on Drug Abuse. Portions of these data were pre- sented at the annual meetings of the College on Problems of Drug Dependence, Palm Beach, Florida, June 1994, and San Juan, Puerto Rico, June 1996. We thank the staff of Addiction Research and Treatment--Camden and Valerie M. Harrington of the Counseling Psychology Program at Temple University for assistance with this study. Correspondence concerning this article should be addressed to Kim- berly C. Kirby, Department of Counseling Psychology, Temple Univer- sity, Weiss Hall (265-63), 1701 North 13th Street, Philadelphia, Penn- sylvania 19122-6085. 761 so that drug use and abstinence are readily detected, drug absti- nence is rewarded through operant reinforcement, drug use re- suits in loss of reinforcement, and reinforcement from nondrug sources is increased to compete with the reinforcing effects of drugs. Higgins and his colleagues (Higgins et al., 1991, 1993) com- pared the treatment outcomes of cocaine-abusing outpatients admitted to a behavioral treatment with the outcomes of outpa- tients admitted to 12-step counseling. Patients assigned to the behavioral treatment received behavioral counseling and pro- vided urine specimens four times weekly. Specimens were tested immediately for cocaine metabolite. Negative specimens earned vouchers that could be exchanged for goods and services. The first voucher earned was worth $1.50, and the value of the vouchers for each subsequent consecutive negative specimen increased on an escalating schedule. Patients assigned to the 12- step counseling condition provided urine specimens on the same schedule as the behavioral group, but no vouchers were pro- vided. Instead, these patients received a token payment for speci- men provision, irrespective of the urinalysis result. Counseling consisted of group therapy sessions or a combination of group and individual therapy following a 12-step model. Treatment outcomes for the behavioral counseling group were consistently better than those observed for patients in the 12- step treatment. Significantly more patients in the behavioral group completed 12 weeks of treatment compared with those given 12-step counseling. Also, patients in the behavioral group achieved significantly longer periods of continuous cocaine ab- stinence compared with the 12-step group. Subsequent investi- gations demonstrated that the voucher incentives were an active component of the intervention (Higgins et al., 1994), and appli- cations by other researchers replicated the effectiveness of voucher incentives delivered on similar escalating schedules (Milby et al., 1996; Silverman et al., 1996; Tusel et al., 1996). We report on two experiments examining voucher incentives. Our first experiment differed from previous studies in two ways. First, the majority of our patients were crack cocaine smokers,