During exposure and response prevention therapy (ERP) addicts are exposed to stimuli associated with their substance abuse histories; the goal is to extinguish their conditioned responses to the cues. These conditioned responses may place addicts at risk for relapse even after completing traditional substance abuse treatments. The current study consists of two retrospective analyses of the outcomes of adult substance abuse patients treated in a residential facility with exposure response prevention (ERP) therapy. In the first study, 47 inpatients with a history of substance received ERP combined with six drug groups and individual psychotherapy as part of their drug rehabilitation treatment. Craving levels were measured using subjects' pulses and self-reported craving levels. As hypothesized, both increased in response to the presentation of the stimuli and returned to a near baseline level by the end of the sessions. Furthermore, the mean pulse and craving ratings were found to be significantly higher during the first two sessions than during the last two sessions. In the second study, 57 patients with a substance abuse disorder were contacted one to two years after completion of treatment; 33 of the patients had participated in ERP therapy sessions in addition to other substance abuse treatments; 24 did not receive ERP. Patients in the exposure and response prevention therapy group were significantly less likely to experience a relapse and were significantly more likely to be in recovery at the time of the follow up than patients in the non-exposure response prevention therapy group. These findings indicate that exposure and response prevention therapy may be an effective adjunct to traditional substance abuse treatments.

Addicts have consistently been found to be particularly vulnerable to relapse when faced with stimuli associated with previous drug use (Wilker, 1965; Siegel, 1975, 1979; Marlatt & Gordon, 1985; Stewart, de Wit & Eikelboom, 1984; Baker, Morse, & Sherman, 1987; Tiffany, 1990; Carter & Tiffany, 1999). DeHaan et. al. (1999) found that even after 6-12 months of treatment, addicts still experienced cravings when presented with stimuli related to their previous drug use. One possible hypothesis for this relationship conceptualizes these reactions in terms of classical conditioning (Tiffany, 1995; Santoro et. al., in press). During drug use, the drug paraphernalia become a conditioned stimulus that, being paired with the unconditioned response to a drug, begins to elicit a conditioned response. Exposure to conditioned or unconditioned stimuli that control substance use trigger two types of behavior: respondent and operant. Sensory stimuli associated with the substance of abuse elicit physiological responses, called respondent behaviors, which are similar to those that are produced by ingestion of the abused substance. These reinforcing consequences make it more likely that the substance use behavior will be repeated in the future. Thus addicts, when confronted with drug paraphernalia, react psychologically and physiologically to the stimulus and experience cravings.

Exposure response prevention (ERP) is a behavior therapy technology that, through a process of extinction, reduces the conditioned response to a set of stimuli. ERP has been used to treat phobias and compulsions by exposing a person to the phobic situation or stimuli set and then preventing him from executing his dysfunctional response. This technology has also been applied to the problem of substance abuse. Blakely and Baker (1980) described how it could be used to treat alcoholism. Studies completed by Dawe et al. (1993) and Powell et al. (1993) provide research data that support the effectiveness of ERP for the treatment of opiate addiction. Despite the promising results of these early studies, both the empirical exploration and the clinical utilization of ERP have been somewhat limited.

Since 1994 ERP has been used as a substance abuse therapy at SLS Health. SLS Health is a private behavioral health treatment facility located in Brewster, New York, for individuals with a variety of psychiatric and substance abuse diagnoses. SLS Health provides both residential and outpatient treatment services. Two retrospective studies were conducted to analyze the clinical data obtained during the course of routine treatment of a sample of residential substance abuse patients.

The first study explored the question of whether ERP therapy effectively reduced patients' craving levels. It was hypothesized that subjects exposed to stimuli associated with their drug histories would react psychologically and physiological which would in turn elicit the subjects to experience cravings. These cravings would initially increase during each session, peaking at the middle of the session. Furthermore, it was hypothesized that over the course of several sessions, subjects' cravings would significantly decrease so that they no longer react to the drug paraphernalia and would therefore be less likely in the future to use.

The second study explored the question of whether participation in ERP reduced the relapse rate of patients who received ERP therapy in addition to SLS's standard chemical dependency treatment protocols. It was hypothesized that subjects who had been treated using ERP would be significantly less likely to experience a relapse after receiving treatment and would be significantly more likely to be in recovery at the time of the follow up than subjects who had not been treated with the exposure response prevention therapy.

Subjects in the first retrospective analysis consisted of 47 psychiatric inpatients with a substance abuse disorder. There were 29 males in the analysis and 18 females. The mean age of the subjects was 27.7 years, with a range of 18 to 62 years and a standard deviation of 9.5 years. Forty subjects were Caucasian, three were African American, and four were Hispanic (see Table 1). Eighteen of the subjects had abuses marijuana, 16 were alcoholics, nine were cocaine users and four were heroin users. Subjects' length of treatment ranged from 34 days to 961 days, with a mean of 260.0 days, median of 156.0 days, and a standard deviation of 246.3 days. Thirty-six of the subjects had never been married, six were divorced, three were married, and two were legally separated from a spouse. Twenty-one of the subjects earned less than $20,000 a year; 14 earned between $20,001 and $30,000; four earned between $30,001 and $40,000; five earned between $40,001 and $50,000; two earned more than $50,001.

Individual ERP sessions took place in a recreational room decorated to resemble a natural drug-using environment. Cravings levels were measured in two manners: a pulse rating was taken in order to determine physiological reactivity and subjects self-reported craving levels in order to determine psychological reactivity. Craving levels were reported using the Here and Now Craving Scales (See Alcholol Cravings Scale - Drug Cravings Scale), which were found to have a test-retest reliability of .983, p=.01, in a separate study (Santoro, DeLetis, & McNamara, 2000). During the cue exposure, alcoholics poured drinks and held the glass during the session; marijuana users rolled and lit joints; cocaine abusers cut up lines of cocaine and cooked up crack; heroin users cooked up heroin.

Subjects participated in exposure response prevention therapy as a routine part of their treatment for substance abuse. In addition to exposure response prevention therapy, subjects attended six substance abuse groups per week and participated in individual psychotherapy. Thirty-four of the subjects also regularly attended community supports groups such as AA or NA.

Before the exposure response prevention therapy sessions began, subjects were taught relaxation techniques. At the start of each session, subjects were instructed to use these relaxation techniques; a baseline measure of both pulse and craving levels were taken. Subjects were then exposed to drug or alcohol paraphernalia corresponding to their addictions. Subjects then repeated 18 individualized cognitive scripts describing the benefits of recovery as well as negative consequences they encountered during their drug use. After each exposure, a pulse measure was taken and subjects self-reported craving levels. This procedure was repeated four times per session.

The number of sessions for each subject was dependent upon length of stay in treatment. For the current study, the number of sessions ranged from 11 to 93. The average number of sessions was 31.1, the median number of sessions was 27.0, and the standard deviation was 15.9.

Data was collected as part of the normal course of treatment and was retrospectively analyzed about 1-2 years later using SPSS 8.0 for windows. For each subject, cravings and pulse measures were averaged for both the first two sessions and their last two sessions (Figure 1 - Figure 2). The subjects' average craving and pulse measures at the peak of the exposure response prevention therapy during the initial two sessions was compared to the mean craving and pulse measures at the peak of the final two sessions using a t-test for paired samples.

The data was further analyzed, again using a t-test for paired samples, in order to determine whether the cravings did in fact increase, as hypothesized, after the subjects were exposed to the drug paraphernalia. Subjects were then separated for analysis into drug groups corresponding to their primary addiction. The same statistical analyses were done for the marijuana subgroup as it contained the greatest number of subjects (Figure 3). A Bonferouni correction was used in order to protect against the possibility of Type I errors resulting from the high number of t-tests. As a result, an alpha level of .005 was required for the results to be statistically significant.

As hypothesized, subjects appeared to respond the drug stimuli when exposed to them. The average craving ratings for the initial two sessions was significantly higher at time 2 than it was at time 1 of the same sessions (t(46) = -4.166, p = .0001). Furthermore, the biometric rating at time 2 was again significantly higher than the biometric rating at time 1 (t(18) = -8.681, p = .0001) for the initial sessions.

Repeated exposure to these stimuli, as hypothesized, appeared to reduce the psychological and physiological responses of the subjects. The average of self-reported cravings at the peak of the initial two sessions was significantly higher than the average of the subjects' self-reported cravings during the peak of the final two sessions (t(46) = 5.717, p = .0001). When the average pulse measure at the peak of the first two sessions was compared to the average of the peak of the last two sessions, the rating during the initial sessions just failed to be considered significantly greater than the rating during the final two sessions (t(18) = 2.554, p = .020).

Finally, for all subjects the craving at time 2 of the initial sessions was found to be significantly greater than the craving at time 4 of the final sessions (t(41) = 6.509, p =.0001). The same was true when the mean pulse rating at time 2 of the initial sessions was compared with the mean biometric rating at time 4 of the final sessions (t(18) = 4.192, p = .0001). This indicates that, as a result of exposure to stimuli associated with drug use, addicts' psychological and physiological reactions to stimuli were reduced.

Similar results were found for craving ratings when the marijuana subgroup was analyzed. There was an increase in self-reported craving ratings from time 1 to time 2 of the initial sessions that approached statistical significance (t(17) =-2.461. p= .025). The mean craving rating at the peak of the initial two sessions was significantly greater than the mean craving rating at the peak of the final two sessions (t(17) = 5.769.p = .0001).

Finally, the mean craving rating at time 4 of the final sessions was significantly lower than the mean craving rating at the peak (time 2) of the initial sessions (t(15) =6.846, p = .0001).

Subjects in this study consisted of 57 individuals with substance abuse histories who had been discharged from the residential treatment center. Thirty-eight of the subjects were male and 19 were female. Forty-nine were Caucasian, two were Hispanic, four were African-American & two were Asian-American (see Table 4). Forty-seven had never been married, six were married, two were divorced, and two were legally separated from a spouse. Thirty-two of the subjects earned less than $20,000 per year; 11 earned between $20,001 and $30,000 per year; four earned between $30,001 and $40,000 per year; five earned between $40,001 and $50,000 per year; and five earned more than $50,000 per year. Subjects ranged in age from 19 to 58 years with a mean of 27.9 years and a standard deviation of 9.24. The mean length of stay in treatment was 180.3 days, while the median length of stay was 107.0 days, and standard deviation was 204.21 days. Lengths of stay ranged from 8 days to 866 days.

Eighteen of the subjects had abused marijuana, 18 were alcoholics, 11 had abused cocaine, and 10 had abused heroin. While in treatment, all subjects attend six substance abuse groups per week and participated in individual psychotherapy. Thirty-seven of the subjects also chose to attend community support groups such as AA or NA while in treatment. Thirty-three of the fifty-seven subjects contacted had participated in ERP sessions while in treatment in addition to the traditional substance abuse treatments. For those that had participated in ERP, the average number of sessions was 23.0, while the median number of sessions was 18.0 sessions, with a standard deviation 17.0 sessions and a range of 2 sessions to 93 sessions. Of the 33 subjects in the ERP group, 21 attended community support groups while twelve did not. In the non-ERP group, 16 people attended community support groups while 8 did not.

A follow-up questionnaire designed specifically for the study was administered to all subjects (See Table 5). The questionnaire consisted of a series of questions about substance use after discharge, including whether or not subjects had experienced a relapse and whether or not they were in recovery at the time of contact.

Subjects were contacted by phone one to two years after being discharged from the residential treatment center. The questionnaire was administered via phone. Additionally, subjects who had participated in ERP while in treatment were asked to rate its effectiveness using a five point Likert-like scale.

Data was analyzed using SPSS 8.0 for Windows. Yates' continuity correction for the chi-squared statistic was used in order to assess the effectiveness of ERP in reducing the psychological and physiological responses to drug stimuli.

Statistical analysis indicates that subjects who participated in exposure response prevention therapy as part of drug treatment were less likely to experience a relapse after treatment (c2(1) = 5.542, p = .019). In addition, subjects in the ERP group were significantly more likely to be in recovery at the time of the follow up (c2 (1) = 13.831, p = .0001). Sixteen subjects in the ERP group reported that ERP was very effective; eight reported that it was moderately effective; one reported that it was somewhat effective; one reported that it was slightly effective; and seven reported that it was not effective at all in preventing relapse (See Table 6).

The current study examined the effectiveness of exposure response prevention (ERP) in reducing the psychological and physiological response to drug stimuli. Unlike traditional treatment modalities that leave the patients vulnerable to the effects of cue reactivity after treatment, ERP extinguishes the conditioned physiological and psychological responses that occur when addicts are confronted with drug cues.

Subjects in the current study experienced a significant increase in both biometric ratings and self reported cravings after being exposed to stimuli related to their drug use. When the average craving at the peak of the initial two sessions was compared with the average craving at the peak of the final two sessions, the cravings at the final two sessions were found to be significantly lower. This trend was found for the members of the marijuana subgroup. This indicates that the subject's experienced expected reactions to drug stimuli and that those reactions decreased over a number of sessions. Similar results were found when the subject's biometric ratings were examined, indicating that exposure response prevention therapy is also effective in reducing the physiological response of addicts to drug related stimuli. In addition, subjects who had participated in ERP in addition to traditional drug treatments were found to be significantly less likely to experience a relapse after treatment than subjects who had participated in traditional drug treatments without exposure response prevention therapy. Subjects in the ERP group were also significantly more likely to be in recovery at time of the follow up than subjects in the non ERP group. This indicates that the reduction of psychological and physiological responses to drug stimuli may in fact aid in substance abusers in recovery.

This was an uncontrolled retrospective study. These results suggest that exposure response prevention therapy can be an effective adjunct to traditional substance abuse treatment and that it appears to work as expected. The study also indicates that ERP may also be practically applied in a clinical setting.

Based on these findings a randomized study design that explores the number of ERP sessions required to achieve a clinically significant reduction in cravings and to produce a lower rate of relapse is warranted.

Baker, T. B., Morse, E., & Sherman, J. E. (1987). The motivation to use drugs: A psychobiological analysis of urges. In Rivers, C. (Ed.). Nebraska Symposium on Motivation, 34, 257-323.

Blakely, R., & Baker, R. (1980). An exposure approach to alcohol abuse. Behaviour Research and Therapy, 18, 319-325.

Carter, B. L. & Tiffany, S. T. (1999). Meta-analysis of cue reactivity in addiction research. Addiction, 94(3), 327-340.

Dawe, S., Powell, J., Richards, D., Gossop, M., Marks, I., Strang, J., & Gray, J. (1993). Does post-withdrawal cue exposure improve outcome in opiate addiction: A controlled trial. Addiction, 88, 1233-1245.

DeHaan, H. A., Franken, I. H. A., van der Meer, C. W., Handriks, V. M., & Haffmans, P. M. J. (1999). Cue exposure. Gedragstherapie, 32(3), 205-215.

Marlatt, G. A. & Gordon, J. R. (1985). Relapse Prevention. New York: Guilford Press.

Oei, T. P. S., & Baldwin, A. R. (1994). Expectancy theory: A two-process model of alcohol use and abuse. Journal of Studies on Alcohol, 55(5), 525-534.

Powell, J. H., Bradley, B., & Gray, J. A. (1993). Subjective craving for opiates: evaluations of a cue exposure protocol for use with detoxified opiate addicts. British Journal of Clinical Psychology, 32, 39-53.

Santoro, J., DeLetis, R. & Bergman, A. (2001). Kill the Craving. How to Control the Impulse to Use Alcohol and Drugs. New York: New Harbinger Publications.

Santoro, J., DeLetis, R. & Bergman, A. (2004). Kill the Craving. How to Control the Impulse to Use Alcohol and Drugs, 2nd edition. New York: SLS Press.

Santoro, S., DeLetis, R., McNamara, M. (2000). Test-retest reliability of the Here and Now Craving Scale. Unpublished manuscript.

Seigel, S. (1975). Evidence from rats that morphine tolerance is a learned response. Journal of Comparitive and Physiological Psychology, 89, 498-506.

Siegel, S. (1979). The role of conditioning in drug tolerance and addiction. In Keehn, J. D. (Ed.) Psychopathology in Animals: Research and Treatment Implications (pp.143-168). New York: Academic Press.

Stewart, J., DeWit, H., & Eikelboom, R. (1984). The role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants. Psychological Review, 91, 251-268.

Tiffany, S. T. (1990). A cognitive model of drug urges and drug-use behavior: The role of automatic and non-automatic processes. Psychological Review, 97, 147-168.

Tiffany, S. T. (1995). Potential functions of classical conditioning in drug addiction. In Drummond, D., C., Tiffany, S. T., Glautier, S., & Remington, B. (Eds). Addictive Behavior: Cue Exposure Theory and Practice (pp. 47-71). Chichester: John Wiley & Sons.

Wilker, A. (1965). Conditioning factors in opiate addiction and relapse. In Wilner, D. I. & Kassenbaum, G. G. (Eds). Narcotics (pp.85-100). New York: McGraw-Hill.