Journal of Substance Abuse Treatment 23 (2002) 351 – 360
`
`Regular article
`
`Behavioral naltrexone therapy: an integrated
`treatment for opiate dependence
`
`Jami L. Rothenberg, Ph.D.a,b,*, Maria A. Sullivan, M.D., Ph.D.a,b,
`Sarah H. Church, Ph.D.c, Angela Seracini, Ph.D.a,b, Eric Collins, M.D.a,b,
`Herbert D. Kleber, M.D.a,b, Edward V. Nunes, M.D.a,b
`
`aThe New York State Psychiatric Institute, Division on Substance Abuse, Substance Treatment and Research Service,
`1051 Riverside Drive, New York, NY 10032, USA
`bColumbia University, College of Physicians and Surgeons, Department of Psychiatry, 630 West 168th Street, New York, NY 10032, USA
`cMontefiore Medical Center, University Behavioral Associates, 4113 White Plains Road, Bronx, NY 10466, USA
`
`Received 28 December 2001; received in revised form 27 June 2002; accepted 5 July 2002
`
`Abstract
`
`Treatment of opiate dependence with naltrexone has been limited by poor compliance. Behavioral Naltrexone Therapy (BNT) was
`developed to promote adherence to naltrexone and lifestyle changes supportive of abstinence, by incorporating components from empirically
`validated treatments, including Network Therapy with a significant other to monitor medication compliance, the Community Reinforcement
`Approach, and voucher incentives. An overview is presented of the BNT treatment manual. In an uncontrolled Stage I trial (N = 47), 19%
`completed the 6-month course of treatment. Retention was especially poor in the subsample of patients who were using methadone at
`baseline (N = 18; 39% completed 1 month, none completed 6 months), and more encouraging among heroin-dependent patients (N = 29; 65%
`completed 1 month, 31% completed 6 months). Thus, attrition continues to be a serious problem for naltrexone maintenance, although further
`efforts to develop interventions such as BNT are warranted. D 2002 Elsevier Science Inc. All rights reserved.
`
`Keywords: Heroin; Behavior therapy; Naltrexone; Treatment
`
`1. Introduction
`
`Opiate dependence has been a significant public health
`problem since the turn of the century, with substantial
`morbidity, mortality, and social costs, including its role in
`recent decades in the spread of HIV and hepatitis B and C. Its
`prevalence may be increasing in the wake of greater inter-
`national production, abundance, and purity of heroin (Hamid
`et al., 1997; NIDA 1995). Thus, improvement in the ability to
`attract opiate-dependent individuals into treatment and
`improvement of treatment effectiveness remain critical goals.
`Methadone maintenance is a highly effective treatment,
`which has allowed many heroin addicts to reduce illicit drug
`
`* Corresponding author. New York State Psychiatric Institute,
`Substance Treatment and Research Service (S.T.A.R.S.), 600 West 168th
`Street, 2nd Floor, New York, NY 10032, USA. Tel.: +1-212-923-3031;
`fax: +1-212-923-4372.
`E-mail address: rothenb@pi.cpmc.columbia.edu (J.L. Rothenberg).
`
`use and improve their occupational and social functioning
`(Ball, Corty, Bond, Myers, & Tommasello, 1988). However,
`the success of methadone treatment is limited (Rounsaville
`& Kleber, 1985; Shaffer & LaSalvia, 1992). Many patients
`continue to abuse illicit drugs or alcohol. Further, metha-
`done treatment availability is limited in many communities,
`and opiate addicts often avoid or refuse this option. New
`agonist maintenance agents, LAAM (leva-alpha-acetyl-
`methadol), and buprenorphine, may have advantages over
`methadone in certain respects, such as reduced frequency of
`dosing and of mandatory clinic visits. However, there is no
`clear data that these will be more efficacious or acceptable
`to patients than methadone. Thus, alternatives to agonist
`maintenance are needed to help attract a wider range of
`heroin addicts into effective treatment.
`Antagonist maintenance with naltrexone is one such
`alternative, but it has not, to date, lived up to its potential
`(Callahan et al., 1980; Kosten and Kleber, 1984). An ideal
`pharmacotherapy in many respects, naltrexone blocks the
`
`0740-5472/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved.
`PII: S 0 7 4 0 - 5 4 7 2 ( 0 2 ) 0 0 3 0 1 - X
`
`APOTEX EXHIBIT 1037
`Apotex v. Alkermes
`IPR2025-00514
`
`

`

`352
`
`J.L. Rothenberg et al. / Journal of Substance Abuse Treatment 23 (2002) 351–360
`
`intoxicating and reinforcing effects of heroin and other
`opiates, but itself has no opiate-like effects. When taken
`regularly, it helps to extinguish opiate-seeking and opiate-
`taking behavior. However, in practice, treatment outcome
`with naltrexone has been poor, as most patients drop out
`and resume opiate use. Despite the problems,
`there are
`compelling reasons to pursue improvements in naltrexone
`maintenance. Since naltrexone is neither an opiate agonist,
`nor a controlled substance, it has no abuse potential and
`offers increased flexibility both for patients and for treat-
`ment settings.
`interventions for
`Early studies assessing psychosocial
`promoting adherence to naltrexone (Callahan, Rawson,
`Glazer, McCleave, & Arias, 1976; Callahan et al., 1980;
`Grabowski, O’Brien, Greenstein, & Ternes, 1979; Resnick,
`Washton, & Stone-Washton, 1981; Stone-Washton, Resnick,
`& Washton, 1982) showed promise. In the past
`two
`decades,
`there has been an expansion of well-controlled
`research on psychotherapy and behavior therapy for drug-
`dependent patients. Rounsaville (1995) recently called for a
`renewed effort to bolster the efficacy of naltrexone by
`applying advances in behavioral therapy, and such efforts
`have begun to appear (Carroll et al., 2001; Preston et al.,
`1999). This is consistent with the recent call from the Institute
`of Medicine (1998) for increased efforts to translate research
`advances into the clinical practice of addiction treatment.
`Here, we report the development of Behavioral Naltrex-
`one Therapy (BNT), an integrative behavioral
`therapy
`intended to improve the outcome of naltrexone maintenance
`for opiate addiction. BNT seeks to embed naltrexone main-
`tenance in a strong psychotherapeutic context, based on
`Network Therapy (Galanter, 1993) and the Community
`Reinforcement Approach (CRA; Hunt & Azrin, 1973;
`Meyers & Smith, 1995) while addressing three specific
`problem areas impeding naltrexone maintenance: (a) dif-
`ficulty transitioning to naltrexone; (b) poor compliance; and
`(c) possible dysphoric effects.
`in opiate-
`Firstly, naltrexone precipitates withdrawal
`dependent patients, so it normally cannot be given until
`7 to 10 days after detoxification, by which time most
`patients, unless continuously hospitalized, will relapse.
`Protracted withdrawal symptoms and immediate relapse to
`opiates contribute to high dropout from naltrexone treatment
`(Callahan et al., 1980). Therefore, BNT begins by rapidly
`transitioning patients onto naltrexone in the hospital, using
`an established regimen of buprenorphine, clonidine, and
`other ancillary medications found to reduce withdrawal
`discomfort (Collins and Kleber, 2000; Stine and Kosten,
`1992). Motivational interviewing techniques and procedures
`to promote continuity of care between inpatient and out-
`patient treatment are also applied.
`Secondly, naltrexone maintenance has been plagued by
`poor long-term medication compliance. Patients can easily
`stop their naltrexone for a few days, after which opiate
`blockade wears off and patients rapidly become re-depen-
`dent. Once re-dependent, they must be detoxified again to
`
`resume naltrexone. Therefore, BNT requires involvement of
`a significant other to monitor medication ingestion and
`positively reinforce compliance (Azrin, Sisson, Meyers, &
`Godley, 1982), and further, provides voucher incentives
`contingent on adherence to naltrexone (Bickel, Amass,
`Higgins, Badger, & Esch, 1997; Grabowski et al., 1979;
`Higgins, Budney, Bickel, & Badger, 1994; Silverman et al.,
`1998; Silverman, Chutuape, Bigelow, & Stitzer, 1999).
`Lastly, some reports suggest opiate antagonists may
`produce dysphoria, even in long-abstinent opiate addicts
`(Crowley et al., 1985) and depression has been found to be
`common in opiate addicts, and to be associated with poor
`outcome (Kosten, Rounsaville, & Kleber, 1986; LaPorte,
`McLellan, O’Brien, & Marshall, 1981; Magura, Siddiqi,
`Freeman, & Lipton, 1991; Rounsaville, Weissman, Crits-
`Christoph, Wilber, & Kleber, 1982; Rounsaville, Kosten,
`Weissman, & Kleber, 1986) and craving (Childress et al.,
`1994). Therefore, BNT incorporates cognitive techniques to
`address management of dysphoric moods and guidelines for
`use of antidepressant medication based on experience treat-
`ing depressed methadone patients (Hamilton, Nunes, &
`Klimchak, 1998; Nunes, Quitkin, Brady, & Stewart, 1991;
`Nunes, Quitkin, Brady, & Koenig, 1994; Nunes, Quitkin,
`Donovan, & Deliyannides, 1998).
`In what follows, we present an overview of the BNT
`treatment manual with therapist training and adherence
`procedures and results of an uncontrolled trial to assess the
`feasibility and efficacy of BNT for outpatient opiate addicts.
`
`2. Materials and method
`
`2.1. Overview of BNT
`
`Behavioral Naltrexone Therapy is delivered over a six-
`month period in weekly individual and network therapy
`sessions. The major therapeutic components included in
`BNT are Relapse Prevention, Community Reinforcement
`Approach and Network Therapy (NT). The goals of BNT
`are for patients to take naltrexone continuously and to
`abstain from opiates. These goals are reinforced by
`vouchers and the support of significant others who
`monitor medication ingestion and attend network therapy
`sessions. Individual treatment sessions continuously assess
`and challenge motivation for treatment and abstinence and
`integrate cognitive-behavioral
`techniques to facilitate all
`of these goals.
`Behavioral Naltrexone Therapy is divided into three
`phases, Induction, Stabilization, and Maintenance, based
`on suggestions of Kosten and Kleber (1984) for establishing
`an effective treatment for opiate dependence with naltrexone.
`These phases serve as a conceptual guideline for BNT
`therapists; all areas of focus are cumulative and can be
`reviewed as often as deemed clinically important. The
`components of the three phases are listed in Table 1 and
`guidelines for conducting each phase are summarized below.
`
`

`

`J.L. Rothenberg et al. / Journal of Substance Abuse Treatment 23 (2002) 351–360
`
`353
`
`Table 1
`Treatment components emphasized in the three phases of Behavioral
`Naltrexone Therapy (BNT)
`
`Treatment
`components
`
`Behavioral Naltrexone
`Therapy phases
`
`Induction
`
`Stabilization Maintenance
`
`ments are made for a network member to escort the patient
`from the hospital to the first outpatient session and then
`safely home to reduce the risk of immediate relapse.
`Because we have found that some withdrawal discomfort,
`mainly anxiety and insomnia, may continue into the first
`week of outpatient treatment, upon discharge patients are
`offered a 5 – 7 day schedule of certain pain medications that
`they have been receiving in hospital (mainly clonidine for
`residual withdrawal symptoms and trazodone or zolpidem
`for insomnia).
`
`2.1.2. Stabilization phase (first month of outpatient
`treatment)
`The first month of naltrexone maintenance is marked by
`a high risk of dropout. Thus, the primary focus of the
`Stabilization Phase is to keep the patient in treatment, on
`naltrexone, and abstinent from opiates. This phase is addi-
`tionally devoted to developing a strong therapeutic relation-
`ship, encouraging the support of the network, and
`identifying goals for lifestyle change. Relapse prevention
`techniques and rewards contingent on compliance with
`naltrexone and abstinence are introduced. Also, skills for
`coping with dysphoria are emphasized and patients with
`persistent depressive syndromes are offered antidepressant
`medication. The Stabilization Phase is planned to last one
`month, but it may be lengthened for patients who continue
`to struggle with ambivalence around naltrexone adherence
`and heroin abstinence.
`In the initial two weeks of treatment, patients with their
`significant other attend three sessions per week (typically
`on Monday, Wednesday, and Friday) and ingest naltrexone
`under direct staff observation (e.g. 100 mg Monday,
`100 mg Wednesday, 150 mg Friday). The therapist trains
`the patient and his/her significant other to monitor naltrex-
`one ingestion during network sessions. Urine is collected
`under staff observation at each visit and tested on site for
`1
`(Roche Diagnostics; Indian-
`opioids with the Accutest
`apolis, IN) method, providing immediate feedback to
`patients and clinical staff.
`After the first two weeks, naltrexone is dispensed and
`ingested at home daily (50 mg of naltrexone with 50 mg
`of riboflavin), under observation of the monitor. In BNT,
`patients continue to attend one individual and one network
`therapy session per week, during which motivational
`techniques and relapse-prevention skills are fostered
`and rehearsed.
`
`2.1.3. Voucher reinforcement program
`Behavioral Naltrexone Therapy provides vouchers con-
`tingent on abstinence from heroin and adherence to naltrex-
`one. Each day of abstinence and each pill taken are rewarded
`with one voucher point ($2), totaling a maximum of 14 points
`or $28 per week, or $672 total if all vouchers are earned over
`six months. Naltrexone compliance is confirmed by both
`recording sheets completed by the monitor and by the
`presence of riboflavin marker in urine samples. Network
`
`X
`X
`X
`
`X
`
`X
`X
`X
`
`X
`
`X
`X
`Xa
`X
`
`Education and support
`Motivational techniques
`Involvement of
`significant others
`Integrated pharmacotherapy
`(naltrexone)
`Behavioral analysis
`Coping with triggers, cravings
`Coping with dysphoria
`Contingent rewards,
`positive incentives
`Social skills
`X
`Social, other non-drug rewards
`X
`Relationship change
`X
`a Antidepressant medication is offered for patients with a depressive
`syndrome persistent during stabilization phase.
`
`X
`X
`X
`
`X
`
`X
`X
`X
`X
`
`2.1.1. Induction phase: assessment and engagement
`The goals of the Induction phase are to recruit, educate,
`motivate, and support the patient
`through the screening
`process and hospitalization for rapid transition to naltrex-
`one. Opiate-dependent patients seeking treatment in BNT
`are initially evaluated by a Master’s or doctoral
`level
`clinician and a psychiatrist to determine psychiatric and
`medical eligibility. Because a central feature of BNT is the
`involvement of at
`least one significant other who will
`monitor naltrexone compliance and support continued
`adherence, potential network members are evaluated and
`are excluded if they abuse substances or are involved in a
`physically abusive relationship with the patient. A patient’s
`network may contain one or more members. One member is
`chosen to be the medication monitor.
`Eligible patients undergo a 7 – 10 day hospitalization for
`detoxification and transition to naltrexone which is mod-
`eled after the clonidine-naltrexone technique (Charney,
`Heninger, & Kleber, 1986). Buprenorphine is utilized ini-
`tially to reduce the severity of withdrawal symptoms;
`thereafter naltrexone is administered in ascending doses
`until a 50-mg dose is tolerated (Collins and Kleber, 2000;
`Stine and Kosten, 1992). Hospital staff is instructed in
`techniques for dealing with fluctuating motivation and
`supporting patients through the considerable physical dis-
`comfort of withdrawal.
`To smooth the transition to outpatient treatment, the BNT
`therapist meets with the patient and significant others at the
`end of the hospital stay to once again orient them again to
`the goals and parameters of BNT and the importance of
`adherence to naltrexone. Managing protracted withdrawal
`symptoms and addressing ambivalence toward abstinence
`and treatment are reviewed in the first outpatient network
`session that is scheduled on the day of discharge. Arrange-
`
`

`

`354
`
`J.L. Rothenberg et al. / Journal of Substance Abuse Treatment 23 (2002) 351–360
`
`members are also reinforced with vouchers, specifically, $1
`for each pill recorded as monitored; a maximum of $168 may
`be rewarded for consistent monitoring.
`Voucher points are exchangeable for goods and services
`chosen by the patient, approved by the therapist and
`purchased for the patient by a program staff member or
`by the significant other, as in the system developed by
`Higgins and colleagues for treating outpatient cocaine
`dependence (Higgins et al. 1991, 1993, 1994). Goods and
`services are chosen to further the goal of developing social
`and recreational outlets to compete with drug use.
`
`2.1.4. Maintenance phase (second through sixth months
`of BNT)
`Once abstinence and naltrexone adherence are success-
`fully established, the focus of BNT expands to address
`broader life-style changes and goals. Network members
`continue to monitor medication and attend weekly network
`sessions. Patients continue to attend one individual and one
`network therapy visit per week. In keeping with the CRA
`model, individual and network therapy sessions are devoted
`to establishing competing/ alternative reinforcers to drug use.
`These reinforcers could include improved marital satisfac-
`tion, a better vocational situation, improved social relation-
`ships and involvement in enjoyable recreational activities.
`During the Maintenance Phase, CORE modules to be
`reviewed include: (a) Orientation to the Program, Rapport
`Building; (b) Functional Analysis of Opiate and Other Drug
`Use; (c) Coping with Cravings; (d) Monitoring Thoughts
`about Drug Use; (e) Problem-Solving Skills; (f ) Drug-
`Refusal Skills; (g) Planning for Emergencies; (h) Seemingly
`Irrelevant Decisions (i) Building a Supportive Network; (j)
`Assessment and Support of Medical Monitoring; and (k)
`Termination. There is no set order of presentation; however,
`all information from these modules should be addressed. It
`is important to note that Managing Negative Moods and
`Depression is a core module for all patients who have
`elevated depression scores at baseline or who present with
`significant depressive symptoms or thinking. For non-
`depressed patients, this module is an elective.
`The BNT therapist will also choose elective modules that
`complement
`the core material and reflect
`the patient’s
`strengths, limitations, and individual goals of treatment.
`These modules include: (a) Social Skills and Relationship
`Enhancement Training; (b) Management of Mood and
`Emotions, which incorporates treatment modules for Aware-
`ness of Anger, Anger Management, Awareness of Negative
`Thinking, Managing Negative Thinking, and Managing
`Negative Moods and Depression; (c) Increasing Pleasant
`Activities; (d) Enhancing Social Support Networks; and (e)
`Job-Seeking Skills.
`
`2.1.5. Handling of lapses and relapses
`Although taking naltrexone provides a safety net for
`patients while they are in BNT, patients may slip, relapse,
`or discontinue the medication. Because we have encoun-
`
`tered a variety of complex situations that demand difficult
`decision-making, the BNT manual provides clinical guide-
`lines to troubleshoot problems that arise. Such situations
`may include use of, or relapse to, opiates during treatment,
`non-compliance with naltrexone or attendance, use of other
`drugs, breakdown of the network, unmasked psychopathol-
`ogy once detoxified, and need for additional medication.
`With regard to naltrexone maintenance and potential
`relapse to opiate use, supervised urine samples are col-
`lected at all visits and tested for opiates on site with an
`Accuttest providing immediate feedback. Naltrexone 50-mg
`pills are packaged in gelatin capsules with riboflavin,
`which is excreted in urine and fluoresces under ultraviolet
`light, providing a check on medication compliance. Nal-
`trexone may also be safely resumed if the specimen tests
`opiate-negative but does not fluoresce, indicating a lapse
`in medication compliance but continued abstinence. How-
`ever, when non-fluorescent urine samples are positive for
`opiates,
`the patient
`is offered a naloxone challenge
`(0.8 mg), which must be negative (i.e. produce no with-
`drawal symptoms) for naltrexone to be safely resumed. If
`the challenge is failed or re-dependence on heroin is
`presumed on clinical grounds, the patient is unable to restart
`naltrexone. Clinical efforts are made to promote abstinence
`with relapse-prevention and motivational enhancement tech-
`niques and then a challenge is re-attempted. If a second
`challenge is failed, then the patient has relapsed and is
`removed from BNT and referred for inpatient detoxification
`or methadone maintenance.
`
`2.2. Therapist training procedures
`
`Similar to manualized approaches upon which BNT is
`based, the following criteria for prospective therapists are
`preferred: (a) a master’s degree or equivalent in psychology,
`counseling, social work, or a closely related field; (b) at least
`3 years of experience working with a substance-abusing
`population; and (c) some familiarity with and commitment
`to cognitive-behavioral theory and approach. All therapists-
`in-training must complete a didactic seminar introducing an
`overview of cognitive-behavioral theory and the therapy
`approaches inherent to BNT. This forum includes role-plays
`and discussions of clinical vignettes and is supplemented
`with review of training tapes and reading materials.
`In order to be certified as a BNT therapist, candidates in
`training are assigned cases that are closely monitored by the
`supervising psychologist. Further, an independent team of
`clinicians trained in BNT must review an audiotaped ther-
`apy session and complete the BNT Therapist Skillfulness
`Form, an original assessment tool which monitors therapist
`adherence, effectiveness, ability to establish treatment
`boundaries, and overall clinical skill. Once the supervising
`BNT psychologist has certified a new clinician, the taped
`sessions are continually monitored in weekly individual and
`group supervision to ensure the therapist’s fidelity to the
`treatment delivered and to minimize technique drift.
`
`

`

`J.L. Rothenberg et al. / Journal of Substance Abuse Treatment 23 (2002) 351–360
`
`355
`
`2.2.1. Measure of therapist adherence
`In order to reliably monitor adherence to BNT, a 33-item
`BNT Session Checklist has been established that integrates
`core elements of Network Therapy, CRA, Motivational
`Enhancement, RPT, and contingency management. These
`checklists are completed after each therapy session by the
`therapist and provide a clinical tool to assess adherence to
`BNT as they are reviewed in weekly supervision meetings.
`The BNT Checklist requires that
`the therapist endorse
`whether an event or a particular topic was covered in that
`session (e.g., yes, no) or how extensively he/she reviewed
`the selected manualized material (1 = not at all; 5 = very
`extensively). In general, measures of adherence to, and
`competence with BNT as well as training procedures have
`been developed according to the guidelines set forth by
`Carroll, Nich, and Rounsaville (1998).
`
`2.3. Pilot trial
`
`2.3.1. Design and methods
`Prospective participants were evaluated with a psychi-
`atric history, a modified SCID interview which delineates
`substance use and onset of psychiatric symptoms (Nunes
`et al., 1996), and physical and laboratory evaluations and
`were eligible if they met DSM-IV criteria for current opiate
`dependence, were voluntarily seeking treatment, and had a
`significant other who could commit
`to participating in
`treatment. Unstable medical or psychiatric disorders, which
`might make participation hazardous, were exclusionary.
`After providing informed consent, consents that were in
`accord with the standards of our institution, patients were
`hospitalized for detoxification and transition to naltrexone.
`Those able to tolerate two daily doses of the full dose
`(50 mg) of naltrexone at the end of a 7 – 10 day hospitaliza-
`tion were discharged to outpatient treatment with BNT.
`Doctoral level psychologists conducted BNT according to
`the working treatment manual. Selection,
`training, and
`ongoing monitoring and supervision of therapists were
`conducted as described above over the subsequent 24 weeks.
`Patients were monitored weekly for drug use by self-report
`and urine toxicology, naltrexone compliance, psychiatric
`symptoms, and adverse events.
`
`3. Results
`
`3.1. Participants
`
`More than 150 potential subjects were initially eval-
`uated, with approximately 70% (105/150) returning for
`second visits. Of these individuals, 78% (82/105) were
`eligible for treatment; the remaining applicants were found
`to be ineligible secondary to either medical (e.g. signifi-
`cantly elevated liver enzymes) or psychiatric exclusion
`criteria, or failing to have an available or appropriate
`monitor. Of those who were deemed eligible, 47 subjects
`
`completed the detoxification and transition to naltrexone
`and entered outpatient
`treatment with BNT. The 35/82
`participants who were eligible but did not enter treatment
`either failed to enter due to ambivalence or accepted
`referrals to alternative treatment settings when the delay
`to enter was experienced as too long. Enrolled participants
`averaged 33.6 ± 9.3 years of age (range: 20 – 54) and
`included 36 (77%) males, 11 (23%) females, 30 (64%)
`Caucasians, 12 (25%) Hispanic-Americans and 5 (11%)
`African-Americans. 17 (36%) were currently married and
`32 (68%) were employed. 18 (39%) met criteria for a
`Depressive Disorder, 16 (35%) for an Anxiety Disorder,
`and 7 (15%) for Antisocial Personality Disorder. Concurrent
`drug use in addition to opiates was common (13 (27%)
`abusing Cocaine, 14 (30%) Marijuana, 7 (15%) Alcohol,
`and 25 (51%) Nicotine. Seventeen (37%) used heroin
`intravenously, and 29 (63%) used heroin intranasally, and
`one patient used exclusively methadone. On average, par-
`ticipants in this sample were using 5.42 (SD = 5.89) number
`of bags of heroin per day and using opiates regularly
`8.62 years (SD = 8.14). Seventeen (36%) were using illicit
`methadone regularly, and one was in methadone mainte-
`nance. Five (11%) patients attended adjunct
`treatment
`opportunities such as AA or NA while participating in BNT.
`
`3.2. Reliability of BNT session checklist
`
`In order to assess the reliability of this 33-item instru-
`ment,
`inter-rater reliability coefficients were computed.
`Thirty audio-taped therapy sessions were rated by both the
`therapist who conducted the session and by a master’s level
`clinician who was introduced to the manual and trained on
`the session checklist and rating task. The thirty tapes, all
`from different patients, were randomly selected from early
`sessions (i.e., sessions 2, 3, or 4) to reduce bias and other
`complications including early dropout from treatment.
`Kappa coefficients were computed for nominal items (e.g.,
`yes, no) and intraclass correlations were computed for
`continuous items. The majority of Kappa and ICC estimates
`fell within the good (0.60 – 0.75, 27% (9 of items) or
`excellent ( > .75, 52% (17) of items) range and reflected
`core clinical ingredients of BNT. The seven remaining items
`which demonstrated weaker reliability are being modified
`and re-evaluated.
`
`3.3. Treatment outcome
`
`On average, 7.8 (SD = 10.3) individual sessions and
`4.9 (SD = 6.7) network sessions were attended. The
`session checklists reflect
`that
`in 45-min sessions, on
`average, 29.5 (SD = 9.6) minutes were devoted to BNT
`skill building elements.
`The principal outcome measure was retention in treat-
`ment. Participants were removed from the trial and counted
`as dropouts if they ceased naltrexone and resumed opiate
`use with physiological dependence as indicated by failure to
`
`

`

`356
`
`J.L. Rothenberg et al. / Journal of Substance Abuse Treatment 23 (2002) 351–360
`
`Fig. 1. Retention in treatment.
`
`pass a naloxone challenge and restart naltrexone. This is the
`most clinically meaningful outcome in this population since
`at
`this point
`treatment has failed, naltrexone cannot be
`resumed, and rehospitalization for detoxification or admis-
`sion to agonist maintenance is warranted. Overall, retention
`in treatment was low, with 55% (26/47) completing at
`least 4 weeks of treatment, 40% (19/47) completing
`8 weeks, and only 19% (9/47) completing the entire 24 week
`treatment course.
`As the pilot trial progressed, it became clear clinically
`that patients using methadone regularly at baseline (either
`dependent use of street methadone or methadone main-
`tained) had markedly poorer outcome. Fig. 1 presents
`survival curves for retention in treatment over time in
`patients with methadone use at baseline (n = 18) vs. those
`with heroin only at baseline (n = 29). The difference
`between the groups is significant (Log rank (1) = 8.47,
`p < .01). As can be seen in Fig. 1, retention was very poor
`among patients with methadone use at baseline, among
`whom only 39% (7/18) were retained to 4 weeks, 16%
`completed 8 weeks (3/18), and 0% completed the 24 week
`program. This finding suggests that satisfactory methods for
`transitioning these patients from methadone to naltrexone do
`not yet exist, and is consistent with other studies that have
`noted this difficulty (Charney et al., 1986).
`Results from subjects using heroin only (no methadone)
`at baseline (n = 29) are more encouraging. Sixty-five per-
`cent (19/29) were retained to 4 weeks, 55% completing 8
`weeks (16/29), and 31% (9/29) completing the 24 week
`program. Additional findings reflect promise for BNT. A
`significant positive correlation was found between length of
`time in treatment and both percentage of opiate-free urine
`samples (r = .65, p < .001) and adherence to naltrexone (r =
`.75, p < .001). Specifically, for individuals who remained in
`treatment beyond 8 weeks, on average, 88.0% of urine
`samples were free of opiates; 88.6% for those who remained
`in treatment beyond 16 weeks. For those who completed
`
`beyond 24 weeks, 93.4% of urine samples were free of
`opiates and naltrexone was adhered to 94% of the time
`during their participation. Further, mean total vouchers
`earned during study participation by heroin dependent
`patients, which is based on both providing clean urine
`samples and adhering to naltrexone, was $292 (SD = 299).
`Comparing participants using only heroin to those using
`methadone in addition to heroin, there was a nonsignificant
`trend in the direction of symptoms of withdrawal during the
`first outpatient week being milder in those using heroin
`only. Specifically,
`the mean score on the Short Opiate
`Withdrawal Scale obtained in the first outpatient week
`was 40.4 (SD = 14.1) in patients with methadone use com-
`pared to 33.1 (SD = 9.7) in those with heroin alone (t = 1.60,
`df = 26, p = .13).
`
`3.4. Adverse events
`
`Patients removed from the trial were treated support-
`ively and appropriate referral was arranged. There were no
`serious adverse events during BNT treatment. However,
`one patient who had dropped out of treatment two weeks
`previously and avoided efforts to arrange another treatment
`referral subsequently died of a heroin overdose. No sub-
`jects were removed because of side effects from naltrexone
`or psychiatric worsening.
`
`4. Discussion
`
`Naltrexone maintenance, while a theoretically ideal treat-
`ment, has been of limited effectiveness and not well
`received by patients (Preston et al., 1999). Behavioral
`Naltrexone Therapy was developed to improve the effec-
`tiveness of naltrexone maintenance by integrating empir-
`ically-validated cognitive and behavioral
`techniques,
`including reinforcing adherence to naltrexone with vouch-
`
`

`

`J.L. Rothenberg et al. / Journal of Substance Abuse Treatment 23 (2002) 351–360
`
`357
`
`ers, monitoring by a significant other, and encouraging
`positive lifestyle changes. A treatment manual for BNT
`was developed together with procedures for therapist selec-
`tion and training, and measures of therapist competency and
`adherence. An uncontrolled pilot trial was conducted with
`47 opiate-dependent participants. Overall, retention in treat-
`ment was poor with only 19% of patients completing the
`six-month course of treatment. Methadone use was found to
`carry a particularly poor prognosis, while results were more
`encouraging for the subgroup of heroin addicts not using
`methadone. In the latter group, 65% of heroin addicts were
`retained beyond four weeks and 31% were retained to
`six months.
`The largest obstacle to maintaining abstinence was
`retaining patients in the first few weeks of treatment. Green-
`stein et al. (1981) found that 47% of opiate addicts who
`were started on naltrexone dropped out of treatment in the
`first few weeks. Similarly, Tennant, Rawson, Cohen, and
`Mann (1984) found considerable dropout (27.5% (44/160))
`after only a few days when detoxified opiate addicts were
`provided with naltrexone. Our observation of very elevated
`early attrition rates in those patients regularly using illicit
`methadone, suggests that the transition from a long-acting
`opiate to naltrexone is poorly tolerated physiologically, and
`improved methods for detoxification and transition are
`needed for such patients. This was supported by the non-
`significant trends towar