Disclaimer: The Health Technology Inquiry Service (HTIS) is an information service for those involved in planning and providing health care in Canada. HTIS responses are based on a limited literature search and are not comprehensive, systematic reviews. The intent is to provide a list of sources and a summary of the best evidence on the topic that CADTH could identify using all reasonable efforts within the time allowed. HTIS responses should be considered along with other types of information and health care considerations. The information included in this response is not intended to replace professional medical advice, nor should it be construed as a recommendation for or against the use of a particular health technology. Readers are also cautioned that a lack of good quality evidence does not necessarily mean a lack of effectiveness particularly in the case of new and emerging health technologies, for which little information can be found, but which may in future prove to be effective. While CADTH has taken care in the preparation of the report to ensure that its contents are accurate, complete and up to date, CADTH does not make any guarantee to that effect. CADTH is not liable for any loss or damages resulting from use of the information in the report.
Copyright: This report contains CADTH copyright material. It may be copied and used for non-commercial purposes, provided that attribution is given to CADTH.
TITLE: Naltrexone for the Treatment of Alcohol Dependence: A Review of the Clinical and Cost-Effectiveness
DATE: 08 April 2009
CONTEXT AND POLICY ISSUES:
Alcohol dependence is a chronic disease that has significant consequences on the public health care system.1 The primary goals of treating alcohol dependence include maintaining abstinence, increasing the duration of the interval before a relapse, and reducing the intensity of drinking if a relapse occurs.1
Naltrexone (ReVia®) is a mu-opioid receptor antagonist that is approved for use in Canada for the treatment of alcohol dependence.2 Naltrexone is believed to modulate the reward effects associated with alcohol intake.3 Acamprosate (Campral) is an N-methyl-D-aspartate (NMDA) and metabotropic glutamate receptor antagonist that is approved in Canada for the
maintenance of abstinence from alcohol.4 In 2008, the Canadian Expert Drug Advisory Committee recommended that acamprosate be listed for patients in which naltrexone is contraindicated.4 This report will review the evidence of the clinical and cost-effectiveness of naltrexone for the treatment of alcohol dependence and where possible, make comparisons to acamprosate.
RESEARCH QUESTIONS:
1. What is the clinical-effectiveness of naltrexone for the treatment of alcohol dependence? 2. What is the cost-effectiveness of naltrexone for the treatment of alcohol dependence? METHODS:
A limited literature search was conducted on key health technology assessment resources, including OVID MedLine, OVID Embase, OVID PsycINFO, The Cochrane Library (Issue 1,
2009), University of York Centre for Reviews and Dissemination (CRD) databases, ECRI, EuroScan, international health technology agencies, and a focused Internet search. Results include articles published between 2004 and March 2009, and are limited to English language publications only. Regular alerts were established on MedLine, Embase, and PsycINFO and information retrieved via alerts is current to April 6, 2009. Filters were applied to limit the retrieval to health technology assessments, systematic reviews, meta-analyses, randomized controlled trials (RCTs), and economic studies.
Included in this report are systematic reviews and RCTs that evaluated naltrexone, alone or in combination, for the treatment of alcohol dependence. Systematic reviews were excluded that used non-systematic review methodology (i.e. fewer than two databases searched and less than two individuals screening for inclusion and performing data abstraction). Systematic reviews and RCTs were excluded if the evaluation of the effectiveness of naltrexone was not a main objective of the study, the outcomes assessed did not include either time to first drink (or days of continuous abstinence) or time to relapse to heavy drinking, and if a study was based on total or partial duplication in publication of results that was published in full elsewhere. Unless otherwise stated the dose of naltrexone was 50 mg per day and trial participants were over the age of 18 years, were abstinent from alcohol for a minimum of three to 30 days prior to randomization (regardless of method of detoxification), were not previously treated with study drugs within three months of randomization, had no drug dependencies other than nicotine, met the criteria for alcohol dependence as detailed in the American Psychiatric Association’s
Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV)5 or in the revised text version (DSM-IV-TR),6 and did not have severe psychiatric disorders or liver disease. In the included studies, heavy drinking was defined as ≥ 5 drinks per day for men and ≥ 4 drinks per day for women. RCTs meeting inclusion criteria, but were previously included in a systematic review, were not appraised separately in this report.
HTIS reports are organized so that the higher quality evidence is presented first. Therefore, health technology assessment reports, systematic reviews, and meta-analyses are presented first. These are followed by RCTs and economic evaluations.
SUMMARY OF FINDINGS:
Our literature search identified one health technology assessment (HTA), three systematic reviews, one analysis, 11 RCTs that were not included in the systematic reviews or meta-analysis, and two economic studies.
Health technology assessments
In 2003, the Health Technology Board for Scotland conducted a HTA of pharmacological and psychosocial treatments for alcohol dependence which was updated in 2005.7 A major objective of the report was to determine the most effective treatment for alcohol dependence. For
inclusion, studies had to be RCTs evaluating a pharmacological or a psychosocial intervention with a minimum treatment duration of three months. The updated report included data from 19 trials of acamprosate and 21 trials involving naltrexone in their analysis. The outcome of interest was either abstinence or controlled drinking. The combined success rates for abstinence or controlled drinking at the end of the trial period was 51% for naltrexone (41% for placebo) and 26% for acamprosate (15% for placebo). It was not stated whether these rates were significant. The authors concluded that both naltrexone and acamprosate were effective in preventing relapse. Limitations of the report included the lack of available information on the primary
studies including patient characteristics, dosages, co-dependencies, and average treatment duration. In addition, outcomes were presented as combined and it was not clear how the authors defined controlled drinking.
Authors of the HTA performed an assessment of cost-effectiveness. The incremental cost or savings per additional abstinent patient was determined. Few details regarding the model the investigators used were available. The costs associated with seven therapies including
naltrexone and acamprosate were compared to a standard care package (details not provided). Relevant disease outcomes included alcoholic psychosis, liver cirrhosis, epilepsy, chronic pancreatitis, cancer, stroke, and death. One thousand treatment-compliant patients were modeled. Acamprosate produced a net savings per abstinent patient. Although naltrexone treatment was associated with a net cost (exact costs were not provided) per abstinent
participant, it was still considered to be cost-effective. The economic evaluation had a number of limitations including lack of information regarding the model used, the perspective taken, and the assumptions made.
Systematic reviews and meta-analyses
A meta-analysis was conducted by Rösner et al. (2008)8 with the objective of comparing the effectiveness of naltrexone with acamprosate. Included in the analysis were placebo-controlled RCTs that had a minimum treatment duration of four weeks and with at least 10 participants in each treatment group. Results from forty-one published and unpublished studies were included in the meta-analysis. Outcomes of interest included return to any drinking and relapsing to heavy drinking. The risk of returning to any drinking while taking naltrexone was reduced by 93% compared to the control group [Relative Risk (RR) = 0.93, 95% confidence interval (CI): 0.88 to 0.99; based on 18 studies]. Treatment with acamprosate also significantly reduced the risk of having a first drink compared to placebo (RR = 0.84, 95% CI: 0.78 to 0.91; based on 20 studies). The relative risk of relapsing to heavy drinking was 0.88 for naltrexone (95% CI: 0.80 – 0.96; based on 16 studies). Acamprosate treatment did not have a significant effect on reducing the risk of relapsing to heavy drinking (based on 20 studies). Authors concluded that naltrexone had a significant effect on the maintenance of abstinence as well as the prevention of relapsing to heavy drinking. Acamprosate treatment was shown to support abstinence only. Authors further suggested that the choice of drug may be dependent on the treatment goal, i.e. abstinence or controlled drinking without heavy drinking. Two of the authors had received funding from the manufacturers of acamprosate.
A systematic review by Roozen et al. (2006)9 evaluated the evidence regarding the
effectiveness of naltrexone in the maintenance treatment of opioid and alcohol dependence. To be considered for inclusion studies had to be either RCTs or controlled clinical trials (CCTs) conducted in the out-patient setting. Seventeen placebo-controlled studies related to alcohol dependence were included. The methodological quality of the trials was assessed using a criteria list from the Cochrane Drugs and Alcohol Group. Quality was assessed as high. Outcomes of interest included relapse rate, time to first drink or days remaining abstinent, and time to first relapse to heavy drinking. Pooled difference in relapse rates calculated from 14 studies was 13% (95% CI: 7% to 18%) in favour of naltrexone compared to placebo. The pooled difference from seven studies in the proportion of participants remaining abstinent was not significant. Data regarding time to first relapse to heavy drinking were not reported with standard deviations in the majority of studies evaluating this outcome therefore pooling was not possible. However, five of nine naltrexone studies evaluating this outcome did report a statistically
significant increase in time to first relapseto heavy drinking for treated patients. The majority of the 17 reports evaluated effectiveness following 12 to 13 weeks of treatment. There was
insufficient data to draw conclusions regarding longer-term effectiveness. The authors concluded that naltrexone is clinically effective for the treatment of alcohol dependence. A limitation of this systematic review was that the authors did not contact study authors to receive additional study information. Statistical analyses were sometimes limited due to missing and unreported data. No conflicts of interest were reported by authors.
In 2005, Srisurapanont and Jarusuraisin published a systematic review of RCTs assessing opioid antagonists for alcohol dependence.10 The objective of the review was to determine the effectiveness of opioid antagonists in decreasing or preventing relapse compared with placebo. To be considered for inclusion, studies had to be RCTs involving alcohol-dependent
participants. Eighteen RCTs investigating short-term outcomes (up to three months following naltrexone treatment) met inclusion criteria. Compared to placebo, treatment with naltrexone significantly decreased relapse to heavy drinking (RR = 0.64, 95% CI: 0.51 to 0.82) and decreased the proportion of participants returning to any drinking (RR = 0.87, 95% CI: 0.76 to 1.00). Naltrexone had no significant benefit on relapse rates beyond 12 weeks of treatment. Authors concluded that naltrexone is effective as a short-term treatment for alcohol
dependence. Two additional studies evaluated nalmefene; however, only one study reported on relapsing to heavy drinking as an outcome. The single study was a 12-week, double-blind, placeb-controlled trial involving 102 participants. Participants received doses ranging from 20 to 80 mg per day. Nalmefene significantly reduced the rate of relapse to heavy drinking compared to treatment with placebo [RR = 0.62, 95% CI: 0.41 to 0.93). Outcomes did not differ between the 20- and 80-mg dose nalmefene groups. Trial authors concluded that nalmefene was effective at preventing relapse in alcohol-dependent adults. Authors of the systematic review stated that this trial did not provide details regarding randomization. The authors declared no conflicts of interest.
A systematic review in 2004 by Carmen et al.11 evaluated the efficacy of naltrexone and
acamprosate in the treatment of alcohol dependence. For inclusion, studies had to be RCTs or CCTs involving alcohol-dependent adults, comparing naltrexone or acamprosate to placebo, to each other, or to a reference group not receiving medication or placebo. The literature search identified 33 trials: 19 RCTs with naltrexone versus placebo or reference group; 13 RCTs with acamprosate versus placebo; and one RCT comparing naltrexone with acamprosate. Outcomes of interest included abstinence rate and rate of relapse to heavy drinking. Acamprosate
treatment was associated with a significant improvement in abstinence rate [Odds Ratio (OR) = 1.88 (95% CI: 1.57 to 2.25); P<0.0001] and had no effect on relapse to heavy drinking.
Naltrexone treatment reduced the rate of relapse to heavy drinking significantly [OR = 0.62 (95% CI: 0.52 to 0.72); P<0.0001], but did not improve abstinence rate. Authors concluded that both drugs showed some clinical-effectiveness and that acamprosate may be useful if the therapeutic goal is to achieve abstinence and naltrexone may be most effective in promoting controlled drinking. No conflicts of interest were reported by authors.
Randomized controlled trials Placebo-controlled trials
Farren et al. (2009)12 examined the hypothesis that the co-treatment of naltrexone and the selective serotonin re-uptake inhibitor (SSRI) sertraline would be more effective in the treatment of alcohol dependence than naltrexone alone. This rationale was based on pre-clinical evidence that supported a synergy between the opioid and serotonin systems in modulating alcohol consumption. The placebo-controlled trial evaluated 111 alcohol-dependent, non-depressed participants. Participants were randomized to receive either naltrexone combined with sertraline
at a dose of 100 mg (n=57) or naltrexone combined with placebo (n=54) for 12 weeks. The mean age was 43.2 years and approximately 82% of participants were male. The two primary outcomes of interest were time to first drink and time to relapse to heavy drinking. There was no significant difference between groups in terms of the primary outcomes. Authors concluded that the study did not provide sufficient evidence that supported their hypothesis that sertraline would augment the effectiveness of naltrexone. Two of the study authors declared a conflict of interest with pharmaceutical companies.
O’Malley et al. (2008)13 also evaluated the effectiveness of naltrexone alone and in combination with the SSRI sertraline (100 mg). A total of 101 participants were randomized to receive either placebo (n=34), naltrexone plus placebo (n=34), or naltrexone plus sertraline (n=33) for 16 weeks. The mean age was 40.0 years and 66% of the sample population was male. Outcomes of interest included time to first heavy drinking day and total abstinence. Monotherapy
naltrexone treatment resulted in a significantly higher proportion of participants reporting abstinence (35%) versus placebo-treated participants (12%) (P=0.027). No significant differences in rates of relapse to heavy drinking were reported nor were any significant differences observed for either outcome between monotherapy naltrexone or the combined therapy. Authors concluded that naltrexone was an effective treatment for alcohol dependent individuals and that combined naltrexone-sertraline treatment did not offer an advantage over naltrexone alone. Analysis was performed on the intention-to-treat (ITT) population; however, it was not stated how data from those that did not complete treatment were incorporated into the analysis. Of note, the authors performed a sub-analysis based on native and non-native participants within the study population; however, these results are beyond the scope of this report. Several of the study authors declared a conflict of interest with pharmaceutical companies.
A RCT by Olsin et al. (2008)14 evaluated three types of psychosocial interventions in
combination with naltrexone or placebo. The three psychosocial interventions were cognitive behavioural therapy (CBT) + medication management (which consisted of five to ten minute sessions with a research physician wherein information regarding alcohol use, medication use, illness, or side-effects experienced with treatment medications), BRENDA (Bio-psychosocial evaluation; Report to patient on assessments including linking drinking to health status; Empathetic understanding of the patient’s situation; Needs that must be addressed; Direct advice to patient on how to meet these needs; and Assess reaction/behaviours of patients and adjust as necessary for best care) which is an intervention that provides motivational
enhancement and promotes adherence to pharmacotherapy + medication management, and thirdly, medication management alone. The average age in the study was 41 years and 72.9% of participants were male. The subjects were randomized to the following groups: naltrexone + CBT (n=40), placebo + CBT (n=40), naltrexone + BRENDA (n=39), placebo + BRENDA (n=40), naltrexone + medication management (n=41), and placebo + medication management (n=40). Naltrexone was given at a dose of 100 mg. Outcomes of interest included percent abstinence and percent any heavy drinking. Results demonstrated that naltrexone was not effective over placebo at modulating the percentage of participants who were abstinent or who engaged in any heavy drinking. There was also no effect when naltrexone was combined with any psychosocial intervention. Authors noted that treatment adherence in the study was low at 50% across all conditions although adherence was based only on pill counts. Limitations of the study included a lack of blinding. Participants were blinded to the medication they received, but neither the
participants nor research staff was blinded to the type of psychosocial intervention. No conflicts of interest were reported by authors.
A double-blind RCT was conducted by Morley et al. (2006)15 to compare the effectiveness of naltrexone and acamprosate. A total of 169 alcohol-dependent participants were randomized to receive either naltrexone (n=53), acamprosate (n=55), or placebo (n=61) for 12 weeks. The mean age of the sample was 45.9 years and approximately 70% of the population was male. Outcomes of interest included time to first drink and time to first relapse. No differences were reported for either outcome between treatment groups. Although depressed individuals were not targeted specifically for this study, subjects were stratified based on depression scores (no depression, mild, moderate, or severe) according to the Depression, Anxiety, and Stress Scale. Subjects with ‘no depression’ who were treated with naltrexone had a significantly longer time to first relapse to heavy drinking (P<0.01). Participants were also ranked according to level of alcohol dependence (low, moderate, or severe). Participants who had a ‘low dependence’ on alcohol had a significantly longer time to first relapse to heavy drinking (P<0.05). There was no effect on time to first drink with naltrexone treatment in those with low dependence. No effect by acamprosate was noted for either outcome in any of the subpopulations. No conflicts of interest were reported by authors.
The COMBINE study conducted by Anton et al. (2006)16 evaluated the effectiveness of pharmacotherapies, behavioural interventions, and their combinations for the treatment of alcohol dependence. A total of 1383 participants (mean age = 44 years; 70% male) were randomized to receive either placebo, acamprosate (3,000 mg) naltrexone (100 mg), or acamprosate combined with naltrexone. These four treatment groups were combined with or without cognitive behavioural intervention (CBI; consisted of up to 20 sessions 50 minutes long integrating aspects from CBT, 12-step programs, and motivational interviewing). A ninth group received CBI alone and all groups received medication management. Outcomes of interest were percent days abstinent and time to first heavy drinking day. Outcomes were assessed at end of treatment period (16 weeks) and at one-year follow-up. Only two of the nine groups
demonstrated significant differences in outcomes. With regards to heavy drinking, naltrexone monotherapy reduced the risk of heavy drinking (RR = 0.72, 97.5% CI: 0.53 to 0.98; P=0.02) and a significant increase in percent days abstinent was observed when naltrexone was combined with CBI (P=0.009).. Acamprosate, alone or in combination, had no effect on either outcome. There were no significant differences in alcohol-related outcomes at one-year post-treatment; however treatment with naltrexone was still associated with a decreased risk of returning to heavy drinking [Hazard Ratio (HR) = 0.77, 97.5% CI: 0.58 to 1.02; P=0.04],
although not significant. Several of the study authors declared conflicts of interest with regards to funding through the pharmaceutical companies.
Kiefer et al. (2004)17 randomized 160 participants to receive either placebo, naltrexone, acamprosate, or naltrexone in combination with acamprosate for 12 weeks. The mean age of the study population was 46.2 years and 74% of the sample was male. Outcomes of interest included maintenance of abstinence and relapse to heavy drinking. Significant decreases in the percentage of those relapsing to heavy drinking were noted in the naltrexone treatment group (35.3%) versus placebo (75%) (P<0.05); acamprosate (50%) versus placebo (75%) (P<0.05); and naltrexone plus acamprosate (27.5%) versus placebo (75%) or naltrexone alone (35.3%) (P<0.05). No significant differences were observed with respect to abstinence in any of the groups. Authors concluded that the combined treatment of naltrexone and acamprosate may be more beneficial than treatment with naltrexone alone. Limitations of this study included that monitoring of intake of alcohol relied of self-report rather than urinalysis. In addition, the sample group was recruited from an inpatient setting, therefore these participants may represent a more clinically challenging population to treat compared with those presenting to an out-patient setting and may be less likely to remain abstinent. No conflicts of interest were reported by authors.
Anton et al. (2005)18 examined the effectiveness of naltrexone when combined with CBT or with a motivational enhancement program (MET) that involved limited professional contact. One-hundred and sixty participants were randomized to receive either naltrexone or placebo and either CBT or MET [naltrexone + CBT (n=39); naltrexone + MET (n=41); placebo + CBT (n-41); placebo + MET (n=39)] for a 12-week period. The average age was 44 years and 75% of the study population was male. Outcomes of interest included percent drinking abstinence and percent relapsing to heavy drinking. Naltrexone had an overall significant treatment effect of decreasing percent relapsing to heavy drinking (P=0.022). The OR of relapse to heavy drinking, regardless of whether naltrexone was combined with either CBT or MET, was 0.48 (95% CI, 0.25 to 0.90). Naltrexone in combination with CBT also significantly increased the percent days abstinent from 79% in the placebo + CBT treatment group to 91% in the naltrexone + CBT treatment group. Authors concluded that data from this study support the use of CBT with naltrexone for the treatment of alcohol dependence. No conflicts of interest were reported by authors.
Direct comparison trials
A double-blind randomized comparison trial was conducted by Martinotti et al. (2009)19 comparing naltrexone with aripiprazole, a dopamine/serotonin system stabilizer, for the treatment of patients with alcohol dependence. A total of 57 participants were randomized to receive either naltrexone (n=28) or aripiprazole (n=29) at a dose of 5-15 mg for 16 weeks. The mean age of the study participants was 40.3 years and 80% of the sample was male. This study targeted participants who had a goal of total abstinence. Detoxification was via benzodiazepine treatment. The outcomes of interest were maintenance of abstinence and relapse to heavy drinking. There were no significant differences observed for either of the outcomes between treatment groups. From these results authors concluded that aripiprazole had the same level of effectiveness as naltrexone. Limitations of this study included the small sample size and lack of a placebo group. In addition, baseline characteristics were not given in full. The authors
declared no conflicts of interest.
A study by De Sousa and De Sousa (2008)20 compared naltrexone with disulfiram for preventing relapse in adolescents between the ages of 15 and 18 years with alcohol dependency. A total of 58 adolescents were randomized to receive either naltrexone (n=29) or disulfiram (n=29) at a dose of 250 mg. Mean age of the sample was 17.3 years and the breakdown by gender was not specified. Outcome was days to first intake of alcohol which was assessed at the end of the six-month treatment period. Authors reported that significantly more disulfiram-treated participants had been abstinent for the treatment period (79.3%) versus naltrexone-treated participants (51.7%; P=0.0001). The percentage of participants who completed the study was 93.1%. All patients had a stable family environment to promote maximal treatment compliance. Authors concluded that disulfiram performed significantly better than naltrexone in preventing relapse to any drinking. Limitations of this study included the small sample size and lack of a placebo group. The authors declared no conflicts of interest.
A comparative trial of disulfiram (100-400 mg, twice weekly), naltrexone, and acamprosate was conducted by Laaksonen et al. (2008).21 During the first 12 weeks of the study, medication dosing was supervised and during weeks 13 to 52, participants took medication when they perceived their propensity to drink to be high. Two-hundred and forty three patients were
randomized equally to each of the three treatment arms. The mean age across groups was 43.1 years and approximately 70% of the sample was male. Outcomes of interest included time to first drink and time to first heavy drinking day. During the supervised treatment period, disulfiram treatment resulted in a significantly (P=0.0002) longer interval to first drink (30.4±27.8) days
than either naltrexone (16.2±20.2) or acamprosate (11.4±17.0). Disulfiram treatment also resulted in a significantly (P<0.0001) longer interval to time to first heavy drinking day at 46.6±27.5 days as compared to naltrexone and acamprosate at 22.0±22.0 and 17.6±22.0, respectively. Authors concluded that during the supervised treatment period, disulfiram was superior to either naltrexone or acamprosate for the treatment of alcohol dependence. During the targeted medication period, there was no significant difference between treatment groups for either outcome; however, there was a significant increase in the mean number of abstinence days per week in the disulfiram group compared to naltrexone (P=0.0006) and acamprosate (P=0.0015). Authors did note that pill counts were not performed during this period and the only information regarding medication adherence was from patient diaries and suggested that this limited conclusions. No conflicts of interest were reported by authors.
De Sousa and De Sousa (2004)22 compared the effectiveness of naltrexone and disulfiram in treating alcohol-dependent men in an open label trial. One-hundred men (mean age, 44.4 years) were randomized equally to receive either naltrexone or disulfiram (250 mg per day). Family members supervised medication dosing with the intent of enhancing compliance. Outcomes of interest included percent abstinence, percent relapsing to heavy drinking, and number of days abstinent. After one year, 90% of disulfiram-treated participants were abstinent versus 44% of naltrexone-treated participants (P=0.0002). The numbers of days of abstinence was also greater with disulfiram treatment versus naltrexone (243 days versus 180,
respectively; P=0.03). The relapse rate to heavy drinking was significantly lower in participants receiving disulfiram at 14% versus naltrexone at 56% (P=0.0009). Authors concluded that disulfiram was superior to naltrexone for the maintenance of abstinence and for preventing relapse to heavy drinking in alcohol-dependent men. This study was an open-label trial and the authors remarked that bias could have been introduced had the investigators noted an
improvement in one arm over the other. No conflicts of interest were reported by authors. Economic evaluations
Zarkin et al. (2008)23 reported the prospective costs and cost-effectiveness for the study interventions evaluated in the COMBINE study.16 The main outcome measures included incremental cost per percentage point increase in percentage of days abstinent and per participant avoiding heavy drinking. The evaluation used a micro-costing approach from the perspective of the treatment provider. The cost of each intervention was based on the sum of space, labour, medication, and laboratory costs. The daily drug cost of naltrexone was US$ 2.74 and US$ 5.76 for acamprosate. The incremental cost-effectiveness ratio (ICER) of medication management in combination with naltrexone is US$ 42.24 per percentage point increase in percent days abstinent and US$ 2,846.85 per participant avoiding heavy drinking. With the addition of acamprosate to the regimen, the ICER increased to US$ 8,095.12 per participant avoiding heavy drinking. Limitations of this evaluation included the study type in that a cost-utility analysis would have provided additional information pertaining to cost per quality adjusted life year (QALY), the clinical-effectiveness data were derived from a single study, the time horizon was limited to the treatment period, and a probabilistic sensitivity analysis was not performed. In terms of generalizability, the clinical and cost inputs for the model may be different in the Canadian health care system.
A cost-utility analysis of interventions for problem drinking and alcohol dependence was performed by Mortimer and Segal (2005).24 Authors used a time-dependent state transition model to estimate QALY gained for each intervention as compared to usual care. Incremental costs and benefits were estimated from a societal perspective. The assumptions included a cycle length of three months for naltrexone and modeled full-life expectancy. This analysis
excluded possible cost-saving due to a reduction or delay in treatment costs associated with liver cirrhosis because there was a high level of uncertainty with these estimates. The estimated cost per QALY gained for naltrexone versus placebo was AU$ 12,996. Compared to placebo, naltrexone is estimated to result in 0.0528 QALY gained per completer at an incremental cost per completer of AU$ 685. The evaluation looked at interventions for problem drinking, mild to moderate dependence, and patients with severe dependence. Authors concluded that
interventions for problem drinking were of best value compared with those for severe physical dependence; however, naltrexone was the only intervention evaluated for the treatment of severe dependence. A limitation of this evaluation includes the use of a univariate sensitivity analysis rather than a probabilistic analysis. In terms of generalizability, the clinical and cost inputs for the model may be different in the Canadian health care system.
Limitations
The evidence for clinical-effectiveness of naltrexone included in this report comes from systematic reviews, a meta-analysis, and RCTs, all of which are considered to be high-quality evidence. This report did not assess the methodological quality of the included trials, although the inclusion criteria were strict and selected for studies in which the methods were clear and had adequate randomization, blinding, and allocation methods. Many of the included trials also had strict inclusion criteria for participants and excluded those from inpatient settings who may have had more severe alcohol dependence15 and participants with dependencies or co-morbidities. This approach may not be reflective of the entire alcohol-dependent patient population. Participants from outpatient settings or who answered an advertisement for trial accrual perhaps could be considered treatment-seeking and may respond to treatment differently than those being referred from inpatient settings. In general, the level of alcohol dependence of trial participants may have varied within studies and between studies included in this report.
The main weakness of the included studies was the lack of details provided regarding how study authors handled withdrawals and dropouts. Only a few of the studies stated that for the purposes of analyses, participants that were lost to follow-up were assumed to have relapsed one day after their last assessment.
A challenge with the available literature related to the ambiguity in the definitions of outcomes. Not all of the studies analyzed the same outcomes or they were defined differently between studies. In addition, several of the studies included in the meta-analysis by Rösner et al.8 did not describe their findings quantitatively and data points had to be estimated from figures.
The majority of the trials were of relatively short duration, typically not extending beyond the treatment period. A chronic illness such as alcohol dependence may require studies with a longer follow-up duration. Most of the trials relied on self-reporting from participants regarding their alcohol intake rather than routine urinalysis. There could be inaccuracies in self-reporting which may influence treatment effect.
Two economic evaluations were identified by our literature; however, neither were Canadian evaluations. The types of costs included in the studies may differ than those that would be expected in the Canadian health care system. Moreover, only one economic evaluation was a cost-utility analysis while the other was a cost-effectiveness study. Details regarding the time horizon were missing from the cost-effectiveness analysis and both studies did not perform probabilistic sensitivity analysis. For these reasons, it is difficult to predict if the findings can be generalized to the Canadian health care system.
This review had several limitations. A limited literature search was conducted and it is possible that studies not published in the databases searched were omitted. Inclusion was assessed exclusively from methodological details in the published article and additional information was not sought from study authors. Studies that had been previously reviewed as part of an included systematic review were not appraised separately for this report.
CONCLUSIONS AND IMPLICATIONS FOR DECISION OR POLICY MAKING: Overall, the majority of the included studies that assessed the clinical-effectiveness of naltrexone demonstrated that it was capable of improving clinical outcomes of alcohol
dependence. More specifically, naltrexone was most effective at decreasing relapse to heavy drinking. Drinking below this level was characterized as controlled drinking. All four of the systematic reviews and meta-analysis concluded that naltrexone was effective for preventing relapse to heavy drinking, while none of these studies reported effectiveness of acamprosate for the same clinical outcome. The findings of the systematic reviews and meta-analyses
suggested that acamprosate was more effective than naltrexone at maintaining abstinence or preventing relapse to any drinking. Several of the reports suggested that acamprosate may be the treatment of choice if the goal of treatment is abstinence and naltrexone may be used if the treatment goal is controlled drinking. Of note, several of the recent RCTs that were not included in the systematic reviews or in the meta-analysis reported that naltrexone also significantly improved abstinence rates. Further RCTs are required to confirm the effectiveness of naltrexone in maintaining abstinence. Three direct-comparison studies compared naltrexone to either acamprosate or disulfiram or both. All three of these studies reported significant improvements in the disulfiram arm in alcohol-related outcomes; however, all of the studies involved a
supervised medication dosing program with the intent of improving compliance. At least one of the studies that used a supervised medication dosing program reported non-significant
differences in the alcohol-related outcomes once the supervised period had finished.21
Supervised medication administration may not be suitable to all patients requiring treatment for alcohol dependence. One additional direct-comparison randomized trial compared naltrexone with aripiprazole and concluded that the two drugs had similar rates of clinical-effectiveness although further studies would be needed to confirm this finding.
The clinical study limitations and the limited Canadian cost-effectiveness information, coupled with the treatment goal of the patient and factors that may influence patient compliance (e.g. level of alcohol dependence, treatment-seeking behaviour versus inpatient referral and the presence of family support) and treatment effectiveness (e.g. severity of dependence and presence of co-morbidities or co-dependencies) may wish to be considered when deciding whether to use naltrexone for the treatment of alcohol dependence.
PREPARED BY:
Michelle Mujoomdar, BSc, PhD, Research Officer Carolyn Spry, MLIS, Information Specialist Health Technology Inquiry Service Email: htis@cadth.ca
Tel: 1-866-898-8439
REFERENCES:
1. Mason BJ. Rationale for combining acamprosate and naltrexone for treating alcohol dependence. J Stud Alcohol Suppl 2005;(15):148-56.
2. Drug product database. In: Drugs and health products [database online]. Ottawa: Health Canada; 2008. Available:
http://www.hc-sc.gc.ca/dhp-mps/prodpharma/databasdon/index-eng.php (accessed 2009 Mar 25).
3. Pettinati HM, O'Brien CP, Rabinowitz AR, Wortman SP, Oslin DW, Kampman KM, et al. The status of naltrexone in the treatment of alcohol dependence: specific effects on heavy drinking. J Clin Psychopharmacol 2006;26(6):610-25.
4. Acamprosate Calcium (Campral - Prempharm Inc.): CEDAC final recommendation on reconsideration and reasons for recommendation. Ottawa: Canadian Agency for Drugs and Technologies in Health (CADTH); 2008. Available:
http://cadth.ca/media/cdr/complete/cdr_complete_Campral_March-2008.pdf (accessed
2009 Mar 25).
5. Diagnostic and statistical manual of mental disorders: DSM-IV. 4th ed. Washington: American Psychiatric Association; 1994.
6. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. 4th ed., text revision. Washington: American Psychiatric Association; 2000.
7. Prevention of relapse in alcohol dependence: update December 2005. Glasgow (UK): Health Technology Board for Scotland; 2005. Available:
http://www.nhshealthquality.org/nhsqis/files/Update%20to%20Alcohol%20report%20(to%2
0web)%20151205.doc (accessed 2009 Mar 25).
8. Rösner S, Leucht S, Lehert P, Soyka M. Acamprosate supports abstinence, naltrexone prevents excessive drinking: evidence from a meta-analysis with unreported outcomes. J Psychopharmacol 2008;22(1):11-23.
9. Roozen HG, de Waart R, van der Windt DA, van den Brink W., de Jong CA, Kerkhof AJ. A systematic review of the effectiveness of naltrexone in the maintenance treatment of opioid and alcohol dependence. Eur Neuropsychopharmacol 2006;16(5):311-23.
10. Srisurapanont M, Jarusuraisin N. Opioid antagonists for alcohol dependence [Cochrane review]. Cochrane Database Syst Rev 2005;(1):CD001867.
11. Carmen B, Angeles M, Ana M, Maria AJ. Efficacy and safety of naltrexone and acamprosate in the treatment of alcohol dependence: a systematic review. Addiction 2004;99(7):811-28.
12. Farren CK, Scimeca M, Wu R, Malley SO. A double-blind, placebo-controlled study of sertraline with naltrexone for alcohol dependence. Drug Alcohol Depend 2009;99(1-3):317-21.
13. O'Malley SS, Robin RW, Levenson AL, GreyWolf I, Chance LE, Hodgkinson CA, et al. Naltrexone alone and with sertraline for the treatment of alcohol dependence in Alaska
natives and non-natives residing in rural settings: a randomized controlled trial. Alcohol Clin Exp Res 2008;32(7):1271-83.
14. Oslin DW, Lynch KG, Pettinati HM, Kampman KM, Gariti P, Gelfand L, et al. A placebo-controlled randomized clinical trial of naltrexone in the context of different levels of psychosocial intervention. Alcohol Clin Exp Res 2008;32(7):1299-308.
15. Morley KC, Teesson M, Reid SC, Sannibale C, Thomson C, Phung N, et al. Naltrexone versus acamprosate in the treatment of alcohol dependence: a multi-centre, randomized, double-blind, placebo-controlled trial. Addiction 2006;101(10):1451-62.
16. Anton RF, O'Malley SS, Ciraulo DA, Cisler RA, Couper D, Donovan DM, et al. Combined pharmacotherapies and behavioral interventions for alcohol dependence: the COMBINE study: a randomized controlled trial. JAMA 2006;295(17):2003-17.
17. Kiefer F, Andersohn F, Otte C, Wolf K, Jahn H, Wiedemann K. Long-term effects of pharmacotherapy on relapse prevention in alcohol dependence. Acta Neuropsychiatr 2004;16(5):233-8.
18. Anton RF, Moak DH, Latham P, Waid LR, Myrick H, Voronin K, et al. Naltrexone combined with either cognitive behavioral or motivational enhancement therapy for alcohol
dependence. J Clin Psychopharmacol 2005;25(4):349-57.
19. Martinotti G, Di Nicola M, Di Giannantonio M, Janiri L. Aripiprazole in the treatment of patients with alcohol dependence: a double-blind, comparison trial vs. naltrexone. J Psychopharmacol 2009;23(2):123-9.
20. De Sousa A, De Sousa A. An open randomized trial comparing disulfiram and naltrexone in adolescents with alcohol dependence. J Subst Abuse 2008;13(6):382-8.
21. Laaksonen E, Koski-Jännes A, Salaspuro M, Ahtinen H, Alho H. A randomized,
multicentre, open-label, comparative trial of disulfiram, naltrexone and acamprosate in the treatment of alcohol dependence. Alcohol Alcohol 2008;43(1):53-61. Available:
http://alcalc.oxfordjournals.org/cgi/reprint/43/1/53 (accessed 2009 Mar 25).
22. De Sousa A, De Sousa A. A one-year pragmatic trial of naltrexone vs disulfiram in the treatment of alcohol dependence. Alcohol Alcohol 2004;39(6):528-31. Available:
http://alcalc.oxfordjournals.org/cgi/reprint/39/6/528 (accessed 2009 Mar 25).
23. Zarkin GA, Bray JW, Aldridge A, Mitra D, Mills MJ, Couper DJ, et al. Cost and cost-effectiveness of the COMBINE study in alcohol-dependent patients. Arch Gen Psychiatry 2008;65(10):1214-21.
24. Mortimer D, Segal L. Economic evaluation of interventions for problem drinking and alcohol dependence: cost per QALY estimates. Alcohol Alcohol 2005;40(6):549-55. Available: http://alcalc.oxfordjournals.org/cgi/reprint/40/6/549 (accessed 2009 Mar 25).
