|Home | About | Journals | Submit | Contact Us | Français|
Extended-release naltrexone (XR-NTX) is a once-a-month injectable formulation for the treatment of alcohol dependence previously shown to reduce drinking and heavy drinking relative to placebo (Garbutt et al., 2005). A 24-week, randomized, double-blind, placebo-controlled study established the efficacy and safety of XR-NTX in this patient population. In this report, the effect of XR-NTX on quality of life (QOL) was examined.
Alcohol dependent patients were randomly assigned to receive XR-NTX 380 mg (N = 205), XR-NTX 190 mg (N = 210), or placebo (N = 209), combined with a standardized psychosocial intervention. Quality of life (QOL) was assessed using the Medical Outcomes Study 36-item short-form health survey (SF-36), administered at baseline and at 4–week intervals during 24 weeks of treatment.
Compared with U.S. population norms, patients showed initial impairment in the health-related QOL domains of mental health, social functioning, and problems with work or other daily activities due to emotional problems. Adherence to all 6 injections was 65% for XR-NTX 190 mg, 63% for XR-NTX 380 mg, and 64% for placebo. Generalized estimating equations analyses using an intention-to-treat sample revealed that XR-NTX 380 mg was associated with significantly greater improvements from baseline in mental health (p = 0.0496), social functioning (p = 0.010), general health (p = 0.048), and physical functioning (p = 0.028), compared to placebo. Linear regression analyses revealed that reductions from baseline in drinking (percentage of days drinking and percentage days of heavy drinking in past 30 days) were significantly (p < 0.05) correlated with improvements in quality of life.
XR-NTX 380 mg in combination with psychosocial intervention was associated with improvements in QOL, specifically in the domains of mental health, social functioning, general health, and physical functioning.
Alcohol dependence is a chronic and disabling disorder, with a prevalence of nearly 4% in the US adult population (Grant et al., 2004). Worldwide, alcohol is the fourth leading cause of disability, and is associated with considerable morbidity and mortality (McGinnis and Foege, 1999; Midanik et al., 1996; Murray and Lopez, 1996; Rehm et al., 2003; Room et al., 2005).
As with a variety of medical and psychiatric disorders, there is increasing interest in the quality of life (QOL) of patients with alcohol use disorders and the effect of treatments on QOL. There have been a variety of approaches to the study of QOL, with the term alternatively referring to general well-being, life satisfaction, or health-related quality of life (Katschnig, et al. (2006). Among individuals who drink alcohol, heavy drinking is associated with broad impairments in health-related QOL, with the largest impact typically found for mental health and social functioning (Donovan et al., 2005; Foster et al., 1999). Factors that influence the QOL of patients with alcohol dependence include marital status, income, job loss, physical illness, drug dependence, and severity of alcohol dependence (Frone, 2006; Johnson et al., 1999; Romeis et al., 1999; Volk et al., 1997).
Controlled studies examining the impact of treatment of alcohol dependence on QOL outcomes have primarily examined psychosocial treatments, and have found that these interventions are associated with improved health-related QOL (Donovan et al., 2005). Moreover, the degree of reduction in alcohol consumption has been found to be correlated with the degree of improvement in QOL over the course of treatment (Foster et al., 2000). Little has been reported in the literature on the effect of pharmacotherapy on QOL among alcohol-dependent patients. One open-label study of acamprosate (a structural analog of γ-aminobutyric acid) in combination with psychosocial support showed that health-related QOL improved over the course of 3 months of treatment (Morgan et al., 2004). A nonrandomized comparison of oral naltrexone in combination with cognitive-behavior therapy (CBT) versus CBT alone found no differences with regard to improvement in health-related QOL between the study groups (Feeney et al., 2004). The only randomized, placebo-controlled study to date has shown topiramate (an anticonvulsant) to be superior to placebo in improving QOL outcomes, specifically physical health/activities, subjective feelings, general activities, and overall life satisfaction (Johnson et al., 2004).
Recently, extended-release naltrexone (XR-NTX; Vivitrol®), in conjunction with psychosocial support, was approved in the U.S. for the treatment of alcohol dependence in patients who are able to abstain from alcohol prior to treatment initiation. A previous article reported on the primary drinking outcomes of a 24-week, randomized, multicenter, double-blind, placebo-controlled trial designed to evaluate the efficacy and safety of XR-NTX in patients with alcohol dependence (Garbutt et al., 2005). XR-NTX 380 mg, injected once every 4 weeks, was superior to placebo in maintaining abstinence and reducing the rate of heavy drinking, particularly in patients who have achieved abstinence prior to treatment (Garbutt et al., 2005). XR-NTX was developed to address the poor patient adherence with oral naltrexone (Harris et al., 2004; Hermos et al., 2004; Kranzler et al., in press) and problems with oral alcohol dependence medications in general that have limited their clinical use in treating alcohol dependence (Bouza et al., 2004; Mark et al., 2003; Volpicelli et al., 1997; Weiss, 2004).
Using data from the multicenter efficacy trial, the goal of the current article is to report on the effects of XR-NTX on a secondary efficacy measure, the health-related QOL of alcohol-dependent patients. Three questions were addressed: (1) which areas of QOL are most impaired among these alcohol-dependent patients prior to treatment; (2) which areas of QOL improve significantly over the course of treatment with XR-NTX (380 mg) compared to placebo; and (3) are changes in drinking patterns associated with improvements in QOL? It was hypothesized that QOL would improve in conjunction with reduced drinking.
A complete description of the methodology and analyses of the drinking outcomes and tolerability are given in the prior publication (Garbutt et al., 2005). Briefly, men and women, aged 18 or older with a diagnosis of alcohol dependence, as defined by the Diagnostic and Statistical Manual of Mental Disorders 4th Edition (DSM-IV), and at least two days per week of heavy drinking (≥ 5 drinks per day for men, ≥ 4 drinks per day for women) in the 30 days prior to detoxification (if needed) and screening, were randomized to a single intramuscular injection of XR-NTX 380 mg or XR-NTX 190 mg, or to placebo. Patients were excluded if they had any clinically significant medical condition that might adversely affect safety or study participation; major depression with suicidal ideation, psychosis, or bipolar disorder; or dependence on benzodiazepines, opiates, or cocaine within the past year. Patients were also excluded if they had more than 7 days of inpatient treatment for substance abuse during the 30 days prior to screening.
Injections of XR-NTX began on the day of randomization and continued at 4-week intervals for a total of six injections. These injections were administered by research staff not involved in the assessments or psychosocial intervention (to preserve the blind). In addition to pharmacotherapy, all patients were provided 12 sessions of a concurrent psychosocial intervention based on the BRENDA model (Volpicelli et al., 2001). BRENDA is a low-intensity intervention with the overarching goal of facilitating feedback to the patient about addiction-related consequences. The BRENDA model involves the following six components: (1) a Biopsychosocial evaluation of information (history, physical examination, laboratory studies) as it relates to the patient’s alcohol dependence, (2) a Report based evaluation, (3) clinician Empathy, (4) identification of each participant’s unique Needs that are related to the goals of treatment, (5) Direct advice focused on strategies to achieve treatment goals, and (6) an Assessment of the participant’s response to the advice.
QOL was assessed using the Medical Outcomes Study 36-item short-form health survey (SF-36) version 2 (Ware et al., 2000) at baseline and at 4-week intervals thereafter for 24 weeks. The SF-36 consists of eight subscales (physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health) and two summary scales (overall physical health and overall mental health) that are generated from the same set of items as the eight subscales.
Alcohol use was assessed at randomization, at scheduled weekly visits for the first 4 weeks, and biweekly thereafter for the next 20 weeks using the timeline follow back (TLFB) method, a recall-enhancing technique that has demonstrated reliability and validity (Sobell and Sobell, 1992). The TLFB method uses calendars and recollections of drinking on specific days to record the frequency and pattern of drinking since the previous assessment, thereby yielding a continuous record of drinking for the 24-week study period. Abstinence prior to randomization, percentage of drinking days per month, and percentage of heavy drinking days per month were obtained from the TLFB.
This paper focuses on the 380 mg dose of XR-NTX relative to placebo, since this was the dosage that was shown to be superior to placebo in reduction of heavy drinking and that is approved for clinical use by the US Food and Drug Administration (FDA). Results for the 190 mg dose were also analyzed and are briefly summarized.
Patients who received study medication composed the intent-to-treat population (ITT) and were included in the analysis. Analyses were performed on the ITT sample using available data. Norm-based scores (with a population mean of 50 and standard deviation of 10, using 1998 U.S. population norms) were used to permit the interpretation of differences across the SF-36 scales (Ware et al., 2000).
Comparisons of the mean change from baseline to post-baseline (all six post-baseline monthly assessments) in the norm-based SF-36 scores over time were conducted using generalized estimating equations (GEE) (Liang and Zeger, 1986) with visit and study group as independent variables. Data from all 3 study groups, and all available SF-36 assessments, were included in the analyses, with a contrast specified to examine the hypothesized difference between XR-NTX 380 mg and placebo. Additionally, GEE analyses were conducted using the subset of patients who were abstinent for 4 days immediately prior to randomization (first injection). This subgroup was examined for the sake of generalizability because many studies of oral naltrexone have recruited samples with four days or more of initial abstinence and four days is a standard detoxification length of stay. Furthermore, the greatest reduction in heavy drinking days was previously found for those XR-NTX 380 mg-treated subjects in this subgroup (O'Malley et al., 2007). The FDA labeling for XR-NTX specifies use in patients who are initially abstinent and this is a criterion that has been used in previous research on medications to treat alcohol dependence (Anton et al., 2006). A further analysis examined change from baseline to month 6 using t-tests to compare the two treatment groups on the change scores.
The relationship between residual change in drinking behavior and change in QOL was examined using separate multiple regression analyses for each of the SF-36 scales (dependent variable = percent drinking days in last 30 days; independent variables = percent drinking days in the 30 days prior to study enrollment, baseline norm-based SF-36 score, study group, and SF-36 scale scores at the week 24 assessment). Similar analyses were conducted using percent of heavy drinking days (in past 30 days) as the dependent variable. Although the focus of this paper is the comparison between XR-NTX 380 mg and the placebo control group, all three study groups (placebo, XR-NTX 190 mg, and XR-NTX 380 mg) were included in this analysis because it was expected that improvements in drinking, regardless of study group, would be associated with improvement in QOL, and because a dose-response effect was possible.
All data were analyzed using SAS version 8.2 statistical software (SAS Institute Inc, Cary, NC). A two-tailed alpha level of 0.05 was used to declare statistical significance.
A total of 899 patients were screened for eligibility for the study; 627 met inclusion/exclusion criteria and were randomly assigned to one of the three study groups (Garbutt et al., 2005). Because three randomized patients did not receive their first injection based on the decision of the investigator, the final ITT sample consisted of 624 patients. Of these, 205 received XR-NTX 380 mg, 210 received XR-NTX 190 mg, and 209 received placebo. All but three 3 of these patients (two from the XR-NTX 190 mg group and one from the placebo group) had baseline SF-36 data. The XR-NTX 380 mg and placebo groups were similar in terms of demographic and baseline drinking variables (Table 1).
At baseline, the alcohol-dependent patients enrolled in this study showed impairments on several of the SF-36 scales relative to U.S. population norms (Figure 1). In particular, the mean mental health summary scale value was one standard deviation below (ie, more impaired) than the U.S. population norm. The physical health summary score fur the current sample was similar to the U.S. population norm value, which is consistent with screening to exclude individuals with substantial medical illness. Among the individual subscales, mean values on the social functioning, role-emotional, and mental health scales showed the greatest impairments relative to the population norms.
At the monthly assessment visits, data were available on the SF-36 for the following percent of patients (months 1 to 6 [or last visit], respectively): XR-NTX 190 mg: 90%, 83%, 78%, 72%, 68%, and 74%; XR-NTX 380 mg: 94%, 84%, 80%, 78%, 72%, 76%; placebo: 94%, 83%, 80%, 73%, 69%, 74%.
For XR-NTX 190 mg, 65% received all 6 injections, and 74% received at least 4 injections. Of those treated with XR-NTX 380 mg, 63% received all 6 injections, and 72% received at least 4 injections. Among the placebo-treated patients, 64% received all 6 injections, and 77% received at least 4 injections.
Psychosocial treatment adherence was high. Out of a maximum of 12 sessions, the mean (SD) number of psychosocial BRENDA sessions attended was 9.7 (3.2) for XR-NTX 190 mg, 9.9 (2.9) for XR-NTX 380 mg, and 9.8 (2.8) for placebo.
On the overall summary scales, the XR-NTX 380 mg group improved significantly more than the placebo group on the mental health score (p = 0.044) but not the physical health score (p = 0.51) from baseline to post-baseline during the 24-week study period (Table 2). On the individual scales, XR-NTX 380 mg resulted in significantly greater improvement from baseline on the mental health (p = 0.0496), social functioning (p = 0.010), general health (p = 0.048), and physical functioning (p = 0.028) scales compared with placebo. Between-group effect sizes (differences between the means of the change scores for the two groups divided by the placebo change score standard deviation) for these significant findings ranged from 0.19 to 0.25 (Table 2). The amount of within-group change on these SF-36 scales was meaningful. For example, on the social functioning scale, patients in the XR-NTX 380 group had a mean score of 41.3 at baseline and improved to a mean score of 49.0 during treatment, close to the population norm score of 50 and a within-group effect size (mean change divided by standard deviation of change scores) of 0.715 (Table 2).
No significant differences were apparent for the vitality, role-emotional, role-physical, or bodily pain scales. Results examining change from baseline to month 6 were nearly identical to the GEE analyses that incorporated all post-baseline assessments (Table 2).
For the scales that were most impaired relative to norms at baseline (social functioning, role-emotional, and mental health), treatment with XR-NTX 380 mg improved scores to close to the normative value (Table 2). With the exception of the role-emotional subscale, the improvement was significantly greater in the XR-NTX 380 mg group than in the placebo group.
Among those with four or more days of abstinence immediately prior to randomization, mean changes from baseline were significantly greater (p = 0.026) on the general health scale for XR-NTX 380 mg (mean change = 4.9; N = 26) compared with placebo (mean change = 0.8; N = 27). There were no other significant differences related to XR-NTX in this subgroup when compared with placebo. However, the magnitude of the difference between XR-NTX 380 mg and placebo in mean change from baseline was found to be larger (for general health, social functioning, role emotional, mental health, and mental health summary scale) compared to the full sample, taking into account those scales that were statistically significant in the full sample. For example, in the full sample, the XR-NTX 380 mg group improved 2.4 scale points more than the placebo group on the social functioning scale, but improved 5.2 points more in the subgroup with 4 days of prior abstinence.
Similar analyses were performed on data from patients receiving XR-NTX 190 mg. There were no significant differences on any of the scales when compared with placebo.
Regression analyses examining residual change from baseline to week 24 in drinking behavior (percentage of drinking days and heavy drinking days in past month) was significantly (p < 0.02) related to residual change in QOL for all SF-36 scales (Table 3). By controlling for baseline values of these measures and study group, these analyses reflect the correlation between reductions in SF-36 and reductions in drinking behavior across the study groups.
There were three main findings from this study. First, this alcohol-dependent sample showed impairments in QOL at baseline compared with population norms, especially in regard to mental health, social functioning, general health, and physical functioning. Second, the XR-NTX 380 mg group showed meaningful within-group change and significantly greater change compared to the placebo group in overall mental health, social functioning, general health, and physical functioning domains of QOL. The XR-NTX 190 mg group did not differ significantly from placebo with respect to improved QOL. Third, reductions in drinking (percentage of days drinking and percentage days of heavy drinking in past 30 days) from baseline were significantly correlated with improvements in quality of life.
Despite the inclusion/exclusion criteria for the trial that ruled out individuals with medical and psychiatric illnesses, the impairments found at baseline in levels of mental health functioning, social functioning, and role-emotional functioning are highly consistent with findings from previous studies of alcohol-dependent patients, reported in a recent review (Donovan et al., 2005). The impairments in mental health and social functioning in alcohol-dependent patients found here were similar to the extent of impairment in these domains reported for individuals with rheumatoid arthritis, migraine, and diabetes (Frank et al., 2002). Similar to the findings of other research, alcohol-dependent patients selected for clinical trials generally show a relatively normal QOL in regard to physical functioning (Daeppen et al., 1998).
The domains of QOL that showed the greatest improvements were primarily those that were most impaired at baseline, namely mental health and social functioning. Thus, XR-NTX resulted in improvements in functioning in areas that were most affected by the disorder. Although these effects could also reflect regression to the mean, arguing against this interpretation is the fact that physical functioning also improved with XR-NTX 380 mg treatment compared with placebo, despite the fact that the sample was not below normative values in physical functioning at baseline. In addition, the overall magnitude of within-group change for patients treated with XR-NTX was large, with patients moving from a baseline mean of about 41 to means of 48 (mental health) and 49 (social functioning) during treatment, which were close to the population norm of 50 of the SF-36 scales. The fact that the improvements were obtained during six months of treatment is noteworthy. Furthermore, the magnitude of difference beteeen XR-NTX 380 mg and placebo in the normed scores in this analysis was in the range reported for accepted interventions in a variety of medical treatments (Farivar et al., 2004).
Although previous studies have reported changes in QOL in alcohol-dependent patients receiving medications (Feeney et al., 2004; Johnson et al., 2004), the current study extends those previous findings. First, the QOL data reported here are the first for XR-NTX 380 mg. Second, two of the previous studies of the impact of pharmacotherapy for QOL in alcohol dependence were not placebo controlled (Feeney et al., 2004; Morgan et al., 2004). Finally, one study failed to find differences in QOL between psychosocial treatment alone and psychosocial treatment plus medication (oral naltrexone) (Feeney et al., 2004). The lack of an effect of oral naltrexone in that study may have been a result of the effectiveness of CBT (i.e., producing a “ceiling effect,” leaving little room for oral naltrexone to augment its effects). Treatment adherence has repeatedly been identified as a mediator of outcome in pharmacotherapy trials (Baros et al., 2007; Krystal et al., 2001; Pettinati et al., 2000). As XR-NTX 380 mg was designed to address adherence difficulties with taking daily oral naltrexone (Harris et al., 2004; Hermos et al., 2004; Kranzler et al., in press), it may help to maintain the course of treatment long enough to have an impact on functioning and QOL. In an analysis of an insured patient population prescribed oral naltrexone, less than 10% sustained refills through six months (Harris et al., 2004), whereas in this trial of XR-NTX 380 mg, 64% received six months of injections (Garbutt et al., 2005).
The current study also found that changes in drinking patterns (percentage days drinking and percentage days heavy drinking) were significantly (p < 0.05) associated with changes in QOL from baseline to week 24. Reductions in total consumption of alcohol have previously been associated with improvements in QOL (Foster et al., 2000). It may be that reducing the negative health and functioning effects of heavy drinking (Gastfriend et al., 2007; Midanik et al., 1996; Room et al., 2005) is sufficient to improve QOL in alcohol-dependent patients, even in patients who continue to drink. These correlations between reduced drinking and improved quality of life suggest that the superiority of XR-NTX relative to placebo regarding changes on the SF-36 is likely due to greater improvements in drinking outcomes for XR-NTX compared to placebo (Garbutt et al., 2005).
Several limitations of the current study should be considered. Although hypotheses regarding QOL were specified in the original protocol, these were analyses of secondary efficacy measures and therefore should be confirmed with additional studies. In addition, because patients were seeking treatment, data on impairments in QOL at baseline cannot necessarily be generalized to the population of alcohol-dependent individuals as a whole. The exclusion of individuals with unstable major mental illness or dependence within the past year on benzodiazepines, opioids, or cocaine also limits the capacity to generalize the results to alcohol-dependent patients with these co-morbid conditions. Another limitation is that only 63% to 65% (across groups) of patients received all 6 injections.
In summary, relative to placebo plus psychosocial support, XR-NTX 380 mg in combination with a psychosocial intervention was associated with improvements in QOL in the domains of mental health, social functioning, general health, and physical functioning. Changes in QOL were associated with changes in drinking, and heavy drinking, over the course of treatment. These improvements with XR-NTX may result in patients being more motivated to continue treatment and to better focus on other personal goals.
We wish to thank the following investigators who participated in the study: Robert Anthenelli, MD, Cincinnati, OH; Louise Beckett, MD, Oklahoma City, OK; Michael Bohn, MD, Middleton, WI; Paul Casadonte, MD, New York, NY; Domenic Ciraulo, MD, Boston, MA; David Gastfriend, MD, Boston, MA; Lawrence Ginsberg, MD, Houston, TX; Hisham Hafez, MD, Nashua, NH; Bankole Johnson, MD, San Antonio, TX; Philip Kanof, MD, PhD, Tucson, AZ; Henry Kranzler, MD, Farmington, CT; Sandra Lapham, MD, Albuquerque, NM; Peter Martin, MD, Nashville, TN; Barbara Mason, PhD, Miami, FL; Mary McCaul, PhD, Baltimore, MD; Denis Mee-Lee, MD, Honolulu, HI; Helen Pettinati, PhD, Philadelphia, PA; Robert Riesenberg MD, Atlanta, GA; Ihsan Salloum, MD, Pittsburgh, PA; Jeffery Wilkins, MD, Los Angeles, CA; Mark Willenbring, MD, Minneapolis, MN; Allen Zweben, PhD, Milwaukee, WI. We wish to thank Paul Crits-Christoph, PhD, who provided assistance with the preparation of the manuscript.
Funding for this research and publication was provided by Alkermes, Inc. and Cephalon, Inc.