Case
Clinical Questions
Introduction
Criteria for Alcohol Abuse and Dependence
Does Stopping or Reducing Drinking Reduce Mortality?
What is Alcoholics Anonymous and What Evidence Supports Its Use?
Drug Therapy
Conclusion
Main Learning Points
Appendices
Bibliography

Mark A. Perini, MD
Resident Grand Rounds
Wake Forest University Baptist Medical Center
Internal Medicine Residency Program

September 26, 2000

For a patient with alcohol abuse or dependence will a referral to Alcoholics Anonymous be effective for stopping or significantly reducing drinking and the effects of drinking?

What medical therapies are available for treating alcoholism and what is the clinical data to support their use?

Introduction

• Alcoholism and alcohol abuse is a major public health problem and accounts for significant morbidity and mortality worldwide.
• There are approximately 13 million Americans who clearly or probably need treatment for alcohol use disorders.¹
• Approximately 13.8% of American adults have had either alcohol dependence or abuse sometime in their lives. ¹
• Approximately 100,000 deaths per year are directly related to alcohol abuse and dependence and the financial toll of alcohol abuse accounts for $98 billion per year in the U.S..¹⁴

Internists are confronted with the consequences of alcohol abuse and dependence regularly. Many internists feel uncomfortable treating this form of substance abuse due to lack of training and a perceived lack of treatment options. Furthermore, there is a large element of therapeutic nihilism. This is caused by the frustration of a chronic illness, which often involves many relapses. Patients are labeled as noncompliant or uncooperative, and, there is a perceived lack of effective treatment options. Therefore, internists may focus more on the physical effects of alcohol abuse without adequately addressing the underlying behavioral problem.

In a national systematic mailed survey of internists 88% reported asking new outpatients whether they drink alcohol but only 13% usually use a formal alcohol screening tool such as the CAGE or MAST (The Michigan Alcoholism Screening Test).⁶ Unfortunately, up to 18% usually or always offer no intervention to their problem-drinking patients. ⁶ The most common recommendation by internists for patients diagnosed with alcoholism is to attend a 12-step self-help program such as Alcoholics Anonymous. Just over 60% of internists usually or always make this recommendation. ⁶ Rates of referral to a mental health or social worker, to formal treatment programs or direct counseling of the patient without referral are done less often. ⁶ A recommendation to attend Alcoholics Anonymous meetings often is made to patients who refuse or are unwilling to attend a more formal substance abuse center or psychiatric evaluation. The following will explore the evidence supporting a recommendation to attend Alcoholics Anonymous.

In addition, there is a growing call from the substance abuse treatment community to primary care physicians to assume greater responsibility for the care of those with alcohol abuse and dependence. ¹⁴ There is substantial evidence supporting brief office interventions by physicians and paramedical staff. There is also a burgeoning research movement investigating new pharmacotherapies for alcoholism. The need and market for this are understandably huge. The results of this ongoing work may translate into further therapeutic options for the general internist. If internists are armed with medicine to help treat alcohol abuse, as is the case now for other psychiatric problems such as depression and tobacco abuse, then more enthusiasm in this arena may develop. Also, treatment rates will increase if studies of the treatment of alcoholics evolve into study designs that are more easily interpreted and applied by internists. However, to date, much of the data regarding substance abuse is confined to less universally known substance abuse journals. These studies lack the hard endpoints, such as mortality and significant morbidity, that are needed to alter current practice patterns. In the following article evidence pertaining to the use of disulfiram and naltrexone, currently the only two Food and Drug Administration (FDA) approved medicines for alcohol dependence, will be discussed. Then preliminary data on two newer agents, acamprosate and ondansetron, will be revealed as well.

A maladaptive pattern of alcohol use leading to clinically significant impairment or distress, as manifested by three (or more) of the following, occurring at any time in the same 12-month period:

1. tolerance, as defined by either of the following:
   a) need for markedly increased amounts of alcohol to achieve intoxication or desired effect.
   b) markedly diminished effect with continued use of the same amount of alcohol

2. withdrawal, as manifested by either of the following:
   a) the characteristic alcohol withdrawal syndrome.
   b) alcohol or a closely related substance is taken to relieve or avoid withdrawal symptoms

3. alcohol is often taken in larger amounts or over a longer period than was intended.
4. there is a persistent desire or unsuccessful efforts to cut down or control alcohol use
5. a great deal of time is spent in activities necessary to obtain the substance, use the substance, or recover from its effects
6. important social, occupational, or recreational activities are given up or reduced because of alcohol use
7. alcohol use is continued despite knowledge of having a persistent or recurrent physical or psychological problem that is likely to have been caused or exacerbated by alcohol (such as continued drinking despite recognition that an ulcer was made worse by alcohol consumption)

1. A maladaptive pattern of alcohol use leading to clinically significant impairment or distress, as manifested by one (or more) of the following, occurring within a 12-month period:

1. recurrent alcohol use resulting in a failure to fulfill major role obligations at work, school, or home (e.g., repeated absences or poor work performance related to alcohol use; alcohol-related absences, suspensions, or expulsions from school; neglect of children or household)
2. recurrent alcohol use in situations in which it is physically hazardous (e.g., driving an automobile or operating a machine when impaired)
3. recurrent alcohol-related legal problems (e.g., arrests for substance-related disorderly conduct)
4. continued alcohol use despite having persistent or recurrent social or interpersonal problems caused or exacerbated by the effects of the substance (e.g., arguments with spouse about consequences of intoxication, physical fights)

2. The symptoms have met the criteria for alcohol dependence for this class of substance.

One should specify whether substance dependence is with physiologic dependence (i.e., there is evidence of tolerance or withdrawal) or without physiologic dependence. ¹

Before addressing the question of the efficacy of AA, one may ask the question, does stopping or reducing drinking help? Has this been shown to reduce morbidity and mortality? After all, this is the ultimate goal of treatment. Several studies have looked at this question. In 1992, a study in JAMA by Bullock et al. examined the mortality of alcoholic men who achieved stable abstinence compared to that of relapsed alcoholics. With a mean follow up of almost six years the authors found that relapsed alcoholics had a standardized mortality ratio of 4.96 whereas continuously abstinent alcoholics had a standardized mortality ratio that was no different than nonalcoholic controls. This means that over 6 years there are 5 times as many deaths in relapsed alcoholics as would be expected. The absolute death rate over follow-up was 18% versus 4.1%.² Therefore, achieving abstinence will translate into longer life expectancy.

However, less is known about the degree to which reducing drinking over a long term will result in reduced mortality. It is felt that any reduction or modification in drinking is likely beneficial and while abstinence likely gives the best results, it may be acceptable to have as a goal markedly reduced or "controlled" drinking for some patients.¹ In a population of alcoholic cirrhotics even just reducing drinking a small amount offered significant improvement in longevity. ⁵ Unfortunately, even when treatment works, patients often have recurrent abuse or dependence. The majority of patients who are treated for alcohol use disorders have at least one relapse episode during the first year following treatment. ¹ However, about 70% of all patients who undergo treatment for alcoholism manifest a reduction in number of days drinking and subsequent improved health status within 6 months. This decrease in drinking extends at least several months past the treatment. ¹

When reviewing studies on alcoholism one must recognize that there are inherent difficulties in studying alcoholics. These include difficulty in maintaining good follow-up, high attrition rates out of studies, the need to rely on self-reported measures, and the difficulty in finding good behavioral measures or endpoints for studies on treatment. Also, while the mortality rate for alcoholics is many times that of the remainder of the population it is still relatively low per year (on the order of 3-4% per year in many studies) and large study groups with long follow-up would be needed to prove efficacy. ³ This is often not feasible. Abstinence is the next best criterion used for positive change but drinking behavior and alcoholism can be modified for the better without total abstinence. ³ Furthermore, the natural history of alcoholism often shows periods of abstinence and relapse with some (about 10%) patients eventually achieving long term abstinence without any specific therapy. ³ Taking all of these facts together one can conclude that stopping drinking in an alcoholic will confer a significant survival benefit. In addition, reducing drinking is likely to confer major benefits not only for health but also in many psychosocial parameters. This is definitely true in the short term and it is likely true for much longer periods of time.

According to the American Psychiatry Association Alcoholics Anonymous (AA) is categorized as a self-help group. Alcoholics Anonymous is "a fellowship of men and women who share their experience, strength and hope with each other that they may solve their common problem and help others to recover from alcoholism. The only requirement for membership is a desire to stop drinking." ¹ It is dedicated to helping those with alcohol abuse and dependence. It is based on a spiritual (not religious) approach and follows the "Twelve Steps" and the "Twelve Traditions".^{see appendix I} The effectiveness of AA has been difficult to gauge because there have been few randomized controlled studies investigating this. To be sure, there are some practical and ethical issues involved in attempting to study a group such as this scientifically. ¹

Despite the paucity of randomized controlled trials or even well performed prospective cohort studies there is a body of epidemiological and related literature which supports the efficacy of Alcoholics Anonymous. The General Services Office of AA tracks various statistics relating to membership. This is not published scientific data. However, the main conclusions of their data show:

• About half of those coming to AA for the first time drop out in less than 3 months. So, early attrition is quite considerable.
• About 40% of the members with less than one year of sobriety will remain sober and active in the fellowship for another year.
• About 80% of the members with between 1 and 5 years of sobriety will remain sober and active in the fellowship for another year.
• About 90% of the members with over 5 years of sobriety will remain sober and active in the fellowship for another year. ⁴

Klaus Mäkelä published an article in 1992 examining the rate of attrition of AA members in Finland. He identified 6,315 members (84% men) and noted very similar results to those of the General Services office. It was determined that the sober membership of AA represented between one-third and one-fifth of all abstaining former frequent drinkers in Finland. This speaks to the magnitude of the role currently played by AA in treating those with alcohol problems. As it turns out the sober membership of AA is an exceptionally stable group. ⁴

Mann and Smart have been studying the demographics of alcohol consumption, cirrhosis rates and treatment for alcoholism in the U.S. and Canada for many years. They point out that between World War II and the mid 1970’s the rate of alcohol consumption and alcoholic liver cirrhosis deaths had been increasing steadily. But, since the mid-1970’s, a striking reduction in levels of alcohol-related problems has been noted. The rate of change in cirrhosis death rates (a decrease of 25%) cannot likely be explained solely by decrease in alcohol consumption alone (only a decrease of 3% between 1975 and 1983). They showed that in the U.S. there was a statistically significant correlation between the decrease in alcohol consumption and the decrease in cirrhosis mortality. Also, as AA membership increased (from 294,825 in 1974 to 583,995 in 1983) this was correlated significantly with the decrease in cirrhosis deaths. Interestingly, increases in other forms of alcohol treatment did not correlate as well. ⁴ Of course, causal interpretation cannot be made with this data but according to the authors the simplest explanation to explain these findings is a causal one. Their work suggests the efficacy of AA but only in an indirect and very loosely inferred manor.

Walsh et. al., "A Randomized Trial of Treatment Options for Alcohol-Abusing Workers"

There have been a few randomized trials involving AA. In the September 12, 1991 edition of the New England Journal of Medicine a study entitled "A Randomized Trial of Treatment Options for Alcohol-Abusing Workers" was published. The objective was to determine how much alcoholism treatment is enough for the rehabilitation of employees who have alcohol problems. ⁷ The setting of the study was a large General Electric plant employing 10,000 employees with an established employee assistance program. Clients who were newly identified as abusing alcohol were randomly assigned to one of three alcoholism-rehabilitation regimens. The mean age was 33 years old and 90 percent of the patients were white. The first arm involved an initial period of mandatory inpatient rehabilitation, the second arm required mandatory attendance at Alcoholics Anonymous ("AA only") meetings, and the third gave the subjects a choice of treatment options. Patients were followed for two years and were assessed in terms of 12 measures of job performance and 12 measures of drinking and drug abuse. Two hundred twenty-seven subjects were randomized with 73 assigned to "compulsory hospitalization", 83 to the "compulsory AA only" arm, and 71 subjects were assigned to the "choice" arm in which they planned their own treatment. ⁷

The results of this study showed no significant differences among the groups in percentage of patients who were fired during the two-year follow-up period. All subjects showed substantial and sustained improvement in all aspects of job functioning. Twenty-three percent of subjects reported an unbroken record of abstinence from alcohol during the 24 months of the study. However, the hospital group was significantly more likely to include continuous abstainers (37% vs. 17 % for the choice group and 16% for the "AA-only" group; p = 0.002). When calculated this gives a relative risk reduction of 25% (95% confidence interval 21.5%- 28.5%) for relapse in the hospital group versus the "AA only" group. This is a number needed to treat of 5 which means one would have to treat 5 patients in the early hospitalization group to prevent 1 relapse over 2 years compared to mandatory "AA-only". Of the 227 subjects randomly assigned in the entire study, 42% were hospitalized for additional treatment: 23% of the hospital group and 63% of the "AA-only" group (this difference was significant, p=0.002; RRR 63%; CI 59%-67%; NNT 2.5).⁷

• All the interventions were successful but compulsory hospitalization with AA follow-up addressed drinking problems more effectively than did compulsory AA alone.
• The higher initial cost of inpatient treatment in this group is offset by the increased frequency with which problems resurface in the "AA-only" group.
• If AA alone is mandated in this setting close monitoring is essential as many employees have serious relapses in the first six months. ⁷

This is one of the only randomized trials to study AA and compare it in some way amidst other treatments. There was good follow-up of patients with only 12% of patients entirely lost to follow-up (this is actually very good for a two year study of alcoholics) and those lost were distributed evenly among the three groups. Intention-to-treat analysis was performed. The study was not truly blinded but any study of this nature is inherently very difficult to blind. The study size is not large but the groups were equal in key demographic and other base-line characteristics. Basically, in this study the "AA-only" group can be considered the control group and the arm with hospitalization plus AA is the intervention. This is because those who received initial hospitalization were also required to attend three AA meetings a week. Finally, it seems that in terms of follow-up and assessment these two groups were treated essentially the same as much as possible. Therefore, the results of this study are likely valid. While a difference in rate of job loss and in several key drinking measures were not found the difference in numbers of complete abstainers and in the numbers of those requiring later hospitalization were substantial. While we do not have a true "control" group on which to gauge the efficacy of the "AA-only" group compared to a group without any intervention it does seem that even with only compulsory AA dramatic improvements over baseline were seen. The average number of drinks per day dropped from 6.3 to 1.5 and the average number of drinking days in a month dropped from 19.8 to 3.1. ⁷ While this study does give valuable data about AA as a treatment option, the applicability of the study to patients seen in a clinic setting is quite limited. These subjects are employees in a large plant. The circumstances of placing a subject in an employee assistance program with the incentive of possible job loss as a motivation is not analogous to most physician-patient interactions. Real world results for a physician referring a patient to AA will not match those found in this study.

There has been one recent landmark study which looked very closely at the twelve-step concept of AA. Project MATCH is a massive effort sponsored by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). At the time of this project’s inception (early 1990’s) it was felt that there was enough proof that alcohol treatment worked, however, there was not enough information to know which kinds of individuals would benefit from which type of treatment. The "matching hypothesis" assumes that prescribing specific treatments based on individual characteristics and needs would improve treatment outcomes. Project MATCH consisted of two parallel but independent treatment matching studies, one with clients recruited at five outpatient sites, the other at five sites with clients who received aftercare treatment following an episode of inpatient or intensive day hospital treatment. The overall objective of each study was to determine if various subgroups of alcohol dependent clients would respond differently to three manual-guided (i.e., sessions administered by following a manual) treatments: Cognitive Behavioral Coping Skills Therapy (CBT), Motivational Enhancement Therapy (MET) and Twelve-Step Facilitation Therapy (TSF). There were ten client characteristics selected to be matching variables:

Various predictions were made as to what type of matching variables would correlate with better or worse outcomes depending on the type of treatment given. ⁸

Subjects were recruited at nine clinical research units.

A total of 4481 potential participants were screened. Of the 1726 eligible patients, 952 were entered into the study in the outpatient arm and 774 were in the aftercare arm. Greater than 95% of the patients met criteria for alcohol dependence. There were 75% men and the mean age was about 40 years old. Eighty percent of the patients were white and the baseline number of drinks per drinking day was about 11. Intention-to-treat analysis was performed. ⁸ After randomization, treatment lasted 12 weeks with follow-up assessments every 3 months thereafter. Follow-up was obtained for greater than 92% of the patients at one-year post treatment and for 85% patients three years post treatment. ^{8, 9}

Cognitive Behavioral Therapy (CBT) and Twelve-Step Facilitation (TSF) sessions were delivered weekly for 12 weeks while the Motivational Enhancement Therapy (MET) involved four sessions over 12 weeks.

Clients completed almost 70% of their scheduled treatments. The two primary endpoints were percent days abstinent (PDA) and drinks per drinking day (DDD). The results of this study showed that all groups achieved substantial positive changes in PDA and DDD from baseline to each follow-up month with only slight deterioration at 1-year post treatment.

For the outpatient subjects, 19% maintained complete abstinence throughout the one-year follow-up and approximately 46% relapsed to a heavy-drinking period of three consecutive days. For both arms there were no significant differences between groups overall for percent days abstinent and drinks per drinking day. In the outpatient arm there was a statistically significant difference in the time to first drink (p = 0.0007) with TSF patients doing the best with 24% remaining completely abstinent for one year compared to 15% and 14% respectively for CBT and MET (a 66% RBI, relative benefit increase, 95% CI 62%-70%; NNT = 10). Therefore, one would need to treat 10 patients with an Alcoholics Anonymous directed approach as opposed to Cognitive Behavioral Therapy or Motivational Enhancement Therapy to achieve one more completely abstinent patient at one year. Of note, AA attendance was not controlled in this study but clients in TSF did attend significantly more AA meetings than did those in the other two treatments.⁸ TSF also performed better in the percentage of patients not relapsing to heavy drinking (three successive days of ≥ 6 drinks per day for men and ≥ 4 drinks per day for women) with 53% not reaching this criterion, followed by MET with 49% and CBT with 48% (p = 0.0127). ⁸ As far as the matching hypotheses outcomes were concerned there were very few matching effects found.

In a three year follow-up of the outpatient arm of Project MATCH the treatment effect was preserved with 30% of the subjects reporting total abstinence in months 37-39 of follow-up. ⁹ In addition, subjects who did report drinking were nevertheless abstinent two-thirds of the time on average in months 37-39 of follow-up. ⁹ This is an improvement in abstinent days of about 150% over the 90-day baseline period just prior to study entry. When drinking, they drank an average of between 6 and 7 drinks per occasion, down form 11 at the time of the intake into the study, representing a considerable improvement over baseline. ⁹ Finally, the slight advantage for TSF at the 1-year follow-up was maintained 3 years after treatment. ⁹

Although the efficacy of the three treatments throughout the course of Project MATCH could not be demonstrated directly as there was not a true control group that received no treatment, the striking differences in drinking from pretreatment levels to all follow-up points suggest that participation in any of these treatments will be associated with substantial and sustained changes in drinking. This was especially true in the outpatient arm where the treatments were the only treatments provided. This trial was meticulously performed and the results are likely to be valid. However, generalizing the results is problematic because these treatments were individual (not group), manual-guided, well structured and well performed with excellent follow-up efforts made (a tenacious research staff following up patients). Nonetheless, Project MATCH represents the first demonstration in a randomized clinical trial, controlling for other treatment factors, of comparable (if not slightly better) outcomes from a 12-step-based approach compared to other treatment methods. This study surprised the substance abuse community as many were expecting cognitive behavioral therapy to be superior in treating those with higher alcohol involvement, cognitive impairment, and sociopathy as well as other characteristics.⁸ However this was not confirmed and Twelve-Step Facilitation actually was slightly better overall. This trial is a valid, well-performed study and stands as the largest, statistically most powerful, psychotherapy trial ever conducted. It challenged the existing view that treatment matching by patient characteristics is a key to improving treatment effectiveness. Finally, it offers significant evidence that the AA philosophy and an approach geared at utilizing AA aggressively is likely one of the most efficacious treatment strategies available.

In summary, there is evidence supporting a recommendation to attend Alcoholics Anonymous in the literature. Unfortunately, most of the evidence is weighted in study designs that do not allow us to infer great strength of evidence. This is mainly due to a lack of large, well performed, randomized, controlled trials. Furthermore, the applicability of the few existing studies to general clinical practice is not established. However, one can feel comfortable in stating the following learning points:

• Alcoholics Anonymous (AA) is a safe, low cost, widely available tool of behavioral change that strives to capitalize on the patient’s inner motivation and spirituality.
• AA should be part of any attempt at treatment of alcohol abuse or dependence.¹
• Success with AA can be enhanced by a twelve-step facilitation treatment implemented concomitantly with AA attendance.
• Referring patients to AA groups composed of individuals of similar age, cultural, and occupational status may improve attendance and outcomes as well. ¹

To date there have been very few pharmacotherapeutic options for the treatment of alcohol abuse and dependence. The following will discuss evidence obtained for the two medicines currently approved by the FDA for the treatment of alcohol dependence, disulfiram and naltrexone. Two newer agents may hold promise in the future, acamprosate and ondansetron.

Disulfiram was the only FDA approved medication for alcohol dependence for many years and has been available since 1948. The ingestion of alcohol by a person who has been taking disulfiram results in an unpleasant reaction manifested by nausea, vomiting, flushing, light-headedness, abdominal pain, and tachycardia. ¹¹ The rationale behind its use is that an alcoholic taking disulfiram will not drink because he or she will not want to suffer the consequences of the disulfiram-ethanol reaction. ¹¹ Prior to the 1980’s there was a fair amount of enthusiasm for its use with nearly 100 studies reporting the efficacy of the drug. However, most of the studies were not controlled, blinded, randomized or of adequate sample size to allow for confidence in these positive results.

There are basically two well-designed trials of disulfiram. The first was an evaluation of 128 alcoholic U.S. Veterans in Cleveland. ¹² It was a randomized, partially blinded, prospective trial of one year of treatment with either a 250 mg dose of disulfiram, a clinically insignificant 1 mg dose of disulfiram (patients knew they were getting disulfiram but did not know what dose), or placebo. While 23% of the disulfiram treated patients remained abstinent and only 12% of placebo treated patients did so, this was not statistically significant (p > 0.10) In fact, there were no significantly different primary endpoints at the conclusion of the study and the authors indicated that disulfiram is of limited value in treating this population of alcoholics. ¹²

However, the study size was not large enough to fully rule out a possible small treatment effect of disulfiram. Therefore the same authors conducted a larger study which was published in JAMA in 1986.¹¹ This was a controlled, blinded, multicenter study of 605 Veterans randomly assigned a similar treatment strategy of 250 mg per day of disulfiram, 1 mg of disulfiram, or placebo. Outcome measures were complete abstinence during the year, time to first drink, total number of drinking days during the year, employment status, and social stability. There were no statistically significant differences among the three treatment groups in percentage of men remaining abstinent for the entire year with an overall abstinence rate of 19.2% for the study. In fact, the only significant finding was that those given 250 mg of disulfiram reported fewer days of total drinking over the course of the year (49 +/- 8 days) than those who received either the 1 mg of disulfiram (75 +/- 12 days) or no disulfiram (86.5 +/- 14 days) (p = 0.05). Furthermore the participants were a very select group as 6629 patients were initially screened and of the 1618 patients deemed eligible for the study a full 62% refused to participate. This suggested that the idea of this type of drug therapy is not very acceptable to patients. ¹¹

In 1989 a position paper from the American College of Physicians reviewed the data on disulfiram and concluded that disulfiram had not been shown to alter the long-term course of alcoholism, and that there existed no evidence that prescribing disulfiram will have any beneficial effect in the absence of coincident counseling or other rehabilitative therapy. ¹³ They did state that a small subset of patients, namely those that are compliant, with stable lifestyles who understand the risks of treatment and are receiving concomitant formal rehabilitative therapy may derive some benefit. Also, there is some evidence that disulfiram, when administered by direct supervision under court order can be associated with reduced alcoholism-related criminal behavior. ¹³ Given the totality of our knowledge of this drug it is not surprising that disulfiram is not commonly used by internists currently and is best reserved for special circumstances in the setting of formal substance abuse treatment.

Naltrexone (REVIA TM) became the second drug approved for the treatment of alcohol dependence by the FDA in 1994. It is an opioid antagonist (similar to NARCAN ®). It was first noted that small doses of opiates will increase alcohol drinking in rats and opiate antagonists will decrease such drinking. ¹⁶ It is speculated that the opioidergic

effect of the first drink "primes" the subject for more drinking. It is postulated that an ideal adjunct to alcoholism treatment should have the following three properties:

(2) block the reinforcing properties of alcohol so that, if drinking occurs, there is neither enhancement of pleasant feelings (positive reinforcement) nor reduction of distressing feelings (negative reinforcement), thus leading to little or nor motivation to continue drinking

(3) be free of adverse physical and emotional consequences. ¹⁶

There have been several randomized, double-blinded, placebo-controlled trials of naltrexone as an adjunct to psychosocial treatment programs for alcohol dependence. In 1992 Volpicelli et.al. published a study of 70 male alcohol-dependent patients. The results showed a decreased craving for alcohol in the treatment group (1.41 vs. 3.42 on a 0-9 point scale of craving, p < 0.01) and a decrease in the mean number of reported drinking days (1.6% of the study days vs. 8.3% of the study days , p < 0.25). While complete abstinence rates were not different between groups (54% treatment vs. 43% placebo), subsequent drinking once "lapse" drinking had occurred was significantly less for the naltrexone group (drinking 3.6% of study days for "lapsed" subjects in naltrexone group vs. 14% of study days for "lapsed" placebo group). Furthermore, the relapse rate was significantly less in the naltrexone group (23% vs. 54%, p < 0.01, RRR 57%, 95%CI 45% - 69%, NNT = 3). Therefore, one would have to treat 3 patients with naltrexone for 3 months (compared to placebo) to prevent one clinically significant relapse in that 3 month period. The most significant side effect of naltrexone was nausea with 2 patients (6%) withdrawing from the study due to nausea. ¹⁶

Figure 2: Relapse rates for naltrexone (closed circles) and placebo-treated (open circles)
Groups across the 12 weeks of the study. ¹⁶

This is an intriguing preliminary study but it is limited by the small sample size, high drop out rate, and total reliance on self-report measures of drinking (no collateral informants and no measures of compliance). A second study published in the same month by O’Malley et. al. was titled "Naltrexone and Coping Skills Therapy for Alcohol Dependence". ¹⁷ In this study 104 alcohol-dependent patients were treated for 12 weeks in a double-blind, placebo-controlled study evaluating naltrexone and two manual-guided psychotherapies in the treatment of alcohol dependence. Patients were randomized to receive either naltrexone or placebo and either coping skills/relapse prevention therapy or a supportive therapy (a 2 x 2 design which placed approximately 25 patients in each treatment group with a total of 52 receiving naltrexone and 52 receiving placebo). The supportive therapy was designed to support the patient’s own efforts at abstinence without teaching specific coping skills. There were 26% women, 93% were white and 73% were employed full time. Mean age was 40 years old. On average they drank on 60% of the days during the 60-day pretreatment baseline period and consumed an average 11 standard drinks per drinking occasion. Subjects were given 50 mg per day of naltrexone. Patients were treated for 12 weeks and also had a final follow-up at 6 months after the end of treatment. Intention-to-treat analysis was performed and collateral informants’ reports supported the validity of the subjects’ self-reports of quantity and frequency of alcohol consumption. More patients in the naltrexone group had nausea (32.7% vs. 13.7%), weight loss (24.5% vs. 7.89%), and dizziness (34.7% vs. 15.7%) (all p values <0.05). A total of 5 naltrexone patients withdrew due to side effects (10%) but otherwise compliance was high. ¹⁷

A full 61% naltrexone/supportive therapy patients remained continuously abstinent over the 12 weeks (p = 0.01) whereas 28% of the naltrexone/coping skills group, 21% of the placebo/coping skills group, and 19% of the placebo/supportive group remained abstinent for the 12 weeks (not statistically significant). ¹⁷ Relapse was defined as drinking five or more drinks on an occasion for men and as drinking four or more drinks on an occasion for women. Subjects taking naltrexone and receiving either supportive or coping skills therapy had a relative risk of relapse of about two thirds less than the risk for subjects taking placebo .

    Figure 3a:  Survival plots showing rates of never Relapsing. ¹⁷

The authors conclude that naltrexone was clearly superior to placebo on a number of measures related to alcohol consumption. It is interesting that the patients in the naltrexone/supportive therapy group had higher abstinence rates and less quantity of drinking than those in the naltrexone/coping skills group. Supportive therapy simply involved frequent visits to the therapist with nonjudgmental support and encouragement to try to remain abstinent. This is similar to what would happen in an internist's office. One reason this may have occurred is that the coping skills sessions taught patients to expect lapses and slips with a goal of teaching skills to keep these lapses from becoming full "relapses". Therefore those patients were more likely to experiment with drinking during the study.

In a six month post treatment follow-up study of these patients 82% of the original patients were assessed (however entirely by self-report data in this follow-up). ¹⁸ The results of this follow-up showed that some but not all of the benefits resulting from short-term treatment with naltrexone persist after discontinuation of treatment. Although naltrexone resulted in higher abstinence rates during treatment, this effect was no longer apparent by the second month of follow-up. The percentage of subjects who avoided relapse over the entire follow-up period was higher for naltrexone than for placebo. In addition, the percentage of subjects who had at least one day of heavy drinking over the 6-month follow-up period was significantly higher for placebo-treated patients than for naltrexone-treated patients [78% (31/40) vs 55% (22/40), p < 0.05; RRR = 29%, 95%CI 19% - 39%, NNT = 4]. ¹⁸

Finally, subjects who had received naltrexone were significantly less likely to meet the criteria for a diagnosis of alcohol abuse or dependence over the course of the follow-up period than subjects who had received placebo. Of those who received placebo 24/38 (63%) met criteria for alcohol abuse or dependence for the six-month follow-up period whereas only 12/39 (31%) of the naltrexone-treated patients met these criteria (RRR = 51%, 95%CI 40% - 62%, NNT = 3). In other words, one would have to treat three patients with 12 weeks of naltrexone plus coping skills or supportive therapy to prevent one patient from meeting criteria for alcohol abuse or dependence at 9 months after the start of therapy compared to coping skills or supportive therapy alone. They postulated that continuing naltrexone beyond 12 weeks may be beneficial for some patients. ¹⁸

Figure 3b:  Survival plots showing continuous abstinence.¹⁷

This study is limited by its small sample size which led to some baseline differences in the groups at study entry and which causes one to wonder whether or not the randomization process was successful. Secondly, the six month follow-up portion of the study relied solely on self-reports of drinking and alcohol-related problems. Also there was a significant dropout rate with 82% of the patients being followed for the full six months post treatment. One would assume that those with poor outcomes dropped out of the study however including those "lost’ patients in the calculations may have changed some of the outcomes. Despite the limitations this study seems to show some important positive effects rendered by naltrexone therapy and helped inspire further research.

More recently Anton et. al. published a study in November of 1997 titled "Naltrexone and Cognitive Behavioral Therapy for the Treatment of Outpatient Alcoholics: Results of a Placebo-Controlled Trial". ¹⁹ In this study 131 recently abstinent alcohol-dependent outpatients were treated with 12 weekly sessions of manual-guided cognitive behavioral therapy and either 50 mg/day of naltrexone (N=68) or placebo (N=63). Alcohol consumption, craving, adverse events , and compliance were assessed weekly. Intention-to-treat analysis was performed and there was follow-up data for 98.5% of the patients. Patients were on average 42 years old, 70% male, and 85% white. Nine patients in the naltrexone group and 14 in the placebo group terminated the study early. At the end of the study 62% of the naltrexone patients compared to 40% of the placebo patients had not relapsed (p = 0.01, RRR = 55%, 95%CI = 47% - 63%, NNT = 4.5). ¹⁹ The naltrexone patients did remain continuously abstinent longer than the placebo group (60 days vs. 22 days) but this was not statistically significant. However, naltrexone treated patients had significantly less drinks per drinking day (2.5 vs. 4.2, p = 0.01) and a greater percentage of days abstinent (90% vs. 82%, p = 0.03). Significant adverse side effects were nausea/vomiting (34% vs 14%, p < 0.01), abdominal pain (31% vs. 11%, p < 0.01), daytime sleepiness (46% vs 27%, p < 0.05), and nasal congestion (46% vs. 25%, p < 0.05) but only one subject dropped out due to side effects. ¹⁹ The authors concluded that motivated individuals with moderate alcohol dependence can be treated with greater effectiveness when naltrexone is used in conjunction with weekly outpatient cognitive behavioral therapy. This was a well-performed valid study however the applicability of the results are somewhat limited. This is due to the highly selected nature of the patients who were overall well-educated, motivated, and socially stable. Furthermore, as with most of the studies of naltrexone, patients with other drug abuse, severe psychiatric problems or medical problems were excluded. However, the findings do confirm earlier findings that naltrexone can add to treatment affects and is possibly synergistic with other therapy.

In an effort to expand treatment options for alcoholic patients encountered in primary care settings O’Connor et. al. performed a prospective cohort study titled "A Preliminary Investigation of the Management of Alcohol Dependence with Naltrexone by Primary Care Providers". ²⁰ This is the first evaluation of naltrexone treatment of alcohol dependence by primary care providers. Patients were treated with 50 mg/day of naltrexone for 10 weeks. The primary care management included an initial 45 minute "new patient visit" and 7 follow-up visits for weeks 1, 2, 3, 4, 6, 8, and 10. All subjects were also referred to AA. The outcomes were completion of treatment, attendance at AA meetings, change in alcohol consumption, and change in gamma glutamyl transferase (GGT) level. A "relapse" was defined as one or more days of heavy drinking (>4 drinks for men and >3 drinks in women). ²⁰

Twenty-nine (29) subjects were enrolled of which 83% were male, 97% were white, and 79% were employed. Seventy-two percent of the patients completed the 10-week program. Of the 8 subjects who failed 3 withdrew due to side effects (dizziness and nausea in 2 and headache and malaise in 1). Only 31% (9/29) attended at least 1 AA meeting while in the study. Overall 45% (13/29) remained abstinent, 35% (10/29) relapsed to heavy drinking, and 21% (6/29) reported some alcohol intake but did not relapse. Percent Days Abstinent, percent days abstinent from heavy drinking, drinks per occasion and serum GGT levels all were significantly reduced (see table below). ²⁰

Table 1: Change in Measures of Alcohol Consumption: Baseline Versus During Treatment (n = 29). ²⁰

The authors conclude that the frequency of visits required for this treatment, while high, is not necessarily inconsistent with the level of interaction needed with other medical patients with such conditions as diabetes, HIV disease, and depression. In addition, patient satisfaction scores were high (71% were very satisfied) and while there were three patients who discontinued therapy due to side effects the drug was tolerated well overall. ²⁰ Given all of this together the authors feel that this may be a promising approach to the care of the uncomplicated alcohol dependent patient. Some notable limitations of this study are the lack of any control or placebo group as this was an uncontrolled cohort study. Furthermore the question of duration of therapy remains to be answered. These authors plan further research.

One further study deserves attention concerning treatment with naltrexone. Kranzler et. al. published a study titled "Naltrexone vs. Nefazodone for Treatment of Alcohol Dependence" earlier this year. ²¹ It was a double-blind, placebo-controlled trial of naltrexone versus a serotonergic agent nefazodone. A total of 183 subjects were randomly assigned (61 to naltrexone, 59 to nefazodone, and 63 to placebo). All the treatment groups were similar demographically with approximately 75% male, 95% white and mean age of approximately 40 years. All subjects received weekly coping skills training and treatment medicine for 12 weeks. Interestingly, only 59% of the naltrexone patients completed treatment which was significantly less than the other two groups (79% of placebo patients and 73% of the nefazodone patients). Significantly more naltrexone patients than placebo reported blurred vision, drowsiness, lightheadedness, poor coordination, confusion, decreased appetite, nausea, and vomiting. This gives a relative risk increase of 310% for vomiting (95% CI = 265% - 355%) and a number needed to harm of 7. Therefore, for every 7 patients treated with naltrexone over 12 weeks one case of vomiting will be caused. Furthermore, Kranzler found no positive treatment effects of either naltrexone or nefazodone. ²¹ Therefore, this study did not at all confirm previous reports of naltrexone’s efficacy. It seemed that the severe side effects limited the efficacy substantially. This study showed a much greater degree of intolerability than other studies including an open-label safety study of naltrexone in 570 alcohol-dependent patients which showed the most common adverse clinical events to be nausea (9.8%) and headache (6.6%) and in that study 15% discontinued medication due to side effects. ²¹

In summary, naltrexone has been the best studied medicine for the treatment of alcohol dependence. With the exception of the most recent study by Kranzler (which was a well performed trial) naltrexone has been shown beneficial effects. Furthermore, while this drug certainly has side effects which will cause at least some patients to discontinue the medication, this most recent study stands as an outlier. Also, no major side effects causing serious harm have been described in these trials (although a risk of hepatocellular injury may be caused with doses greater than 5 times normal over significant time frames) (source: Physicians’ Desk Reference). This leaves one to wonder just what is the true tolerability and efficacy of naltrexone? In the final analysis, it is clear that with respect to naltrexone or any postulated medical therapy for alcoholism much further study, with well designed large trials are needed. However, given the current evidence, it would be reasonable for an internist to attempt a trial of naltrexone in a patient with alcohol dependence who is at low risk for complicated withdrawal, and refuses more specialized treatment but earnestly desires treatment. If the patient can achieve 5 days of abstinence before starting the medicine, will agree and comply with close follow-up and brief counseling sessions and attempt to attend AA then this course would be appropriate. Also, this medicine could be prescribed by an internist in combination with referral to a substance abuse counselor.

The next agent which will most likely be approved by the FDA for the treatment of alcoholism is acamprosate. It is to be reviewed by the FDA this year. Acamprosate (calcium acetylhomotaurinate) is a compound that has a similar chemical structure to that

of aminoacid neuromediators such as taurine and GABA. It has been reported to stimulate inhibitory GABA transmission and to antagonize excitatory aminoacids, particularly glutamate. Restoration of the inhibition/excitation balance might be the biochemical basis of its effects. ²² It does not enhance ethanol toxicity, has no abuse potential, and no hypnotic, anxiolytic or muscle-relaxant properties.

Several preliminary trials showed acamprosate to alter alcohol intake. ²² One substantial study investigating acamprosate as a treatment adjunct in alcoholism was published by Whitworth et. al. in The Lancet in 1996. This was a randomized, double-blind, placebo-controlled trial which took place in Austria. It included 455 patients (mean age 42 years old, 79% men) with alcohol dependence. All patients received treatment with psychosocial rehabilitation after at least a 5 day inpatient detoxification. Acamprosate was given to 224 patients (1998 mg/day if body weight > 60 kg or 1332 mg/day if body weight < 60 kg) and 224 patients received placebo. Outcomes were self-reported abstinence, time to first treatment failure (relapse or nonattendance), and side effects. A secondary outcome measure was the sum of all periods of total abstinence. The treatment and placebo groups were similar demographically and in terms of drinking variables and intention-to-treat analysis was performed (n = 448) ²³

Patients who received acamprosate were significantly more likely to remain continuously abstinent for the one year of treatment (41/224 (18.3%) vs 16/224 (7.1%), p = 0.007).This gives a NNT of 9 which means that 9 patients would need to be treated with acamprosate for 1 year (compared with placebo) to have 1 additional patient remain abstinent. Also the mean cumulative duration of abstinence was higher in the acamprosate group (139 vs 104 days, P = 0.012). The only significant side effect was diarrhea (20% vs 12%, P = 0.021). ²³ The authors concluded that acamprosate was a safe and effective adjunct to psychosocial therapy. This study did have a fairly unrestrictive selection criteria, a good sample size and a long duration of follow-up. These factors make the results fairly generalizable to clinical practice.

However, limitations of this study were that 61% of the patients dropped out during the 12 months of treatment so that outcomes could only be evaluated in a minority of the patients (of course, any patient who dropped out was considered a treatment failure so there was definitely no overestimation of the treatment effect). In addition, one of the main outcome measures was time to first relapse (which constituted as taking any drink or missing any follow-up appointment). This ignored the severity of relapse so that theoretically a patient who drank himself into a coma at 2 months would have a "better" outcome than one who had a single drink at one month. ²³ Also the outcome measures were somewhat insensitive. They used self-reported alcohol consumption, physical signs of alcoholism, tremor index, mean corpuscular volume, and GGT values which are not very sensitive or specific markers of alcohol consumption. This study is a valiant pilot study of a new treatment that may be clinically important for internists but studies with better markers of therapy need to be done.²³

One of the newest drugs to be studied is ondansetron which is a selective 5-HT3 (serotonin) receptor blocker. There is evidence that early onset alcoholism is associated with serotonergic disfunction.²⁴ Johnson et. al. recently published an article in the August 23, 2000 release of JAMA titled "Ondansetron for Reduction of Drinking Among Biologically Predisposed Alcoholic Patients". In this double-blind, randomized, placebo-controlled trial the authors tested the hypothesis that patients with early-onset alcoholism compared with late onset alcoholism would experience better drinking outcomes in response to ondansetron treatment due to amelioration of serotonergic dysfunction. Included were 321 enrollees (70% male, 79% white) diagnosed with alcohol dependence. Ondansetron significantly reduced alcohol consumption and increased abstinence among patients with early onset alcoholism but not late-onset alcoholism with, although the effect sizes were small.²⁴ While this is clearly a preliminary study and certainly does not open the door for implementing ondansetron in clinical practice, it supports research that actively attempts to implement rational drug design based on basic science principles. Furthermore, there is a growing body of literature delineating the biological basis of alcohol dependence and its relationship to clinical subtypes such as early or late onset alcoholism. ²⁵

The treatment of problem drinking, alcohol abuse and alcohol dependence remains one of the greatest challenges in medicine. Over the last several decades various psychosocial therapies such as cognitive behavioral therapy, behavioral therapy, psychotherapy, group, marital, and family therapy have been employed by psychiatrists, psychologists and counselors to achieve some meaningful treatment effects in these patients. Alcoholics Anonymous has been an ever-present component in the landscape of alcoholism treatment but has avoided much pure scientific study. Recently, especially with the results of Project MATCH more is being understood about how effective the concept of AA can be. Drug therapy for alcoholism has lagged behind other fields but there is currently a renewed interest in research. ²⁵ Disulfiram represents a fairly primitive form of aversive therapy with a narrow therapeutic index and its use should probably be relegated to the most specialized circumstances. Naltrexone offers more promise but has not become a commercial success in the U.S.. This is likely due to a combination of underdiagnosis of alcohol use disorders and pessimism on the part of physicians over the value of alcoholism treatment.²⁵ Also its side effect profile is significant and duration of therapy and extended long term outcome studies are lacking. Acamprosate will likely be approved soon and is already in widespread use in Europe and newer drugs such as ondansetron are being actively investigated. ²⁵ Other important therapies for the internist include brief physician interventions which hinge on the knowledge of the various stages of change that patients go through when the need for behavior change is confronted. Brief physician interventions have been shown to be effective to reduce drinking in a number of studies. ¹⁴ Finally, the internist can recruit added resources in the management of these patients. In particular, employee assistance programs (EAP), family members, and significant others (if the patient allows) can have a tremendously positive impact on the patient.

For the practicing internist, there is a call for renewed vigor in diagnosing alcoholism and referring patients to the proper treatment. Furthermore, as treatment evolves the internist should consider, at least as an option, a therapeutic trial of one of the newer medicines such as naltrexone, and soon acamprosate, for those patients who refuse referral but wish to work with the internist for a better outcome. This should never be done expecting the medicine alone to solve the problem. Brief, frequent counseling sessions and thoughtful referral to AA are required in this circumstance. We should anticipate larger, well-designed studies with endpoints not just including abstinence rates and drinking quantity but also hard outcomes such as death and hospitalization for alcohol related illness. There is no question that the treatment of the underlying disorder of alcoholism, as opposed to just the medical consequences of the disease, will increasingly migrate into the realm of primary care physicians.

1. Screening and offering specialized referral for alcohol abuse and dependence should be done more frequently. Due to failure to diagnose alcoholism, therapeutic nihilism and lack of training the care of alcoholics is not adequate in most internal medicine practices.
2. Despite high early dropout rates and a low frequency of patients following the recommendation, a referral to attend a well matched Alcoholics Anonymous group is a safe, inexpensive, and potentially efficacious option for those with alcohol abuse or dependence. Continued attendance should be encouraged despite high early drop-out rates.
3. A referral to attend Alcoholics Anonymous should be part of any treatment plan no matter how formal or informal.
4. Drug therapy for alcoholism is still mostly in developmental stages, however:

Selected individuals in formal rehabilitative treatment may benefit from treatment with disulfiram.

Naltrexone in combination with formal treatment may increase abstinence rates and reduce drinking.

The most appropriate duration of naltrexone treatment is not fully known.

There may be selected individuals in a general internal medicine practice who could appropriately be prescribed naltrexone, especially in combination with frequent counseling/brief intervention visits, demonstrated follow-up compliance, and concurrent attendance to AA meetings or other treatment. This is particularly so for a patient who refuses more formal treatment but is still earnestly interested in treatment by the primary care physician. If one chooses this path it is important to make sure the patient is abstinent for 5 or more days prior to starting the drug (as naltrexone has not been studied during alcohol withdrawal) and that the patient does not use opioids as an acute withdrawal syndrome could be precipitated.

Acamprosate will likely soon be the third FDA approved medicine for the treatment of alcoholism

Newer drugs such as ondansetron are actively being investigated.

1. We admitted we were powerless over alcohol – that our lives had become unmanageable.
2. Came to believe that a Power greater than ourselves could restore us to sanity.
3. Made a decision to turn our will and our lives over to the care of God as we understood Him.
4. Made a searching and fearless moral inventory of ourselves.
5. Admitted to God, to ourselves and to another human being the exact nature of our wrongs.
6. Were entirely ready to have God remove all these defects of character.
7. Humbly asked Him to remove our shortcomings.
8. Made a list of all persons we had harmed, and became willing to make amends to them all.
9. Made direct amends to such people wherever possible, except when to do so would injure them or others.
10. Continued to take personal inventory and when we were wrong promptly admitted it.
11. Sought through prayer and meditation to improve our conscious contact with God, as we understood Him, praying only for knowledge of His will for us and the power to carry that out.
12. Having had a spiritual awakening as the result of these steps, we tried to carry this message to alcoholics, and to practice these principles in all our affairs. ¹⁰

Appendix II: Decision Tree from the NIAAA Guideline¹⁵

Mental Health Center (Centerpoint Human Services) 725-777

Alcoholics Anonymous 725-6031

Many programs financed by state and county funds and private donation/volunteerism

ARCA (Addiction Recovery Care Association) 784-9470

Has residential care treatment, family program and detox program.

Alcohol/Drug Council of NC 1-800-688-4232

Information and referral service, public education, and advocacy for public policy

Associates in Christian Counseling 896-0065

Counseling and psychological services for a wide range of problems

Wake Forest University Baptist Medical Center – Substance Abuse 716-4558

Treatment Program – treatment for alcohol, drug, and nicotine addiction

The Coalition for Drug Abuse Prevention 723-3784

Clearinghouse of information, referral information

F.I.R.S.T. – Forsyth Initiative For Residential Self-Help Treatment, Inc. 777-0997

Two year re-educational environment for substance abusers.

First Line – Forsyth Line (Forsyth Information & Referral Service 727-8100

Telephone Line – referral information (Spanish) 748-3288

Friendship House, Inc. 725-1142

Helps rehabilitate alcoholic women. Counsels any women desiring help.

Lighthouse Ministries, Inc. 723-7884

Provides residence for homeless men and alcoholics.

Novant Health – Behavioral Health Resources 718-3550

Provides a full continuum of adult mental health and substance abuse services

Novant Health – Recovery Network 718-3800

Intensive outpatient addiction treatment program.

Prodigals Community – An ecumenical ministry providing recovery 785-0770

ministries for drug-free responsible living, spiritual growth, etc…

Winston-Salem Rescue Mission 723-1848

Ministers to the physical and spiritual needs of persons requesting help.

Samaritan Ministries 748-1962

Soup Kitchen and Inn

Step One, Inc. – Alcohol/Drug Outpatient Treatment Services 725-8389

Programs include medical evaluation, assessment, individual/group counseling

Vocational Rehabilitation Services – NC Dept. of Health and Human 761-2400

Services. Provides services to vocationally disabled individuals.

Numbers From Telephone White Pages:

Alcohol Abuse 24 Hour Addiction Services of Focus Healthcare 1-800-274-2042

Appendix IV: Prices for Disulfiram and Naltrexone (One month supply)

Disulfiram Brand Name (Antabuse ®) $40.29
Generic $8.99

Naltrexone Brand Name (REVIA TM) $165.99
Generic $132.99

Source: CVS Cloverdale Ave. , Winston-Salem, NC

Appendix V: The CAGE Questions

Have you ever felt you ought to Cut down on your drinking?

Have people Annoyed you by criticizing your drinking?

Have you ever felt bad or Guilty about your drinking?

Have you ever had a drink first thing in the morning to steady your nerves or get rid of a hangover (Eye-opener)?

Bibliography

1. American Psychiatric Association; "Practice Guidelines for the Treatment of Patients With Substance Use Disorders: Alcohol, Cocaine, Opioids"; American Journal of Psychiatry. Vol. 152, No. 11, November 1995 Supplement. Pp. 5-59.