Irving M. Maltzman is professor emeritus, department
of psychology, University of California, Los Angeles. He has
written extensively on the controlled drinking controversy.

There is no good evidence supporting the effectiveness of controlled drinking for chronic alcoholics. Why should it be offered as an alternative to abstinence training?

However, it is logically possible that a small number of people suffering from alcohol abuse or alcohol dependence (alcoholism) could choose moderation over abstinence and benefit.

But, self-selection of treatment is currently impossible for many important reasons: (1) Caregivers cannot assist their patients in making an informed, ethical and valid choice in treatment, because they do not offer patients material information concerning the neurological and neuropsychological state of the patient's brain.

The majority of individuals diagnosed as alcohol abusers and alcoholics suffer from brain dysfunction and neuropsychological deficits (see Maltzman, 2000, for references supporting this and other assertions herein).

Patients who practice abstinence may reverse those deficits. But, continued drinking, even in small amounts, prevents reversal and may exacerbate the damage. A valid informed consent requires that such information be provided the patient before proceeding with the selection of a treatment modality.

(2) A related problem is the patient's diminished capacity; his or her inability to fully understand the information presented, integrate it and determine its relevance to self.

Diminished capacity is common in alcohol-abusing patients due to the neurochemical effects of alcohol, either in terms of brain damage and consequent neuropsychological deficits, depression, panic, polydrug use and more.

A patient with diminished capacity is not fully autonomous, and is incapable of providing a valid informed consent and a reasoned treatment self-selection. There are standardized assessment procedures developed by Grisso and Appelbaum for determining whether or not an individual has the capacity to provide an informed consent. As far as I know they have not been used with alcoholics or alcohol abusers.

(3) Self-selection of moderate drinking presupposes that controlled drinking training is equally as effective as abstinence training, and that self-selection increases self-efficacy thereby resulting in more superior outcomes than beneficence, caregiver assigned treatment.

Both are false assumptions based on inadequately designed studies, misinterpretations of results, or both. Rotgers (1996) argues that if there is self-selection with the options of abstinence or controlled drinking, patients who need abstinence will choose it. He claims that a treatment goal prescribed for a patient produces poorer results than one freely chosen as demonstrated in a study by Booth, Dale and Ansari, (l984).

This is a poorly designed and conducted efficacy study with a small number of participants and an unreliable outcome. The study confounds severity of alcoholism with experimental conditions. It is no basis for treatment program innovations.

The other study Rotgers (1996) cites in support of patients self-selection (Sanchez-Craig & Lei, l986) used participants from an earlier efficacy study (Sanchez-Craig, l980), which explicitly excluded participants who believed in the disease concept or who participated in Alcoholics Anonymous. Results from such a biased sample cannot be generalized to the population of people seeking help for problems of alcohol misuse or to the ethical conflict between autonomy and beneficence.

Rotgers (l996) fails to cite the most extensive experiment available comparing choice versus prescribed treatments (Walsh, et al., 1991). Participants with drinking problems in an employee assistance program were randomly assigned to a traditional inpatient treatment program, mandatory AA meetings or a choice group.

The latter group was free to choose either of the other two treatments, a different treatment of their own choice or no treatment as long as they were not drunk on the job.

At the 2-year follow-up the prescribed inpatient treatment group had a significantly higher continuous abstinence rate and significantly lower percentage of heavy drinking days than the choice group which did not differ significantly from the prescribed AA group.

Foy, Rychtarik and their colleagues have conducted the best-designed study comparing abstinence versus controlled drinking treatment for chronic alcoholics. Patients were randomly assigned to two groups. Social workers blind to the design and purpose of the study conducted follow-up interviews. A randomly assigned control group received traditional abstinence training.

The experimental group received abstinence training plus controlled-drinking-skills training. At the 6-month follow-up the abstinence-only group was significantly superior to the experimental group receiving abstinence training plus controlled-drinking-skills training.

A follow-up 5-6 years later found that controlled drinking training had no significant effect beyond abstinence training. There is no good evidence supporting the effectiveness of controlled drinking for chronic alcoholics or controlled drinking training for alcohol abusers.

Studies by William Miller and his colleagues suffer from numerous shortcomings and misinterpretations (Maltzman, 2000). The purported authoritative review of treatment outcomes by Miller (1995) and colleagues is nothing of the sort. They employed a biased and limited sample of journals to be scanned and consistently ignored the best-designed studies of abstinence treatment conducted in the United States, Germany and Australia (Maltzman, 2000).

An evaluation of 9,000 patients primarily from Minnesota Model programs (Harrison, Hoffmann, & Streed, 1991) yielded treatment outcomes far superior to results obtained with cognitive behavior therapy such as reported by Hester and Miller (1995).

(4) Schneider (1998) reports considerable denial in seriously ill people such as patients in need of a liver transplant. They may therefore select a less severe treatment. Standardized procedures for assessing denial of physical illnesses are available (Levine, Rudy, & Kerns, 1994).

Denial or diminished capacity to appreciate that one is ill has also been reported as a diagnostic feature of schizophrenia (Carpenter, Bartko, Carpenter, & Strauss, 1976; Carpenter, Bartko, Strauss, & Hawk, 1978). Denial of illness or its symptoms, a person's capacity to appreciate the nature or severity of their disorder, is common to a variety of disorders. It is a form of self-deception (Myslobodsky, 1997).

Interpretations of denial are that it is a form of avoidance serving as a means of coping with anxiety or that the lack of insight into the severity and characteristics of the disorder typical of denial is a consequence of neuropsychological deficits. The latter theory is supported by evidence from anosognosias, the lack of awareness of deficits due to a specific brain lesion. Mohamed, Fleming, Penn and Spaulding (1999) studied denial in schizophrenics providing evidence supporting the neuropsychological deficit hypothesis.

They obtained significant correlations between performance on measures of frontal lobe executive functions and measures of symptom awareness. Given the extensive evidence (Maltzman, 2000) that excessive alcohol consumption produces significant deficits in brain function, the implication is that denial or lack of awareness of alcoholism symptomology is due at least in part to frontal lobe deficits.

Research of the kind conducted with schizophrenics and other special populations is needed in alcoholism studies. The alcoholic, and alcohol abuser in true denial, must be differentiated from the person with a drinking problem who desires harm reduction rather than abstinence and is not in denial and does not suffer brain damage.

Making this distinction requires standardized assessment procedures and a reliable research derived database. Instruments are available for assessing denial in alcoholics (Goldsmith & Green, l988; Newsome & Ditzler, 1993; Ward & Rothaus, 1991). There is a need to design and conduct studies investigating implications of denial for self-selection. Advocates of self-selection have failed to conduct this essential research as they have failed to conduct essential neuropsychological research and research on the capacity of patients to provide a valid informed consent. Until such research is conducted and results used to guide treatment, advocating self-selection of moderate drinking as opposed to abstinence is promoting an iatrogenic treatment.

Frederick Rotgers, PsyD, is assistant chief
psychologist at Smithers Treatment Center, St.
Luke's-Roosevelt Hospital, New York, N.Y.

That these questions are asked is emblematic of the deep confusion that characterizes our field with respect to the implications of an alcohol abuse or dependence diagnosis. It reflects a continued reliance on folklore and myth in the face of excellent scientific evidence, the continued dominance of an unusual hybrid "disease" model of problem drinking and the concerns of counselors and other professionals that they may make "the wrong decision" with respect to a patient in our litigious society.

The questions also assume that there is a "correct" mode of treatment, that the "correct" treatment must focus on achieving a specific drinking pattern (abstinence) and that clinicians can somehow "prescribe" a life course to a patient without that patient's full compliance and consent. Finally, the second question confuses legal and clinical issues. Competence is a legal issue, not a clinical one.

I will address the second question by means of an analogy. Substitute in the question "has cancer" for the words "abuses alcohol". Cancer is a potentially life-threatening disease that requires both compliance with a treatment regimen for a return to health, and often requires additional lifestyle changes for the patient to maintain that health.

It is a disease, for which, like addictions, there are a variety of viable treatment options, some more drastic than others. Cancer is also more treatable if detected early, yet many people fear it sufficiently that they deny the symptoms until they are quite severely affected.

We would never think to question the competence of a cancer patient to choose his/her own treatment from an array of effective options provided by an oncologist, even when that patient delayed taking action when symptoms appeared, or when that patient chooses an option the doctor sees as less than optimal. We would never question a cancer patient's competence to opt for chemotherapy rather than surgery when surgery is an unacceptable option to that patient.

Only in extreme cases, where there is clear evidence of cognitive confusion does the issue of competence to choose among cancer treatments arise. That we ask it with respect to persons whose drinking creates problems indicates how different our "disease model" is from that characterizing mainstream medicine. Some persons with drinking problems are incompetent to decide on treatment options — but they are rare. Only persons who fail to meet basic legal standards of competence (i.e., ability to know who they are, where they are, understand and respond to basic questions, understand the nature of the situation they are in, etc.) are unable to decide on the means for achieving health.

There is no research or other reliable evidence to the contrary. This first question confuses the goal of treatment with the means to achieving the goal. In mainstream clinical practice, the primary goal of treatment is the restoration of health and well-being.

For people whose drinking creates problems, there are two primary means to reaching those goals (or what the DSM-IV calls "remission"): reducing drinking to zero (abstinence) or reducing it to non-harmful levels (moderation).

If either of these means is used and sustained over time, the individual's diagnosed problem (alcohol abuse or alcohol dependence) will, of necessity, remit. Remission does not depend on abstinence: nowhere in the DSM-IV is not quantity of consumption mentioned, nor is abstinence cited as the only means to achieving remission. We need to begin thinking more about the goal of treatment (helping people restore health and well-being) rather than focusing exclusively on one means of achieving the goal.

By being fixated on one particular means, abstinence, we reduce our ability to help many people for whom the means we advocate are unacceptable. It’s as though we turned away all cancer patients who would not accept surgery as an option, but preferred chemotherapy. I believe that good clinicians always want the best possible outcomes for their patients — a return to full health and well-being and should work with their patients to achieve that goal using whatever legitimate, research-validated means the patient is willing to accept toward that end.

My belief is strongly supported by research. There are at least 10 well-designed studies, dating back to the late 1970s that have examined the choices patients with drinking problems make with respect to means of achieving health and well-being. Uniformly, across the studies, patients entering treatment for drinking problems choose abstinence as the means of resolving those problems. In fact, averaging across these studies shows that 75 percent of patients choose abstinence when offered a choice between abstinence and moderation.

Interestingly, of the 25 percent who initially choose moderation, nearly half, in studies, which have examined this, shift to abstinence by the end of treatment. So, if patients are incapable of making healthy choices about the means of resolving a drinking problem, these data do not show it! What about the effect of encouraging patient choice on the success of treatment?

First, it is clear that most patients want to decide for themselves how to go about resolving a drinking problem. In a study at the addiction Research Foundation (now the Center for Addiction and Mental Health) in Toronto, two-thirds of patients polled wanted to self-select the means to resolving their drinking problems. More striking are the results from another study from the Addiction Research Foundation by Martha Sanchez-Craig and her colleagues.

Sanchez-Craig randomly assigned patients to one of two groups. In one group, patients were allowed to choose between abstinence and moderation programs — that is, they were permitted a choice of means to resolving their problems. In the other group patients were assigned, without choice, to either an abstinence or moderation program.

Both groups of patients were similar in the severity of their drinking problems, and all had sought help for them. The results were clear. Subjects who had no choice did more poorly, regardless of whether the means imposed was abstinence or moderation, than did those subjects who chose the means they preferred.

This is a powerful demonstration of the detrimental effect of imposing means on patients. Finally, let’s look at what Bill Wilson says about choice in Alcoholics Anonymous.

I’ll provide page numbers rather than complete quotes, and encourage the reader to examine the source for him/herself. On Pg. 94-95, Wilson speaks of allowing people to pursue their own path, and encourages patience and self-discovery as the best means for an "alcoholic" to make the decision to come to AA.

On Pg.31-32 he proposes a quick, self-diagnostic "experiment" to help in making this decision, an experiment that can most safely be done in the context of a solid relationship with a counselor.

Finally, he states, at the bottom of p. 95: "We have no monopoly on God: we merely have an approach that worked with us." Wilson understood the diversity of problem drinkers, the capacity of "alcoholics" to decide to change using effective means, and the viability of a variety of means to achieving a healthier life.

Perhaps it’s time we professionals re-read Bill Wilson’s words, and take them to heart! Then we’d spend more time figuring out how to work "with" our patients, rather than "on" them!