Treatment of Alcohol Dependence

Treatment of alcohol dependence

Topics covered:

  • A chronic relapsing disorder
    • Starting treatment
  • The initial interview
  • Medical assistance for withdrawal
    • The setting
    • Medication
    • Treating convulsions
    • Preventing convulsions
    • Treating delirium tremens
    • Preventing delirium tremens
    • Vitamin therapy
  • Interventions to reduce relapse
    • The evidence
    • Abstinence or ‘controlled drinking'?
    • Maintaining motivation and compliance
    • Helping motivation: the social matrix
    • Cognitive-behavioural therapies
    • Relapse prevention therapy
    • Cue exposure
    • Alcoholics Anonymous
  • Help for the family
    • Behavioural marital therapy
  • Deterrent medication
    • Disulfiram
    • Specific neurotransmitter antagonists
  • Helping women with alcohol problems
  • Treatment of coexisting disorders
    • Affective disorder
    • Anxiety and panic disorder
    • Other comorbidities
  • Residential and inpatient treatment
  • Matching patients to treatments
    • The problem
    • Guiding principles
  • Some clinical situations
    • Morbid jealousy
    • The homeless alcohol-dependent person
    • Young people
    • Employment referrals
    • The liver transplant candidate
    • Physicians as patients
  • Follow-up
  • References

A chronic relapsing disorder

Some people repeatedly put themselves or others at risk by drinking. One view is that such people could drink sensibly if they were more considerate and used more will power. Another increasingly accepted view is that many such individuals are in a state, existing in degrees of severity, in which the freedom to decide whether to change their drinking, and to adhere to that decision, is reduced compared with other drinkers. This state partly depends on perceived pay-offs for changing, and on acquired dispositions which are less accessible to conscious control. Such persons become aware of a wish, or urge, to drink which overcomes rational thought. They may then make up an explanation, for example ‘No wonder I feel like a drink, I've had a hard day's.

Such individuals benefit from help to unlearn those patterns, and to learn different approaches to problems. Discussion, care, and encouragement from others can bolster their will to do so. Assistance to set up controls within or from outside themselves may help. Some people can do this without external help, and others with the help of Alcoholics Anonymous (AA) alone. (1)

This approach argues that dependence on alcohol should be managed like other relapsing disorders, such as diabetes and asthma, (2) by using long-term monitoring coupled with intermittent or continuous treatment. However, social and cultural influences are stronger than in relapsing medical conditions and have more effects on outcome.

Starting treatment

The initial interview

Assessment is the first step of intervention; clumsy interviewing alienates an ambivalent patient. The key to success is accepting that the patient is probably in two minds about the interview and about changing his or her drinking habits. Avoid confrontation. The drinking has probably already shown its resistance to deterrence by fear or pain. Gently nudge the matrix of conflicting motivations in the direction of action.

Patients may or may not have been referred for help with alcohol problems. Even if they have, the interview should begin with enquiry into the patient's current concerns. Reflective listening (3) helps the patient to clarify these concerns, conveys empathy, and avoids premature closure. A spirit of collaborative enquiry helps patients to reach their own conclusions about the role of alcohol in their troubles. This will be more convincing than a recitation of medical advice. People are more likely to believe what they hear themselves say than what others tell them. The interview is less likely to slip into confrontation if the doctor conveys recognition that, for the patient, drinking alcohol has been pleasurable. Therefore the assessment should not proceed in a series of closed questions, such as: ‘Do you drink more than you intend to?' ‘Does alcohol make you depressed?' Instead, ask open-ended questions: ‘Tell me about your pattern of drinking. What are the good aspects...and what are the disadvantages?' ‘How does alcohol fit in with these periods of hopelessness you describe?' The patient may want it understood that at times alcohol has dulled pain. Only then will there be a concession that overall it has worsened mood.

A comment such as ‘I'm just a heavy social drinker, not an alcoholic's is not a gauntlet to be seized—an argument about definitions will distract from the work of clarifying and planning how to deal with the current problems. Instead, a response such as ‘I gather you don't like labels's may reveal pertinent fears and prejudices (e.g. that alcoholics get locked up in hospital or are failures).

Denial permits dismissal of unpleasant or unwanted facts and feelings. It hurts to admit that you have lost your family's respect, or that you will have to give up alchol which you enjoy. Alcohol problems still carry disgrace. In Islamic cultures, where alcohol is forbidden, denial from shame may be deepened by fear of punishment from the authorities.

Explain symptoms

Help the patient to understand withdrawal symptoms and how they can abort attempts to reduce consumption. Patients frequently attribute withdrawal symptoms to other causes; for example, waking at 4 a.m. with sweats and anxiety may be attributed to worry, and trembling hands in the morning to stress.


If the partner, a close friend, or a relative is present from the start, the salient points usually emerge more rapidly. However, the patient should also be seen alone because matters to do with the police, an employer, the bank, or a lover may still be unknown to the partner. Relatives should hear the exchange between doctor and patient, otherwise the version they hear later from the patient may be diluted—‘The doctor says I'm not an alcoholic'. This can leave relatives even angrier than before, convinced that no-one understands their distress and that the drinker has once again deceived the doctor.


The use of a breathalyser or saliva test to measure blood alcohol concentration puts alcohol consumption firmly into the objective arena. Use the test before the individual starts to detail recent drinking—there is nothing to be gained from showing that the patient sometimes minimizes the drinking.

Physical signs may be helpful. Heavy drinking may cause excessive capillarization in the conjunctivae or in the skin of the nose and cheeks. The liver may be enlarged. Look for tremor in the outstretched tongue, which is less commonly concealed (or exaggerated) than tremor in the fingers. Tachycardia is another useful sign of withdrawal. In a hyperaroused fearful patient, who has already been without a drink for 24 h, a pulse of over 110 beats/min may presage delirium tremens.

Clinicians vary in how structured an assessment they prefer, but at some point in the first one or two interviews the following should be noted: drinking patterns, history of withdrawal symptoms, previous attempts to stop drinking, use of drugs (prescribed and not prescribed), physical complications including head injuries, police or court involvement (past and current), dwelling arrangements, problems at home, trouble at work, specifying whether the employer has referred to alcohol and/or started disciplinary action, psychiatric illness, family history, previous treatments, and experience of AA.

Medical assistance for withdrawal

Medical assistance to reduce the short-term discomfort of withdrawal can be the beginning of restructuring of thoughts and lifestyle towards long-term abstinence.

If dependence is severe, especially in an unplanned situation where a very heavy drinker is suddenly deprived of alcohol because of an accident, illness, or police arrest, care must be taken to prevent the life-threatening complications of convulsions or delirium. Anticipation is the key.

When dependence is less marked, withdrawal symptoms are mild and the person can stop drinking by gradual reduction, encouraged by the physician or a friend.

When the patient's aim is ‘controlled drinking' (see below), this may also entail an initial stage of withdrawal, as the final goal is more likely to be achieved after abstinence for 2 or 3 months.

The setting

Controlled studies have shown that outpatient withdrawal is safe and effective for mild and moderately dependent alcoholics. (4,5) Advice for patients withdrawing at home is given in Bullet list 1 below. Hayashida et al. (4) randomly allocated 164 mild to moderately affected patients to either inpatient or outpatient detoxification. Completion was successful in 95 per cent of the former and 72 per cent of the latter; inpatient care cost eight times more than outpatient care.

Admission to hospital is indicated when the home social milieu is inimical to abstinence, or when there is a history of withdrawal convulsions or delirium; it is urgent when there are any signs of Wernicke's encephalopathy.


A benzodiazepine (see systematic review by Mayo-Smith et al. (6)) is prescribed for two reasons: first, to reduce the risk of severe withdrawal symptoms with delirium or convulsions (indicated if recent consumption has been more than 15 units/day for more than 10 days); second, to assist the individual whose wish to abstain or reduce drinking is overcome by longing for alcohol (craving), shaking, anxiety, insomnia, or nausea and vomiting.

A typical outpatient regimen would be chlordiazepoxide 20 to 30 mg four times daily, reducing to zero over 5 days, with the larger doses given at night (Table 1). Medication is issued on the understanding that the patient does not also take alcohol. If there is any doubt that this instruction will be followed, medication is issued daily and a check made (ideally by breath or saliva tests) that drinking has not been resumed. Chlordiazepoxide is preferred to diazepam for outpatient use because it has a lower street value and is therefore less likely to be sold on. When managing severe withdrawal symptoms with marked agitation and tremor, or incipient delirium, diazepam (starting at 10 mg four times daily) is preferred because it has a more rapid action and can be given parenterally. A benzodiazepine with one metabolite only and a shorter half-life (e.g. oxazepam, lorazepam) is preferred if liver function is significantly impaired (i.e. there is jaundice, ascites, oedema, low serum albumin, or raised serum bilirubin).

Bullet list 1: Advice to patient on withdrawing from alcohol at home

  • If you have been chemically dependent on alcohol, stopping drinking causes you to get tense, edgy, perhaps shaky or sweaty, and unable to sleep. There can be vomiting or diarrhoea. This ‘rebound' of the nervous system can be severe. Medication controls the symptoms while the body adjusts to being without alcohol. This usually takes 3 to 7 days from the time of your last alcoholic drink. If you did not take medication, the symptoms would be worst in the first 48 h, and then gradually disappear. This is why the dose starts high and then reduces.
  • You have agreed not to drink alcohol. You may get thirsty. Drink fruit juices and water but do not overdo it. You do not have to ‘flush' alcohol out of the body. More than 3 litres of fluid could be too much. Do not drink more than three cups of coffee or five cups of tea. These contain caffeine which disturbs sleep and causes nervousness.
  • Aim to avoid stress The important task is not to give in to the urge to take alcohol. Help yourself relax by going for a walk, listening to music, or taking a bath.
  • Sleep. You may find that even with the capsules, or as they are reduced, your sleep is disturbed. You need not worry about this—lack of sleep does not seriously harm you, starting to drink again does. Your sleep pattern will return to normal in a month or so. It is better not to take sleeping pills so that your natural sleep rhythm returns. Try going to bed later. Take a bedtime snack or milky drink. The capsules may make you drowsy so you must not drive or operate machinery. If you become drowsy, miss out a dose.
  • Meals. Even when you are not hungry, try to eat something. Your appetite will return.

For inpatients, a benzodiazepine such as diazepam 10 mg may be given every hour until symptoms are controlled (symptom-triggered dosing). This procedure leads to lower total prescription of benzodiazepine, less oversedation, and quicker discharge from hospital. (7)

If the patient is vomiting, give metoclopropamide 10 mg intramuscularly 30 min before the first benzodiazepine tablet and/or perhaps choose a benzodiazepine that can be administered parenterally; lorazepam 1 mg is absorbed adequately from the intramuscular site, or diazepam 10 mg can be given intravenously (or rectally).

Treating convulsions

With the aim of preventing further convulsions, the patient who has just had a fit or is in a fit is given 10 mg diazepam. Consider giving 15 to 20 mg in a patient who has been taking benzodiazepines regularly prior to this event, or is much above average weight. It is illogical to commence an anticonvulsant which may take 2 to 3 days to reach a therapeutic serum level. Rather, increase the dose of the benzodiazepines. A convulsion may presage delirium.

Preventing convulsions

Deaths have occurred in hospital, prison, and police cells from repeated alcohol withdrawal fits. When withdrawal is planned in patients with a history of fits of any cause the risk can be reduced by commencing phenytoin (300 mg daily) 4 days before the cessation of drinking. In an acute situation, larger than normal doses of long-acting benzodiazepines are given in the first 36 h without waiting until the blood alcohol level has fallen to zero. The benzodiazepine should be started as soon as the blood alcohol level can be presumed to be falling, even though the patient still smells of alcohol or has a positive breath test, provided that he or she is sober enough to understand and co-operate with the procedure.

Treating delirium tremens

Increasing the dose of the benzodiazepine may be sufficient to control the agitation. If not, the slight epileptogenic effect of antipsychotic drugs should not deter their use, especially if delusions and hallucinations have developed, provided that anticonvulsant protection by a benzodiazepine is in place. Parenteral droperidol plus parenteral lorazepam is usually effective. When a patient's behaviour is uncontrolled or dangerous, transfer to a secure unit may be needed. Authoritative calm nursing reduces the risk of aggression. Hospitals should have an emergency team of sufficient personnel to manage disturbed patients.

Preventing delirium tremens

If confusion and hallucinations develop, this usually occurs 48 to 72 h after the last drink. Sufficient benzodiazepine, given early in the withdrawal, reduces the risk, as does sensitive nursing in a quiet evenly lit environment. Explaining symptoms and orientating the patient reduces anxiety, paranoia, and confusion.

Vitamin therapy

It is reasonable to prescribe thiamine 50 mg orally three times a day for 2 to 3 weeks, as thiamine stores may be depleted because of poor diet and alcohol-impaired gut absorption. Wernicke–Korsakoff syndrome is life-threatening and steps must be taken to avoid it developing. (8) The malnourished patient, or the patient who shows any sign of Wernicke's encephalopathy (confusion, ataxia, ophthalmoplegia, nystagmus—do not wait for the ‘triad' of symptoms), must be given immediate parenteral B vitamins. Anaphylactic shock was a very rare complication of some older preparations. It is less likely with intramuscular than intravenous injection; infusion saline drip, when practicable, is probably preferable to slow bolus injection.

Interventions to reduce relapse

The evidence

With appropriate help, withdrawing from alcohol is not the dependent drinker's main difficulty. The main difficulty is avoiding later relapse into further problematic drinking or dependence.

Until recently no treatments had been tested in a randomized controlled trial. Therapists explored with patients possible personality or psychological causes of their excessive drinking—trying to find out ‘why?'. However, there was no evidence that this reduced relapse. Indeed, it could have had an adverse effect by creating transference problems which triggered drinking and by reinforcing the drinkers's perception of having a need to drink. (9) Similarly, non-directive counselling could act as a confessional, with a sense of absolution allowing further drinking.

Miller et al. (10,11) reviewed 302 controlled trials. A score for each therapy was calculated based on the number of positive and of negative reports and their methodological quality. Some of the apparently most effective treatments, such as systematically helping people anticipate and cope with high risk situations (‘relapse prevention' (12,13)) and motivational enhancement, (14) had been tested partly in less severe groups of patients. Other effective treatment, such as community reinforcement, (15) social skills training, cognitive therapy, behaviour contracting, and behavioural marital therapy, had been tested in patients showing a wider range of severity. (10) Miller's review (10) of disulfiram studies did not separate those where compliance was assured by supervision (where efficacy was demonstrated) from those where supervision was not in place (where the drug was ineffective). (16)

Abstinence or ‘controlled drinking'?

Harmful or hazardous use of alcohol without severe dependence can sometimes revert to risk-free drinking. Patients with social supports (family and job) and without impulsive personalities and many social problems are most likely to succeed. For others, including most of those dependent on alcohol, the goal of abstinence is better. In patients attending specialized outpatient clinics, the proportion who can sustain problem-free drinking for at least 1 year is small—5 per cent is a typical finding. (17,18 and 19) A randomized trial comparing the goals of controlled drinking and abstinence did not favour controlled drinking. (20) However, for patients without established dependence, reduction programmes (whether or not towards abstinence) using FRAMES proved to be more effective than no intervention. (21,22 and 23)

Interventions in primary care are discussed here: Services for alcohol use disorders.

If controlled drinking is the agreed goal, the patient and physician collaborate to monitor the amount and pattern of the drinking as follows.

  1. Limit number of days of drinking and number of drinks on any occasion.
  2. Slow the rate of drinking, and/or reduce alcoholic strength of drinks.
  3. Develop assertiveness skills for refusing drinks.
  4. Design reward system when goals are achieved.
  5. Develop awareness of triggers to overdrinking.
  6. Practise other ways of coping with triggers.
  7. Record pattern and amount of drinking, for example in a diary.
  8. Physician and patient monitor g-glutamyl transferase blood test results.

Maintaining motivation and compliance

Enhancing motivation has a place not only at onset, but throughout the clinical contact. (24) Treatment aimed only at enhancing motivation was for most measures equal to cognitive-behavioural therapy, and intensive intervention aimed at linking patients with AA. (25) Randomized controlled studies have shown the advantage of motivational interviewing over traditional supportive therapy. (10,26) The style of the opening interview using motivational interviewing techniques has already been discussed. The patient is encouraged not to forget the harm that drinking caused and the benefits of abstinence, but the losses and problems of being sober are not denied. Strategies for maintaining abstinence emerge from collaborative dialogue, and are owned by patients rather than offered as advice from the clinician. If medication is part of the treatment plan, unwanted effects are actively enquired into, and are recognized and not dismissed, and remedies are sought. Any discrepancies that patients reveal between their present view of themselves and how they would like to be, or between what patients say they believe and how they actually behave, are used as a fulcrum for shifting attitudes and testing alternative strategies. These techniques were elaborated by Miller and Rollnick (3) and enshrined as motivational enhancement therapy (27) by Project MATCH (see below).

Helping motivation: the social matrix

It is said that the only successful way to change your drinking is to do it for yourself. Nevertheless, research and experience shows that those dependent on alcohol can start on the road to recovery when their reason is pressure from outside. This may be from the court which is seeking evidence, before deciding on sentence, that offenders have taken steps to alter harmful drinking patterns. The driving licence may have been withdrawn following a drink–drive offence and evidence that drinking is under control is required before its return. Perhaps the partner is ready to take a firm line, even to demand a separation or divorce, or the employer has given a warning.

Friends, partners, colleagues at work, and even employers sometimes adopt an approach that they believe to be motivating but which has the opposite effect and enables the drinker to continue drinking. They may cover up, gloss over, make excuses, or even start blaming themselves for what is going wrong. This cushions drinkers from experiencing the harmful consequences of their drinking or allows them to believe that alcohol is not the chief problem, despite evidence that alcohol is in fact the critical common factor in their downward spiral.

A physician can help the parties improve communication so that important messages are not lost. If the message from the employer or partner, or even the children, is clear and positive, it can have a powerful motivating effect: ‘We value our relationship with you. But the way you are drinking is harming that relationship and we will not tolerate it'.

Some physicians are overcautious about confidentiality in this situation. If a doctor is asked by a partner or an employer to comment on the patient's condition, he or she may or may not have permission, or feel it appropriate, to do so. But doctors can usefully help partners or employers clarify for themselves what they want, and then encourage a clear and firm, but positive, message.

Sometimes doctors unwittingly collude in a cover-up. The smokescreen that can be set up by a drinker who is severely dependent and ambivalent about change can be hard to penetrate: ‘It' depression, doctor's; ‘It' stress at work's; ‘If only my wife was more understanding/my sleep was not so disturbed/I didn't get these memory blanks which I think are some kind of stroke's. The doctor may need to wait for that medical moment, perhaps a crisis, to help such an individual. Or, if the doctor has patience, the drip, drip of non-judgemental evidence, and perhaps some social pressure, may bring about the necessary change in the patient's understanding and thus the perceived motivational pay-offs. Understanding may lead to action. However, that action may not be sustained and the process of helping understanding may need to be repeated many times. (28) 

There are few randomized controlled studies allocating patients to different intensities of external motivation. However, the evidence is that alcoholics coerced into treatment do no worse than those attending voluntarily. (29)

Cognitive-behavioural therapies

When incentives are powerful, many newly abstinent patients are able to abstain for short periods. Others lack the skills to cope with the triggers to drinking even when their motivation to abstain has been strong. Cognitive-behavioural therapies seem to improve the coping skills of these patients. If the triggers are in social situations, assertiveness or conversation skills training can help. If the trigger is related to life problems, cognitive therapy may be effective. Other patients are helped by learning to handle frustration and criticism without anger, and to express anger instead of harbouring it. Treatment can be in groups, where the opportunity to discuss these topics with others who have similar problems is appreciated. Groups also enable learning through role playing and by modelling on others.

Relapse prevention therapy

Much of the above is also part of ‘relapse prevention therapy'. (12,13) In this therapy patients are shown how to analyse and modify the causes of relapse. They are helped to identify ‘seemingly irrelevant decisions', in which a sequence that ended in drinking began with apparently unconnected actions. For example, the person chooses to stop at the supermarket to buy fruit, knowing that there is plenty at home already, and comes out with a bottle of wine. The patient is encouraged to identify seemingly irrelevant decisions, and, by recording thoughts and feelings during the days between therapy sessions, to map out risky emotional or environmental situations and to prepare avoidance and coping tactics. By logging success they can build up their sense of mastery.

Relapse may be preceded by a feeling of deserving a drink or of being deprived. Helping the patient to have more satisfaction in other areas of life helps combat these feelings.

Patients are educated about the abstinence violation effect. The suggestion is that a patient who made a strong commitment to abstain, but has taken one drink, is overcome by a sense of failure and counters this by thinking, for example, ‘This shows I'm just a drunk after all!'s—which leads to further consumption. The abstinence violation effect predicts that relapse after taking a single drink would be most frequent in those with the strongest resolution to abstain. However, research has found that greater commitment is associated with less risk of relapse after a single lapse. (30) Nevertheless, it seems helpful to prepare patients for a lapse, and to avoid catastrophic thinking.

Cue exposure

The smell or sight of alcoholic drinks can be a powerful stimulus to drinking. Initial studies (31) have shown that ‘deconditioning' by exposing inpatients to the sight and smell of their preferred drinks in a laboratory setting, without drinking, was associated in the coming 6 months with a longer period without a relapse. The approach is challenged by those who recognize overconfidence as a precursor of relapse. Although patients should not expect that to be in a bar will inevitably result in heavy drinking, they should not court danger. Bars are places where people go to drink. Cue exposure studies need to be replicated.

Couples should decide together whether to have alcohol in the house. However, patients should not be encouraged to ‘test themselves'.

Alcoholics Anonymous

There are many ingredients in the healing process of AA. Newcomers are helped to identify with others as members tell their stories. They see that it is possible to be frank about past errors and the hurt caused to others through the drinking. Telling their own story helps the members not to forget the harm that accrued from drinking. This reduces complacency, which is one of the most common precursors of relapse.

Alcoholism is viewed by AA as a physical, psychological, and spiritual illness, which can be arrested (by avoiding another drink) but cannot be cured. The meetings offer a new social network. Emotional openness is encouraged. Members learn to express warmth, and to accept that they and others have failings. The AA advice on coping with emotions and relationship difficulties has much in common with cognitive-behavioural therapy and relapse prevention therapy. The method has some attractively simple concepts (‘Just don't pick up that first drink's; ‘HALT'—being alert to four of the most common triggers to relapse, i.e. hunger, anger, loneliness, tiredness). There is a deeper aspect which is to replace preoccupation with self by handing over to the group process, or to a ‘Higher Power'. (32)

Accepting that you are ‘powerless' to control your drinking is the ‘first step' in AA. This entails ceasing the struggle and letting the ‘Higher Power' take over. Members vary in their interpretation of the ‘Higher Power', and avowed atheists should not be deterred from sampling AA. Residential, outpatient, and day programmes which teach the AA approach are sometimes called 12-step programmes (Bullet list 2). One of their strengths is linking patients to the AA network.

A psychiatrist can introduce patients to AA through a contact member who will tell the patient how AA works, will not ask personal details, and will extend an invitation to a meeting. Doctors are welcome to attend ‘open' AA meetings to see how it works. A contact number is given in local telephone directories. AA does not work for everyone, but since it is difficult to predict who will be helped it is good practice to offer contact to all patients with impaired control of their drinking.

A warning, often based on personal experience, may be given at AA meetings about transferring dependence from alcohol to other drugs. This usually refers to use of barbiturates or benzodiazepines, or to the danger of relying on a medication instead of adjusting one's way of living. The use of prescribed medication is not formally disapproved of by AA.

Evidence of efficacy

Naturalistic non-randomized studies have shown that treatment programmes using the AA approach are associated with outcomes in drinking and overall functioning similar to those of programmes using the cognitive-behavioural approach. Patients in 12-step programmes improve on self-efficacy and coping skills scores much as patients treated by cognitive-behavioural therapy. (33) There are significant associations between AA attendance and positive outcome. (34)

Bullet list 2: The 12 steps of Alcoholics Anonymous

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

Only two randomized controlled studies of 12-step programmes have been conducted. One compared inpatient treatment (with fewer hours of psychotherapy than many such programmes) with a 12-step inpatient programme (with slightly more hours of therapy). There was a non-significant trend towards a greater total abstinence programme and less relapse in the 12-step programme. (35) In Project MATCH, patients were randomly allocated to cognitive-behavioural therapy, motivational enhancement therapy, or ‘12-step facilitation', which instructed patients in the tenets of AA, and assisted and encouraged them to attend AA meetings. The three treatments resulted in similar outcomes after 1 and 3 years. However, for those who had been relatively free of psychiatric problems at entry to the study, 12-step facilitation was associated with slightly better outcomes after 1 year. After 3 years the 12-step facilitation led to better outcome for patients who, at entry to the study, had family, social, or work environments bringing them into frequent contact with drinking. (36)

Help for the family

Working with the family

The family of someone with a drinking problem may suffer for years without recognition and can benefit from advice and understanding. They are a vital monitor of the patient's progress. Good family cohesion and low expressed emotion predict better outcome, even after controlling for the predictors of demographic variables and severity of alcohol dependence.

Life in the family becomes increasingly restricted. Finances dwindle. The children fear that the parent may be drunk, and so stop inviting friends to visit. They dread that arguments between mother and father will become violent. The drinker's behaviour becomes slovenly. He or she may wet the bed. Despite these hurts, the drinker may still make the family believe that they are the reason he or she drinks.

The invitation to a family member or partner to attend with the patient may be rejected if the drinker is the messenger, and the message is distorted to: ‘The doctor says you're part of the problem's. A direct letter or telephone call from the clinician requesting ‘your views on how I can assist' reduces the partner's fear of being burdened with extra guilt or responsibility.

The clinician can help reduce family behaviours, such as hostility or cover-up, that are damaging to the family and counterproductive for the drinker's recovery. Communication between the drinker and the spouse or children has often broken down. In many countries family groups, such as Al-Anon, provide help to families.

Behavioural marital therapy

When the patient is in a relationship, its quality can be motivating or demotivating. Reciprocal contracts are aimed at making the relationship more rewarding for each partner. Although abstinence is a prerequisite, specific agreements should not be contingent on the drinking; (37) otherwise, a relapse means that the partner ceases to work on the relationship. Another prerequisite might be that physical violence is excluded. Contracting could start thus: ‘Although you are responsible for not drinking, is there anything that your partner could do more of, or less of, that would help you stick to the plan?' Check that the requests are reasonable and available before the partner is asked to agree. The partner makes reciprocal requests and negotiation follows. Even requests for small changes can start the process.

The partners should give clear messages, owning their statements: ‘This is what I would like', ‘It makes me feel good if you...'. They will need to be reminded to state the positives and to practise being good listeners, giving non-verbal signals that they are listening, and not butting in with unsolicited good advice.

Violence in the partnership may require specific attention. If the drinker is intoxicated, the partner is advised to back off and avoid argument. Sometimes each partner is asked to sign an agreement that neither will threaten or hit the other. If they do, time-out in another room is agreed in advance to permit slow-breathing to aid calming down, or one of them will leave the house and go to a designated place for 36 to 72 h.


The partner ‘bringing up the past' can be a major irritant to the drinker. This can be reframed as the partner ‘helping the couple not repeat their past'. A partner who feels heard and understood is more ready to look at other ways of achieving these goals.


Behavioural marital therapy produces better outcomes of drinking and marital relations than individual counselling or similar control conditions. The superior effects last for 24 months after treatment. Outcome at 1 year is better if sessions of behavioural marital therapy continue after the end of treatment to reinforce what has been learnt and rehearse relapse prevention plans. (38) This has been reviewed by Miller et al. (10)

Deterrent medication

Disulfiram (Antabuse)

If taken in a sufficient dose for at least the preceding 3 to 4 days, disulfiram causes an unpleasant reaction to develop 15 to 20 min after alcohol enters the body. The reaction is due to accumulation of acetaldehyde, the intermediate metabolite of ethanol. The reactions includes flushing, headache, pounding in the chest or head, tightness in breathing, nausea, and sometimes vomiting. Hypotension can occur and is potentially dangerous. (Calcium carbimide has a similar action but is no longer available.) The disulfiram–ethanol reaction varies in intensity. It is recognized practice to increase the dose of disulfiram up to 400 mg daily if the patient has tested the alcohol reaction and it has not been severe enough to act as a deterrent.

Disulfiram is an aid, not a cure. The individual can become used to life without alcohol. This allows time for confidence to recover—personally, in the family, and at work. Patients may object that it is weakness to take a deterrent, and they prefer to show that they can use will power. Explain to the patient that will power is not always there when most needed. With disulfiram a decision to drink or not still has to be made, but only once a day or three times a week.

Unwanted effects which occur even when no alcohol is taken include drowsiness, bad breath, and headache. These make the drug unacceptable to some patients. Concerns that disulfiram can harm the liver are based on a few case reports (the risk is about 1 in 30 000 patient-years). It appears to be a hypersensitivity reaction, and if it is to occur it is likely to be in the first month. Overall, disulfiram is associated with improved liver function tests compared with control groups, presumably owing to reduction of drinking. (39) Peripheral neuropathy (almost always reversible) has been reported following several months at doses of over 250 mg. There are a few reports of psychosis induced by disulfiram, and psychotic illness has been a formal contraindication in the licensing in some countries. The risk is so low and the need to help schizophrenic patients with alcohol problems is sometimes so great that in other countries this contraindication has been changed to a ‘caution'. There are many documented cases where improvement has occurred in psychotic patients while taking disulfiram, and in a dose of up to 250 mg daily there are no problems from unwanted actions or interactions with medication for the psychiatric illness. (40)


Disulfiram will only aid recovery if it is taken regularly in a sufficient dose to deter. In randomized controlled studies showing efficacy, a supervisor aided compliance. All reported controlled studies which enhanced compliance in this way showed efficacy. (16) In some of these studies there was a degree of coercion; for example, if the patient ceased taking the disulfiram the partner might withdraw from some agreed item, or disciplinary action at work might be reinstated.

Mode of use

Before prescribing, a physical examination and baseline liver function tests are performed. The patient is encouraged to ask the partner, a nurse or welfare officer at work or at the health centre, or a pharmacist to see that the disulfiram is taken. This can be daily, or three times a week, provided that the total weekly dose is sufficient. The product is in a dispersible form to be taken in water so that it can be seen to be swallowed.

There should be medical follow-up, but there is no consensus as to whether monitoring of liver function tests should be carried out. However, monthly follow-up is appropriate to check for signs of drinking and of liver disease. (39,41)

It is common to prescribe disulfiram for 6 months, but many patients ask to continue for longer and there may be slips when disulfiram is withdrawn, even after long periods of abstinence. Some patients keep a supply to use when they feel an increased risk of drinking, for example on a business trip or at a social event. Occasionally, when abstinence seems stable, a couple agree the opposite, namely that the drinker may have occasional breaks in the disulfiram regimen to permit drinking at a particular event or on a holiday if holidays have not been times of problematic drinking previously. Sometimes this is successful, but at others the wish to drink again regularly can be reawakened. (42) The taking of disulfiram may re-establish an employer's confidence, so that the patient may be reinstated.

Specific neurotransmitter antagonists

Two drugs, sometimes called ‘anticraving agents', have been shown in randomized controlled trials to have modest but useful efficacy. However, some methodological problems concerning the conduct and analysis of these studies have been noted. (43,44 and 45)

Acamprosate (calcium acetyl homotaurinate)

Acamprosate enhances g-aminobutyric acid (GABA) transmission and antagonizes glutamate transmission, probably by antagonizing N-methyl-D-aspartate receptors. It reduces drinking in alcohol-dependent animals, and reduces the reinstatement of drinking behaviour in animals re-exposed to alcohol after a period of abstinence. Animals do not seek out acamprosate as they do addictive substances, and it does not have mood-altering or drug-abuse potential in humans. (46) It has no deterrent or disulfiram-like effect.

Acamprosate is excreted unchanged in the kidney. It has few unwanted effects; diarrhoea and abdominal discomfort are the only ones reported in more than 10 per cent of patients (up to 20 per cent) and these are mild and transient. It does not exacerbate psychomotor impairment caused by alcohol. There are no known drug interactions.


Acamprosate has a dose-related effect of improving abstinence rates in recently detoxified patients. (47,48 and 49) Large randomized controlled studies of acamprosate (50,51) have shown an increase, compared with placebo, in the percentage remaining totally abstinent for 12 months from 10 to 25 per cent to 20 to 50 per cent, a doubling of the time to first relapse, and a halving of total alcohol consumed. There are no studies comparing the advantages of differing lengths of treatment. Systematic follow-up after the end of drug or placebo treatment shows no sudden relapse and no discontinuation symptoms in patients who have received acamprosate for up to 1 year. (51)

Several studies have shown that acamprosate reduces self-reported craving for alcohol; one of these failed to find an effect on drinking. (52) Some newly abstinent patients experience strong craving, but others experience very little.

Acamprosate has only been tested in patients who intend to abstain from alcohol. It has not been shown to assist patients aiming for controlled drinking.

Suggested mode of use

Acamprosate is indicated for patients who have typical severe alcohol dependence requiring medical assistance to withdraw. It is started 2 to 7 days after the last drink. Steady state pharmacokinetics are reached after 5 days. It is given in a dose of 1998 mg daily divided through the day. Patients who relapse while on acamprosate are advised to continue taking the medication and exert effort to limit the lapse. However, acamprosate is not normally continued in patients who relapse more than once despite regularly taking the drug. About 50 per cent or more of patients do not benefit from acamprosate. Those who appear to be benefiting from it should continue the drug for at least 6 months, and up to 1 year if there has been previous relapse in treatment.

Patients taking disulfiram randomly allocated to acamprosate seem to be more successful than patients taking disulfiram and placebo, (53) suggesting that acamprosate's potential to reduce craving in newly abstinent patients may help them to continue taking disulfiram. The reverse procedure, i.e. randomly allocating patients taking acamprosate to disulfiram or control, has not been carried out.


Naltrexone antagonizes endorphins which are released in one of ethanol's many acute actions on the limbic system. It has been suggested that this action contributes to loss of control. (54) Naltrexone reduces ethanol seeking in dependent animals. It does not exacerbate the psychomotor impairment caused by alcohol.


Two double-blind randomized controlled studies in detoxified outpatients showed that naltrexone 50 mg daily reduces the risk of relapse over a 3-month treatment period. (55,56) A subsequent study failed to find a significant treatment effect in the all-patient intention-to-treat analysis, but found a beneficial effect in a predefined subgroup of compliant patients (who had attended regularly for treatment and received 80 per cent of the prescribed study medication). (57) The effect size of naltrexone treatment in reducing the percentage of days drinking was 0.42 in one study and 0.60 in the other. (58) (For comparison, the mean effect size in meta-analyses of studies of fluoxetine in the treatment of depression is around 0.4.)

Follow-up has not indicated rapid relapse on cessation of these drugs. However, during a 6-month follow-up after 3 months of placebo-controlled treatment, patients relapsed gradually after naltrexone to a final rate similar to that seen in placebo-treated patients. (59)

Mode of action

Some patients who drink while taking naltrexone report that they feel less of the ethanol ‘high'. This could lead to less impulse to carry on drinking. (58,60) However, O'sMalley et al. (56) found that more patients reported total abstinence as well as a reduction of drinking overall. It is possible that the reduced craving for alcohol and the reduced likelihood of picking up the first drink occur because the strength of the previous triggers—emotional, cognitive, or environmental—is attenuated.

Reports in the 1970s that naltrexone might cause dysphoria seemed to be supported by statements from heroin addicts given naltrexone to help them abstain from opiates. However, laboratory studies and randomized controlled trials have not found evidence of dysphoria or loss of feelings of pleasure. (61)

Helping women with alcohol problems

It has been said that when a woman has an alcohol problem, there is a man in her life with a similar problem—usually her partner or her father. When the partner also drinks heavily, he should if possible be involved in treatment. Other partners may be overinvolved and have adopted a controlling role, especially if she has been unreliable with the children, money, the car, etc. The woman's resentment at this can fuel the drinking, and it may need months to help her to see how this has come about to help the partner stop checking her behaviour and to restore trust.

Low self-esteem is very common in such women, even in those who were confident before the drinking got out of hand. The partner, while remaining firm about the unacceptability of her drinking, may need help to be more caring and positive, to show interest in what concerns her, and to show appreciation.

Resentments towards family, employer, and partner are common and sometimes lead to depression. Women with alcohol dependence can be helped psychologically to abstain in several ways.

  • Help her to stop feeling taken for granted, and to know that she has a right to set limits on what others expect of her.
  • Although guilt may be proportional to what she has put her family through by her drinking, it may not help. It may prevent her from asking for the conditions at home or work that would make it easier for her to stop drinking.
  • Help her let go of anger.
  • Help her find ways of recharging her batteries by, for example, taking up new interests or exercise.
  • Talk with the partner, both alone and with her present. He may want to know that she acknowledges the strain on him. While accepting complete responsibility for her drinking, she can let him know what he can do to help her.
  • Self-help literature is available in many languages to help women improve self-confidence and self-assertion. (62)

Treatment of coexisting disorders

Affective disorder

Depression is common in patients who are dependent on alcohol. The drinking may have alienated friends, family, or employer, with resulting feelings of hopelessness, guilt, and lack of direction. Alcohol can reduce appetite, energy, and sexual drive. The drinker wakes in the small hours of the night feeling anxious owing to the rebound wakefulness of alcohol withdrawal. Those signs and symptoms suggesting depressive illness commonly clear with abstinence and help in tackling or tolerating personal problems and improving relationships.

Sometimes (more often in women than in men) a depressive episode precedes the alcohol dependence. Alcohol was taken in part as self-medication. Sometimes depressive symptoms continue despite abstinence. In these cases, antidepressants should be offered in the usual way. (63,64) Relapsing alcoholism, secondary to depressive illness, is an indication for long-term antidepressants. Lithium is not a treatment for alcohol dependence itself, but is effective if it is secondary to manic–depressive disorder.

Slightly higher blood levels of antidepressants may result when they are taken concurrently with disulfiram.

Depressed patients who have become dependent on alcohol are susceptible to the nausea and/or agitation which occur with many serotonin-enhancing agents. They may need an antiemetic such as domperidone or metoclopramide for a few days and a slightly longer time on the reducing benzodiazepine regimen given for alcohol withdrawal. Trials in alcohol-dependent patients of serotonergic drugs with greater 5-HT 1A and less 5-HT2 activity, such as mirtazepine, have not yet been reported.

Anxiety and panic disorder

Some patients have had panic attacks for years before discovering that alcohol can end or prevent them. Others have a first panic attack during alcohol withdrawal, but the attacks continue independently even during sustained abstinence. In either case, cognitive-behavioural therapy and/or medication are indicated.

Three studies suggest that the serotonin agonist buspirone can help reduce both drinking and anxiety. (65) Tricyclic antidepressants and selective serotonin-reuptake inhibitors (SSRIs) have been shown to be effective in randomized controlled trials of panic disorders, but alcohol dependence has been an exclusion criterion in these trials. Newly abstaining alcohol-dependent patients seem particularly susceptible to the unwanted effects of serotonergic medication (see above).

Some patients with long histories of alcohol dependence and severe panic disorders fail to respond to psychological or antidepressant treatments. For these patients the risk of complications from repeated prescribing of a long-acting benzodiazepine may be less than those from continued excessive drinking. If prescribed (and to do so is controversial), the benzodiazepine should be dispensed in limited amounts. The prescription should be conditional on abstinence from alcohol, which can be aided by disulfiram if necessary. ‘As-required' use (e.g. for travelling on public transport) helps to limit the development of tolerance, even though in theory it may perpetuate phobic beliefs. This method probably commits the patient to long-term use and an enduring risk of escalation.

Other comorbidities

Self-harm, sometimes in very severe forms such as self-immolation or disfigurement, may occur when personality disorder accompanies alcohol misuse. Such patients are sometimes helped by antidepressants, lithium, or depot antipsychotic drugs, in conjunction with other relapse prevention measures, and supervised living arrangements.

Attention-deficit hyperactivity disorder in adulthood is associated with alcohol dependence in 25 per cent of cases. There are no trials of methylphenidate in alcohol misuse.

Residential and inpatient treatment

It is debatable whether a period of inpatient treatment can improve the eventual outcome. Some studies have compared outcomes after patients have been randomly allocated to either inpatient or outpatient treatment. Usually no difference has been found. However, the interpretation of these results and their extrapolation to clinical reality has been debated. Finney et al.(66) concluded that the studies often lacked statistical power. Furthermore, the more seriously affected patients had sometimes been excluded before randomization.(66,67) While evidence that it is inpatient treatment rather than intensity of treatment which improves outcome is lacking (68,69) admission to hospital can provide valuable respite for the drinker and the family when life is severely disorganized because drinking is out of control. Perhaps such respite need not be offered in a relatively expensive medical environment. However, if the patient has become suicidal as difficulties increase or has developed serious medical complications, then hospital admission may be indicated, ideally to specialized facilities if available. Longer stays in hospital are not supported by research. For example, Trent(70) found no evidence of worse outcome when the United States Navy reduced the length of its inpatient alcoholism treatment programme from 6 to 4 weeks. The role of inpatient treatment is considered further in this article: Services for alcohol use disorders.

Matching patients to treatments

The problem

It is recognized that people with alcohol dependence present a range of problems, come from various backgrounds, and have different personality characteristics. Some have no accompanying emotional disturbance; others have a psychiatric disorder. The poor outcomes of treatment for alcohol dependence have been attributed to their use with unsuitable patient, and better matching of patients to treatments has been sought. A North American study of 1726 outpatients (Project MATCH) set out to test hypotheses about matching treatments to patients. Three treatments were studied, each established in previous randomized controlled trials as more effective than ‘supportive therapy': motivational enhancement therapy, cognitive-behavioural therapy, and instruction in the AA approach with encouragement to take part in AA meetings (‘12-step facilitation').

Few matching effects reached statistical significance. In patients recruited from outpatient clinics, those who scored high on anger at initial assessment averaged 85 per cent of abstinent days if they had been allocated to motivational enhancement therapy compared with 75 per cent if they had been allocated to 12-step facilitation or cognitive-behavioural therapy. (71) In the first year of follow-up, patients with initially less severe psychiatric symptoms had more abstinent days after the 12-step facilitation than after cognitive-behavioural therapy. Patients with critically high psychiatric severity did no better with cognitive-behavioural therapy. (26)

Another marker of who benefits most from AA emerged in the 3-year Project MATCH data. Patients who came from a social milieu where they mixed a lot with other drinkers owing to family, neighbourhood, or work influences did better if they had received 12-step facilitation than with either cognitive-behavioural therapy or motivational enhancement therapy.(37)

There are several reasons for the absence of evidence of other powerful predictors of treatment outcome in the Project MATCH data. Perhaps the key behaviour—not taking the first drink—can be arrived at in different ways.

Guiding principles

Research into alcoholism spanning 50 years has shown that the attitudes of the agency and the therapist influence patients's outcome, as they may do for many illnesses. Showing respect, enhancing dignity, conveying accurate empathy, adopting objective and not moral criteria, involving the family, and reducing hurdles to seeking help have been shown to improve compliance, and often outcome, for alcohol dependence. The therapeutic alliance is a strong predictor of outcome in the treatment of alcohol dependence. (72) Agencies must set limits on drunken behaviour at the clinic and telephone calls when intoxicated. When relapse is recurrent, resumption of treatment can be made conditional on complying with a new initiative, such as supervision of medication. (73)

Some clinical situations

Morbid jealousy

This is discussed in this article: Persistent delusional symptoms and disorders.

The homeless alcohol-dependent person

It is difficult to conduct randomized controlled studies with adequate follow-up to test the efficacy of interventions to reduce drinking and improve social conditions for the homeless, and few answers have been found. A brief hospital admission to ‘dry out' and assessment for transfer to residential care may result in transient improvement in physical health and is more humane than prison. However, supporting evidence is lacking. A structured intensive outpatient intervention, called the ‘community reinforcement approach' has been shown in a North American study to reduce drinking (corroborated by improvement in serum g-glutamyl transferase) and increase the number of clients at work and in satisfactory housing. (74) The community reinforcement approach combined an offer of free housing, a place at a ‘job club' to assist with finding employment, training in problem-solving skills, communication, goal-setting, refusal of drinks, and independent living. Patients had access to an alcohol-free social club. The housing offer was contingent on sobriety and some evidence of saving money. Continuation in the housing was contingent on sobriety checked by breathalyser. Disulfiram had been shown to improve the effects of the community reinforcement approach. (15,75)

Young people

There is a dearth of evaluation of programmes to help young people with alcohol problems. AA groups may have teenage members. When education or employment are in jeopardy, young people may accept disulfiram, supervised perhaps by the family. However, without the support of a non-drinking peer group (which they would have in AA), most young people will try again and again to resume ‘social drinking'. Job or marriage commitments sometimes alter the pay-off matrix sufficiently for recovery to be sustained. Otherwise, it may not be until age 30 that the young person is sufficiently convinced that he or she cannot control drinking and takes serious steps to seek help.

Employment referrals

It is common for individuals to seek help when their drinking has put their job in jeopardy. Having a job helps recovery, and for the person to lose employment while paying only lip-service to treatment is common and disheartening for all. The psychiatrist should find out whether disciplinary procedures are in motion or threatened. It can be helpful if the psychiatrist and the patient are told this directly by the employer. If the consultation is part of an undertaking under a company ‘alcohol and drugs policy', the patient may have given permission for the psychiatrist to answer the employer's request to know whether he or she is attending and following advice.

Patients who are on the point of dismissal may offer to take disulfiram supervised in the company's occupational health or welfare department. This can bring about recovery and employment for as long as the threat of dismissal remains, and sometimes afterwards. (76)

The liver transplant candidate

Some transplant centres require a demonstration of months of abstinence, to show commitment, before offering transplant to a patient with alcoholic liver disorder. Other centres have no such restrictions. From 6 to 80 per cent of transplant recipients, varying between centres, have recommenced drinking and exceeded safe limits by the end of the first year. Their eventual outcome in terms of quality of life and psychiatric health is no worse than in other transplant patients, and evidence obtained to date suggests that demanding lengthy preoperative abstinence does not improve outcome. (77)

Physicians as patients

Doctors have a raised rate of alcohol dependence. Their outcome, once in treatment, tends to be better than average, if they can return to their practice. This is probably partly due to the requirement by the licensing body that the ‘impaired physician' accept monitoring by an independent specialist to corroborate that he or she is following advice and continuing to progress. (78)

Doctors' reluctance to accept help for their illnesses, and their tendency to treat themselves, is well known and is especially true for substance misuse. Initial denial often means that problems escalate until there are disciplinary or court proceedings, and attempts to treat their own alcohol dependence may result in dependence on other substances.

The alcoholic doctor should be treated in the same way as a lay person. The partner should be invited to be involved. If possible, information should be obtained from the employer or from a colleague about the nature of any problems at work or any disciplinary action, actual or threatened.

In some countries there are support groups for recovering doctors and dentists who meet together and are ready to offer advice and encouragement to individuals and their families.


Systematic follow-up has been shown to improve outcome. (17,79) Early detection of relapse is important, and is aided by regular contact with the family or the workplace, a breathalyser test at interview, and tests for blood markers of drinking (g-glutamyl transferase or carbohydrate-deficient transferrin). (80) Objective markers are required when a patient requests a report for a court, the driving licence authority, or an employer. Individualized outcome measures can be used, defined by the patient's own goals. This is particularly relevant when the patient's goal is to avoid further health or family problems, to avoid reoffending, or to stay in employment, rather than abstinence per se. (81) 


1. Vaillant, G. (1997). The natural history of alcoholism revisited. Harvard University Press, Boston, MA.

2. O'sBrien, C.P. and McLellan, A.T. (1996). Myths about the treatment of addiction. Lancet, 347, 237–40.

3. Miller, W.R. and Rollnick, N. (1992). Motivational interviewing. Guilford Press, New York.

4. Hayashida, M., Alterman, A.I., McLellan, T., et al. (1989). Comparative effectiveness and costs of inpatient and outpatient detoxification with mild-moderate alcohol withdrawal syndrome. New England Journal of Medicine, 320, 358–65.

5. Bennie, C. (1998). A comparison of home detoxification and minimal intervention strategies for problem drinkers. Alcohol and Alcoholism, 33, 157–63.

6. Mayo-Smith, M.F., for the American Society for Addiction Medicine Working Group (1997). Pharmacological management of alcohol withdrawal: a meta-analysis and evidence-based practice guideline. Journal of American Medical Association, 278, 144–61.

7. Foy, A., March, S., and Drinkwater, V. (1988). Use of an objective clinical scale in the assessment and management of alcohol withdrawal in a large general hospital. Alcoholism: Clinical and Experimental Research, 12, 360–4.

8. Cook, C.C. and Thomson, A.D.T. (1997). B-complex vitamins in the prophylaxis and treatment of Wernicke–Korsakoff syndrome British Journal of Hospital Medicine, 57, 461–5.

9. Vaillant, G.E. (1991). An alternative to psychotherapy. In The international handbook of addiction behaviour (ed. I.B. Glass), pp. 236–9. Routledge, London.

10. Miller, W.R., Brown, J.M., Simpson, T.L., et al. (1995). What works? A methodological analysis of the alcoholism treatment outcome literature. In Handbook of alcoholism treatment approaches: effective alternatives (2nd edn) (ed. R.K. Hester and W.R. Miller), pp. 12–44. Allyn and Bacon, Boston, MA.

11. Miller, W.R., Andrews, N.R., Wilbourne, P., and Bennet, M.E. (1998). A wealth of alternatives: effective treatments for alcohol problems. In Treating addictive behaviours (2nd edn) (ed. W.R. Miller and N. Heather), pp. 203–16. Plenum Press, New York.

12. Dimeff, L.A. and Marlatt, G.A. (1995). Relapse prevention. In Handbook of alcoholism treatment approaches: effective alternatives (2nd edn) (ed. R.K. Hester and W.R. Miller), pp. 176–94. Allyn and Bacon, Boston, MA.

13. Marlatt, G.A. and Gordon, J.R. (1985). Relapse prevention. Guilford Press, New York.

14. Miller, W.R., Benfield, R.G., and Tonnegan, J.S. (1993). Enhancing motivation for change in problem drinkers: a comparative outcome study of three controlled drinking therapies. Journal of Consulting and Clinical Psychology, 61, 455–61.

15. Azrin, N.H. (1976). Improvements in the community reinforcement approach to alcoholism. Behavioural Research and Therapy, 14, 339–48.

16. Hughes, J.C. and Cook, C. (1997). The efficacy of disulfiram—a review of outcome studies. Addiction, 92, 381–96.

17. Chick, J., Connaughton, J., Ritson, B., Stewart, A., and Chick, J.A. (1988). Advice versus extended treatment for alcoholism: a controlled study. British Journal of Addiction, 83, 159–70.

18. Helzer, J.E., Robins, L.N., Taylor, J.R., et al. (1985). The extent of long-term moderate drinking among alcoholics discharged from medical and psychiatric treatment facilities. New England Journal of Medicine, 312, 1678–82.

19. Vaillant, G.E. (1996). A long-term follow-up of male alcohol abuse. Archives of General Psychiatry, 53, 243–9.

20. Rychtarik, R.G., Foy, D.W., Scott, T., Lokey, L., and Prue, D.M. (1987). Five year follow-up of broad spectrum behavioral treatment for alcoholism: effects of training controlled drinking skills. Journal of Consulting and Clinical Psychology, 55, 106–8.

21. Sanchez-Craig, M., Leigh, G., Spivak, K., and Davila, R. (1989). Superior outcome of females over males after brief intervention for the reduction of heavy drinking. British Journal of Addiction, 84, 395–404.

22. WHO Brief Intervention Group (1996). A cross-national trial of brief intervention with heavy drinkers. American Journal of Public Health, 86, 948–55.

23. Bien, T.H., Miller, W.R., and Tonigan, J.S. (1993). Brief interventions for alcohol problems: a review. Addiction, 88, 315–36.

24. Miller, W.R. (1998). Enhancing motivation for change. In Treating addictive behaviours (2nd edn) (ed. W.R. Miller and N. Heather), pp. 121–32. Plenum Press, New York.

25. Project MATCH Research Group (1997). Matching alcoholism treatments to client heterogeneity: Project MATCH posttreatment drinking outcomes. Journal of Studies on Alcohol, 58, 7–29.

26. Bien, T.H., Miller, W.R., and Boroughs, J.M. (1993). Motivational interviewing with alcoholic out-patients. Behavioural and Cognitive Psychotherapy, 21, 347–56.

27. Miller, W.R., Zweben, A., DiClementi, C.C., and Rychtaril, R.G. (1992). Motivational enhancement therapy manual: a clinical research guide for therapists treating individuals with alcohol abuse and dependence. NIAAA Project MATCH Monograph Series, Vol. 2. DHHS Publication No. (ADM) 92-1894. US Government Printing Office, Washington, DC.

28. Prochaska, J. and Di Clemente, C. (1992). Stages of change in the modification of problem behaviors. In Progress in behavior modification, Vol. 28 (ed. M. Hersen, R. Eisler, and P. Miller), pp. 183–218. Sycamore Publishing, Sycamore, IL.

29. Chick, J. (1998). Treatment of alcoholic violent offenders—ethics and efficacy. Alcohol and Alcoholism, 33, 20–5.

30. Hall, S.M., Havassy, B.E., and Wasserman, D.A. (1990). Commitment to abstinence and acute stress in relapse to alcohol, opiates and nicotine. Journal of Consulting and Clinical Psychology, 58, 175–81.

31. Drummond, C. and Glautier, S. (1994). A controlled trial of cue exposure treatment in alcohol dependence. 62, 809–17.

32. McCrady, B. (1994). Alcoholics Anonymous and behaviour therapy: can habits be treated as diseases? Can diseases be treated as habits? Journal of Consulting and Clinical Psychology, 62, 1159–66.

33. Oiumette, P.C., Finney, J.W., and Moos, R.H. (1997). Twelve-step and cognitive–behavioural treatment for substance abuse: a comparison of treatment effectiveness. Journal of Consulting and Clinical Psychology, 65, 230–40.

34. Emrick, C. (1987). Alcoholics Anonymous: affiliation processes and effectiveness as treatment. Alcoholism: Clinical and Experimental Research, 11, 416–23.

35. Keso, L. and Salaspuro, M. (1990). In-patient treatment of employed alcoholics: a randomised clinical trial of Hazelden-type and traditional treatment. Alcoholism: Clinical and Experimental Research, 14, 584–9.

36. Longabough, R., Wirtz, P.W., Zweben, A., and Stout, R.L. (1998). Network support for drinking: Alcoholics Anonymous and long-term matching effects. Addiction, 93, 1313–34.

37. O'sFarrell, T.J. (1995). Marital and family therapy. In Handbook of alcoholism treatment approaches (ed. R.K. Hester and W.R. Miller), pp. 195–220. Allyn and Bacon, Boston, MA.

38. O'sFarrell, T.J., Choquette, K.A., and Cutter, H.S.G. (1998). Couples relapse prevention sessions after behavioural marital therapy for male alcoholics: outcomes during the three years after starting treatment Journal of Studies on Alcohol, 59, 357–70.

39. Chick, J. (1998). Safety aspects of disulfiram in the treatment of alcohol dependence. Drug Safety, 20, 427–35.

40. Larson, E.W., Olincy, A., Rummans, T.A., and Morse, R. (1992). Disulfiram treatment of patients with both alcohol dependence and other psychiatric disorders: a review. Alcoholism: Clinical and Experimental Research, 16, 125–30.

41. Wright, C. and Moore, R.D. (1990). Disulfiram treatment of alcoholism. American Journal of Medicine, 88, 647–55.

42. Duckert, F. and Johnsen, J.(1987). Behavioural use of disulfiram in the treatment of problem drinking. International Journal of the Addictions, 22, 445–54.

43. Moncrieff, J. and Drummond, C.D. (1997). New drug treatments for alcohol problems: a critical appraisal. Addiction, 92, 939–48.

44. Moncrieff, J. and Drummond, C.D. (1998). The quality of alcohol treatment research: an examination of influential controlled trials and development of a quality rating system. Addiction, 93, 811–23.

45. Addiction (1997). Comments on Moncrieff and Drummond's ‘New drug treatments for alcohol problems: a critical appraisal' (seven comments). Addiction, 92, 949–65.

46. Littleton, J. (1995). Acamprosate in alcohol dependence: how does it work? Addiction, 90, 1179–88.

47. Paille, F.M., Guelfi, J.D., Perkins, A.C., Royer, R.J., Steru, L., and Perot, P. (1995). Randomised multicentre trial of acamprosate in a maintenance programme of abstinence after alcohol detoxification. Alcohol and Alcoholism, 30, 239–47.

48. Pelc, I., Verbanck, P., Le Bon, M., Gavrilovic, M., Lion, K., and Lehert, P. (1997). Efficacy and safety of acamprosate in the treatment of detoxified alcohol-dependent patients: a 90-day dose finding study. British Journal of Psychiatry, 171, 73–7.

49. Garbutt, J.C., West, S.L., Carey, T.S., Lohr, K.N., and Crews, F.T. (1999). Pharmacological treatment of alcohol dependence—a review of the evidence. Journal of the American Medical Association, 281, 1318–25.

50. Whitworth, A.B., Fischer, F., Lesch, O., et al. (1996). Comparison of acamprosate and placebo in long-term treatment of alcohol dependence. Lancet, 347, 1438–42.

51. Sass, H., Soyka, M., Mann, K., and Zieglgansberger, W. (1996). Relapse prevention by acamprosate: results from a placebo controlled study on alcohol dependence. Archives of General Psychiatry, 53, 673–80.

52. Chick, J., Howlett, H., Morgan, M.Y., and Ritson, B. (2000). United Kingdom Multicentre Acamprosate Study (UKMAS): a 6 month prospective study of acamprosate versus placebo in preventing relapse after withdrawal from alcohol. Alcohol and Alcoholism, 35,.

53. Besson, J., Aeby, F., Kasas, A., Lehert, P., and Potgieter, A. (1998). Combined efficacy of acamprosate and disulfiram in the treatment of alcoholism: a controlled study Alcoholism: Clinical and Experimental Research, 22, 573–9.

54. Gianoulakis, C., Krishnan, B., and Thavundayil, J. (1996). Enhanced sensitivity of pituitary b-endorphin to ethanol in subjects at high risk of alcoholism. Archives of General Psychiatry, 53, 250–7.

55. Volpicelli, J.R., Alterman, A.I., Hayashida, M., and O'sBrien, C.P. (1992). Naltrexone in the treatment of alcohol dependence. Archives of General Psychiatry, 49, 876–80.

56. O'sMalley, S., Jaffe, A.J., Chang, G., Schottenfeld, R.S., Meyer, R.E., and Rounsaville, B. (1992). Naltrexone and coping skills therapy for alcohol dependence. Archives of General Psychiatry, 49, 881–7.

57. Volpicelli, J.R., Rhines, K.C., Rhines, J.S., Volpicelli, L.A., Alterman, A.I., and O'sBrien, C.P. (1997). Naltrexone and alcohol dependence: role of subject compliance. Archives of General Psychiatry, 54, 737–42.

58. Volpicelli, J.R., Volpicelli, L.A., and O'sBrien, C.P. (1995). Medical management of alcohol dependence: clinical use and limitations of naltrexone treatment. Alcohol and Alcoholism, 30, 789–98.

59. O'sMalley, S., Jaffe, A.J., Chang, G., Schottenfeld, R.S., Meyer, R.E., and Rounsaville, B. (1996). Six month follow-up of naltrexone and psychotherapy for alcohol dependence. Archives of General Psychiatry, 53, 217–24.

60. Swift, R.M. (1999). Drug therapy for alcohol dependence. New England Journal of Medicine, 340, 1482–90.

61. Doty, P. and de Wit, H. (1995). Effects of naltrexone pretreatment on the subjective and performance effects of ethanol in social drinkers. Behavioural Pharmacology, 6, 386–94.

62. Jeffers, S. (1987). Feel the fear and do it anyway, Century Hutchinson, London.

63. McGrath, P.J., Nunes, E.V., Stewart, J.W., et al. (1996). Imipramine treatment of alcoholics with primary depression: a placebo controlled clinical trial. Archives of General Psychiatry, 53, 232–40.

64. Cornelius, J.R., Salloun, I.M., Ehler, J.G., et al. (1997). Fluoxetine reduced depressive symptoms and alcohol consumption in patients with comorbid major depression and alcohol dependence. Archives of General Psychiatry, 54, 700–5.

65. Kranzler, H.R., Burleson, J.A., Boca, F.K., et al. (1994). Buspirone treatment of anxious alcoholics. Archives of General Psychiatry, 51, 720–31.

66. Finney, J., Hahn, A.C., and Moos, R.H. (1996). The effectiveness of inpatient and outpatient treatment for alcohol abuse: the need to focus on mediators and moderators of setting effect. Addiction, 91, 1773–96.

67. Schuckit, M. (1998). Penny-wise, ton-foolish? The recent movement to abolish inpatient alcohol and drug treatment. Journal of Studies on Alcohol, 59, 5–6.

68. Mattick, R.P. and Jarvis, T. (1994). Inpatient setting and long duration for the treatment of alcohol dependence? Outpatient care is as good. Drug and Alcohol Review, 13, 127–35.

69. Annis, H.M. (1996). Inpatient versus outpatient setting effects in alcoholism treatment: revisiting the evidence. Addiction, 91, 1804–7.

70. Trent, L.K. (1998). Evaluation of a four- versus six-week length of stay in the Navy's alcohol treatment program. Journal of Studies on Alcohol, 59, 270–9.

71. Project MATCH Research Group (1997). Project MATCH secondary a priori hypotheses. Addiction, 92, 1671–98.

72. Connors, G.J., Carroll, K.M., DiClemente, C.C., Longabaugh, R., and Donovan, D.M. (1997). The therapeutic alliance and its relationship to alcoholism treatment participation and outcome. Journal of Consulting and Clinical Psychology, 65, 588–98.

73. Sereny, G., Sharma, V., Holt, S., and Gordis, E. (1986). Mandatory supervised Antabuse therapy in an out-patient alcoholism program: a pilot study. Alcoholism: Clinical and Experimental Research, 10, 290–2.

74. Smith, J.E., Meyers, R.I., and Delaney, H.D. (1998). The Community Reinforcement Approach with homeless alcohol-dependent individuals. Journal of Consulting and Clinical Psychology, 66, 541–8.

75. Azrin, N.H., Sisson, R.W., Meyers, R., and Godley, M. (1982). Alcoholism treatment by disulfiram and community reinforcement therapy. Journal of Behavior Therapy and Experimental Psychiatry, 13, 105–12

76. Robichaud, C., Strickland, D., Bigelow, G., et al. (1979). Disulfiram maintenance employee alcoholism treatment: a three phase treatment. Behaviour Research and Therapy, 17, 618–21.

77. Gledhill, J., Burroughs, A., Rolles, K., Davidson, B., Blizard, B., and Lloyd, G. Psychiatric and social outcome following liver transplantation for alcoholic liver disease: a controlled study. Journal of Psychosomatic Research, in press

78. Shore, J.H. (1987). The Oregon experience with impaired physicians: an eight year follow-up. Journal of the American Medical Association, 257, 2931–4.

79. Ahles, T.A., Schlundt, D.G., Prue, D.M., and Rychtarik, R.G. (1983). Impact of aftercare arrangements on the maintenance of treatment success in abusive drinkers. Addictive Behaviours, 8, 53–8.

80. Reynauld, M., Hourcade, F., Planche, F., Albuisson, E., Neunier, M.-N., and Planche, R. (1998). Usefulness of carbohydrate-deficient transferrin in alcoholic patients with normal gamma-glutamyl transferase. Alcoholism: Clinical and Experimental Research, 22, 615–18.

81. Patience, D., Buxton, M., Chick, J., Howlett, H., McKenna, M., and Ritson, B. (1997). The SECCAT Survey (II): the alcohol related problems questionnaire as a proxy for resource costs and quality of life in alcoholism treatment. Alcohol and Alcoholism, 32, 79–84.