The future of specialised alcohol treatment services: a matter of policy?

Summary

Around 7.1 million people in England drink hazardously or harmfully and a further 1.1 million are dependent on alcohol. Motivational interviewing is widely used to treat people with alcohol problems and is probably the best described example of a brief intervention. However, some recent trials have been disappointing. Specialised alcohol treatment services have also suffered from weakness in the evidence base. Investment in treating alcohol misuse has fallen far behind that for drug misuse. The Department of Health's Alcohol Harm Reduction Strategy for England embraces policies that are high-profile and cheap but are ineffective and ignore many effective measures. It recommends stepped care for alcohol treatment, but unlike the equivalent for drugs treatment, it sets no targets and leaves the small (7%) increase in funding to the discretion of local purchasers. UK spending on specialised treatment for drug misuse is estimated to be around £600 million for 2007 – around three times the estimated cost of treatment for alcohol misuse.

Worldwide, around 2000 million people consume alcoholic drinks. The World Health Organization estimates that, at any one time, 76 million people have an alcohol use disorder (World Health Organization 2004). It reports that alcohol causes 3.2% (1.8 million) of all deaths worldwide each year, 4% of all ill health, and contributes to 58 million premature deaths. Alcohol is also responsible for an enormous burden of social, mental and economic problems, including crime, accidents and domestic violence (Babor 2003). There are enormous regional variations in alcohol dependence. Just over a decade ago, Gossop (1995) reported prevalence rates ranging from 11–12% in Brazil and Poland to 0.2% in Egypt and many Islamic states. Global average alcohol consumption was equivalent to 5.8 litres of pure alcohol per head, although consumption in Europe was higher (12 litres alcohol per head). The highest consumption occurred in former communist countries of the Eastern Bloc, at almost 14 litres per head. Alcohol causes 6% of all deaths in Europe, half of which are related to accidents (Rehm 2006).

In Great Britain, around 90% of adults (aged 16–64) consume alcohol and most do not experience problems. The UK General Household Survey (Office for National Statistics 2001) found that more than 75% of adults in England were either non-drinkers (4.7 million people) or drank within safe limits (26.3 million people). The Alcohol Needs Assessment Project (Department of Health 2005) found that 23% of adults (about 7.1 million people) in England drank at hazardous or harmful levels and estimated that a further 1.1 million were dependent on alcohol.

Clearly, effective treatments for alcohol use disorders (for a review in Advances see Luty 2006) would have great health and economic benefit.

Evidence base for alcohol misuse interventions

Brief interventions

There has been a worldwide movement away from expensive residential treatment services in favour of community-based brief interventions. Primary care can deliver brief interventions inexpensively. They involve up to four sessions using counselling techniques such as motivational interviewing (Moyer 2002), and target moderate and non-dependent drinkers, who account for the greater proportion of alcohol problems and related ill health. In contrast, specialised services tend to focus on a smaller number of severely alcohol-dependent people.

Motivational interviewing has been the cornerstone of psychological treatment of substance misuse problems for some time. It is probably the best described example of a brief intervention ^† (Bien 1993; Moyer 2002). The client, rather than the therapist, gives the reasons for abstinence and provides a list of problems caused by their drinking. Motivational interviewing can be delivered in a small number of sessions (the Matching Alcohol Treatments to Client Heterogeneity trial, or Project MATCH, used four sessions: see below) and can easily be adapted to primary care settings (Miller 2002). Brief interventions seem modestly effective in opportunistic samples of people who did not realise they were drinking too much (Moyer 2002). Moyer and colleagues reported effect sizes of 0.14–0.67 over control conditions, equivalent to a 10–20% increase in the number of people achieving a favourable response. Of 33 brief intervention studies reviewed in the meta-analysis by Bien et al (1993), there was an overall improvement in drinking outcomes with an effect size of 0.7–0.8 over the follow-up period for all groups (control and intervention groups). The brief interventions themselves produce a 0.38 effect size over the control condition in 19 trials, whereas extended treatment produced no significant advantage over brief interventions (effect size 0.06) in 14 other trials. Much research has shown that brief interventions, delivered in a variety of settings, are effective in reducing alcohol consumption among people who drink to a hazardous or harmful level. However, there is no evidence that opportunistic brief interventions are effective among people with more severe alcohol problems and levels of dependence – that is, who are moderately or severely dependent. Such individuals should be encouraged to attend specialised treatment services.

More research is needed into the longevity of the effects of brief interventions and their efficacy in reducing alcohol-related mortality.

Intensive treatment and specialised treatment services

In the era of evidence-based medicine, specialised alcohol treatment services have also suffered from a weak evidence base (Tucker 2006). The diminishing funding for research into treatment for alcohol use disorders is a major problem for policy development (Drummond 1997). Similarly, economic analyses of alcohol treatment are uncommon and notoriously poor, rarely being based on randomised trials and often with no control group (McCollister 2003).

Cynicism about the effectiveness of specialised alcohol treatment has been expressed for some time. In Britain, Edwards' classic randomised controlled trial of a brief intervention (a 3-hour assessment session) compared with a comprehensive package of treatment showed no difference even after 10 years of follow-up (Edwards 1983). In New Zealand, a randomised controlled trial involving 113 alcohol-dependent people confirmed the results of the British study (Chapman 1988). A 6-month prospective study reported that supervised administration of disulfiram decreased drinking significantly (Chick 2000), but subsequent reviews of disulfiram treatment (Hughes 1997) concluded that unsupervised administration of the drug is of limited utility and associated with poor outcome.

Acamprosate and naltrexone

However, in recent times a few studies have spurred a new wave of optimism. Although the two largest trials of acamprosate and naltrexone showed no significant benefits (Chick 2000; Krystal 2001), a meta-analysis of studies (Bouza 2004) has demonstrated a beneficial therapeutic effect.

The meta-analysis included 13 acamprosate studies (n = 4000) and 19 naltrexone studies (n = 3205). Except for four studies involving naltrexone, all participants were detoxified before treatment. These studies typically lasted between 3 months and 2 years. Acamprosate was found to increase the continuous abstinence rate with a calculated number needed to treat of 10 (95% CI 7–15). The cumulative abstinence (total periods of abstinence regardless of relapses) is significantly increased both in treatment and during the follow-up period. Acamprosate may thus be more useful in therapeutic approaches that target alcohol abstinence. The evidence for naltrexone is less convincing because of small sample sizes and short-term studies. Comparisons between acamprosate and naltrexone are often difficult as many trials of acamprosate report total periods of abstinence, whereas trials of naltrexone report relapse rates. It is also unclear how the type of psychosocial therapy used along with the medication, patient characteristics and the duration of therapy might affect the outcome measures.

Bouza and colleagues concluded that both acamprosate and naltrexone work (although differently) in treating alcohol dependence and that they have a place in such treatment. There are unconfirmed suggestions that the two may act differently and may therefore act additively if combined.

The COMBINE study: acamprosate, naltrexone and cognitive–behavioural intervention

The influential COMBINE study in the USA reported on a 16-week randomised controlled trial involving 1383 recently abstinent alcohol-dependent volunteers (Anton 2006). The participants were randomised into nine groups to receive naltrexone, or acamprosate, or naltrexone plus acamprosate or placebo (all with or without a ‘combined behavioural intervention’ based on cognitive–behavioural therapy), or the behavioural intervention alone. Participants were followed for up to 1 year after treatment. All groups showed substantial improvements. However, acamprosate showed no evidence of efficacy above placebo, with or without the behavioural intervention. Naltrexone plus the behavioural intervention produced the best efficacy. Placebo pills and a meeting with a healthcare professional had a more positive effect than the behavioural intervention alone. Note that because of safety concerns, naltrexone is not licensed in the UK to treat alcohol dependence syndrome.

Project MATCH: motivational enhancement, twelve‐step facilitation and cognitive–behavioural therapy

Project MATCH was an older multicentre US trial involving two groups of patients with alcohol dependence (Project MATCH Research Group 1997). The aftercare group comprised 774 patients, who were followed after in-patient treatment. The out-patient-only group involved 952 participants who entered out-patient treatment. The trial randomised the participants in each group to three forms of manualised psychotherapy: 4 sessions of motivational enhancement therapy, 12 sessions of twelve-step facilitation or 12 sessions of cognitive–behavioural therapy (CBT); there was no control group.

At 1 year, 35% of aftercare participants (who had undergone detoxification) remained completely abstinent compared with 20% of the out-patient participants, regardless of which treatment they received. Motivational enhancement therapy over four sessions was generally as clinically effective as the two more intensive treatments – twelve-step facilitation and CBT delivered over 12 sessions. This equivalence in effectiveness applied across both aftercare and out-patient arms of the trial and in a population of people with relatively severe levels of alcohol dependence and problems.

UKATT

Project MATCH took place in the USA, which has a different healthcare system from that of the UK. In the UK, where an abstinence-based philosophy prevails over moderation, the UKATT study enrolled 742 people seeking treatment for alcoholism at seven sites around the UK (United Kingdom Alcohol Treatment Trial Research Team 2005). The trial randomised participants to social behaviour network therapy or motivational enhancement therapy.

The mean number of days that participants abstained from alcohol increased from 30% at baseline to 46% at 1 year, and average alcohol consumption on a drinking day fell from 27 units to 19. Both social behaviour network therapy and motivational enhancement therapy produced statistically significant improvements in alcohol consumption, alcohol dependence, alcohol-related problems and aspects of general functioning. The trial therefore confirmed the effectiveness of motivational enhancement therapy and found that the novel treatment, social behaviour network therapy, is as effective.

Analysis of effectiveness

A review of the effectiveness of treatments for people with alcohol problems that evaluated Project MATCH, among other research, concluded that even people with severe dependence should be offered motivational enhancement therapy as the first stage in a stepped programme of care in specialist services, provided there are no sound reasons for immediately offering a more intensive treatment (Raistrick 2006).

The review also mentions results from the Mesa Grande project (Miller 2003), a large systematic review of treatment outcomes research before 2001, which summarised 381 randomised controlled trials of different types or intensities of treatment. A cumulative evidence score was calculated, with higher scores indicating large amounts of evidence in favour of each trial and lower scores indicating unfavourable evidence. Although the Mesa Grande project has had many criticisms, brief interventions and motivational enhancement therapy received ranks of one and two respectively.

Cost-effectiveness of alcohol treatments

Studies have demonstrated both short-term and longer-term cost savings. The UKATT looked at the short-term public sector resource savings in healthcare, other alcohol treatments, social care and the criminal justice system. Comparing the use of resources 6 months before the start of treatment with 6 months before the 1-year follow-up interview, the conclusion was that for every £1 spent on treatment, the public sector saved £5 (United Kingdom Alcohol Treatment Trial Research Team 2005). When this result is extrapolated to just 10% of alcohol-dependent people in the UK, it is thought that it will reduce annual public sector resource costs by between £109 million and £156 million (National Treatment Agency for Substance Misuse 2006a).

Specialised alcohol services are also useful for advice regarding harm reduction (for example, in dealing with episodes of alcohol withdrawal), cognitive impairment, treating suicidal crises and other complications of alcohol misuse. Many specialists would point out that specialised alcohol services should also be judged in respect of these outcomes and not just by reduction in alcohol consumption, although analyses of effectiveness often neglect these other activities.

MoCAM and the stepped-care treatment model

Models of Care for Alcohol Misusers (MoCAM; National Treatment Agency for Substance Misuse 2006b) is a UK government publication designed to assist health service managers to commission local alcohol treatment services. It advocates a stepped model of care (Table 1) and identifies four main categories of people who misuse alcohol: hazardous drinkers (who drink over medically recommended limits or binge drink but show no obvious alcohol-related problems); harmful drinkers (who are experiencing alcohol-related damage to health but are not clinically dependent); moderately dependent drinkers; and severely dependent drinkers. Individuals who misuse alcohol may move in and out of different categories over the course of a lifetime. The MoCAM model envisages two main components: brief interventions for people who drink excessively but do not require treatment for physical dependence, and specialised treatment for people with moderate or severe dependence. In stepped care, people new to such treatment should be assessed and should initially receive the least intensive or least prolonged intervention suitable for their level of need. If the person's response to such a limited initial intervention is inadequate, a more intensive or prolonged package of care may be needed. Someone who is moderately dependent and has other complex needs might require the more intensive interventions from the outset.

Alcohol policy and treatment in England

The UK spends around £13 billion on buying alcohol each year and, on average, each UK adult consumes the equivalent of 9.1 litres of alcohol (Raistrick 2005). The recommended safe limits are 21 units a week for men and 14 for women, but 27% of men and 17% of women drink more than this. In addition, 7% of men and 3% of women drink over the danger limit (50 and 35 units a week respectively). There are at least a million alcohol-dependent people in the UK (Department of Health 2005). Alcohol misuse in England costs the National Health Service (NHS) £1.7 billion each year and chronic alcohol-related diseases cause 11 000–18 000 deaths (Prime Minister's Strategy Unit 2004). Alcohol costs the UK £7.3 billion through crime and public disorder and up to £6.4 billion through lost productivity. However, investment in treatment for people with alcohol problems has fallen far behind that for drug problems (Raistrick 2005), presumably because the British government ‘is determined to be toothless with respect to alcohol policy’ (Room 2004).

Many national alcohol policies also emphasise demand-reduction techniques such as restriction on advertising, age limits on purchasing alcohol, stringent drink-driving levels, random driver breath testing and taxation on alcohol. Although the effectiveness of these methods is proven, their application and enforcement vary considerably (Gossop 1995; Plant 2004; Room 2004; Hall 2005). Demand-reduction techniques are relatively inexpensive and flexible (they can be changed or abolished easily by government edicts). They also do not require the significant, long-term investment in health infrastructure that is required with specialised alcohol treatment services.

The Alcohol Harm Reduction Strategy and the Licensing Act 2003

The Alcohol Harm Reduction Strategy for England is a spectacular example of how the drinks industry can subvert alcohol policy (Plant 2004; Prime Minister's Strategy Unit 2004). In summary, the strategy embraces all the policies that are high-profile, cheap and ineffective, such as school education and voluntary advertising codes. However, the strategy dismisses virtually all the policies that are known to be effective in reducing alcohol consumption (reduced licensing hours, increased excise duty, reducing the drink-driving limit). Changes to the English Licensing Act 2003 (which regulates sales of alcohol) are similarly emasculated (Room 2004). Drummond (2006) describes MoCAM as ‘a light touch’. Unlike its equivalent for treating drug misuse, there are no targets and a derisory 7% increase in funding. The decisions to implement the recommendations (or not) are left to the discretion of local purchasers. Specialised treatment services are hardly mentioned in the commissioning sections, which focus almost entirely on screening and brief interventions in primary care (National Treatment Agency for Substance Misuse 2006b).

Shifting agendas: from healthcare to public order

The NHS currently spends about £217 million a year on specialised alcohol treatment services (Prime Minister's Strategy Unit 2004; Department of Health 2005). These are provided by around 300 advice and counselling services, 100 day programmes and nearly 200 residential programmes. There has been a striking move away from residential treatment towards community-based facilities following an influential government report and the increasing popularity of home detoxification and brief interventions in primary care (Edwards 1967, 1977; Centre for Reviews and Dissemination 1993). Hence, healthcare commissioners in the UK have progressively withdrawn funding from established treatment centres and diverted it into screening programmes and brief interventions in primary care.

Recent government policy documents mean that the dissolution of specialised alcohol services in England is likely to continue. It is clear that the healthcare agenda now takes second place to the public order agenda in guiding investment in treatment for substance misuse problems in England (Raistrick 2005). Consequently, the total spending on specialised treatment for drug misuse is estimated at around £600 million for 2007 (National Treatment Agency for Substance Misuse 2006b), which is three times the estimate for alcohol treatment. This is made worse by the fact that the NHS has directed its spending on substance misuse towards treatment for illicit drug use.

A European alcohol strategy

The situation in England is remarkably similar to that in Germany, where political apathy in the presence of a heavy-drinking culture has produced relatively ineffective attempts to face up to the nation's alcohol problems (Buhringer 2006). Just over 20% of German adults screen positive for alcohol use disorders. Alcohol policy has only recently been devised in Germany and tends to be limited to demand reduction. Regulations concerning sale of alcohol to minors are rarely enforced. For most of the 20th century the German approach to treating people with alcohol problems was long-term residential treatment in specialised centres. This is now financed by public pension insurance schemes and there has been a dramatic reduction in the availability and duration of residential alcohol treatment over the past two decades. Although Germans have a legal right to treatment without cost, there has been a huge increase in people seeking out-patient treatment for alcohol problems but little extra investment to meet these needs.

A European alcohol strategy is currently being developed (McKee 2006). Unfortunately, so far this appears to be full of bland platitudes regarding ‘underlining’, ‘stressing’ and ‘endorsing’ various aspects of the alcohol strategies of the member states. The strategy contains no concrete recommend ations regarding targets for treatment or prevention of alcohol-related problems, nor any specific measures (such as taxation) to reduce alcohol-related harm. However, some mention is made of irresponsible advertising, preventing alcohol-related accidents and preventing alcohol misuse in young people. The strategy reiterates the view that alcohol policies should ensure balanced representation for the various stakeholders, including the alcoholic beverage industry. There appears to be little so far in the European alcohol strategy that has not already been adopted by the more affluent member states.

Conclusions

Specialised alcohol treatment services are an endangered species. The crime agenda means that there is great political enthusiasm in high-income countries for treating people with drug problems. In countries where alcohol is seen as part of the overall substance misuse problem, such as Spain and Australia, alcohol services have been enhanced. However, in countries such as England, where alcohol problems are of secondary political importance, drug treatment services may even starve resources from alcohol services. National alcohol policies, where these are emasculated by the political influence of the drinks industry, concentrate on demand-reduction regulations rather than treating alcohol problems. Evidence for treating people with alcohol problems has also moved away from specialised, residential facilities towards screening and brief interventions in primary care. Although there is emerging evidence in favour of specialised treatment, there are limitations in the evidence base for treatment of people with severe alcohol problems and more research is needed. The future for specialised alcohol services in England looks bleak.

MCQs

1. 1 It is estimated that:

   1. a 7 million people throughout the world have an alcohol use disorder at any one time

   2. b the highest per capita consumption of alcohol is in North America

   3. c in the UK 2% of the population (aged 16–64) drink hazardously or harmfully

   4. d around 1.1 million people in England are dependent on alcohol.

2. 2 As regards interventions for alcohol misuse:

   1. a there has been a world-wide move away from community-based brief interventions towards expensive residential treatment

   2. b brief interventions are designed to target severely dependent drinkers

   3. c recent trials favour brief interventions

   4. d specialised alcohol treatment services suffer from weakness in the evidence base.

3. 3 The UK Alcohol Harm Reduction Strategy:

   1. a embraces cheap, effective strategies

   2. b embraces cheap, ineffective strategies

   3. c sets targets

   4. d proposes funding increases similar to those for treatment of drug misuse

   5. e proposes a stepped-care approach.

4. 4 Models of Care for Alcohol Misusers:

   1. a proposes a stepped-care approach

   2. b offers all new patients intensive prolonged treatment from the start

   3. c proposes new ways of commissioning care

   4. d is published by the alcohol drinks industry.

5. 5 In England:

   1. a the NHS spends about £1 billion a year on treatment for alcohol misuse

   2. b government policy emphasises demand-reduction techniques such as restriction on advertising, age limits on purchasing alcohol and stringent drink-driving levels

   3. c the NHS currently spends more on specialised alcohol treatment services than on drug treatment services.

   4. d purchasers have progressively increased funding on specialised alcohol treatment services rather than on treatment in primary care.

MCQ answers

1		2		3		4		5
a	f	a	f	a	f	a	t	a	f
b	f	b	f	b	t	b	f	b	t
c	f	c	f	c	f	c	f	c	f
d	t	d	t	d	f	d	f	D	f

TABLE 1 The model of treatment delivery envisaged by Models of Care for Alcohol Misusers

Tier description	Settings
Tier 1 interventions include: identifying people who drink to hazardous or harmful levels or who are alcohol-dependent; providing information on sensible drinking; simple brief interventions to reduce alcohol-related harm; and referring those with alcohol dependence or harm for more intensive interventions.	Can be delivered by a range of agencies and in a range of settings, the main focus of which is not alcohol treatment (for example, primary healthcare services, psychiatric services and acute hospitals such as accident and emergency (A&E) departments).
Tier 2 interventions include providing open-access facilities and outreach that offer: alcohol-specific advice, information and support; extended brief interventions to reduce alcohol-related harm in people who misuse alcohol; and assessment and referral of people with more serious alcohol-related problems for care-planned treatment.	May be delivered in the following settings, if the agencies have the necessary competence: specialised alcohol services, primary healthcare services, psychiatric services and acute hospitals such as A&E and liver units.
Tier 3 interventions include providing community-based specialised alcohol- misuse assessment, and alcohol treatment that is care-coordinated and care-planned.	Normally delivered in specialised alcohol treatment services with their own premises in the community (or sometimes on hospital sites). Other delivery may be by outreach (peripatetic work in generic services or other agencies, or domiciliary or home visits). Tier 3 interventions may be delivered alongside Tier 2 interventions.
Tier 4 interventions include providing residential, specialised alcohol treatments that are care-coordinated and care-planned to ensure continuity of care and aftercare, including residential rehabilitation units.	People with complex alcohol-related and other needs requiring in-patient interventions may require hospitalisation for their other needs (e.g. liver problems or pregnancy). This other hospital setting may provide the person with the best overall care, when combined with specialised alcohol liaison support.