< Previous | Contents | Next >

EFFECTIVE AND COST-EFFECTIVE MEASURES TO REDUCE ALCOHOL MISUSE IN SCOTLAND: A LITERATURE REVIEW

EXECUTIVE SUMMARY

Aim of the Study

0.1 The aim of this study is to review evidence on the effectiveness and cost-effectiveness of interventions aimed at reducing alcohol misuse.

Methods

Effectiveness Literature

0.2 This study presents a review of existing reviews of the effectiveness literature. This reflects the sheer volume of literature available and the short time-scale available for the study. A comprehensive and systematic search of electronic databases has been undertaken to identify all relevant reviews. The quality of the reviews has been assessed using various criteria ( e.g. whether a systematic search had been undertaken, whether explicit inclusion and exclusion criteria had been employed and whether the review authors had carried out an assessment of the quality of the studies).

0.3 While the effectiveness reviews, and the studies that they report, have used a variety of different outcome measures, most have failed to report health outcomes. This study presents an overview of the outcome measures that have been reported (by type of intervention) and describes the statistical methods used in reporting and summarising results within the reviews.

Cost —Effectiveness Literature

0.4 A search of electronic databases for the period 1990-2001 has been conducted for the review of cost-effectiveness literature. The cost-effectiveness literature is smaller than the effectiveness literature and of variable quality. All economic evaluations have been included, even if they are incomplete in their coverage of costs or benefits. The studies identified have been quality assessed using a standard checklist.

0.5 The basic methods of cost-effectiveness analysis are explained and the potential costs and consequences to be considered are summarised. A description is also given of the different forms of economic evaluation; cost-offset studies, cost-effectiveness analysis, cost-utility analysis and cost-benefit analysis.

Effectiveness review - Main findings

0.6 There are seven broad categories of interventions. The main findings presented in the literature are summarised below by type of intervention.

Policy and legislative interventions

0.7 There is good and consistent evidence that fiscal policy (taxation) is effective in reducing total alcohol consumption, although estimates of the size of the effect are variable. What is clear is that fiscal policy affects all drinkers, not just problem drinkers, in terms of the higher prices to be paid for alcohol. Evidence suggests that the alcohol consumption of the heaviest 10% of drinkers is not responsive to price increases but problem drinkers below this level do respond. The evidence relating to under-age and youth drinking is unclear.

0.8 A range of legislation has been enacted in the US to reduce drink driving:

• lower permitted blood alcohol levels for younger drivers and for persons with previous convictions for drink driving;
• reductions in permitted alcohol levels for all drivers;
• additional sanctions for drink driving, such as mandatory licence suspension and ignition interlock devices.

Inter-state comparisons and general trends suggest that this legislative action has been successful. However, the results may not translate directly to the UK. Trends in total alcohol consumption have also been falling in the US and this may reflect a different cultural attitude towards both drinking and drink driving. Reductions in permitted blood alcohol levels for all drivers took the level down to the current UK level. The impact of reductions below this level may be less.

0.9 Evidence relating to licensing controls is mixed. Some studies from other countries have suggested that longer licensing hours increase alcohol related problems but UK evidence is unclear. There is some evidence from the US and other countries that higher outlet density is associated with higher sales and increased fatal crashes but again the UK evidence is mixed. The type of outlet is also a relevant factor. There is mixed evidence about the impact of outlet type on alcohol-related problems but some evidence of factors that reduce the likelihood of sales to minors. In the US, there is some evidence that raising the legal drinking age to 21 has reduced alcohol related crashes and injuries and may reduce consumption. However, these results may not transfer to a UK setting because of different attitudes to drinking amongst young people in the UK.

0.10 Studies of advertising and alcohol consumption over time have failed to find a significant association, although this may be due to the limited variation in advertising expenditure. Studies of advertising bans across countries have found an effect but this may be due to countries with low consumption being more likely to ban advertising. There is stronger evidence to support the effect of advertising on children.

Enforcement

0.11 The best evidence of effectiveness relates to random breath testing, which has been shown to be effective in both Australia and the US. The evidence relating to the effectiveness of mandatory licence suspension is mixed and ignition interlock devices appear to be effective but only while fitted.

Prevention

0.12 Most of the effectiveness evidence relates to school-based interventions and provides relatively weak evidence of effects on knowledge rather than behaviour. These studies are not of particularly good quality; in particular, the studies have poor controls. Characteristics of programmes which appear to contribute to success are interactive delivery, parental or community involvement and peer involvement.

0.13 Mass media campaigns relating to alcohol, tobacco or illicit drugs show some effects on knowledge and attitudes but little on behaviour. Evaluation of three community prevention programmes provided mixed results.

Screening and detection

0.14 Although a number of screening questionnaires are available to detect alcohol misuse, their performance is extremely variable. Some are better at detecting certain levels of alcohol misuse than others. For general screening purposes, AUDIT is more effective in detecting at risk, hazardous or harmful drinking whilst CAGE is superior for detecting alcohol abuse and dependency. These two tests perform better than other screening tests. T-ACE and TWEAK are more sensitive and specific than CAGE for screening pregnant women. MAST may be useful in psychiatric settings. Laboratory tests do not perform well as screening instruments.

Brief interventions

0.15 The majority of studies have shown brief interventions to be effective in changing drinking behaviour and reducing alcohol consumption for at least 12 months in patients who are not alcohol dependent. The evidence from these research studies is consistent but there are concerns about generalising these results outwith the research setting.

Detoxification

0.16 The literature supports the use of benzodiazepines as the first choice therapy on the basis of safety and effectiveness but the quality of studies is not very high. Outpatient treatment is safe and effective for patients with mild to moderate symptoms. Where inpatient treatment is required, longer stays have not been demonstrated to increase effectiveness.

Relapse prevention

0.17 Despite a lack of randomised-controlled trials, psychosocial interventions are considered to be effective. In a large US study, the total percentage achieving abstinence or controlled drinking was 56% to 60% compared with an estimated spontaneous remission rate of 33%. Pharmacological treatments are effective as adjuncts to psychosocial interventions. Both Naltrexone and Acamprosate have been shown to delay a return to drinking. Co-existing psychiatric problems should be appropriately treated.

Cost-Effectiveness review - Main findings

Enforcement

0.18 A modelling study of the costs and benefits of random breath testing using US data provides evidence to suggest that this intervention is cost-effective. A cost-benefit study of a server training programme in the US showed benefits in excess of costs. As with all studies based on US cost data, these results are not directly transferable to a UK setting. Avoided use of health care services forms a large part of the benefits and service usage in the US is likely to higher in both volume and price.

Prevention

0.19 A cost-effectiveness study of implementing office-based preventive services for adolescents in the US suggests that this is not cost-effective. The cost of preventing a death due to a motor vehicle crash was $12 million (£8.17 million). This compares with Department of Transport estimates of the value for preventing a fatality of approximately £1 million. An Australian study of the cost-effectiveness of thiamine-supplementation alternatives in preventing the Wernicke-Korsakoff syndrome found that the most cost-effective strategy for preventing WK encephalopathy was fortifying full strength beer, rather than wine or bread-making flour. The cost per case averted was AUS$662 (£235).

Screening and detection

0.20 There is little evidence about the cost-effectiveness of screening. Telemarketing has been shown to be a cost-effective strategy for promoting the Drink-less screening and brief intervention package to GPs in Australia. A study of alternative staff carrying out screening in a UK general hospital, concluded that a specialist worker was most cost-effective but noted that nurses could be used more flexibly.

Brief interventions

0.21 Three economic studies have shown brief interventions to be relatively cost-effective, due to fairly high levels of effectiveness and low costs. Modelling results using UK cost data suggests that the cost per life saved is in the range £1446-£2628 if no savings in resource use are taken into account. If resource savings are considered then the benefits exceed the costs of the intervention.

Detoxification

0.22 Economic studies have shown home detoxification and outpatient detoxification to be cost-effective but these were small and rather limited studies.

Relapse prevention

0.23 The health care cost savings for psychosocial interventions are dependent upon the key patient characteristics of alcohol dependence, psychiatric severity and the level of network support for drinking. Cost-effectiveness can be improved by matching patients to treatment. Two studies have modelled the cost-effectiveness of Acamprosate. In a Belgian study, a cost saving to the health care provider was found of 22,000 BEF (£337) per patient over 24 months. The results were sensitive to the probability of relapse and under different conditions there may not be a cost saving. A German study found an overall cost saving to society of DM2,602 (£822) per additional abstinent patient. Modelling the results using UK data confirms that a resource saving of over £600 per patient will result.

0.24 Two studies of inpatient versus outpatient care after detoxification found outpatient care to be more cost-effective. Two small studies of behavioural marital therapy have produced opposite conclusions regarding cost-effectiveness.

Other issues

0.25 Some cross cutting reviews look at settings and populations rather than specific interventions. These studies do not provide comparative effectiveness data. There are also reviews providing information about education and training for professional groups.

Scottish evaluations

0.26 There has been considerable activity in Scotland in implementing initiatives to reduce alcohol misuse and its consequences. Some examples of the evaluation studies have been reviewed but not all of these are focussed on effectiveness. The most useful findings reported here relate to brief interventions by health visitors and home detoxification services. Counselling services are clearly effective but research is required to establish whether the current pattern of provision is more effective than a briefer intervention. The role of complementary therapy requires larger studies to be carried out. A culture of evaluation that is focussed on outcomes needs to be fostered at all levels from policy making to service delivery.

Conclusions

Summary of available evidence

0.27 The main findings from the review of the literature on effectiveness and cost-effectiveness are summarised in table 0.1. In terms of the effectiveness review, there is a strong and relevant evidence base to show that:

• the use of price increases, via taxation, and brief interventions will reduce the number of problem drinkers;
• effective screening tools to detect problem drinkers are available (CAGE and AUDIT); and
• detoxification services and relapse prevention, through appropriate psychosocial and pharmacological treatments, are effective.

0.28 The cost-effectiveness review found evidence to support the cost-effectiveness of:

• brief interventions;
• home and outpatient detoxification;
• outpatient treatment for relapse prevention; and
• the use of acamprosate as an adjunct treatment in relapse prevention.

Quality, coverage and relevance of the evidence

0.29 The quality of the effectiveness reviews and the underlying primary studies is variable. The evidence base of economic studies is weak. The best evidence of effectiveness, in terms of both quality and coverage, relates to brief interventions and relapse prevention. There are a number of economic studies in these areas, of reasonable quality, but of limited relevance to the UK. Prevention, particularly in schools, has been widely researched but the poor quality of many of the studies means that the evidence base remains weak. Policy evaluation is not of high quality and particularly difficult to relate to the UK.

Lessons about methods of evaluation

0.30 In order to improve the evidence base in the UK, more emphasis needs to be placed on evaluating alcohol initiatives in the UK and on modelling UK results based on international studies. Evaluations should be planned at an early stage in the development of an initiative to ensure that the data required to conduct a meaningful evaluation are available. Better methods of policy evaluation are required to provide robust evidence where there is no access to random controls and these need to be developed and applied in a UK context to provide relevant information.

Implementation issues

0.31 The results of the review will have to be interpreted alongside information about interventions that are already taking place in Scotland. Both the impact which interventions will have upon strategic targets and the resources required to implement interventions, will depend upon the extent to which they have already been deployed. The effectiveness of these interventions has been demonstrated in research settings and arrangements for auditing or monitoring the effectiveness of interventions in routine practice will need to be put in place. A review of policy strategies in other countries has found that reducing alcohol consumption in the general population or in high risk populations are equally effective in preventing alcohol related problems. The costs of the alternative strategies are not reported.

Recommendations

0.32 It is recommended that the Plan for Action should support:

Population measures

• the introduction of a brief intervention programme where this is not already provided; and
• the development of monitoring arrangements to assess the impact of these interventions.

Treatment measures

• improved access to treatment and relapse prevention; and
• the use of cost-effective alternatives, such as home detoxification, where appropriate.

Research measures

• the prioritisation of policy evaluation and prevention as the areas requiring most development of the evidence base;
• better co-ordination of research effort with resources concentrated on fewer larger studies with longer term follow up;
• better knowledge management in terms of access to relevant research results;
• the fostering of an evaluation culture amongst those responsible for delivering services, focussed on outcomes and the monitoring of effectiveness; and
• guidelines for evaluation to assist in this process.

Table 0.1 Summary of Findings

< Previous | Contents | Next >