The avoidable costs of alcohol abuse in Australia and the potential benefits of effective policies to reduce the social costs of alcohol
5.4 Brief interventions for reducing hazardous alcohol consumption
Studies have shown that brief interventions have proven effectiveness in reducing hazardous alcohol consumption (Bien et al., 1993). Brief interventions have also been shown to be cost-effective (Wutzke et al., 2001) and therefore clearly represent a useful strategy to reduce the rates of alcohol abuse.
Brief interventions usually, but not always, refer to advice and information provided to ‘at risk’ drinkers in the context of a consultation by a primary care physician. Such information is initially conveyed verbally, and usually in the context of a primary care consultation for a different purpose. The initial screening may be accompanied by a range of additional support, including the provision of printed information, follow-up telephone calls, and diaries to record and monitor alcohol consumption. Cost studies of brief interventions include the costs of recruitment and training of general practitioners, the provision of appropriate materials, and the expanded cost of the consultation to include counselling (Gomel et al., 1998).
In some countries, physicians receive program grants or practice-enhancement grants for such interventions. Some interventions may be undertaken by staff employed in primary care practices such as practice nurses. In the Australian context, screening and brief counselling by a general practitioner increases the consultation from Level B to Level C (lasting at least 20 minutes), thus incurring a small additional Medicare cost for every patient who is counselled.
Potential cost savings, including reduced alcohol-related morbidity and mortality, have been estimated to demonstrate the cost-effectiveness of well designed brief intervention programs if they were implemented nationally. Assumptions included that much of the impact of such interventions would not be realised for several years, and therefore studies have discounted the life years saved.
Within any brief intervention program undertaken by GPs, there is considerable flexibility regarding screening, intervention and screening intervals, which may be influenced by factors such as the type and busyness of the practice, numbers of new patients, and financial incentives. It is therefore difficult to calculate the costs of wide implementation.
As there are demonstrably effective interventions which have been developed and tested in Australia, it is evident that these brief interventions should be included in analysis of policies which if implemented would reduce the avoidable costs of alcohol consumption.
Top of Page
In an Australian context, Wutzke et al. (2001), following-on from a study by Gomel et al. (1998), estimate the cost-effectiveness of different general practitioner training and support strategies for reducing hazardous and harmful alcohol consumption, related to the
Drink-less program. A comparison is made between a do-nothing scenario and three training and support strategies with general practitioners implementing the program.
- control—no initial training and no ongoing support on program implementation;
- no support— five minutes of initial training together with the provision of advice to receptionists on data collection offered on a fortnightly basis;
- maximal support—five minutes of initial training plus alternate telephone contact and personal visits every two weeks.
Table 24. Average and marginal costs of three implementation strategies for the Drink-less program (1996 prices)
|Strategy||Average cost per life year saved $||Marginal cost per life year saved|
Source: Wutzke et al. (2001), Tables 2 and 3.
In the above table, average costs are calculated on the basis of a comparison with a do-nothing strategy. Marginal costs are calculated on the basis of a comparison with the previous strategy.
The possible outcomes of brief interventions have been considered in a range of papers. Chisholm et al. (2002, p. 785) report that:
- Efficacy reviews of brief interventions reveal an estimated 22 per cent net reduction in consumption among hazardous drinkers … which would have the effect of shifting the entire distribution of hazardous drinking downwards if applied to the total population at risk (a reduction in overall prevalence of 35–50 per cent, equivalent to a 14–18 per cent improvement in the rate of recovery over no treatment at all).
- The estimation of post intervention alcohol consumption was based on the results of phase 2 of the Australian arm of the WHO collaboration (Saunders et al., 1991). This found that baseline alcohol consumption fell by 28 per cent on average in hazardous and harmful drinkers following exposure to the intervention. This result was consistently robust across international health-care settings and sociocultural groups …
Top of Page
Table 25. Number of lives potentially saved by brief interventions
|Strategy||Number of lives saved|
To put these figures in context, the estimated number of alcohol-attributable deaths in 2004/05 was 3,494 (Collins & Lapsley, 2008a, Table 10).
5.4.1 The impact of brief interventions on the overall social costs of alcohol abuseFew cost-effectiveness studies of brief interventions produce output in a form which is suitable for use in calculation of the potential benefits of this type of intervention in Australia. Those mentioned above, which are considered to have some suitability, produce estimates either of a reduction in hazardous/harmful alcohol consumption or of the number of deaths averted.
A complication in dealing with the social costs of alcohol is the problem of the treatment of the protective health benefits of non-hazardous consumption. As indicated above, the approach adopted in this study is to assume that it would be possible to reduce risky consumption without prejudicing any protective health benefits.
At one end of the spectrum of Australian estimates is a reduction of 28 per cent in the average consumption of hazardous and harmful drinkers. At the other end is a predicted reduction of approximately 56 per cent in the number of alcohol-attributable deaths. The ‘best’ estimate adopted here represents a midpoint between these two extremes. The estimates of savings in social costs are presented in Table 26.
Table 26. Estimated reduction in social costs resulting from the implementation of acomprehensive program of brief interventions (2004/05 prices)
|Best estimate $m||Minimum $m||Maximum $m|
|Total cost savings|
Given that the estimated social costs of alcohol in 2004/05 were over $15 billion, this range of estimates represents a very substantial range of estimated potential reductions in the social costs of alcohol. However, it is very clear that, even on the most pessimistic assumption, the cost-effectiveness of brief interventions for reducing hazardous alcohol consumption is very high.
Top of Page