International Guidelines for the Estimation of the Avoidable Costs of Substance Abuse

7 Policy implications of avoidable cost estimates

7.1 Complexities of policies to reduce substance abuse

In comparison with alcohol and illicit drugs, the design of policy to minimise tobacco-attributable harm is relatively straightforward, for three reasons:

• Since any non-trivial consumption of tobacco is harmful to health, the objective of public policy should be the simple one of reducing the use of tobacco by the maximum amount possible.
  
  
• Tobacco is a relatively homogeneous product. Admittedly, as well as cigarettes/cigars, it can take the form of oral or chewing tobacco. However these latter two uses represent, in most countries, a small proportion of total tobacco consumption and their chronic impacts are similar to those of smoking.
  
  
• Since tobacco is usually a legal product ( though not necessarily when used by minors), excise taxes and advertising bans (both powerful tools) can be used to reduce consumption.

Illicit drugs represent a more complex policy issue, for two main reasons:

• A whole range of products with diverse health and other effects is subsumed within the title of illicit drugs. Policy appropriateness can vary from drug to drug.
  
  
• As a result of the illegality, by definition, of these drugs, a range of polices, including taxation and regulation of product quality, are precluded, and the underlying data are much less robust.

Alcohol, too, is not a homogeneous product, with the three major product categories of beer, wine and spirits. A further consideration is that, under certain conditions, the consumption of alcohol has been proven to have protective effects.

7.2 Dealing with the protective effects of alcohol

In calculating both the aggregate and the avoidable costs of alcohol use, a complication arises which does not exist in the cases of tobacco or illicit drugs. There is significant evidence that, for some medical conditions, alcohol consumption at appropriate levels can have a protective effect, that is, alcohol consumption can reduce the risk of illness or death. In these circumstances, abstinence-based attributable fractions (as opposed to attributable fractions based on hazardous or harmful consumption) are negative. With very minor exceptions, there is no evidence of any analogous health benefits from consumption of tobacco or illicit drugs.

This is an issue that has been addressed by, inter alia, Ridolfo and Stevenson (2001) and Rehm et al (2004). They conclude that low-to-moderate levels of alcohol consumption can, given appropriate drinking patterns, confer health benefits in relation to ischaemic heart disease, supraventricular cardiac dysrhythmias, cholelithiasis (gallstones), ischaemic strokes, haemorrhagic strokes, hypertension and type II diabetes.

The existence of protective effects of alcohol raises the issue of whether these protective benefits (which can be considered as negative costs) should be incorporated in estimates of the aggregate social costs of alcohol, and therefore also in avoidable cost estimates. Some aggregate cost studies (for example Collins and Lapsley, 2002) have taken these benefits into account, while for other researchers (for example Easton, 1997) the inclusion of the benefits of alcohol "misuse" was considered inappropriate.

This issue is complicated by the fact that it appears that even so-called "responsible' levels of alcohol consumption can be dangerous in relation to certain medical conditions, for example female breast cancer. Furthermore, even risky/high risk drinking can prevent some deaths (while causing far more). There is, thus, a difficulty in defining the term alcohol "misuse". How can alcohol consumption which saves some lives (while also causing deaths) be considered to be misuse in all cases. Conversely, how can some level of alcohol consumption be considered to be "safe" when it causes some deaths (while preventing others)?

This point is illustrated in a paper by Chikritzhs, Stockwell et al (2002) who estimate the numbers of lives lost and lives saved in Australia in 1998 due to low risk and risky/high risk drinking, compared with a baseline of complete abstinence. Their results are presented in Table 9 below.

Table 9 - Estimated numbers of lives lost and saved due to low risk and risky/high risk drinking when compared to abstinence, Australia, 1998

	Low risk drinking	Risky/high risk drinking	All drinking
Lives lost	1,505	3,294	4,799
Lives saved	(6,605)	(557)	(7,162)
Total	(5,100)	2,737	(2,363)

Note: figures in brackets represent numbers of lives saved.
Source: Chikritzhs, Stockwell et al (2002), Table 1

They conclude that:

It is recommended that for future reports on alcohol-caused morbidity and mortality, there would be value in presenting a more detailed picture that identifies both the costs and the benefits of low risk drinking and risky/high risk drinking. In order to do this, an abstinence-based contrast must be adopted.

If protective benefits are taken into account in aggregate cost estimates they should also be incorporated in avoidable cost estimates. However, policies designed to minimise the costs of alcohol use may also reduce its benefits. The existence of protective benefits throws doubt on the usefulness of the prevention paradox ( Kreitman, 1986) as a guide to alcohol policy. Policies aimed at minimising the costs of alcohol misuse (for example, high excise taxes on alcohol) may, in the process, reduce the number of low or moderate alcohol consumers, and so reduce the total benefits of low or moderate alcohol consumption.

Thus, measures of the Arcadian normal appropriate for estimating avoidable costs may differ completely from those appropriate for alcohol benefits. In practical policy terms, the best outcome to be hoped for may be to reduce alcohol costs while maintaining alcohol benefits unaffected.

7.3 Policies available to reduce substance abuse costs

A very broad range of measures is available to minimise the social costs of substance abuse. Loxley et al (2004), in surveying the effectiveness of potential prevention policies, categorized available measures under the following broad headings:

• Interventions for children (0-11 years);
  
  
• Interventions for young people (12-24 years);
  
  
• Broad-based prevention;
  
  
• Demand reduction;
  
  
• Regulation and law enforcement: licit drugs;
  
  
• Regulation and law enforcement: illicit drugs;
  
  
• Judicial procedures; and
  
  
• Harm reduction interventions.

Appendix E provides greater detail of the available interventions and of the authors' judgment of the effectiveness of these interventions. Further information about the effectiveness of interventions to reduce drug-attributable crime is presented in Appendix D.

Reviews such as that provided by Loxley et al (2004) provide an important basis for the development of effective intervention strategies.