AMA submission to the Draft Australian Alcohol Guidelines for Low-Risk Drinking - 2007

The amount of national and international research and information regarding patterns of alcohol usage and associated harms is rapidly increasing. It is important in such circumstances to ensure that public guidelines for recommended drinking levels keep pace and adequately reflect the current evidence. Along with prominent and informative drink labelling, and public education about the consequences of drinking, appropriate guidelines are an important tool in reducing the incidence of alcohol related harms. The AMA recognises the efforts of the NHMRC in systematically bringing together and analysing the recent evidence, to review and refine the Australian Alcohol Guidelines.

The AMA supports the three guidelines proposed in the draft Australian alcohol guidelines for low-risk drinking (Draft Guidelines), subject to certain qualifications and concerns which the AMA invites the NHMRC to consider and address in finalising its recommendations.

The guidelines proposed in the Draft Guidelines are a significant revision of the ones published in 2001 in the Australian Alcohol Guidelines Health Risks and Benefits (2001 Guidelines). The Draft Guidelines are much simpler in three broad respects:

(a) they do not include guidance for levels of consumption above those constituting 'low risk';

(b) they refer only to daily, rather than weekly, consumption levels and make no reference to alcohol-free days or patterns of drinking generally; and

(c) they include 'health advice and precautions' rather than separate guidelines for specific situations and groups.

While there is merit in simplified guidelines which are easy to understand and recall, this should nonetheless be balanced against the importance of providing people with the amount and type of information and advice they need to appropriately regulate their drinking behaviour in a range of possible situations. At some points, the proposed guidelines could include fuller guidance to professionals and the public, in the service of reducing the risk of alcohol related harm. There are also respects in which the notion of 'low-risk' adopted in the Draft Guidelines needs a stronger rationale. These issues are discussed below:

The notion of low-risk drinking

The Draft Guidelines are based on a conception of what counts as a 'low risk' in relation to the likelihood of a certain level of harms occurring. The Draft Guidelines characterise low-risk alcohol consumption as that which keeps the risk of a person acquiring alcohol related injuries and disease to a tolerably low level, and keeps the lifetime risk of dying from an alcohol related injury or disease to less than 1 in 100. Neither what constitutes a 'tolerable' level of injury and disease nor the rationale for the adoption of a <1 in 100 chance of death as a standard for 'low risk' is sufficiently discussed in the Draft Guidelines and supporting material. This is particularly problematic in the context of quite different standards and thresholds for low or acceptable risk for other health effects. For example, risk levels below 1 in 10 000 are often considered those for which regulatory intervention is not warranted; the range 1 death/10 000 - 1 death/1 million is often taken as 'generally acceptable risk', with the option that 1 in ten thousand may be exceeded in some cases (eg, as applied in the USA for environmental contaminants). Health initiatives that result in adverse complication rates of between 1:500 and 1:10 000 are regarded as likely to be 'too high'.¹ Section 2 of the Draft Guidelines states that most people accept a level of risk much higher than 1 in a million for 'lifestyle issues where they have some personal control' (p.23). Even given this, a lifetime risk of death from alcohol-related injury or disease of 1 in 100 seems particularly high and would benefit from further discussion

Differences in interpretation of low or tolerable risk have led some bodies to nominate consumption levels different to those proposed by the NHMRC for low risk drinking. For example, the World Cancer Research Fund's 2007 report on cancer prevention has recommended, in view of the increased risk of breast cancer for women, a limit of two drinks per day for men but one for women. The discussion in the Draft Guidelines of cancer risks associated with alcohol is limited and may need to be updated.

The AMA believes that the Draft Guidelines would benefit from a fuller and more comparative discussion of the risk threshold upon which the guidelines are based.

Confining guidance to low-risk drinking

Unlike the 2001 Guidelines, the Draft Guidelines provide information and advice relating to low risk, but not 'risky' or 'high risk' drinking. To an extent, this may be a reflection of the evidence that the risks of injury, disease and death increase significantly after the nominated two drinks level (the risk curve being continuous and steep thereafter). However, a stated aim of the Guidelines is 'to inform policy-makers, planners, decision-makers, and those responsible for the provision of alcohol, who have a broader responsibility to the community and whose decisions may influence the health of communities'. The Guidelines could go further than simply stating that 'the increase in risk of harm depends on the extent to which intake exceeds the guideline level of consumption'. It would be useful for the Guidelines to provide certain other information beyond this one notion of increased risk, as follows:

(i) It is not implausible to suppose that a significant number of people are likely to exceed the recommended 'low risk' level, at least on some occasions. If alcohol guidelines are to be a tool for decision-making, to allow people to control the risks of their own and others' consumption, then information in the guidelines as to when consumption is definitely at high risk levels as opposed to merely risky, would seem to be essential.

The 2001 Guidelines provided this sort of information, and the AMA invites the NHMRC to consider revisiting this issue.

(ii) There is another important point at which the Draft Guidelines remain silent because of the exclusive focus on low-risk drinking, and this is in relation to draft Guideline 3. The AMA recognises that there is no safe drinking level for women who are pregnant, planning a pregnancy or who are breastfeeding. Guideline 3 is well-founded in advising that no drinking is the safest option.

However, the AMA is concerned that the guideline's advice to women in these circumstances includes nothing further. Despite the best of intentions and efforts, a zero alcohol intake may be difficult for some people to maintain, including some women who are pregnant, intending to be, or who are breast-feeding. It is possible also that some women in these circumstances may simply decide that they will from time to time drink some alcohol, or recognise it is likely they may. Some women may also find that they have consumed alcohol after they have unknowingly become pregnant. It is critical that women in all these cases should receive sound advice that is appropriate to their individual circumstances.

The AMA would urge the NHMRC to include as part of Guideline 3, that pregnant, intending and breastfeeding women should seek information and advice from their doctor as soon as possible if there is a chance that they have consumed or will consume alcohol in those circumstances.

(iii) The draft Guideline 2 for young people under the age of 18 does not acknowledge that there are adolescents who are very likely to consume alcohol in situations other than when they are under parental supervision. Many will experiment with alcohol in the company of their peers. No information is provided about how risks can be minimised in these situations.

The AMA believes the NHMRC should consider the inclusion in Guideline 2 of appropriate information and guidance for young people in these situations, and their educators.

Patterns of consumption (including binge drinking)

While the Draft Guidelines acknowledge the concept of 'patterns of drinking', they do not appear to use this concept in any meaningful way in discussing short-term or long-term risks. If patterns of consumption such as recommended weekly consumption, alcohol-free days, etc., are irrelevant to risk, then the Guidelines should make this clear. Otherwise, there is an expectation that they are relevant and meaningful, at least in terms of how people think about their own consumption and those of others. The Draft Guidelines define and dismiss 'binge drinking' in a few lines (p.19), whereas New Zealand's low-risk alcohol guidelines include recommended maximum consumption levels for a single drinking session.

It is also unclear why weekly consumption guidelines (provided in the 2001 Guidelines) have been omitted from the Draft Guidelines, especially given the currency of such levels in survey research and in the popular conception of how people think of their levels of alcohol consumption and may try to regulate them.

The AMA believes the Draft Guidelines would benefit from including more information about patterns of consumption and how these relate to various types of risks.

Specific groups and situations

There is question as to whether it is sufficient to provide 'health advice and precautions' rather than guidelines for particular risk situations (eg., alcohol in conjunction with other drugs) and especially for at-risk groups (eg., people with a mental health problem). The NHMRC has not, in our view, presented a convincing case for providing 'health advice and precautions' only, particularly in view of the risk that this advice will be seen as a footnote to the more weighty and authoritative 'universal' guidelines.

The AMA believes the information and advice for specific groups and situations is of sufficient importance to warrant its inclusion in, or as, specific guidelines.

The AMA also believes that the information provided to at-risk groups, and to those who should seek professional advice, should be a recommendation that they consult their doctor .

Further comments

There are a number of other observations that the NHMRC may wish to consider in refining the Draft Guidelines. These are as follows:

• The 2001 Guidelines contain a discussion about the term 'standard drink', noting concerns about the ease with which this term can be confused with a (usually larger) serving of a drink. The Draft Guidelines repeat many of the observations noted in the 2001 Guidelines in relation to the concept of a 'standard drink' being of questionable meaning when alcohol is not served in standard drink measures and when drinks are 'topped up'. The 2001 Guidelines note that 'this issue should be pursued and reconsidered next time these guidelines are reviewed, if not before.' However, this matter has not been taken up in the current draft. At the very least, any new Guidelines should include, and publicise, illustrative information of the type printed on the inside back cover of the 2001 Guidelines.
• The Draft Guidelines may benefit from the following:
  • more information and discussion about the relationship between alcohol consumption and the increased risks of certain cancers, as discussed in the World Cancer Research Fund's (WCRF) 2007 report on cancer prevention;
  • more information and discussion about the synergistic relationship between alcohol consumption and tobacco smoking for certain cancers, and
  • reference to the findings from the most recent (2005) Australian Secondary School Alcohol and Drug Survey (ASSAD); and
• It may be worthwhile for the NHMRC to consult with relevant organisations, including the Australian Institute of Health and Welfare and the Australian Bureau of Statistics, concerning the impact on surveys, data analysis and uniformity of data sets, as these are likely to be affected by the focus on and definition of 'low risk', to the exclusion of other categories, in the proposed guidelines..
• An evaluation strategy should be established which includes monitoring the impact of the Guidelines on the incidence of alcohol related harms.

Use of the Guidelines

Appropriately evidence-based alcohol guidelines can play a significant role in reducing the harms of alcohol use. As indicated earlier, however, the AMA believes that even the best guidelines will not achieve their full harm-reduction potential without accurate and informative labelling on alcohol. The AMA's view is that the required standard drink label is too small and can be easily hidden or at least hard to find. Understanding how much one is drinking is clearly fundamental to being able to follow any guidelines, so improving the visibility of these labels should be an important part of the education process.

The AMA also sees reason to have mandatory health warning information provided on alcohol containers as part of a comprehensive, multi-faceted public education strategy in the same way that health warning labels on tobacco products are mandatory, rotated, and supported through information campaigns.

References

1. FA Macrae , 2005, Screening for colorectal cancer: virtually there, Medical Journal of Australia; 182 (2): 52-53