Critique 071: Forum comments on proposed new dietary guidelines for Australia — 16 February 2012
Critique 071: Forum comments on proposed new dietary guidelines for Australia — 16 February 2012
Comments by the Forum on the draft critique of the chapter on alcohol intake of “A review of the evidence to address targeted questions to inform the revision of the Australian Dietary Guidelines”
Australian Government Department of Health and Ageing and National Health and Medical Research Council (NHMRC) November 2011, ISBN Online 1864965304
Summary: In 2008, the NHMRC commissioned the Dieticians Association of Australia to undertake systematic literature reviews to support the revision of the Dietary Guidelines for Australians. The primary aim was to undertake a series of systematic reviews of the national and international literature from the year 2002 on the food-diet-health-disease inter-relationship for different population subgroups. One of the 29 sections in the report (pp 613-678) covered the evidence for the risks and benefits of alcohol drinking. This critique is only of the alcohol section.
The systematic reviews were primarily conducted using the methods described in the NHMRC publication “How to use the evidence: assessment and application of scientific evidence”, and have resulted in body of evidence statements. Databases searched were CINAHL, PREMEDLINE, MEDLINE, EBM REVIEWS, DARE, COCHRANE, PUBMED, PSYCHINFO, ERIC and SCIENCE DIRECT. All reviewers were dieticians. The search considered only evidence published from 2002, to provide an update on literature published since the last edition of the Dietary Guidelines for Australians. The searches were mostly carried out to April 2009, so more recent publications are generally not included unless specifically requested by the NHMRC.
The report noted that when examining diet-health relationships, there is a notable dearth of evidence from Level I and Level II studies, and much of the scientific evidence is observational, especially from prospective cohort studies. It is rarely possible to conduct blinded intervention studies with whole foods or diets, and very few trials are conducted for long enough periods to assess long term health outcomes. Therefore in the review authors’ view, Level III prospective cohort studies often provide more important evidence for the development of dietary guidelines than Level I evidence summarising small short-term randomised controlled trials.
The review did not use cross-sectional epidemiological studies because of their low rating in the NHMRC evidence hierarchy (Level IV evidence) and because correlation cannot be used as evidence of causation. Only studies that provided total alcohol intake in grams, or enabled its calculation, were included.
Search results: The initial search of the databases revealed 1363 articles for alcohol and the specified disease outcomes. In all, 70 references concerning alcohol had data extracted and 53 of these papers met the selection criteria. Sufficient evidence was found to make body of evidence statements for alcohol and cardiovascular disease, type 2 diabetes, mental health and a range of cancers including breast, colorectal, oesophageal, oral and nasopharyngeal, renal, liver, ovarian, pancreatic and Non Hodgkins lymphoma. The search conclusions were:
|Condition||Quality grade of Evidence||Evidence statement|
|Cardiovascular disease||B||Consumption of alcohol regularly at an intake of 1 standard drink per day for women and 1.5-2 per day for men increases HDL cholesterol.|
|Type II diabetes mellitus||D||Alcohol intake of 1-3 standard drinks per day is not associated with risk of type 2 diabetes.|
|Mental health||C||Consumption of alcohol at the level of 1 standard drink per day for women and 1.5-2 per day for men, with a maximum intake of 4 standard drinks per day, is associated with reduced risk of dementia in older adults.|
|Breast cancer||B||Consumption of alcohol, even from low (10-15 g/d), is associated with increased risk of breast cancer.|
|Colorectal cancer||C||Consumption of alcohol, even at low levels (10g/day) of consumption, is associated with an increased risk of colon cancer and rectal cancer.|
|Oesophageal cancer||B||Consumption of alcohol is associated with increased risk of cancer of the oesophagus.|
|Oral cavity, pharynx and larynx||C||Consumption of alcohol is associated with an increased risk of cancer of the oral cavity, pharynx and larynx.|
|Renal cancer||D||Consumption of alcohol is associated with a reduced risk of developing renal cancer.|
|Liver cancer||C||Consumption of alcohol, even at low levels (10 g/d) isassociated with increased risk of liver cancer in some populations.|
|Pancreatic cancer||D||Consumption of high intakes of alcohol is associated with an increased risk of pancreatic cancer.|
|Non Hodgkins lymphoma||D||Consumption of alcohol is associated with reduced risk ofNon Hodgkins lymphoma.|
|Ovarian cancer||D||Consumption of alcohol is not associated with risk of ovarian cancer.|
The alcohol section of the report referred to the “NHMRC Guidelines to Reduce Health Risks from Drinking Alcohol”, Commonwealth of Australia, 2009. The conclusions of the former NHMRC review and the body of evidence for the current review are consistent for: cancer of the breast, oesophagus, oral cavity, pharynx and larynx, and liver, but were found to be inconsistent for heart disease and renal cancer; the former NHMRC review did not include studies on the positive effects of alcohol, and the current review found alcohol had a positive effect on both these disease outcomes although the introduction contained a statement suggesting that this is being challenged, and may not be observed in all populations. The conclusions of the former NHMRC review were also inconsistent with this review’s findings for colorectal and pancreatic cancer which were not listed as having increased risk with alcohol consumption in the former NHMRC review.
The alcohol section also referenced WCRF/ACIR, Food, Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective”, American Institute for Cancer Research, 2007.
The WCRF data contributes strongly to this review for specific cancers.
The search methodology used to create this review is well done and fits with best practice. The authors need to be commended on adopting a balanced view toward alcohol by considering both positive and negative health effects of alcohol.
- The lack of overt bias may have been due to the fact that all reviewers appear to have been dieticians. That may of course have introduced an inherent unknown bias from the fact that no other health disciplines had input into the process.
- An unintended bias may have been introduced by the reliance on the 2007 WCRF/ACIR study. Although the methodology and process may have appeared the same, without the reviewers of this study’s direct involvement there is always a risk that that assumption is incorrect.
- The review only considered literature from 2002 to 2009. Although this mostly applied to all conditions and should be applauded for consistency purposes, it has ignored seminal studies from outside that time frame that alter the quality of evidence grade and evidence statement.
- Evidence quality grades of D and possibly C suggest that the evidence for those statements may be so unreliable that it would be better to make no evidence statement at all for those conditions.
- We note that only health benefits and risks were considered. Alcohol is a complex subject and this review misses commentary on social benefits and risks.
- Some of the evidence statements are too simplistic. They ignore the fact that some benefits or risks of alcohol may only become relevant at high alcohol use levels, that there is a balance between risk and benefit that for any condition has to be considered in the context of any individuals non alcohol based risk factors, and that in some conditions the benefit or risk is very small in absolute terms. Indeed, throughout the text there is no clear delineation between the effects of light-to-moderate and heavy alcohol consumption, especially regarding increased risks of adverse health effects – the text refers throughout to “alcohol use or alcohol consumption”, where amount and pattern are not considered.
- A dual relationship exists between alcohol consumption and diabetes mellitus. Light to moderate drinking may be beneficial while heavy drinking is detrimental. (Baliunas et al., 2009 and cited in the World Health Organisation Global Status report on Alcohol and Health 2011). The grade of D for the evidence statement about alcohol and diabetes assigned in this review appears incorrect in light of these references and may reflect the fact that predominantly the review does not cover literature reported after 2009.
- The authors appear to have assumed any cardioprotective effect of the alcohol component was due only to the reduction in the concentration of HDL cholesterol observed following alcohol consumption. This could argue their search strategy for the cardioprotective effect of alcohol was too narrow as alcohol is also observed to produce a cardioprotective effect by a reduction in platelet stickiness and in the production of blood clotting proteins that reduces the potential for blood clots to form, as well as facilitating the breakdown of blood clots that have already formed. In addition alcohol has an endothelial anti inflammatory effect on for example, C-reactive protein (CRP). (Rimm E et al 1999. Wannamethee S et al 2003. Booyse F et al 2007. Pai J et al 2006. Imhof A et al 2001.)
- The evidence quality for risk of oesophageal cancer, especially in heavy alcohol drinkers appears correctly rated as B. The evidence quality for risk of oropharyngeal cancer, again especially in heavy alcohol drinkers, appears to be incorrectly rated as C, and should be rated as B. Again this review has not considered more recent literature, (World Health Organization International Agency for Research on Cancer (IARC). 2012)
- There is a balance between the risks and benefits of alcohol for any individual. The benefits appear to outweigh the risks for light to moderate drinkers older than the mid 50s. The risks (predominantly due to accidents and binge patterns of drinking both of which were not considered in this review) outweigh the benefits in youth. (Connor et al 2005).
On balance the authors should be commended on their work. Our criticisms are primarily about the restriction in the time frame of the review, and minimal consideration of key nuances, such as quantity of alcohol consumed and its relation to risk or benefit, patterns of drinking, the balance between risks and benefits, and making statements at all when the evidence is too weak. More details (e.g. considering cardiovascular mechanisms to widen the search) would have added weight to the conclusions.
References for Forum Comments
1 “A review of the evidence to address targeted questions to inform the revision of the Australian Dietary Guidelines”. Australian Government Department of Health and Ageing and National Health and Medical Research Council (NHMRC), Commonwealth of Australia 2011, http://consultations.nhmrc.gov.au/files/consultations/Evidence%20Report-%20For%20Public%20Consultation-%20Dec%202011.pdf
2 “How to use the evidence: assessment and application of scientific evidence”. National Health and Medical Research Council (NHMRC), Commonwealth of Australia 2000, http://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/cp69.pdf
3 “Guidelines to Reduce Health Risks from Drinking Alcohol”, National Health and Medical Research Council (NHMRC), Commonwealth of Australia, 2009, http://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/ds10-alcohol.pdf
4 WCRF/ACIR, “Food, Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective”, American Institute for Cancer Research, 2007
5 Baliunas DO, Taylor DO, Taylor BJ, et al. Alcohol as a risk factor for type 2 diabetes: A systematic review and meta-analysis. Diabetes Care 2009 Nov;32(11)2123-32.
6 Global Status Report on Alcohol and Health. World Health Organisation, Switzerland, 2011 http://www.who.int/substance_abuse/publications/global_alcohol_report/msbgsruprofiles.pdf
7 Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects on lipids and haemostatic factors. BMJ. 1999 Dec 11;319(7224):1523-8.
8 Wannamethee SG, Lowe GD, Shaper G, Whincup PH, Rumley A, Walker M, Lennon L. The effects of different alcoholic drinks on lipids, insulin and haemostatic and inflammatory markers in older men. Thromb Haemost. 2003 Dec;90(6):1080-7.
9 Booyse FM, Pan W, Grenett HE, Parks DA, Darley-Usmar VM, Bradley KM, Tabengwa EM. Mechanism by which alcohol and wine polyphenols affect coronary heart disease risk. Ann Epidemiol. 2007 May;17(5 Suppl):S24-31.
10 Pai JK, Hankinson SE, Thadhani R, Rifai N, Pischon T, Rimm EB. Moderate alcohol consumption and lower levels of inflammatory markers in US men and women. Atherosclerosis. 2006 May;186(1):113-20.
11 Imhof A, Froehlich M, Brenner H, Boeing H, Pepys MB, Koenig W. Effect of alcohol consumption on systemic markers of inflammation. Lancet. 2001 Mar 10;357(9258):763-7.
12 World Health Organization International Agency for Research on Cancer (IARC). (2012). Consumption of alcoholic beverages. In A review of human carcinogens: Personal habits and indoor combustions (IARC Monographs on the Evaluation of Carcinogenic Risks to Humans Vol. 100E, pp. 377–503). Lyon, France
13. Connor J, Broad J, Rehm J, Vander Hoorn S, Jackson R. The burden of death, disease, and disability due to alcohol in New Zealand. N Z Med J. 2005 Apr 15;118(1213):U1412
Comments prepared by selected members of the International Scientific Forum on Alcohol Research.
Ross McCormick PhD, MSC, MBChB, Associate Dean, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand.
Input from the following members of the Forum:
Creina Stockley, MSc, MBA, Clinical Pharmacology, Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia;
Harvey Finkel, MD, Hematology / Oncology, Boston University Medical Center, Boston, MA, USA;
Andrew L. Waterhouse, PhD, Marvin Sands Professor, Department of Viticulture and Enology, University of California, Davis; Davis, CA, USA; and
R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA.
Additional input from
Helena Conibear, Executive and Editorial Director, Alcohol-in-Moderation (AIM), Dorset, United Kingdom, and Co-Director, International Scientific Forum on Alcohol Research: