All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts
BMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h384 (Published 10 February 2015) Cite this as: BMJ 2015;350:h384
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
The following is written in the Rapid Response by Dr. Saripanidis: “In Russia, resulting alcohol addiction and abuse claims the lives of almost half of the entire adult male population.” [1] This statement, wrong in my opinion, is corroborated by links to non-professional websites (with one exception), most of them 10 years old, but the above statement does not follow even from these sources. One of the non-professional articles is titled “Alcohol blamed for half of ’90s Russian deaths” [2], which is doubtful and not equivalent to “In Russia, resulting alcohol addiction and abuse claims the lives of almost half of the entire adult male population” [1].
Some papers create an impression that consumers deliberately purchase surrogates for drinking [3,4]; discussed in [5]. According to our observations and generally known facts, drinking of technical liquids and lotions has decreased abruptly after the end of the anti-alcohol campaign in 1989, when vodka, beer and other beverages have become easily available and relatively cheap. The only major exceptions are alcohol-containing liquids from the pharmacy. Many alcohol-dependent people go to the pharmacy not because of the lower price but as they hope to obtain quality alcohol i.e. better purified than vodka from the bottle store. It is a well known secret that diluted technical alcohol has been sold in vodka bottles through legally operating shops and eateries, added to beer, wine and other beverages since 1990. Converted to absolute alcohol, the tinctures from pharmacies are currently more expensive than the cheapest vodka [6]. In fact, the alcohol consumption and heavy binge drinking in Russia are decreasing; exaggeration of the alcohol-related mortality is sometimes used to veil shortages of public health and assistance. More details and references are in [5-7].
References
1. Saripanidis S. Latent alcohol mortality. BMJ Rapid Response 22 January 2017. http://www.bmj.com/content/350/bmj.h384/rr-7
2. http://www.nbcnews.com/id/31544292/ns/health-addictions/t/alcohol-blamed...
3. Khaltourina D, Korotayev A. Alcohol control policies and alcoholrelated mortality in Russia: reply to Razvodovsky and Nemtsov. Alcohol Alcohol 2016;51:628-9.
4. Khaltourina D, Korotayev A. Effects of specific alcohol control policy measures on alcohol-related mortality in Russia from 1998 to 2013. Alcohol Alcohol 2015;50: 588-601.
5. Jargin SV. Alcohol abuse and toxicity of alcoholic beverages in Russia: Recent history. ARC J Addiction 2016;1:21-29.
6. Jargin SV. Pine tree tapping in Siberia with special reference to alcohol consumption. J Addiction Prevention. 2017;5(1):3. (in press)
7. Jargin SV. Alcohol and Alcoholism in Russia: Insider’s Observations and Review of Literature. J Addiction Prevention. 2016;4(1): 6. http://www.avensonline.org/medical/addiction-prevention/archive/
Competing interests: No competing interests
I would like to remind distinct Colleagues that alcohol carries an inherent addictive potential that leads to ever increasing consumption.
In Russia, resulting alcohol addiction and abuse claims the lives of almost half of the entire adult male population.
References
https://www.theguardian.com/world/2007/jun/15/russia.science
https://www.newscientist.com/article/dn12071-cologne-and-cleaning-agents...
http://www.ibtimes.co.uk/former-soviet-republic-wants-ban-large-bottles-...
http://www.nbcnews.com/id/31544292/ns/health-addictions/t/alcohol-blamed...
http://www.thelancetnorway.com/journals/lancet/article/PIIS0140-6736(07)60922-2/fulltext
Competing interests: No competing interests
I concur with points made by others, and apply them to the design of a randomized trial of ethanol, as follows:
1) The pattern of ethanol consumption is important. In France, moderate amounts of wine are consumed with most meals. In England, ethanol is typically consumed in periodic weekend "binges", consisting of one or two extremely high spikes of serum ethanol concentration, separated by a 5 or 6 day "dry" period of no ethanol. Which pattern is likely to be more beneficial: platelet function slightly impaired by ethanol most of the time (French pattern) or severely impaired only when the likelihood of falling down drunk and sustaining a head injury is also maximized? Beneficial increases in HDL should also be sustained with less variation if the constant consumption pattern, rather than bingeing, is tested.
2) Benefit is more likely to be seen in patients at higher baseline risk of cardiovascular death, as with statins. A recently published meta-analysis of ethanol consumption in hypertensives confirmed that low-to-moderate intake was significantly associated with reduced overall and cardiovascular mortality for this population, in the J-shaped pattern. [1]
3) Persons who flush when they drink small amounts of ethanol should be excluded from an ethanol randomized trial, since they often harbor mutations in their alcohol dehydrogenase genes, and exhibit excessive risk and reduced benefit from moderate ethanol consumption. [2]
4) The dangers of ethanol addiction can be mitigated by excluding all persons with known or suspected alcoholism, or a family history of alcoholism, or any pre-existing liver or esophageal diseases. By the age of 40, when cardiac risk rises, most people who need to avoid ethanol are probably well aware of their tendency to abuse it.
5) A proposed experiment would administer capsules at bedtime containing an appropriate dose of a safe concentration of ethanol, or placebo capsules filled with water. [3] Both types of capsule would also contain mint oil or anise extracts to preserve blinding. Subjects would be non-drinking persons with elevated lipids and other cardiac risk factors, but no history of MI or alcohol abuse. They would be encouraged to continue to avoid other ethanol consumption for a year, or to record it in a diary.
Only with a randomized trial will we learn the answer to the question: Is low-dose ethanol safe and effective for reducing cardiac and all-cause mortality?
[1] Huang C, Zhan J, Liu YJ, Li DJ, Wang SQ, He QQ. Association between alcohol
consumption and risk of cardiovascular disease and all-cause mortality in
patients with hypertension: a meta-analysis of prospective cohort studies. Mayo
Clin Proc. 2014 Sep;89(9):1201-10. doi: 10.1016/j.mayocp.2014.05.014. Epub 2014
Aug 1. PubMed PMID: 25091872.
[2] Holmes MV, et al.; InterAct Consortium. Association between alcohol and
cardiovascular disease: Mendelian randomisation analysis based on individual
participant data. BMJ. 2014 Jul 10;349:g4164. doi: 10.1136/bmj.g4164. PubMed
PMID: 25011450; PubMed Central PMCID: PMC4091648.
[3] Keller DL. Ethanol should be subjected to a randomized controlled trial. Mayo
Clin Proc. 2015 Jan;90(1):160. doi: 10.1016/j.mayocp.2014.10.013. PubMed PMID:25572202.
Competing interests: No competing interests
Reading the article by Knott et al [1], it is clear that authors try to exercise prudence in the message they want to release to the general population. However, perhaps they are not supporting their message in the results they obtained (that in fact did show an apparent risk reduction among moderate drinkers).
On the one hand, it is widely accepted that alcohol intake is one of the leading risk factors accounting for morbidity and mortality. And it is also well-known that this is the result of an unbalanced risk-benefit equilibrium. Therefore, as in many other exposures, the effect of alcohol on health is very likely dependent on the dose. At certain alcohol intake levels the balance might incline to the benefit, but in some other levels it might tend to the harmful side of the balance. So caution is always needed when recommending to the population about drinking alcohol.
However, the solution is not to conservatively conclude from the data, but to conservatively give the recommendation to the population.
An undesirable conservative interpretation of their data would be to conclude that they found "evidence of absence" when they in fact may more likely be in the field of "absence of evidence", because they repeated their analyses within many small subgroups, thus reducing the statistical power for each of these comparisons [2].
Notwithstanding, the effect may be in fact different across subgroups of population. And in this respect, author’s contribution is remarkable. So, the message to give to the general population must be segmented and adapted to different subgroups. If there is an apparent and consistent risk reduction among moderate drinkers, a recommendation to heavy drinkers to reduce their alcohol intake to at least moderate levels is fully warranted. A prudent contention before recommending any alcohol consumption to abstainers is perhaps also warranted.
However, we are stuck in assessing the effect of the quantity or the frequency, but both are two dimensions of the same phenomenon. And moreover, the type of beverage, the consumption during meals or not… are some other important aspects of alcohol drinking. In short, to measure the whole pattern of consumption is a better approach for this problem, and few studies are currently using this approach [3].
In nutritional epidemiology, research moved from nutrients, to foods, and to the overall dietary pattern as the most appropriate approach to capture diet. This approach has proven to be tremendously advantageous [4].
Therefore we should move on to assess the overall alcohol drinking pattern to better understand the role of alcohol on health.
References:
[1] Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ 2015;350 doi:http://dx.doi.org/10.1136/bmj.h384.
[2] Altman DG, Bland JM. Statistics notes: Absence of evidence is not evidence of absence BMJ 1995; 311:485.
[3] Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 2002;13:3-9.
[4] Gea A, Bes-Rastrollo M, Toledo E, et al. Mediterranean alcohol-drinking pattern and mortality in the SUN (Seguimiento Universidad de Navarra) Project: a prospective cohort study. Br J Nutr 2014;111:1871-80.
Competing interests: No competing interests
Reading the recently published article: "All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts“, written by Knott and colleagues, reminded us of gravity of the situation in Croatian General Practice (GP) regarding mental and behavioural disorders due to use of alcohol. Looking at the morbidity patterns from the Croatian Health Service Yearbooks, Croatian Institute of Public Health, 1995 to 2012, we found a double increase in the total number of diagnoses registered in GP2. Almost a double increase was also observed regarding mental disorders (F00-F99, ICP-10). But, during the same period of time, the number of diagnoses of mental and behavioural disorders due to alcohol use (F10, ICD-10) decreased from 29 870 cases registered in 1995 to 21 077 cases registered in 2012. The share of alcohol related disorders in the total number of mental disorders decreased from 9% observed in 1998 to 3.1% observed in 2012. A decreased trend was observed in all age-groups; in age-group 7-19 years a decrease of 66.9%, in age-group 20-64 years a decrease of 20.5%, and in age-group over 65 years a decrease of 33.6%.
At the same time, the research results brought different picture. In the 1980s it was found excessive drinking in 15% of the male population and 6% of them were alcohol dependant3. Taking the absolute numbers, it was found that 250.000 people were excessively drinking in 1983 while in 2012 it was registered in GP only 21.077 cases3. Excessive drinking was observed in high school children (77.3% were temporarily or permanent alcohol users) as well as in the student population (15.1% excessive drinking) 4,5. Furthermore, hospitalisation due to the alcohol problems increased from 155.9/100 000 hospitalised patients in 1995 to 205.5/100 000 patients in 2012 [2].
Therefore, we can make a conclusion that alcoholism is a forgotten diagnosis in Croatia GP and that “no diagnosis means no treatment”. According to the mortality study results of Knott and colleagues, this conclusion should be seriously taken into account1. A reason for the Croatian results could be related to the phenomenon called "medical fashion“, meaning that some problems are in the focus of medical interest for some period of time and then the interest decreases. In the 1970s and 1980s, it was very strong movement in Croatia toward the establishment of "anonymous-alcoholic groups“, worldwide within the communities, initiated by professor Hudolin6. After his retirement, the number of groups rapidly decreased and very few of them are still working. It seams that the awareness of alcohol problem in GP were followed by this decrease. Furthermore, our experience emphasises once again the important roles professional leaders play in medicine.
References
01. Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ 2015;350 doi: http://dx.doi.org/10.1136/bmj.h384.
02. Croatian Institute of Public Health. Croatian Health Service Yearbooks, 1995-2012. Croatian Institute of Public Health: Zagreb, 1996 to 2013.
03. Hudolin V. Psychiatry and Neurology. (Psihijatrija i neurologija). Školska knjiga: Zagreb, 1983.
04. Samardžić S, Bujšić G, Kožul K, Tadijan D. Drinking in Adolescents. Coll Antropol 2011; 35(1):123-126.
05. Zloković J, Vrcelj S. Risk drinking in children and adolescents. (Rizična ponašanja djece i mladih). Odgojne znanosti 2010; 12 (1): 197-213.
06. Maljković I. Prevalence of alcoholism in Zadar County and experineces in leading the anonimus-alcoholic's groups. (Raširenost alkoholizma u Zadarskom području i njegovo suzbijanje). AL-klub 1988;7(8):8-14.
Competing interests: No competing interest at all!
Caution over association or absence of association in observational studies
We read the paper entitled “All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts” by Knott and colleagues with great interest.[1] The attenuation of protective effects of alcohol across most use categories on adjustment for a range of personal, socioeconomic, and lifestyle factors as well as after the exclusion of former drinkers highlighted an important and possibly intrinsic imitation of observational studies and a vital need to control or adjust for all known potential confounders before any attempt to infer on possible cause-effect relationship.
The authors are to be applauded for their attempt to control for “healthy survivor effect“ by choosing never-drinkers as the reference group: drinkers perceiving themselves as healthy are more likely to continue to drink and the generally observed higher mortality risks observed among former drinkers or those drinkers who did not drink within last 12 months across all the tables appear to support their choice.[1] However, alcohol use may be associated with smoking [2] and substance abuse [3], which could heavily confound the use of never drinkers as the reference group. Indeed, from Table 3 using never drinkers as the reference group, the additional adjustment for smoking and other potentially confounding variables in model 2 accentuates the apparent protective effects of drinking across most weekly consumption categories as compared to those found in model 1 (adjusted only for age) among males aged 65 or above, in sharp contrast to the general attenuation after adjustment for confounders using (current) non-drinkers as reference group (Table1). It would therefore appear that smoking and / or other potentially confounding variables could either mask a true effect or cause a spurious effect, depending on choice of the reference group. This, together with possible uncontrolled confounders (say substance abuse), residual confounding and interaction, may cast a serious doubt on the ability of the complex statistical analysis in overcoming the inherent limitations of the observational data.
If we look at biological plausibility, there is perhaps little reason to believe that moderate alcohol use could protect against all causes of mortality. To avoid searching a needle in a haystack, it might be a good idea to look at specific cause of mortality, say cardiovascular mortality, related to the controversy [4]. Furthermore, the division of the key variable of interest (weekly alcohol consumption) into excessive number of subcategories might also have incurred a heavy cost to the statistical power, despite the large datasets. For example, in Table 3, despite best estimated hazard ratios ranging from 0.49-0.72 across ALL of the seven weekly consumption categories (from < 1 occasion per month to > 20 unit per week) among those aged 50-64 in model 2, which would normally regarded as highly substantial risk reductions for mortality, NONE of the P value reaches statistical significance. With the highly consistent direction of effect across the subcategories, it might be prudent to merge some of the subcategories to dispel concern that the apparent loss of statistically significant association results not from the attenuating effects of confounders, but rather from excessive inflation in the degrees of freedom in the complex multivariate analysis.
While randomized controlled trials might overcome the various limitations of observation studies, including “confounding by indication”, ethical and logistic considerations are likely to preclude the conduction of such a human trial in the near future. Public health policy decision has, however, to be based on robust research to ensure that the potential benefits are not over promoted and risks downplayed. Alcohol has been classified as one of the carcinogens by the World Health Organization,[5] and evidences for the personal and social harms of alcohol misuse are overwhelming.[6] Caution must therefore be exercised against advocacy on the controversial cardiovascular health and mortality benefits of moderate alcohol consumption, particularly for the older population. Commercial health claims based on inconclusive medical evidence should be similarly regulated.
Dr Chi Chiu LEUNG, Consultant Chest Physician, Department of Health, Hong Kong, China
Professor Eliza Lai-yi WONG, Associate Professor, JC School of Public Health and Primary Care, The Chinese University of Hong Kong
Reference
1: Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ.2015; 350:h384.
2. Leung CC, Li T, Lam TH, Yew WW, Law WS, Tam CM, Chan WM, Chan CK, Ho KS, Chang KC. Smoking and tuberculosis among the elderly in Hong Kong. Am J Respir Crit Care Med. 2004;170:1027-33.
3. Jones CM, Paulozzi LJ, Mack KA; Centers for Disease Control and Prevention (CDC). Alcohol involvement in opioid pain reliever and benzodiazepine drug abuse-related emergency department visits and drug-related deaths - United States, 2010. MMWR Morb Mortal Wkly Rep. 2014;63:881-5.
4. Daube M. Alcohol's evaporating health benefits. BMJ. 2015;350:h407.
5. Cancer prevention by World Health Organization. Accessed at http://www.who.int/cancer/prevention. on 14th Feb 2015.
6. World Health Organization. Global strategy to reduce the harmful use of alcohol. World Health Organization 2010.
Competing interests: No competing interests
Multiple prospective studies have demonstrated that long-term moderate alcohol consumption is associated with a decreased risk for overall- as well as cardiac mortality (Camargo et al. 1997; Gaziano et al. 2000; Fuller 2011; Pai et al. 2012). This consistent J-shaped mortality curve observed for the relationship between alcohol intake and mortality not only confirms the deleterious effects of excessive consumption, but suggests that the implementation of cut-off limits for safe drinking habits could assist in improving health outcomes over time (White et al. 2002; Costanzo et al. 2010).
It is therefore noted with interest that a number of media sources have recently reported that these well-established benefits are “overblown” (Health Day; 10 February 2015) and “overestimated” (Medical News Today; 11 February 2015). These conclusions are derived from the interpretation of findings from a recent study conducted by Knott et al. (2015) titled “All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts” published in the British Medical Journal (BMJ).
In this article, the authors utilized data from the Health Survey for England pertaining to over 18 000 individuals in order to investigate the relationship between alcohol consumption and risk for overall mortality. In this study, mortality rates for non-drinkers and never-drinkers were compared to those for drinkers across different levels of alcohol consumption. Results were presented separately for study participants aged 50-64 and those aged 65 or older. The authors constructed two hazard models, the first adjusting only for age, and the second for other relevant covariates as well.
Compared to non-drinkers, the relative risk for mortality was <1 for both male and female drinkers across almost all levels of alcohol consumption. In considering the comparison between never-drinkers and drinkers, again, the relative risk for mortality in the majority of cases was still <1 for both males and female drinkers. In the latter analysis, a number of non-significant results are reported: the authors over-interpret the importance of these nominal differences in statistical significance for several reasons.
Firstly, it is emphasized that the lack of significant findings in some cases could be attributed to there being so few never-drinkers compared to non-drinkers considered as reference groups in this study. This leads to very wide confidence intervals and uncertainty regarding the interpretation of the true relative risk for mortality related to alcohol consumption. Secondly, due to the high number of separate groups defined according to different levels of alcohol consumption, the number of participants per group was small, which decreased the statistical power necessary to assess modest effects on mortality risk. Again, this complicates the correct interpretation of data reported, as only extreme effects on risk would achieve significance in an underpowered study. It is felt that the authors over-interpret a lack of significant findings in cases where stratification based on sex, age and different levels of alcohol consumption divide the total study group into a large number of small subgroups. It is indeed intuitive that sub-division into increasingly smaller groups would cause significant associations to disappear. In addition, the observation that the authors report a few groups with “significant effects” and claim that others show “little to no effect” is a classic example of statistical error which leads to incorrect conclusions drawn from the data.
The statement that “compared with never drinkers… [a] beneficial dose-response relations between alcohol consumption and all-cause mortality may be largely specific to women drinkers aged 65 years or more, with little to no protection present in other age-sex groups” is simply not supported by the data provided in this study. Rather, findings reported in Knott et al. [2015] are in accordance with those from previous prospective studies showing that moderate alcohol consumption decreases overall mortality risk, across boundaries for age and different categories of intake. It is difficult to understand how the media could have made concluded as stated in the press that “alcohol consumption has no health benefits” based on these results: findings from Knott et al. [2015] clear support the well-evidenced notion that people who consume moderate levels of alcohol live longer than non-drinkers and never-drinkers.
It is emphasized that emerging evidence supports the notion that genetic factors modify the cardio-protective effects attributable to moderate consumption of antioxidant-rich alcoholic beverages including red wine (Kotze et al. 2014). These observations suggest that certain genetic subgroups in the general population will not derive the same benefits from moderate alcohol consumption, highlighting the need for further investigation into the potential role of genetic testing in the development of personalized cut-off limits for safe drinking habits at the individual level. The translation of such insight into standardized healthcare policies could ensure that high-risk genetic subgroups are not exposed to the well-known deleterious effects of excessive alcohol intake, while still maximizing the potential benefits attributable to moderate consumption.
ACKNOWLEDGEMENTS
We gratefully acknowledge the financial support from Winetech and the Technology for Human Resources and Industry Program (THRIP).
REFERENCES
Camargo CA Jr, Hennekens CH, Gaziano JM, Glynn RJ, Manson JE, Stampfer MJ. Prospective study of moderate alcohol consumption and mortality in US male physicians. Arch Intern Med 1997; 157(1):79-85.
Costanzo S, Di Castelnuovo A, Donati MB, Iacoviello L, de Gaetano G. Alcohol consumption and mortality in patients with cardiovascular disease: a meta-analysis. J Am Coll Cardiol 2010; 55(13):1339-1347.
Fuller TD. Moderate alcohol consumption and the risk of mortality. Demography 2011; 48(3):1105-1125.
Gaziano JM, Gaziano TA, Glynn RJ, Sesso HD, Ajani UA, Stampfer MJ, Manson JE, Hennekens CH, Buring JE. Light-to-moderate alcohol consumption and mortality in the Physicians' Health Study enrollment cohort. J Am Coll Cardiol 2000; 35(1):96-105.
Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ 2015; 350:h384. doi: 10.1136/bmj.h384.
Kotze MJ, Marnewick JL, Kidd M, Fisher LR, van Velden DP. Assessment of the impact of hereditary factors on biochemical parameters of cardiovascular risk in relation to moderate alcohol consumption. Nutr Aging 2014: 189-195.
Pai JK, Mukamal KJ, Rimm EB. Long-term alcohol consumption in relation to all-cause and cardiovascular mortality among survivors of myocardial infarction: the Health Professionals Follow-up Study. Eur Heart J 2012; 33(13):1598-1605.
White IR, Altmann DR, Nanchahal K. Alcohol consumption and mortality: modelling risks for men and women at different ages. BMJ 2002; 325(7357):191.
Competing interests: No competing interests
It is good to see even the slightest support for the seemingly hopeless effort of making science prevail when commerce decides otherwise. But one small concern with this article was the reference to the association of lower mortality among older light drinkers in western nations as the 'protective effect'. This is what is in question and I am not sure we can appear to endorse that formulation or seemingly take it as a given, in an article that adds further doubt to the validity of the proposition.
The alleged causality beneath the association is of course not the subject of the study. But it cannot be assumed as established particularly because 'abstainers' are such a small minority in the countries where the studies are conducted over and over again (and highlighted in the mass media). This renders the ‘abstainer’ group abnormal to start with – and not only in ways controlled for in our traditional list of potential confounding factors.
The idea that alcohol is good for the heart is now widespread. If it indeed is, and we want to consume alcohol as a medicine, we should seek to take the minimal dose at the cheapest price. A generic medicinal alcohol should be allowed on the market if the medical profession is honestly convinced that a little alcohol is good for us (or at least for elderly westerners). And the profession should tell us whether a teaspoon a day is enough or not for the presumed benefits. Doctors should not collude with the alcohol trade to make us swig unnecessarily large quantities of expensive branded commercial beverages in the belief that we are taking medicine
Refs:
Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ 2015;350:h384
Competing interests: No competing interests
We welcome the study by Knott et al (2015). However, we are concerned over the misinterpretation of the suggestions regarding alcohol and harm in the 2011 Royal College of Psychiatrists Report (Crome et al, 2011a). Closer scrutiny of the text would have revealed that there were no specific recommendations, but suggestions for modifying UK safe drinking limits that were in line with US guidance (U.S. Department of Health and Human Services, 2005). The only ‘recommendation’ was that further consideration to be given to further examining the relationship between alcohol and harm.
In our publication on alcohol limits in older people (Crome et al 2011b), we recommended interrogation of large international databases as a method of identifying the relationship between alcohol and harm. We stated that any recommendation ‘is based on good clinical practice’ and that acquisition of the scientific base for developing globally uniform guidelines should be prioritised. In addition, we suggested a review of thresholds for harmful drinking; noting that for some older people, the best medical advice might be not to drink at all.
We note that the data source from the survey reported in this paper by Knott and colleagues is the non-institutionalized general population. The sample may also be biased by survival and selection factors: people who died prematurely from alcohol use are excluded. People with serious medical or psychiatric conditions are also excluded (some may be related to alcohol use).
As it stands, we are apprehensive about the authors’ interpretation of their findings. It is paramount that further refinement of biological and epidemiological research (as outlined in the limitations of the paper from Knott and colleagues) needs to be carried out before further advice is cautiously and prudently delivered.
REFERENCES
Crome I, Dar K, Janikiewicz S, Rao T, Tarbuck A. Our Invisible Addicts: First report of the Older Persons’ Substance Misuse Working Group of the Royal College of Psychiatrists. RCPsych, 2011.
Crome I, Li TK, Rao R, Wu, L-T. Alcohol limits in older people. Addiction 2012;107:1541-1543.
Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ 2015;350:h384
U.S. Department of Health and Human Services. Helping Patients Who Drink Too Much: A Clinician’s Guide. National Institute on Alcohol Abuse and Addiction, 2005
Competing interests: No competing interests
Reply to Dr Sergei V. Jargin
Trustworthy news agencies and newspapers, I cite as references, investigate, scrutinize, review all evidence before publication, and assume full responsibility for every disclosure they print.
Anyhow, it does not seem that Russian Federation ever officially doubted these alcohol mortality numbers, or filed a lawsuit against any one of these newspapers, for false publications.
Competing interests: No competing interests