June 1, 2022

expert reaction to conference presentation on alcohol consumption in older surgery patients and quality of life

A poster abstract presented at Euroanaesthesia 2022 looks at alcohol consumption in older surgery patients and quality of life.

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This is a good example of why it’s impossible to say much that’s meaningful about a conference presentation on the basis of what’s usually released before the conference. This might be a good, reasonably informative piece of research (though it certainly has important limitations) or it might not be good, and there’s really no way to tell from what has been released so far.

“A specific issue is that this appears to be an observational study, even though the patients involved were part of a clinical trial. There are differences between the groups of people who reported drinking different amounts of alcohol, apart from the amounts of alcohol they drank. Those differences could be the cause of the observed differences in quality of life, partly or completely, rather than quality of life having anything to do with the actual effects of alcohol. It’s possible to make statistical adjustments to take into account some of these other differences, though you can never be sure that everything relevant has been taken into account, so a single observational study can never really establish what’s causing what. So in assessing what a study might have shown, it’s important to look at what factors have been adjusted for, and how the adjustments were done.

“But we can’t do that, because the information hasn’t been made available. The press release tells us that some adjustments were made, and lists a few of the factors, but says that a ‘range of factors’ was involved. It does also point out that some potentially relevant factors, like socioeconomic status or levels of physical activity, couldn’t be taken into account at all, because the researchers had no data on them. And that’s all. The abstract of the conference presentation doesn’t even mention that any adjustment was done, and the abstract is the only part that would have been reviewed before acceptance of the research for the conference. The press release tells us that the paper describing the research has not yet been submitted to a journal, so there won’t have been the more thorough peer review that happens before a paper is published in a journal. There’s no mention at all of these adjustments in the poster from the conference. So I just can’t tell what the researchers did on this crucial aspect. I suppose I should be grateful that adjustments are at least mentioned in the press release, though there’s not enough there to assess things.

“The press release does also give a link to the test (AUDIT-C) that was used to assess alcohol use, and the abstract and poster say what cut-offs were used to distinguish between low (or no) alcohol consumption, and higher levels. A woman who scored 3 or more on the test, or a man who scored 4 or more, was classified as having “medium to potentially hazardous alcohol consumption.” To get a score of 3, a person could have reported that they drank alcohol typically two or three times a week in the past year, typically drank only one or two drinks on the days when they did drink alcohol, and had never had six or more drinks on one occasion in the past year. So a woman who has just one alcoholic drink twice a week would be in the group with “potentially unhealthy alcohol intake” – as indeed would be a woman who drinks far more than that. I think it’s hard to say how useful this division is, in this context.

“So we don’t know how good the study was in looking for patterns of cause, and the higher-risk alcohol category that they used was very broad indeed. Maybe the rather speculative suggestion in the press release, that “higher alcohol consumption may lead to elevated mood, enhanced sociability and reduced stress,” is indeed what’s going on, in part at least. Or maybe the cause and effect runs in the opposite direction – people who have a better quality of life in their later years might possibly be more likely to use alcohol, perhaps because they are in rather better health to begin with, or have a different kind of social life, that makes a certain level of alcohol consumption more likely. Possibly they mainly use alcohol in fairly small quantities but above the cut-offs that the researchers used. Or maybe the explanation is different – it could even have nothing at all to do with alcohol consumption. You just can’t tell, and there’s no point in me speculating further on the basis of such limited information.” 

Dr Tony Rao, Visiting Clinical Research Fellow, Institute of Psychiatry, Psychology and Neuroscience, said:

“This study is reported in a poster presentation and there does not appear to be a peer-reviewed study. The conclusion that higher alcohol consumption may lead to elevated mood, enhanced sociability and reduced stress is flawed.  It does not consider the possibility of “reverse causality” in that higher self-reported quality of life may be the cause of unhealthy alcohol use because those in this group can not only socialise more to drink greater amounts but may also be generally healthier and able to drink at these levels without consequent harmful effects on health and wellbeing. The study has borne this out with the finding that the group with unhealthy alcohol use had a lower average body mass index, higher level of education, better overall physical health and – for men – were younger than those drinking at low risk levels. Adjustment for these factors is reported but there is no statistical analysis to back this up. Cut-off points used to differentiate low risk and risky drinking are based on established research in this area in older populations. However, a major flaw in this study is that even low risk drinkers may be those “sick quitters” who have stopped drinking but still live with long term ill-health and poor quality of life.

“The study confirms what we already know about drinking in older people, which is that average alcohol consumption is higher for people in higher – compared with lower – socio-economic groups. Those living in more deprived areas have higher levels of poor general health and “deaths of despair” from accompanying physical and mental disorders. As this study does not assess deprivation, individual physical and mental disorders, and the impact of other substance use, it does not drive home the clear scientific message that alcohol is a direct cause of better health.” 

Poster abstract 13AP05-10 at Euroanaesthesia 2022. There is no paper at this stage and the research has not yet been submitted to a medical journal for publication.

Declared interests

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  My quote above is in my capacity as an independent professional statistician.”

Dr Tony Rao: “My potential COI is as a previous member of The Alcohol Health Alliance, current member of the Society for the Study of Addiction and have lobbied for a comprehensive government alcohol strategy. I am also Editor-in-Chief of the journal Advances in Dual Diagnosis.”