Science and Technology Committee - Minutes of EvidenceHC 1536 Alcohol Guidelines

Back to Report

Oral Evidence

Taken before the Science and Technology Committee

on Wednesday 12 October 2011

Members present:

Andrew Miller (Chair)

Gavin Barwell

Stephen McPartland

Stephen Metcalfe

Stephen Mosley

Pamela Nash

Graham Stringer

Roger Williams


Examination of Witnesses

Witnesses: Sir Ian Gilmore, Royal College of Physicians (by video-link), Dr Richard Harding, Member of the 1995 Interdepartmental Working Group on Sensible Drinking, Professor Nick Heather, Alcohol Research UK, and Dr Marsha Morgan Institute of Alcohol Studies, gave evidence.

Q1 Chair: Good morning, everyone-particularly to Sir Ian, who I hope can hear us in Australia.

Sir Ian Gilmore: Yes, Chairman, very clearly thank you.

Q2 Chair: We want to probe some issues around the current alcohol guidelines. First, what is their purpose? What do the Government seek to achieve? Who wants to start?

Dr Harding: It seems to me that the purpose of the alcohol guidelines is to give general advice to adults on the level of drinking at which no significant harm accrues, or the level at which there is likely benefit, based on an analysis of the most recent scientific evidence, in such a way that it is meaningful to them as individuals.

Professor Heather: First, the public have a right to information about what level of drinking increases the risk of harm. Of course they have a right to ignore that advice, but they certainly have a right to know what those levels are.

Drinking guidelines are also essential for a concept of something called hazardous drinking, which forms the basis of screening and brief interventions, a way of helping people who are drinking excessively to cut down; it is carried out by general practitioners and other generalists. The point is that, without what I might call official drinking guidelines, there would be no way of describing the concept of risky or hazardous drinking. They are essential for that purpose.

Finally, there is no research on whether drinking guidelines will lead directly to people cutting down on their drinking. There is no answer one way or the other on that, but that is not the point. Even if they did not directly lead to people reducing their consumption-I would be surprised if it did not-there would still be a role for drinking guidelines in reducing alcohol- related harm.

Chair: We can come back to that point in a moment.

Dr Morgan: I echo what my two colleagues have said. The Government have an obligation to provide, on the basis of the best evidence, information about the risks of alcohol intake so that the general public can make informed decisions. If we were working in a nudge framework, we would assume that given decent information, a reasonable man drinking outside those guidelines would moderate them, but that is not necessarily the purpose of the guidelines. Their purpose is to inform.

Picking up on Nick’s point, when you have identified low risk levels of consumption, then health care workers can use levels of consumption above that to classify drinking behaviour into hazardous and harmful. However, that is a different process. That is not a guideline; it is using it to determine classification of drinking behaviour that is above the recommended levels.

Q3 Chair: Do you wish to add anything, Sir Ian?

Sir Ian Gilmore: Just this: as well as using the best evidence available, the guidelines have to be understandable. They have to be a tool that is useful to health professionals in conveying the information, but the information must be presented in such a way that the general public have a reasonable chance of understanding it. It is up to individuals whether they act on it, but the guidelines must be in a form that allows a reasonable prospect of people engaging with them.

Q4 Chair: Are the guidelines set because of the need to improve the general health of the population, giving people advice about looking after their own health, or are they influenced by the social behaviour of people who drink excessively?

Sir Ian Gilmore: The aim of the guidelines must be to improve health, and that includes the health of the population as well as of individuals. Yes, I would agree with your summary.

Q5 Chair: Professor Heather, you said that there is no specific research about whether the guidelines influence behaviour, but you assume that they do. Would it be a valuable piece of research to see whether there is a cause and effect?

Professor Heather: It would be valuable but difficult. It would be hard to prove that there was a direct effect, but you could start by simply asking people whether they think that their drinking has been influenced. You could start from there and then find other ways.

Q6 Chair: Over the years, the various medicals that I have undertaken have included questions about alcohol, but there is a varied way in which doctors collect that information. Is there some merit in finding a consistent way of posing the questions and seeing whether there could be some interpretation of that information?

Professor Heather: There is a strong impetus to standardise that form of questioning, using particular screening instruments and particular forms of questioning. That is already under way. It would also be advantageous if people answered questions in the same way.

Dr Morgan: One of the difficulties is that it would have to be a two-tiered approach. If our basic premise is to provide guidelines in order to inform the public, we would first have to see whether they are actually informed. In other words, you would have to look at a scenario whereby you questioned a group of people, provided information on the guidelines and then revisited the matter. Running in parallel, or even sequentially, you would then look at individuals’ drinking behaviour and see whether the acquisition of knowledge had changed it. It would be a two-step procedure; whether it was done in parallel or sequentially would be up to the individuals designing the studies. It could be done, but it would be a difficult piece of research.

Q7 Stephen Mosley: Different countries take different approaches to this. For instance, Canada takes a zero net risk approach but Australia uses absolute risk. In the UK, we use the J-shaped curve to assess risk. Is there a best approach?

Professor Heather: The Canadian approach uses the J-shaped curve as well. Its zero net risk is assessed relative to abstainers, whereas the 1995 committee that Dr Harding was on-he will confirm this-took advice from Sir Richard Doll that the reference point should be the lowest point on the J-shaped curve. That is a technical difference. The Australian and Canadian exercises have their merits, and both are impressive pieces of work, but there is no need to go into an analysis of the underlying research again, because it has been done for us. How to interpret it would be different, but the analysis has been done both in Australia and Canada. I much prefer the Canadian approach, and my advice to the Committee is that the Canadian approach is the better way to go. I could tell you why if you want.

Stephen Mosley: Yes please; it would be useful.

Professor Heather: The Australian absolute risk approach, although it was very logical and although its time may come, has resulted in advice to the public that I think the public are not ready to accept. What it amounts to is that the guidelines come down to advising that people, both men and women, should never exceed 1 pint of normal strength beer or the equivalent a day. My view, and I hope yours too, is that the British public would find that incredible and derisory, however impeccable the logic.

The Canadian guidelines come up with much more credible advice. That takes us back to what Sir Ian said at the beginning-that the credibility of guidelines is most important. By the way, they are not so different from the guidelines that we have had since 1987, with the Royal College of Physicians report of 21, 14 and so on. Rather than them being plucked out of the air, those guidelines have stood the test of time. After the huge amount of research that has been done since, we have arrived at guidelines that are roughly similar to those of 1987-the Canadian ones.

Dr Morgan: There is a discrepancy between what the public want and what policy setters want. If you speak to epidemiologists-I am certain that you will in your further ferreting for information-you may hear that the advantage of absolute risk is that it is a cleaner piece of evidence on which policy makers can set policy, as opposed to the ease with which the population can be given sensible advice. Without going into the nuances, as Nick said, there is merit in both approaches. One approach may suit one group more readily than it suits another. There is merit in both, and running in parallel would probably not be a bad idea.1

Dr Harding: As you have seen from my submission, I am rather sceptical about the extent to which work in other countries is relevant to problems here. The reasons are that the problem that they are trying to solve might be different; they may be entirely focused on the reduction of misuse rather than general advice to the population; the target population might be different; you do not know what evidence they are taking into account and you might know the rationale of their analysis; and sometimes political judgment is involved, which introduces other factors. My view is that plenty of competent people in Britain do a very competent job.

Professor Heather: We obviously have competent people in Britain, but the evidence base is the same.

Sir Ian Gilmore: I am in Australia, and therefore feel duty-bound to comment that, as Marsha Morgan said, the absolute risk is the cleanest one. In many other areas of life, people accept that a risk of less than one in 100 of dying from something such as an environmental factor is reasonable.

The Australians have been careful not to say that it is an upper limit. They have said, "You should be aware that, if you drink more than two standard units a day, your risk of dying from an alcohol-related cause will rise." The risk can rise quite sharply above that level. They are not saying that you should never do it, but there is a good evidence base that people know that they are taking greater risks-and risks greater than they would accept in other walks of life-if they go above that level. We should not overplay the difference between the Canadian and Australian approach. The matter will be discussed in a forthcoming edition of Drug and Alcohol Review, and I believe that there will be a reasonable consensus.

Professor Heather: The evidence base was the same; if we had our own exercise in this country we would be looking at the same papers and the same meta-analyses. They are mostly from English-speaking Anglo-Saxon cultures around the world. That is where most of the research comes from, so it is biased to that extent, but it is not a different evidence base. We have experts in this country, but they would be looking at the same stuff.

The other point is that, if one decided on a relative risk approach for deciding guidelines for the general public, one could at the same time have a website where interested people could examine the absolute risk, based on gender, age and other characteristics, which would contain more detailed information. But for simplicity and credibility, in my opinion the relative risk approach is far better.

Q8 Graham Stringer: What are the key difficulties in assessing the risks of alcohol consumption? A GP who was an MP until the last election said that, when people came to his surgery who told him that they drank 10 pints a week, he wrote down 20. What are the difficulties in sussing the risk, given that aspect?

Dr Morgan: There are intrinsic difficulties in collecting data. In particular, you have highlighted the possibility of people under-reporting. That will always be a problem. However, the odd individual will over-report, usually young men.

There are also difficulties in the fact that you are collecting data over a relative period. A lot of the questioning is normally about what you are drinking now and takes no account of what you may have drunk in the past and therefore what burden of illness you may already be carrying. There are intrinsic difficulties not only in under-reporting but in the way in which the data are collected. As our Chairman mentioned earlier, there is the way in which you are questioned and the way in which the information is put down-for instance, whether it is frequency or amount. There are intrinsic difficulties in collecting the data and being sure of the data that you have collected.

Equally, there are difficulties in collecting information on the alleged harms. Unless someone has florid evidence of liver disease or pancreatic disease, you may not know that they are already damaged in that respect. I have spoken of the intrinsic overall difficulties of the collection of the data, but then we come to the interpretation of the data-the way in which the studies are done, whether they are case-controlled studies or cohort studies. Those are fairly well entrenched and well respected, in terms of the large epidemiological studies that have been done. The actual statistical analysis of the data, once collected, is fairly robust, but there are difficulties with its collection.

Dr Harding: There are hundreds of studies that show a correlation between alcohol consumption by individuals and chronic disease outcomes. Some of those diseases are underpinned by observations of changing biological markers as well, so there is a credible hypothesis as to why those effects actually happen. With the relationship between alcohol consumption and coronary heart disease there is, as you know, a decrease in risk, but it eventually goes up with consumption. When I was on the 1995 committee, there seemed to be two ways in which alcohol might do that. One was an increase in HDL cholesterol, and the other was the anti-clotting factor effect that alcohol has. There are now many ways in which researchers see that moderate consumption could have a beneficial effect, which is then seen in population studies. With other diseases, you see the relationship between consumption and disease outcome, but there is not yet the underpinning biological hypothesis to support it.

Sir Ian Gilmore: Some of us are less taken with that link, with the protection from cardio-vascular disease given by alcohol. There probably is an effect, but it does not affect the main age group that is damaged by alcohol. The peak deaths from alcohol are among 45 to 65-year-olds, who are in the most productive phase of their lives. Certainly young people damaged by alcohol get no cardio-vascular benefit whatsoever. There are serious scientists who still believe that the apparent cardio-vascular benefits are spurious. I am not going down that line, but I believe it is overplayed as a benefit.

Professor Heather: To answer your question, the main difficulty for the researcher is what are known technically as confounds-that is unknown factors that might influence the relationship between drinking and morality, heart disease and so on.

In relation to the cardio-protective effect, when the "Sensible Drinking" report was written in 1995, which was the last time that the Government addressed this problem, there was much more confidence in the cardio-protective effect, which is reflected in the report by the committee of which Dr Harding was a member, and also in the report of the three royal colleges that came out at roughly the same time. That consensus has now largely disappeared, which is the result of more careful research. For example, it shows that people who were classified as lifetime abstainers were not really lifetime abstainers. It is known as the "sick quitter" hypothesis; in other words, the objection is that people saying that they are abstainers may be abstaining because they have an alcohol problem or are unwell. The other kind of possible confound, which is now quite easy to understand, is that the moderate drinkers at the bottom of the J-shaped curve are in fact healthy, well-living people, and that low alcohol consumption is not so much a cause of longevity as a marker of a health lifestyle.

Q9 Graham Stringer: Is it a correlation rather than a causality?

Professor Heather: It is a confounder to the relationship. It is a difficult area of research, but it is clear now-I agree with Sir Ian on this-that there is far less confidence in the cardio-protective effect than there was a few years ago.

Dr Harding: I would not agree with that. It seems to me that the evidence has got stronger over the years. The "sick quitter" hypothesis is that the abstainers are unwell and therefore have a higher rate of disease. However, some studies have been large enough to take them out, yet when you omit the sick quitters and lifetime abstainers you still see the effect. In many studies, the confounding factor has been taken care of.

Chair: Shall we rest there, as there is clearly a difference of opinion? Any supportive evidence you have to support your views would be welcomed in further submissions.

Q10 Pamela Nash: The British public over the past few years have been bombarded with various news reports on the benefits and risks of alcohol consumption. What credible evidence has appeared since 1995 on the risks and benefits? As a result of any such evidence, should the guidelines be changed?

Dr Morgan: I do not want to revisit the cardio-vascular risk again, but may I explain that the biggest body of evidence on the potential beneficial effects of alcohol relates to the cardio-protective effect. However, as you can see, there is enormous contention among the team here and in general. Again, I emphasise that, if there were to be a cardio-protective effect, it would selectively be found in middle-aged men and post-menopausal women, and you do not gain that protective effect in middle life by drinking at a younger age. You cannot justify someone in that age group who does not drink starting to drink for the cardio-protective effect. After all, it is still within guidelines at between one and two units.

The two other areas where there have been alleged protective effects are in the development of diabetes and possibly, from the recent Million Women Study, on the development of cancers. However, the evidence is very thin. There is no body of evidence like that for the cardio-protective effect. Much more important, since the guidelines were last considered in 1995, is that the major body of evidence has been on the detrimental effect of alcohol and the cancer risk, particularly for breast cancer in women, and that the risk levels are not far off the top end of the current guidelines. Although there have been some reports in the press for a protective effect about diabetes and some types of cancer, there is not a strong evidence base.

Equally, there is much more important evidence that we did not have in 1995 which suggests a quite significant risk of cancer of the oropharynx, larynx and oesophagus and cancers among people who already have liver damage, and there is evidence on breast cancer and to a degree some early evidence on bowel cancer. As far as I am concerned, those detrimental effects overwhelm any potential benefit that there might be on diabetes.

Dr Harding: In my submission, I list a number of categories of evidence that have emerged since 1995. I shall go through them briefly.

The first was the finding that frequency of drinking is as important as the amount consumed, within the moderate consumption band. It is clear that drinkers get more benefit from not bingeing in any way, and from keeping consumption small and moderate. A nice study came out a couple of years ago comparing drinking levels in Northern Ireland and France. I believe that in France it was an average of 30 grams a day, and in Belfast it was 20 grams a day. However, because the pattern of drinking was less frequent in Belfast, the overall health outcome was worse there than it was in France, where they had a healthier pattern of drinking. The second was the evidence of the cardio-protective effect, which has already been mentioned.

Type 2 diabetes has been mentioned; the evidence was pretty clear in 1995, but it has been reinforced by further studies. There seems to be less osteoporosis in older people for moderate drinkers compared to abstainers, leading to an increase in bone mineral density and fewer fractures-an improvement of about 20%. Among older consumers there seems to be a reduced risk of dementia. A lot of work has been done on the effect of alcohol on cognitive function; again, there is about a 20% reduction for moderate drinkers over abstainers.

My colleague is correct that quite a lot of work has been done on cancer. The working party referred the whole question of cancer to the Committee on Carcinogenicity, mentioning that there seemed to be some evidence of an increased risk of breast cancer and that it should be kept under review. There has been quite a lot more work since. The Million Women Study that my colleague referred to showed an increased risk of breast cancer, and cancers of the oral cavity, the oesophagus and the larynx, which was expected, but a decreased risk of non-Hodgkin’s lymphoma, thyroid and renal carcinoma. The overall risk of cancer to women in that group was lower in moderate consumers than in abstainers. When it got to between seven and 14 drinks a week, however, it was about the same.

I urge the Committee to keep a sense of perspective. With breast cancer there seems to be an increasing risk of about 5% to 10% per drink per day. If a woman aged 20 has a lifetime risk, let us say, of 20%-I do not know the exact figure-of getting breast cancer before the age of 75, a drink per day would increase that risk to 21% or 22%. Up to the menopause, the risk is much lower-say 5% or 10%. If it was 5%, the increased risk would be 5.5%. If it was 10%, the increased risk would be 11%. After the menopause, women are exposed to the risk of cardio-vascular diseases; before that they are protected by their hormones. The benefit that they gain from moderate consumption after the menopause would outweigh any increased risk of cancer. They may not get cancer after the menopause even though there is more cancer then. It is not risk-free. There is a risk, but on balance the data show that if they wish to consume alcohol for health reasons the net effect is beneficial.

Q11 Pamela Nash: Would you advocate the guidelines being reinforced by the evidence base?

Dr Harding: I would not change them.

Professor Heather: In answer to your question, and as a remark to the Chairman, may I say that I know this is a question and answer session but I was hoping to make a number of points during the sitting? One of them is a direct answer to your question.

My advice is that under no circumstances should the Committee recommend that the guidelines are increased. That would be inimical to the health of the nation and wrong on the basis of scientific evidence. There is no case for increasing them. I mention this because, as you probably know, there was a press report saying that that would be the likely outcome of the Committee’s deliberations. I do not know where it came from, and I am not interested, but it was most unhelpful and wrong.

Chair: We, too, were fascinated, as nobody had bothered to ask us-and we have not yet heard the evidence.

Dr Morgan: I would be happy to submit a report from the International Agency for Research on Cancer. It was produced by 30 scientists from 10 countries meeting to consider the risk of alcohol consumption. It includes extensive evidence and the statement that there is clear evidence of about a 10% increase in the risk of cancer for every 10 grams drunk per day. I shall be very happy to submit that report.2

Q12 Pamela Nash: That would be very much appreciated. Sir Ian, do you have anything to add?

Sir Ian Gilmore: No, except that you are getting two different perspectives. I am one of the three independent public health and clinical opinions here. As someone who looks after patients with liver disease who sees hospital admissions rising year on year and now topping 1 million-that is not just presentations to A and E but over the whole hospital-I would very much echo Professor Heather’s comments that a recommendation to increase the limits would be swimming against the tide of harm that we see in our hospitals every day.

Q13 Pamela Nash: Thank you for that answer, Sir Ian. Do you have an opinion on how to improve the way that we communicate with the public about the various health benefits and risks if we advocate keeping the guidelines as they are? We cannot deny that confusing messages are being put out there. What would you suggest to improve that?

Sir Ian Gilmore: Although there is no strong sign or evidence that education and information have improved the situation with regard to alcohol, there are many areas where they have changed behaviour. Indeed, the drinks industry would not be spending all the money that it does on advertising if it did not work. The problem with alcohol is often that we have not gone about things the right way or with sufficient funding. It is a great disappointment to me that the present Government’s policy seems to be against funding public health information; they are devolving it to other organisations, including those funded by the drinks industry. We should not give up on education and information. It needs to be much better resourced, as we need a better evidence base on what works. We know that sending men in grey suits into schools does not work, but we must remember that peer pressure and peer influence does work. We should invest a lot more in independent advice and information to the public.

Professor Heather: Could I come back once more? I agree that the limits should stay roughly as they are, but only roughly. In other words, the way in which the limits are expressed could be improved. They say that men should not regularly exceed three or four daily and women two or three daily. I do not know whether this is an appropriate time to say so, but I shall quickly tell the Committee what I believe should be the case. There should be two kinds of limits. That is reflected in the Canadian guidelines. One is what people should usually drink on average, and that is relevant to chronic illnesses and is based on a particular set of evidence; there is another set of evidence that is relevant to harm from injury-

Chair: We will deal with binge drinking in detail in a moment.

Professor Heather: This will be relevant to that aspect, as it is related to intoxication. The limit should stipulate an amount that should never be exceeded3. To cut a long story short, my advice is that the guidelines should take this form. For example, men should not drink more than X units a week, probably 21, and never more than Y units in a day, whatever that might be-perhaps eight units, as at present, or a bit lower4. As well as that, there should be at least two days’ abstinence. We should revert to the old weekly limits of 21 and 14 for the average guideline, and have another daily limit that should never be exceeded on any day. That would help communication.

Dr Morgan: One of the other problems is that we are focusing-rightly, because it is the most important message-on the weekly or daily limit, whichever you want to take. However, the Canadian guideline model does not start with the amount that you should or should not drink. It starts by saying that there are circumstances when you should not drink-when using machinery or driving, if you are physically or mentally unwell, or if you have important decisions to make. It gives other guidance also to put into context the levels being suggested for weekly and daily consumption, for example that women should not drink if they are pregnant and that young people should limit their intake. One of the problems with the current UK guideline is that they just stipulate an amount that you may or may not be able to drink with low risk. There need to be more caveats, taking the Canadian as a model. That would serve people better as it will cover bases that are not covered by simply giving an amount.

Q14 Pamela Nash: I appreciate that, but do you think that it would be possible to do it without making things even more confusing?

Dr Morgan: There are always difficulties in messages that are longer. If you want to get one message over-Sir Ian has touched on this-correct labelling on bottles and information at places where alcohol is bought would be one good thing for pushing the message home. An informed public are possibly better able to understand what is going on.

Chair: Going a bit further into this aspect, I call Stephen.

Q15 Stephen Metcalfe: You have raised some interesting points. Do you not think that the public know full well what the safe limits are and that they choose not to abide by them because they do not like them? Most people have heard about 14 and 21, and about two or three units a day, but decide that it is not for them. My concern is that there is no connect between the public’s understanding of the risk and the message being communicated.

Professor Heather: I am sure that you are right that people know them but choose to ignore them. They have a perfect right to do so, of course, in a free society as long as they do not harm others.

Stephen Metcalfe: Yes.

Professor Heather: There might also be some confusion, for example, about the three and four units. A recent article in a British medical journal suggested that elements of the alcohol industry had put over the three and four units as being what the Government recommended people to drink. That, of course, is iniquitous. There is room for confusion.

Q16 Stephen Metcalfe: We have moved towards expressing the amount of alcohol in a daily way. Is that the right way to proceed? You suggested that it should be daily, I think, but there is still some debate on that.

Professor Heather: I am for a weekly average figure. At the moment, the guideline says that men, for example, should not regularly exceed three or four drinks. Unfortunately, as I wrote a couple of years ago, the word "regularly" in information given out by health authorities is sometimes dropped, so that it appears as an absolute maximum upper limit, which it was not intended to be. As I keep saying, it is intended as guidance on the average amount of consumption. For example, if your daughter was getting married you might want to exceed four units on that particular day-and on other fairly infrequent occasions. Why not? There is nothing wrong with it. The evidence suggests that the sky would not fall in if you did, and there must be some way of reflecting that. The weekly limit is a much more convenient way of doing that.

Dr Morgan: You asked whether the public were aware. In fact, a survey was carried out by the ONS in 2009, and the results were quite interesting. Basically, it showed that about 90% of the public have heard of the guidelines; that compares with about 79% in 1995. However, there was an enormous range between social demographic groups and age groups. In fact, the most poorly informed were those in the older age groups, but the managerial and professional classes obviously knew more. People have an awareness that there are limits out there; whether it goes beyond that to what the limits are and what they actually mean was not picked up in the survey, but there was 90% awareness in 2009 among the general public that there are limits for drinking.

Dr Harding: The 1995 report had a strong anti-intoxication theme. The guidance says don’t get drunk, don’t drink and drive, don’t drink and do sport, and don’t drink and operate machinery or whatever. That is one reason why it moved from a weekly advisory level to a daily one; under the weekly level, it was theoretically possible to put away a lot of alcohol in a couple of days and still be within the limit. That is clearly undesirable. Evidence that has come out since on the importance of frequency rather than amount also supports the concept of a daily advisory level. The committee did not like the idea of a limit because the evidence cannot support a limit, so it gave people a range. Also, a weekly limit is difficult to follow, because you have to remember what you drank last Thursday or whenever. There are lots of advantages for having an advisory daily level.

Dr Morgan: I was saying that you do not preclude it; you could do both.

Sir Ian Gilmore: I would go with Professor Heather and the weekly level if you had to go for one or the other, but the most important thing is that we should have a single, consistent message not only for the general public but for the health professionals, so that they know what message they should be conveying when they see their patients.

The pertinent point was made that most people think that these recommendations are for other people rather than for themselves. It is about getting clear guidelines but also backing them not only with public information and education but with GPs and other health care workers every time they come into contact with patients, taking a recent alcohol history and pointing out that it is possible to steer people back from drinking at hazardous levels by simple procedures such as brief interventions. Then there are the wider Government responsibilities over the most important drivers such as price, availability and marketing. It has to be taken in the wider context, and the most important thing is the clear, concise message that health professionals can share with the general public. The detail over weekly versus daily is less important. However, considering the burden of harm, it is clearly vital that the levels are not increased at this point.

Dr Harding: I make a general point here. It is important-it is the very first question that was posed-to decide the purpose of the drinking message. Is it there to help reduce alcohol-induced harm? The 1995 committee decided that measures should be put in place to help people drinking an excess to reduce it. That aside, what is the sensible level of drinking for the general population? Measures to reduce harm are one thing, but the basis of the recommendations is what is sensible to consume for the general population.

Q17 Stephen Metcalfe: I take that on board; you have made that fairly clear.

Professor Heather, everyone would agree, from what you have said, that it is less worse to partake regularly in a relatively small amount of alcohol than to binge drink, which is why Dr Harding is arguing for daily guidance. Would you confirm which is worse for your health, or which presents more risk-long-term drinking, your total consumption, or the pattern in which you drink?

Professor Heather: Both; they are different types of harm. Long-term average drinking is related to chronic illnesses. Binge drinking, if you want to use that term-I do not-leads to intoxication-related harms such as accidents and violence. There is also recent evidence that it contributes to hypertension, but that is a complication. What I said is roughly true. There are two types of harm. In my view, therefore, there should be two types of guideline.

Q18 Stephen Metcalfe: We touched upon this earlier, so we have already covered much of what I wanted to ask about the way in which the guidance was communicated in the past for those who have taken it on board. Is there any evidence that it has changed drinking patterns? Is there any authoritative evidence that it has changed the way in which people view alcohol and the way that they use it?

Professor Heather: Not directly, no. I said at the beginning that it is not evidence of absence but absence of evidence. There has not been the research.

Q19 Stephen Metcalfe: Even in your view.

Professor Heather: I would think it very unlikely that some people did not respond to guidelines by saying, "Oh dear, I am drinking above those guidelines. I must cut down." They may well be middle-class Guardian readers, if you like, but it is a matter for research.

Q20 Stephen Metcalfe: Does anyone else wish to comment on that?

Dr Morgan: It would be very difficult to look at trends, saying that we started with guidelines in 1987, tiddled around with them in 1995, and then say, "Look what an effect it has had", as so many other forces are driving total consumption. As Sir Ian mentioned, that is much more likely to be driven by policies on taxation, licensing and otherwise. Even if there were a trend within that relating to the guidelines, you would not see it.

Q21 Gavin Barwell: I have some questions about the specific guidance for women. We have been provided with a copy of the J-shaped curve, which shows a much larger confidence interval for women. The 1995 report that Dr Harding was involved with said that it was harder to produce an authoritative statement. What specific evidence has emerged since 1995 in relation to women and alcohol? You touched earlier on the question of cancer, but has it become easier to provide advice specifically for women?

Dr Harding: There are lots of studies on men and women. It is appropriate for women to be advised to drink less than men for two very good reasons. The first is that, on average, they have a lower body weight. Secondly, they have proportionately more fat in the body than water, and alcohol resides in the water, so at any given level of intake the alcohol concentration will tend to be greater. Indeed, in population studies that show benefit from moderate consumption and harm at high consumption, the tendency is for the beneficial effects to kick in earlier for women and the harmful effects to kick in earlier too.

Q22 Gavin Barwell: Dr Morgan, it seems that you might disagree with all three of those points.

Dr Morgan: Only a man would say that women have more body fat. The fact is that women have less body water. From a physiological point of view, the difference in how the body is made up between fat and water means that, if a woman of 70 kg drinks a double gin, a man of 70 kg would have to drink a triple gin to match her blood alcohol level. Her blood alcohol tends to be about a third higher on a weight-for-weight basis. The tissue dose of alcohol that she receives is clearly higher. Overall, the propensity for her to develop harm therefore kicks in earlier, after seemingly less alcohol. That is beautifully demonstrated in studies of the 1970s from Germany, where they looked at the risk of developing cirrhosis of the liver, which kicked in at as low as 20 grams of alcohol per day for women and at about 40 or 50 grams for men. There is a physiological basis to it, and there is epidemiological evidence showing that the risk of harm is higher. That was very much behind the 21:14 differential that the working party decided on in 1987.

Other factors clearly play a role for women, particularly in terms of liver injury. That may include hormonal status, which is why to a degree it is possible that if there is a protective effect it does not kick in until post-menopausal times. If you have a look at the study by White-you will have been provided with those curves-you can see that in both men and women you have an exponential increase in all-cause mortality in younger ages, and a suggestion of the J-shaped curve in men in later life, and later again in women. The most interesting thing, however, is the overall notch down in the women. There is a physiological basis for assuming that women are at a different risk, and there is epidemiological evidence that clearly shows that that is the case.

Q23 Gavin Barwell: Unless you have anything further to add-I am pressed for time-I have two further questions, and I shall ask them together. May I start with Sir Ian and Professor Heather?

First, given your agreement that there is evidence, I am interested to hear why a number of countries for which we have been given them have the same limits. We have been given data on the limits for a range of countries. Australia, Sweden, Spain, France, Italy and Romania all have the same limits for men and women. I am interested to hear the reasons for an international similarity in that regard.

Secondly, a report published in June by the Royal College of Psychiatrists suggests a much lower limit for older people and the same limit for men and women in later life. I would like to hear your comments on that.

Sir Ian Gilmore: Whether you differentiate between men and women depends on where you set your recommended levels. The higher the consumption, the greater the harm curves diverge between men and women. That is why, for example, using the Australian absolute risk of one in 100, with two standard drinks a day, they do not distinguish between men and women; they are quite happy if you go up to three drinks a day, but the risk of harm in women increases very much more than in men. That is one of the factors in determining whether or not you should set different levels for men and women. If we stick with the sort of levels used in the UK, it is important for the reasons that you have heard that we should distinguish the two.

The question on older people is interesting. There are many reasons why, in principle, one would expect older people to be more sensitive to the adverse effects of alcohol; that includes their propensity to fall, the fact that they have other diseases and may be on medication that changes their body water-fat balance. Yes, there is a rationale in setting lower limits, but you come back to the question of complexity. If you start saying that it should be different for men and women, different for people under 65 and over, different for pregnant and not pregnant women, and different for under age and over 18, you run the risk of getting to a level of complexity that will not be understood by the public. That is why many would argue that the Australian system’s relatively low level compared to some countries has the virtue of simplicity, but it also includes women and older people in the same guidelines.

Dr Morgan: Very quickly, we have considerable concerns about the report by the RCPsych. It was not an officially sanctioned report. We can provide you with a detailed critique of it. In essence, its findings were based on data for two years-2003 and 2008-and it said that there was a slight increase in people over 65 drinking above the recommended levels. However, you need at least three points for a trend, and it did not back up any of its data by showing differences in hospital trends for falls or problems in older people. It also misquoted the American units, so there are some real problems with that report. If you go back to White, you will see that, if anything, the units in the older people could be increased rather than decreased. I have definite concerns about the recommendations for older people.5

Chair: We would welcome that critique.

Dr Harding: I have already said that it will be difficult to draw anything from those comparisons unless you know the evidence base, the nature of the analysis, the type of groups and so forth. I was puzzled by the Royal College of Psychiatrists report because, looking at what I regard as the beneficial effects of moderate consumption on coronary heart disease, through to dementia and osteoporosis and so on, which are the diseases of ageing, it seems that there are considerable public health benefits to be gained in that age group by sensible consumption.

Q24 Roger Williams: I have a few questions about drinking during pregnancy and when breast feeding. Would it be your view that women should abstain absolutely during this period, and are there different risks at the different stages of pregnancy and during breast feeding?

Dr Morgan: There is no doubt that alcohol is a teratogen-in other words, it damages the foetus. The bottom line, in terms of the time span, is that a body of evidence shows that drinking pre-conception and at the time of conception is more harmful to the infant than drinking in later pregnancy, which is why it was originally suggested that you should not drink in the first three months of pregnancy but that later perhaps one or two drinks once or twice a week would possibly do no harm. We do not have a secure evidence base for that. That is why the Canadians in particular, who have a huge public health campaign on the subject, have suggested that the safest thing would be not to drink during pregnancy. It is a less secure evidence base, but it is a pragmatic and sensible approach that women should, by and large, be advised not to drink when pregnant. A lot of women do not need that advice, because losing the taste for alcohol is one of those things that alerts them to the fact that they are pregnant. It is the safest option rather than necessarily the strongest evidence base.

Professor Heather: That is, of course, our own Chief Medical Officer’s advice too.

Q25 Roger Williams: Would you like to say something about breast feeding?

Dr Morgan: From the practical point of view, a woman who has been drinking who breast feeds conjures up all sorts of difficulties. I would not want a woman who has been drinking to be actively breast feeding. In terms of the practical approach, however, the amount of alcohol that would get into the breast milk and into the baby is quite small, because it is heavily diluted within the woman’s system. The amount that goes through in the breast milk is similar to the quantity in blood, and that is then diluted within the baby’s system. On a theoretical basis, the amount of alcohol that the baby would receive is probably going to be relatively small, but for me there is no reason for saying other than that if you are breast feeding it would be better not to drink.

Q26 Roger Williams: In the generality, rather than referring to pregnancy, is there any way in which we can identify individuals who are more susceptible to harm from alcohol? Individuals find the concept of risk difficult.

Dr Morgan: A great deal of work is ongoing on the genetic factors that may predispose people to the development of alcohol dependence and alcohol-related harm such as liver disease and pancreatitis. We know that the hereditability of alcohol dependence within families is as high as 50%. We know that if you have a first-degree relative who is an alcohol misuser, you have possibly about a one in six chance of developing it yourself. It may be possible eventually, using genetic markers, to predict, "If you drink you will get liver disease, but you won’t get pancreatitis", but what the value of that is in the bigger picture I do not know. If you have a family history, particularly if you have affected first-degree relatives, being aware that you yourself are more susceptible is a possible strong message.

Professor Heather: There are lots of risk factors-individual personality, and genetic and social factors. For example, socio-economic status is a big risk factor for alcohol-related harm. Recent research shows that middle-aged men in the lowest quintile had a four times higher rate of alcoholic liver cirrhosis than those in the highest socio-economic status quintile. That cannot be explained by differences in consumption. There are lots of risk factors, but they cannot all be incorporated into guidelines, as it would make them immensely complex.

Q27 Stephen McPartland: Sir Ian, do you believe that the Government are using evidence from a wide enough range of scientific sources?

Sir Ian Gilmore: They do not use a sufficient evidence base when it comes to developing alcohol policy. That evidence base can come from a wide range of sources, whether it is social sciences, clinical sciences or basic sciences. The problem that I have is that the evidence is out there on what will reduce alcohol-related harm, but it is true that we need to persuade the Government to use that evidence. I understand that there are problems for Governments in terms of things that may not be popular with voters. It is interesting that where the greatest harm is seen in the UK, in Scotland, is where the Government are furthest ahead in taking evidence-based action.

Q28 Stephen McPartland: Do the rest of you agree with that?

Dr Morgan: Yes.

Professor Heather: Entirely.

Q29 Chair: This has been a helpful session. Clinicians such as yourselves will doubtless be asked when you go to parties whether the guidelines or the research should be considered as the most important. Are you personally affected by research that you have seen? Has it influenced your own behaviour? At the end of the day, the Committee trusts the science community, and it is interesting to see how you respond to research that you have seen. Has any particular piece of research influenced your personal behaviour?

Dr Morgan: I have never been a particularly heavy drinker; even at university I was not a particular drinker. The only thing that would have influenced me as a woman would have been the data on breast cancer. My own drinking levels were below that level, but what most influenced my friends was the data on breast cancer in women.

Professor Heather: I drink less than I used to when I was a young man, but I am not sure why. Perhaps it is because of my research, but it may be because I am older.

Chair: Older and wiser.

Professor Heather: Is it possible to make a couple more points?

Q30 Chair: We are very pressed for time. If you wish to make any additional points I should be grateful if you were to follow them up in writing, as we are conscious that another panel is waiting. I ask for a final comment from Australia.

Sir Ian Gilmore: It is fair to say that the medical profession does not have a good reputation with alcohol, from medical school onwards. It is interesting that doctors used to be near the top of the list of occupations suffering from alcohol-related deaths, but the most recent statistics show that they have dropped almost to the bottom. I hope that it means that doctors are beginning to take the lead, rather as they did in stopping smoking. I hope that it shows that doctors are beginning to moderate their consumption in line with the evidence. There are other explanations of why doctors’ mortality from alcohol has dropped, but I shall leave it at that.

Dr Harding: I take advantage of the evidence of moderate consumption.

Chair: We all consume moderately. I thank you all very much for your frank answers. This has been a very helpful session. Sir Ian, I understand that you are staying online in order to listen to the rest of the evidence; I hope that the line stays properly connected. Thank you all for joining us.

Examination of Witnesses

Witnesses: Jeremy Beadles, Chief Executive, Wine and Spirit Trade Association, Professor Averil Mansfield, British Medical Association, and Chris Sorek, Chief Executive Officer, Drinkaware, gave evidence.

Q31 Chair: I welcome the three of you. For the record, would you introduce yourselves?

Jeremy Beadles: My name is Jeremy Beadles. I am chief executive of the Wine and Spirit Trade Association. I am here to speak on behalf of four drinks industry trade associations-ourselves, the National Association of Cider Makers, the British Beer and Pub Association and the Scotch Whisky Association. Between us, I think we represent a substantial part of the alcohol drinks industry.

Professor Mansfield: I am Averil Mansfield. At core, I am a surgeon but I am here as chairman of the science board of the British Medical Association. I am not all that familiar with the world of alcohol but I shall do my best to be as up-to-date as possible.

Chris Sorek: My name is Chris Sorek. I am the chief executive of Drinkaware, an independent charity fully funded by the drinks industry. It acts as a blind trust for the money that comes in. We are governed by an equal number of public health professionals and people from the industry, along with independents, to ensure that the governance model of the charity is totally independent.

Q32 Chair: I want to probe your perspective on the purpose of Government guidance on alcohol consumption. Do you think that the public generally understand the guidelines?

Jeremy Beadles: The guidelines were introduced to provide some guidance to the public about safe and responsible levels of drinking. There is a growing understanding of what they mean and what units mean. If you track ONS data over the last 15 years, the data indicate that more and more people do understand. As we understand better how to pass on that message, particularly in the work of Drinkaware and what is being done by the Government, we seem to be getting more understanding and acceptance of the guidelines.

Professor Mansfield: So far as I can judge it, I share the view that the guidelines are understood. Their availability may not be quite as good as it should be, but they are certainly better understood.

Chris Sorek: I would agree. The guidelines are there to give consumers an idea of what their responsible drinking limit should be. They act as good guidance, but the ability of people to take them up and use them still needs to be discussed.

Q33 Chair: What are the most important factors in driving consumer behaviour in this area?

Professor Mansfield: From my point of view, it is the health factors that you were discussing earlier; I very much enjoyed listening to that first session. Most consumers are now aware of and are concerned about the effect that alcohol will have on their health. There is an increasing awareness of that among the general public, but the knowledge is not always followed that they need to reduce their drinking because it is dangerous. None the less, the knowledge is there.

Chris Sorek: Following on from that, we know that consumers, especially adults of 30-plus, consider health risks to be the most important reason to moderate their drinking; hence, it is the message that resonates highest with them. We also know that health risks resonate very well with parents of young children. Those messages that we communicate to adults and parents on health are the most important of what we try to communicate. Getting to them and targeting them on health issues is a different matter, but talking to them about health is what seems to resonate best.

Jeremy Beadles: I agree with my colleagues. The only additional factor is that the availability of the information is increasing at a considerable rate because both industry and Government understand that if you want the public to take the message on board it cannot be delivered only through doctors’ surgeries and hospitals. The message needs to be delivered in places that they visit regularly, in places where they buy alcohol and places where they consume it. In terms of understanding how better to deliver that message, the industry and the Government and Drinkaware work together to ensure consistency of approach and consistency of message. We need a breadth of delivery into the marketplace that we all use day in day out, rather than restricting it to hospitals and doctors’ surgeries, places where we hopefully do not go too regularly.

Q34 Chair: I am curious that the three of you, particularly having listened to the previous session, have not said that price affects behaviour. If it does not, I shall talk to the Chancellor; he has a big income stream from it.

Jeremy Beadles: My perspective on price is obviously that there is a huge amount to be said about alcohol pricing. Without doubt there is a link between the price at which you set a product and the sales of that product. If you put the price up the sales will drop. That is basic economics, and I would not disagree with that.

Where we take a different view is on how it impacts on harmful and hazardous drinkers or problem drinkers, and whether they are more or less affected by the price. I understand that price mechanisms have been used in other markets, particularly in places such as Sweden, where all the alcohol is sold through a monopoly structure. The national statistics there recording alcohol consumption may have gone down considerably, but our experience of that market is that a huge amount of alcohol comes across the border and that there is a huge amount of illegal brewing and distilling. Indeed, many people go out on to the Baltic sea on ferries at the weekend to get drunk. People who have visited Sweden see on those streets a lot of the issues that we have with regard to price correlation in Europe. Many of the countries with the fewest problems have the lowest price of alcohol. We in the UK market have an average price point for a bottle of wine at £4.84. In France, it is about £1.42 and it is less in Spain and Portugal. Price correlation is much more complex when it comes to tackling harmful drinkers rather than just acting on the consumption of the moderate consumer.

Professor Mansfield: The BMA stance would be that we need to take every possible step to reduce the amount of alcohol that people are taking. This dangerous situation, with so many people being made sick by the consumption of alcohol, needs to be tackled on many fronts, one of which must be pricing. That may not be the complete answer, and I am sure that what Jeremy Beadles has said is correct, but one of the prongs of the attack that we should be taking is to make alcohol more difficult to buy, especially for young people.

Chris Sorek: In terms of pricing and policy issues, we are a charity set up by Government and industry, and the public health community with Sir Ian Gilmore was at the front end of the creation of Drinkaware. We cannot comment on policy and policy-related issues. We provide information and education for consumers. There are two points to be made. There are issues around the supply side of the equation, which include pricing availability and licensing. We work on the demand side, which is when consumers go out to have a good time or may have too much to drink. That is what we work on.

Q35 Stephen McPartland: Is the use of alcohol units the best way to convey the alcoholic content of drinks, or do you think that the Government should promote the concept of litres of alcohol or something similar as a way of getting across how much alcohol is contained in drinks?

Professor Mansfield: It is pretty clear that the units that we have at present are as good a way as any of describing the amount of alcohol that we consume. A lot of effort has been put into making them understood by the general public. For better or worse, the message should be retained because it is now fairly widely understood. What is not quite so well appreciated is what each item of alcohol contains in units. One aspect that could certainly be improved is the availability of information about how much it contains. However, it is a good standard way of describing alcohol. The other ways, in milligrams or millilitres, are rather complicated, and we need something simple and straightforward.

Chris Sorek: The Government has put a lot of money into making sure that units is the best way in for people to understand how much they drink, and also the amount of alcohol that they are drinking on a regular basis.

As for the point that Averil made, one of the things that we, the British Beer and Pub Association, the Wine and Spirit Trade Association and the retailers, have come up with is a way to easily show people-this is going to appear across the country in pubs and the off trade-how many units go into people’s favourite units. The favourite drinks of the British public are a pint of beer, a 330 ml bottle of premium lager, a 175 ml glass of white wine and a 25 ml spirit measure. That will give people an idea of the units they consume. Using the mnemonic device "2-2-2-1", consumers will then immediately be able to say, "Let’s see what my favourite drink is." If they can figure that out against the unit guidelines-two to three for women and three to four for men-they can immediately make the connection between the two. That is the critical juncture that we seem to have been missing for the longest time. People talk about units but do not necessarily know what a unit looks like. We have seen this in research-and not only in our own; I believe that Government research said the same thing.

Jeremy Beadles: I agree with what has been said. The important thing is that we stick with what we have. Changing now would set us back a long way. There are issues of consistency with units of alcohol around the world, with different things in different markets. There is a consistency problem even in Europe, but we are where we are and the key is to ensure that people understand what is in their drinks. It is not just about point of sale communications; it is what is on labels as well that helps people define it. Progress was slow in the last part of the previous decade, but it has been kicked forward very fast in the last couple of years through the work of Drinkaware in partnership with the industry. Things are changing quite dramatically.

Professor Mansfield: May I add something? It would be very helpful if the Government were to say what should be happening with regard to the public’s knowledge of what is contained in every alcoholic drink. Instead of the inevitable influence of the industry-for better or worse, they have their own vested interests in making alcohol available to people-the standard should be set by the Government and not by the BMA or the industry. You should be the people who take on this problem, declaring what is needed on labels in pubs and on bottles wherever you go, but it is not there at the moment.

Jeremy Beadles: All of the stuff that we are doing is in agreement with the Department of Health. We would not put out messages on unit information or the chief medical officer’s advice without the Department’s agreement. It has been part of the process, so it is up to them.

Chris Sorek: The information that people get from the website-we regularly exceed 235,000 unique visitors to our website every month-is based on the guidance, and it is vetted by an independent medical panel of members of the British Medical Association.

Q36 Stephen McPartland: In previous answers, you mentioned the weekly and daily limits. Are those limits about right, or should they be increased or even lowered? Do you have any ideas on that aspect?

Chris Sorek: From Drinkaware’s perspective, whatever the guidance is and whatever Government provides is what we use and what we communicate to consumers.

Professor Mansfield: I listened to Ian Gilmore’s evidence and I share that view. Please do not put the limits up.

Q37 Graham Stringer: Do you have a view on whether advice to consumers should be based on weekly or daily consumption? Do you have any evidence that, when the advice changed in 1995 from weekly to daily, it changed consumer behaviour in any way?

Jeremy Beadles: As I said, it is not up to us to decide whether it should be weekly or daily guidance. That has to be a decision for the Government, based on the evidence available to them. I am not aware of any data that correlate to changing consumer patterns from that date.

Q38 Chair: The industry gave evidence to the Government during their consultations.

Jeremy Beadles: Not about the medical impact of drinking, no.

Professor Mansfield: With regard to daily or weekly limits-I would come down in favour of daily rather than weekly-what matters most is that the message should not be that you should drink two to three units a day. Somehow, we have to get the message over that you do not have to drink at all, and that you certainly should not drink at all on a couple of days a week. It almost gives the green light to go ahead and drink two, three or four units a day; the Government guidelines seem to indicate that that is okay. We need to tone that down so that people know it is the maximum and not something that is desirable every day, and it will not give you added health, but if they do consume that amount there will inevitably be a health risk.

Chris Sorek: The one thing I would say to that question is that whatever we put out-this echoes what we heard from the first panel-has to be clear, consistent and accessible. Those are the three things we need, whatever guidance is given. We have also heard this morning that some people are saying that there is a protective element to certain types of alcohol, but others were saying no. Consumers are looking for direction on which is right and which is wrong. It is the evidence that we communicate that people take on board and use. As long as it is clear, consistent and accurate, I doubt whether there will be a wrong conclusion.

Q39 Graham Stringer: Should advice be directed to demographic groups? In particular, we have heard that the behaviour of young people between 18 and 28 is very different from that of people between 35 and 50. Should advice be directed to those groups?

Professor Mansfield: I suggest that you need to monitor the advice. You need to modify it according to where the particular groups at risk are most to be found. For example, if you are talking about binge drinking that is a particular age group, is it not?

Q40 Graham Stringer: I do not know. Is it?

Professor Mansfield: I think so; I think it is very much the younger end of the spectrum, but I am no expert.

Chris Sorek: To the question about whether the message should be different, the answer is absolutely yes. We have done research across the United Kingdom that shows that adults of 30-plus respond better to the health-harm issues that need to be addressed. However, for younger people of 18 to 24, research shows that once they start drinking at 18 it is almost as if the lid has come off and they start drinking very quickly, and a lot of everything. We need to bring down and reduce the acceptability of being drunk and change people’s culture. Another of the things that we try to do, is to change their behaviour. We give them coping tips now. As we work through this year’s campaign and consider what we have done over the last two years, we are trying to give them information to change their behaviour. It is not necessarily just to be aware of issues but to change their behaviour so they do not go out and let a good night go bad. They should take a look at how much they are drinking and ensure that they do not go over that limit and all of a sudden get into trouble.

Q41 Graham Stringer: Can public information campaigns really compete with the subliminal messages you get in films such as "The Hangover", in which young people completely losing their memory is seen as having a good time? Can you compete with that?

Chris Sorek: Social marketing shows that there are some places where you can do that. It takes time; it is not going to happen overnight. It will continue to build up as you go through a number of campaigns to find out what messages resonate best with the target audience. I do not think that there is any one magic bullet that will solve the problem of episodic drinking at the weekends by young adults.

We have two campaigns going on right now. One is "Why let good times go bad?" The other is a campaign in which we are working with the Welsh Assembly Government to look at social norming. Basically, as we heard earlier, we are looking at peer group pressure to see whether there are ways in which we might be able to get young people to understand that their friends are not drinking as much as they think they are. What is really going on is that they really do not drink as much as their friends. This is clearly a big thing that we need to address, and that is one way of doing it. We are testing a number of methods of social marketing to change people’s attitudes to drinking and also to change their behaviour.

Q42 Roger Williams: What role should the drinks industry play in communicating alcohol guidelines? Perhaps we can break it up between manufacturers, retailers and hospitality outlets such as pubs, clubs and so on.

Jeremy Beadles: What role they should have or do have?

Roger Williams: What role they should have.

Jeremy Beadles: I start with the supply perspective. I recognise that retailers, too, are suppliers; 30% of the wine sold in the UK is own-brand wine from the supermarkets as is 30% of spirits. The role of suppliers in the first instance is to ensure clear labelling on the product and to ensure that unit information and the chief medical officer’s guidance is on the product.

Moving forward to the off-trade and the on-trade, there are different mechanisms for delivering messages in those two environments, and Chris is the expert on that. However, we think that a consistent unit message can be delivered across both. Three years ago, the WSTA worked with Drinkaware to put a campaign called "Know your drink" in supermarkets and off-licences. It got coverage in more than 10,000 stores, and evolutions of it have been running in most of the major retailers since then.

At the end of last year, with the British Beer and Pub Association and Drinkaware, we looked at the materials that they were producing for pub trade and the on-trade. That is a very different environment, and different guidance and different thought patterns are needed on how to influence behaviour for those sitting in the pub with a drink. We saw the work that they had done together on that, and our view is that it was better than the work that we had done previously on the off-trade. We felt that there was a lot more credibility in coming up with a consistent message across both environments, so that when people walk into a shop, a pub or a club, they see exactly the same message wherever they go. It becomes a message that they can understand, and "2-2-2-1" will be the strapline of the whole thing. We are working at this moment in time on rolling them out in both the off-trade and the on-trade, so you should shortly start seeing them everywhere, from beer mats to wine displays.

Beyond that, there are roles in terms of how we can use our websites, our digital communication routes and so on to communicate that message. The key thing is that all of those messages come through Drinkaware. That is why you will see the Drinkaware brand on pretty much every product in the UK marketplace, on every piece of advertising and on every piece of sponsorship. It is the most recognised charitable website brand in the UK marketplace, because it is seen on more than 3 billion products. The key is to drive a consistent message and to ensure that the drinks industry is delivering best practice; and it is delivered by Drinkaware and signed off by its medical panel and the Department of Health.

Professor Mansfield: I stress what I said earlier. It is important, and I am hopeful about it, that the Science and Technology Committee is seeking evidence from the medical profession and the drinks industry. I hope that, at the end of the day, you will not be dictated to by the drinks industry-whether neatly by the gentleman on my right or diluted through charity by the gentleman on my left. We need the Committee to consider the evidence and to declare what is the appropriate way to label things so that the public will know what they are getting. I promise you that, at the moment, it is not out there.

Chris Sorek: Echoing what we have heard so far, we need to make sure that consumers get clear, concise and accurate information, and that it comes well-evidenced. It should come from the chief medical officer and the Government, and that is what should be communicated.

Q43 Roger Williams: Should the Committee be concerned that in the public health Responsibility Deal that has been produced, the Government are putting more responsibility on the industry and less on statutory or public bodies?

Professor Mansfield: I have made my position pretty clear on that. That is not correct; that is not the right way to go about it.

Jeremy Beadles: As one of the co-chairs of the Alcohol Responsibility Deal, I have been intrinsically involved in the process since the start. My view is that it is not about setting or dictating Government policy; it is about the alcohol industry and other organisations finding ways of delivering things that the Government wish to have delivered, such as unit labelling and point of sale information. To be frank, it would be extremely time-consuming and costly putting it through Europe and getting the legislation out on the other side, and frankly impossible in terms of providing unit information in a pub environment. The evidence base would be extremely difficult to put together, and the cost of administering a scheme of that nature would be disproportionate. If the industry is prepared, willing and happy to do this stuff and can roll it out through its mechanisms, I am not sure that I see a problem.

This and the previous Government have been talking to industry about delivering these things. Labelling, point of sale information and community alcohol partnerships are all positives that everyone wants the industry to do, as far as I am aware, and we are doing it within the parameters set out in the Responsibility Deal. The Responsibility Deal does one thing above all else that has not been done in this sphere before: it requires individual companies to sign up and make a commitment about what they are going to do as businesses. It is not asking the industry to deliver; if it did, at that point you would always get three or four companies that did lots and lots of stuff while everyone else lagged behind. In this instance we have everyone doing it.

In terms of the labelling commitment, I have here the label that will be on products across the UK. If you turn around a bottle of Bollinger or Veuve Clicquot or Moët and Chandon on the shelf, you will find this label on it. One of the more difficult negotiations that I have had to do during my period in this job was persuading the French champagne houses that they should put that information on their bottles, but they have done so, and we now see the rest of the industry following. It is a large and diverse industry with many small companies, but particularly in the wine sector many tiny companies are involved. Getting to them all takes some time, but the Responsibility Deal now has more than 220 businesses signed up to it. It is one of the largest voluntary agreements ever put together.

Chris Sorek: As a charity, Drinkaware is basically a channel for communications for what might be done under the Responsibility Deal, and we co-ordinate closely with the Department of Health, the Home Office and the Department for Education on all of our communications.

Q44 Roger Williams: Finally, Professor Mansfield, you have made your position very clear. We have a BMA quote saying that industry self-regulation has at its heart a conflict of interest. Do you have any specific examples of that? Has its work been influenced by a concern for profitability or sustainability?

Professor Mansfield: I cannot say that I have direct evidence, but from what I what have heard it seems quite wrong. The word "voluntary" has been used repeatedly this morning. In the first session, you heard about the enormous harm that this drug, alcohol, is doing to the population. Most of us like it, but we have to know that there is a limit. Passing on information in a voluntary way is not strong enough.

Q45 Stephen McPartland: Mr Sorek, I am interested in what information should be printed on alcoholic beverage labels. You mentioned earlier the different things that apply, with different reactions for different socio-demographic groups. I wonder how many calories there are in a bottle of wine for those younger people who are weight conscious.

Chris Sorek: We were talking about that specifically, and earlier this year we did a mail drop of a unit calculator to 2.3 million households across the United Kingdom. The reaction was extremely good. It shows exactly how many units and how many calories are in your favourite drink. I am happy to supply you with one.

Stephen McPartland: Yes please.

Chris Sorek: Either that, or you can go to MyDrinkaware, a digital tool on our website. People can put in their favourite drink and find out whether they are drinking at a low level, an increasing level or a harmful level that is a high risk. People put in their drink of choice and can find out whether or not they are drinking at a limit that is above the guidelines, and whether it is at a rate that could harm them. It also tells them how much money they are spending-and, to answer your point, how many calories they are taking on board. You can find it out very quickly.

Q46 Stephen McPartland: I wonder whether you think that for some socio-demographic groups the number of calories would resonate more than one or two units.

Chris Sorek: Absolutely. Some people-for example, young women-respond very well to calorific content. They will do that almost immediately. Weight watchers will do the same. We find that those types of messages work extremely well with them, but that always leads them into a conversation about units and guidelines. Calories could be a way in for some people, and it definitely works.

Q47 Stephen McPartland: Mr Beadles, are those calorific amounts displayed on labels at the moment?

Jeremy Beadles: No; at this moment in time European legislation does not permit calorific information. Some businesses display it, but it is not a legal requirement and the means by which you have to present that information are not yet signed off by the Government. Don’t get me wrong; we believe that there is a route to providing calorie information. We support Chris in the work that he is doing, and one of the pilots that the Responsibility Deal is considering is how to convey that information to consumers in a way that is helpful to them. For instance, is it helpful to present consumers with a calorific content for an entire bottle of wine? Is that a helpful way of delivering the information? Is it helpful to say to consumers that one glass of wine has 110 calories and that another has 130 calories, or is it better to say that a standard glass of wine has so many calories as we do for units? When you start getting into the detail of calorific information, it is not a problem talking about calories. What matters is doing it in a way that improves consumer health.

Concerns have been raised by people around the Responsibility Deal table, the public health community included, about how that drives behaviour. You can reduce your calorie intake and increase your alcoholic intake at the same time; a neat spirit product is less calorific than a pint of beer or cider. The question is whether that is the sort of behaviour that you are seeking to bring through. What impact will it have, particularly on young people on a night out, if they know that there are lots of calories in the alcohol; should they eat before they go out or while they are out? We would like to see them doing those things. Should it be an either/or thing? The industry is certainly not opposed to the provision of such information. We already support its provision in some formats, but we would like to understand a little better how to standardise it. I am not sure, from a public health community perspective-this view has been put to me quite forcibly-that we would wish to get into a diet drink promotion and the idea that it is okay to have a Bacardi and Diet Coke rather than a Bacardi and Coke. For instance, would you be promoting something else by giving such calorie information?

Q48 Stephen McPartland: The label that you showed us earlier seemed to contain an awful lot of information in a very small space. Do you think that that is in the best format? Are we putting too many messages on that label?

Jeremy Beadles: The label here, which I hope is in your pack, has the three bits required by the Government as part of the Responsibility Deal; they are the UK units, the chief medical officer’s guidance and the pregnancy message. The Drinkaware logo and the "drink responsibly" message are optional, although most businesses include them. There is already a huge amount of information on the back of all bottles or cans. Simplifying it is always the absolute key. This is the format that the Government developed and have signed off with the Portman Group, and it is already to be found on tens of millions of products. To move away from it would be extremely problematic to the industry.

Q49 Stephen McPartland: How do you think the industry would react if it said on the front of the product, "Drinking this could be bad for your health"?

Jeremy Beadles: We would like to see the evidence before using it.

Q50 Stephen McPartland: If the Department of Health provided the scientific evidence to say that drinking it was bad for your health, would the industry go along with putting it on the front of their products?

Jeremy Beadles: On the front of their products? Voluntarily? I could not say at this point in time.

Q51 Stephen McPartland: My final question is to Professor Mansfield. In a written submission, you suggest that labels should contain warnings about the risks of exceeding guidelines. What evidence do you have on the impact?

Professor Mansfield: I did not hear the question.

Stephen McPartland: The BMA has made a written submission suggesting that labels should contain warnings about the risk of exceeding guidelines. What evidence is there that that would have an impact on drinking behaviour?

Professor Mansfield: I have no evidence to support the idea that that would change behaviour, but at least it would give them the opportunity to change, perhaps with a little more knowledge than is currently available. Although we have heard from the industry that the knowledge is on every bottle, it has not yet achieved that aim. As I said earlier, it has to be pushed upon the industry-by you.

Jeremy Beadles: May I clarify that? I was not saying that it had been achieved on every bottle. We are working towards a target of 80% of products on shelves by 2013. There will always be some products in this market-small volume products-that do not have it under a voluntary agreement, as it would put small wine merchants out of business. You cannot require Italian or French producers, who may produce only a couple of thousand bottles of wine a year, to label specifically for our market. If you do that, the product will never come here. It is about the majority of products having it, and certainly the vast majority that most consumers see.

Chair: You have provided some graphics in evidence, and screen dumps from the website; it would be helpful to have copies of those and of the cards that you circulated to households.

Q52 Stephen Metcalfe: I have some questions about drinking during pregnancy and understanding the risks. We have heard slightly mixed messages about whether it is safe to drink during pregnancy. Has that confused the public?

Professor Mansfield: I am certain that it has confused them. Yes, it may be possible to have a little alcohol, but the fundamental message that the medical profession would like to send is this. For women who are contemplating pregnancy-after all, as far as I can judge, the most important time is right at the very beginning, so it has to be the women who are thinking of becoming pregnant and not only those who are already pregnant-the simplest message is that if you are thinking of becoming pregnant or are already pregnant, alcohol is bad for you and for your baby. It should be completely avoided.

Chris Sorek: We followed the same guidance. On our website, the page on pregnancy and alcohol has received 55,000 hits over the last 12 months. I know that breast feeding came up in the previous panel discussion. We have had 19,000 pages views on breast feeding and alcohol. Both of these are fairly substantial increases-15% for the first and 49% for the second.

Q53 Stephen Metcalfe: Do you communicate the actual risk on the website, or does the guidance just say, "Do not drink when breast feeding"? Perhaps for some the guidance is fine; they just need to know that guidance is in place and that the advice is not to drink during pregnancy. Others, however, might want to understand the actual risk-what they are playing with, why there is a risk and what are the margins. We probably tend to understand that much better with smoking, but I am not sure that we understand about the risks to health of our drinking. I do not know what the risks are of my drinking. I know what the guidance says, but I do not know what my potential health impacts are. Perhaps we need to work a little more on that aspect.

Chris Sorek: On our website, we tell people what the potential risks may be, but we also signpost them to organisations and charities that have information and more detail; that is where they need to go for that. We also suggest that they might want to talk to their GP or their obstetrician.

Q54 Stephen Metcalfe: Do you accept that, in the wider world, there is no understanding of the actual risk but only of the fact that there is a risk?

Professor Mansfield: I am not sure about that. We need clarity in the message. People who are intelligent will delve deeper into the evidence, but overall we just need a clear message.

Chris Sorek: I concur 100%, yes.

Jeremy Beadles: We were asked by the Government to put out the message on our products that people should not drink when pregnant or trying to conceive. That is what we have done.

Q55 Stephen Metcalfe: Do you think that it is working? You said that you have had 55,000 hits.

Chris Sorek: We know that there is interest. Whether or not it is working is another question. These are all things that we will have to keep on considering and then go back to consumers and ask them.

Jeremy Beadles: The Department of Health research shows a high level of knowledge among women of issues around pregnancy and alcohol.

Q56 Stephen Metcalfe: Does anyone know what other countries do and how they handle the question of drinking during pregnancy?

Jeremy Beadles: I believe that we were following France with our message and with our pregnancy logo. I believe that Australia is now following us.

Q57 Chair: Earlier, we touched tangentially on what is the most effective way of communicating with the public. Has any work been done using new technologies to communicate some of these messages?

Chris Sorek: Yes, there has been. We also have low technology solutions. For example, we have a drinks cup, to help you figure out how many units are in the cup. We gave out close to 600,000 of these across the country, although most of them went to Scotland in support of Alcohol Awareness Week.

There are a number of others. In one of them, we produced a primary and secondary school resource based on "Unplugged", an EU-evaluated educational programme that has been running over the last seven years in 10 countries that showed substantial reductions of 25% to 35% in substance misuse by secondary school students. We took the same concept of developing a life-skills based approach; educationalists and academics created the programme and it is now being made available to schools across the country. The idea is not necessarily to talk to young people specifically about alcohol but to give them an understanding of how to do deal with issues in a way that builds up their confidence and their decision-making capabilities. Most of this is done in a digital format, although there is a low-tech version as well.

The work that we are doing in delivering messages through the internet is very targeted. It goes to parents’ websites and adult websites-I say "adult", but I must be careful in what I say; I mean websites that are oriented towards adults and their interests. We can target those specifically. For example, with the screenshots that I showed you a few moments ago, we had 80,000 people sign up to MyDrinkaware in a very short period literally by going online. We were delivering the message, "Are you interested in something like this? Please go to the website." They took it down, went through it and then signed up to the programme.

There are many ways to use the internet to reach out to people, through digital campaigns as well as through other information sources. We also use parent bloggers and parent networks to do that. Next week, we will announce a new parents campaign aimed specifically at parents of under-18s on how to give them the information that they need to be able to do so, rather than telling them what they are supposed to be doing as parents. As it turns out, Mumsnet-I am sure that most of you will have heard of it-along with Family Lives and other sites, are meeting us next week as part of the round table, when we shall talk about the best way to communicate via the internet to parents across the country.

Q58 Chair: Do you use social media?

Chris Sorek: Absolutely. We use Twitter, Facebook and all social media to get the message out, as well as blogs.

Q59 Chair: Have you made an analysis of the effectiveness of that kind of investment?

Chris Sorek: Indeed, and MyDrinkaware is a prime example. The campaign cost us approximately £90,000. We saw an immediate uptake, with about 50,000 people signing up to it. That gives you an idea of how fast it can work. Over the last few years, we have gone further and further with getting such things as "Why let good times go bad?" into the digital format, and also working with other applications. In the same way, we have seen general uptakes in people taking up our tips. Yes, it does work, but we are at the front end. As we go through the next few years and we evaluate things more and more, I am sure that we will find that it has an impact-but that is not to say that we should not be using low-tech versions as well.

Q60 Chair: Mr Beadles, that was an interesting series of answers from Mr Sorek, but some of these effective tools run counter to the interest of your industry. They will impact upon volumes of consumption.

Jeremy Beadles: The industry takes its social responsibility very seriously. We promote social and responsible drinking, and we do not encourage binge drinking. We would like all people to drink within Government guidelines. As for what we can do to support Chris’s work, Drinkaware does the work and it knows what works and what does not, and it asks us to amplify that using the routes to market that we have through our businesses. There is a huge level of investment beyond the direct funding of Drinkaware in how we amplify its messages, using all the routes open to us through out promotional organisations, producers, retailers and suppliers.

Professor Mansfield: These public health messages that we hear about, which are excellent, are inevitably linked with the drinks industry. Surely they should be public health coming from you and not from the industry.

Q61 Chair: We are Parliament, not Government. You are arguing that the public health campaigns that the Government run periodically on different topics should include a consistent one on alcohol.

Professor Mansfield: Yes please.

Q62 Chair: How would you handle that? How would you communicate the message? You have heard what Mr Sorek is doing in his use of modern tools. How would you get your message to the broadest base of the population?

Professor Mansfield: It would be foolish of me to suggest that I am an expert on that. I am not, but I have listened to good pieces of material that are being put out. I believe that it should be a public health message coming out rather than one from the drinks industry. There are many ways of doing it, but, as I say, I am not an expert and cannot speak to that subject. However, public health is really important.

Q63 Stephen Metcalfe: Do I take it that you agree with the message but that you are not keen on the messenger?

Professor Mansfield: The message is linked to alcohol. At all times, that potentially legitimises the use of alcohol by all these groups. I would like to see it completely dissociated from the drinks industry, in whatever form it takes, so that we get the clear message that it is your health that we are concerned about. There is no financial benefit to be gained, although the national health service would obviously benefit hugely if we did not have so much alcohol-related harm, but that is the aspect that concerns me.

Chair: Thank you very much indeed for your insight into this problem. Any further documents you have would be happily received. Thank you for your attendance this morning.

[1] Note by witness: See Attachment 1 [Ev 78 ] for a nalysis of the approaches used to derive the drinking guidelines in Canada and Australia; important clarification of the approach used by the Canadian workers which is said to be based on relative risk but does take baseline risk into account; both methods produce roughly the same end point

[2] Note by witness: See Attachment 2 [Ev 79 ] for a summary only as the full report is awaiting publication .

[3] Note by witness: On any one occasion or day.

[4] Note by witness: Within the context of the weekly limit.

[5] Note by witness: See Attachment 4 [Ev 79] for a critique of the RCPsych report in which a fundamental and important mistake was made in extrapolating data from the American NIH guidance for older people. The authors of the report recommend, based on this incorrect extrapolation, that daily alcohol limits in > 65 year olds should be reduced whereas correct extrapolation would support the opposite---in line with data produce by White et al BMJ 2002 .

Prepared 5th January 2012