 Mae'r gweithio ddweud wedi cyd-dweud yn ymweld, a fyddwn i chi'n cael eu cael'r marran gyda'r newfgredd a'r llai'r hunain a'r llai'r hunain a'u'r hunain. Mae'n cael ei ddweud am gyllid yn ymweld yw'n cael eu hunain a'u'r hunain a'u'n cael eu hunain. Mae'n cael eu hirio'n ei chengen. Mae'n ddyn nhw'n ymdigech chi'n ei hwn i'r Unedigol Sinoesol a'r Unedigol Sinoesol a'r Unedigol Fylltwyd yn ymweld, Felly, mae gweithio i mi i gweithio i gweithio. Mae'r gweithio yn gweithio yn ei gweld. Mae'r gweithio yn gweithio yn gweithio. Felly, os ydych chi'n meddwl i chi ar y cyd-ddiweddau yma sy'n byddai'n gweithio gweithio gyda'r cyfriffau yng nghymru o gweithio yn fawr ei wnaeth gyda'r cyfrifos yn fawr ymlaenig i gdweithio i'r cyfrifos mewn gwahanol ar hyn o rhan o'i gweld. Roeddwn i chi'n gweld yn gweithio oherwydd, I hope I'm a bit of an expert in global health, as professor of international nutrition I ought to be, but I'm no expert in dairy, so it's been an interesting process for me to pull together the data that I'll show you in order to prepare this presentation. So I guess what you're going to get is a sort of generalist's view, someone who's relatively confident looking at PubMed and other relevant sources, rather than a deep specific expert view. So my disclosures are as follows, I sit on the Taneeta Medical Advisory Board, I teach pediatric courses for the Nestle Nutrition Institute, and I'm speaking today on behalf of Danone Institutes International. So let's think about dairy products in the human diet. We can track this back many, many thousands of years, and we can partially do that through these cave paintings and there are many, many more of these which give us quite a good description of the time course, although some of these are very difficult to date, but the time course of the process by which humans domesticated, milk producing animals and learnt that this was a pretty good trick to use the products of those animals and incorporate those into our diet. And I wanted to start with a few lessons from history. And I think this is really fascinating, this issue of the evolutionary origins of lactose digestion, because as most of you will know, lactose is a special sugar. We're not inherently designed to deal with it. The galactose, moiety of lactose is of course deemed to be important in terms of brain development, so it's potentially very important in the diet. And of course initially we get that from breast milk. So we've got domestication of milk yielding animals. Then we've got a parallel domestication. We don't quite know the dates of this, but we do know that the technology for fermenting milk products was probably around at least 8,000 years ago. So there are potsards, there are strainers that were deemed to be used for making yoghurt and cheese and analysis of residues on those show that indeed they do have milk fat products. So that in a way is the domestication of the lactic acid bacteria, which have special properties in terms of allowing us to keep milk longer without it spoiling. There are other aspects of that, which is the fact that they produce beta-galactosidase, which will help with the digestion of the lactase. But what to me as an amateur geneticist is really fascinating, is this issue of the evolution of lactase persistence within the human genome. So many of you will be familiar with this story. I think it's such a strong story that it's probably incorporated in many of the basic classes to any of you who have done a biological degree. So here we have one of many possible maps of, at the top there, the phenotype distribution of lactase persistence. So this is the ability to continue to digest lactose into later childhood and adult life. I'm sorry that America doesn't appear on this, you're not very important. But this is the old world distribution. And you can see that if we just look at the phenotype, there's a very strong gradient, particularly in north-west Europe, where lactose persistence is very, very strong. If we then map that onto just this, at the bottom there is just one particular genotype, one SNP variant, you can see that that maps really very strongly. But if you look at Africa and the Middle East and northern India, Pakistan, you'll see that there is a mismatch between those. And that's because the lactase persistence there is contributed by other polymorphisms. And here the story gets really, really interesting because we have very strong convergent evolution of lactase persistence. This is one of the strongest signals of recent genetic selection. We can show that this selection has occurred within about the last 200 to 300 generations. And it is penetrating the whole of the rest of the world. What's even more fascinating is that this lactase persistence has been independently evolved at least four different times. And that is shown by this famous paper from Sarah Tishkoff here, which shows you those four different variants in the different colours, A, B, C and F there, and shows you the different SNPs. Now these SNPs are all on a region of the promoter for the lactase gene. And they are on chromosome 2 and associated, incidentally, with some other genes that help to metabolise cholesterol effectively. And recent work from the Maasai shows that again these co-evolve and are under very strong recent selection. We can tell that because they're in vast haplotypes, so the haplotype hasn't had a chance to recombine. And so that allows us to date when these have occurred. So this to me is amazing. This is telling us that the ability to digest lactase into older childhood and adulthood has been incredibly important in our evolutionary history. Now of course there's controversy as to what the reason for that, what is behind this selection trait. And I think incidentally that remains a research need if we could understand what is the cause of that, why it is so beneficial, then we may be able to modulate that further to the benefit of human health. So we start off of course with breast is best, so the young child has well induced lactase in its villi in the gut. There's a biological clock though that allows this or causes this lactase to disappear over time. And we believe that that is important because the mum can't continue to feed the infant throughout its childhood that has to become a time when it's appropriate for her to move on. And some people argue that the disappearance of the lactase gene is actually or the lactate, not the disappearance of the gene, the disappearance of its induction is important in mediating weaning. And so what that does is if the child can then be moved on to other forms of milk then it emancipates the mothers. It means that they can wean the child and that will generally in traditional societies initiate a new reproductive cycle. So to me this is probably a very important part of the selection. Many of you probably think about survival of the fittest in terms of death and individual dying. In fact that is not the optimum way of looking at it, it's been shown for many many years in fact from Fisher in 1932 that evolutionary selection by fertility selection as opposed to viability selection is much much faster and stronger. So to me anything that is going to enable a mother to produce more offspring successfully is going to be a strong candidate on the causal pathway for this genetic adaptation. So what are some contraindications to dairy consumption and of course those are intimately related to the point we've just been making about at least the second of those about lactase persistence. So the two main issues would be cow's milk protein allergy and lactose intolerance and as I say I'm the warm up act you'll be hearing more about those later in the day. Let's have a few thoughts about the role of dairy products in meeting nutrient needs. Now of course it's true to say that it is possible to live without dairy products and many people do that. But many people consume dairy products. Now the first wonderful thing about them and of course this is a no brainer is that they are designed to both eggs and milk are designed to provide the entire nutrient requirements of young animals. And we as humans have done this clever trick of domesticating these and taking advantage of these wonderful sources of nutrients. So here we have one of many possible figures of the contribution of dairy foods to some key nutrient intakes and this is plotted for a population who would consume about 450 grams of milk per day or equivalent products. And what it shows is that milk that consumption of milk can make a very significant contribution to the overall daily needs. And particularly important is the fact that if you look at the energy there it provides only 10% of energy so they are very nutrient dense compared to the caloric intake. And in populations where we're so worried about obesity this becomes a very important attribute. Now of course in the west we would be very interested in the calcium concentration. Also the vitamin D and as you know many milks are fortified with vitamin D. But to me who works most of my time in rural Africa, riboflavin stands out as massive. We have huge levels of riboflavin deficiency in our population who consume very little milk. Milk is an enormous contributor to riboflavin and can really change between a mother being very riboflavin deficient or not. So Adam Dranowski has shown this, drawn up this schema which in a western society puts milk and milk products in a beneficial quadrant here where there's relatively low cost compared to the nutrient richness. That's not necessarily the case in other societies. This is a picture from the Gambia where I work. This chap Marlamin in the back here is struggling to bring up his family. He has very limited financial resources. He's a subsistence farmer. And this incidentally is true of many people in China as well. So milk and milk products are high aspirational in China but they're also very expensive. And under those circumstances we really have to move milk and milk products into a different quadrant. They are very high cost under these circumstances. And that actually has quite a significant impact on the way that people treat milk products. So here's a young boy early in the morning milking some of his cows. You would look at that and you'd think dairy produce no problem. There are problems though. First of all these cows are very very low yielding. Secondly they are enormously costly and there's only a few people in the village who own these cows. And thirdly it's very hot. So within an hour or two of him collecting this milk there's no refrigeration. That milk will go off unless he does something to it. So they ferment it of course and then having fermented it they sell it. They sell it elsewhere. They transport it on the local buses down to the coast and they use it as income. So in fact these rural populations are having a very low milk intake. And I give you that example because I think it's reflective of what will be happening in many many populations around the world. And so we see here that there are enormous variations in the intake of milk around the world. So industrial societies on the left there. These are the projections or well the projection to 2030 in green there. And you'll see that we consume on average large amounts in industrial societies not surprising much smaller amounts in developing societies. Sub-Saharan Africa is low. Latin America considerably higher and rising fast. South Asia rising fast. And one aspect of that is that milk is very is heavily consumed in India. And there is the penetration of lactase persistent gene into India. But in East Asia where there was that desert of lactase persistence and I don't know whether these two things are correlated. You'll see there are very very low intakes of dairy products. And the question that I would be interested in if I were a market here involved in this would be are we ever going to really penetrate that market or does the genetic background of that market mean that they will always be low consumers. So just looking at one of those nutrient needs we can look here at calcium. So here's a measure of calcium intake versus RNA of dairy product based on dairy produce intake. And you can see that there are vast variations with some surprises in terms of for instance the United Kingdom and New Zealand looking quite low. Now of course this is entirely predicated on the value that we put for the RNA this 1000mg per day set by WHO. And I think a very big question is is that appropriate around the world and that will be discussed later this afternoon. Is it really appropriate that Chinese people have an RNA of 1000mg per day and I would postulate that it's not. So let's have a little look at some of the health implications of dairy intake. Well of course the first thing that I think comes to mind is in terms of growth and skeletal health. We know that many children around the world are stunted and that's a major issue in terms of concern in global health. This little picture on the right here of Miro, our head of station boy standing next to a boy born on exactly the same day as him living in Kenabar gives a graphic display of the differences in their height. But we don't know that that is related of course to dairy intake. There are many many other contributors to the fact that Amadou is stunted. And there are complexities here. For instance we would ask the question what in this population in the Gambia that has extraordinary low calcium intakes may be down at somewhere between 200 and 300mg per day. What are the implications of that in later life? What is the bone density of these people? What is their susceptibility to fractures? And the answer is that their bone density is very good, probably contributed because they work so hard in the fields throughout their life and they have almost no fractures at all. And this led Amprentis to postulate this calcium paradox where if you do ecological comparisons you find this rather strange completely inverse association between calcium intake and fracture risk. And so that's giving us a wake up call. It's not a simple matter that the more calcium you take the better your bones will be. So we can summarise the knowledge in this field rather superficially as follows. There are ecological associations between dairy intake and height. Norwegians, Finns, Swedes and Danes and particularly the Dutch tend to be very high and have very tall and have very high dairy intakes. It's been speculated that there are some additional growth factors in milk apart from just the high quality protein. That speculation I don't think is at all resolved and I think there's work to be done in that field. I've already suggested to you that there are adaptations to low calcium intake which can protect bone health. And the work we've been doing in the Gambia shows that we perhaps interfere with these at our peril. We have got some very surprising results from randomised clinical trials in which we have given calcium. If we give calcium to women in pregnancy paradoxically they lose more bone rather than less bone in lactation. If we give calcium to boys at the age 8 to 10 before puberty they do grow a little bit faster in puberty, but they finish puberty earlier and they end up shorter rather than taller. So there are many complexities here which I think are important research needs for the future. I've put in grey there that high dose vitamin D and calcium are beneficial, there's randomised controlled trials to show that, but this is not a dietary issue, these are pharmacological doses of these nutrients. And then I just checked this with Connie a little earlier because my reading of the literature was that there are no existing metronalysis on dairy foods and osteoporosis. So I checked with Connie, have I made a complete mistake here and she was not able to refute and say Andrew you fool, you've missed such and such. So to me this is a big surprise that actually there seems to be nothing out there in terms of that. Of course it would be very difficult to do a decent randomised controlled trial on that but there's no metronalysis of observational studies either. So let's think a little bit more broadly about the potential health implications. There's a huge literature, if you go to pub bed and put in dairy and diabetes or whatever there's an enormous literature and of course as usual this literature is very mixed and confusing. I think even to anyone who is working specifically on dairy and one of these outcomes it's going to be a confusing literature. To the rest of us in this audience who want to get a sort of global picture of is dairy beneficial or not, it's enormously confusing. Why is that the case? Well there are many reasons of course. First of all if you try to separate out yoghurt, the purpose of today's symposium from other milk products, it's very difficult. It's a very complex exposure. Many of you have had a mixture of yoghurt and milk and other dairy products already this morning. How do we separate that out as an exposure in epidemiological studies? Very difficult. There's residual confounding. People who eat a lot of yoghurt are much more likely to wear open-toed sandals and socks, this kind of thing. And then there's the substitution. If you're not eating dairy products, where are you getting your calcium? What are you eating instead of those dairy products? And that also makes it very difficult to separate out cause from effect. But I think by and large there are specific benefits of dairy. These are often apparent and me as a sort of lay interpreter of this would say that I find it quite persuasive. You're going to hear more later today about dairy products and obesity, which of course has been a major area of controversy. So I've skirted around that one to leave space for other speakers later. There is a meta-analysis of randomized controlled trials of dairy substitution in terms of weight loss. You'll see details of that later today. And there's a marginal benefit of dairy. Dairy products and diabetes, there's quite a lot of studies in this area. A recent meta-analysis shows that diabetes risk is reduced, especially by low-fat dairy intake and by yoghurt intake in particular. High-fat dairy seems to be neutral. That's why there's a horizontal arrow there. It seems to be neutral. So quite strong. Now these are effect sizes of about a risk ratio of about 8, 10, 12, 15% between the top to the bottom quartile of intakes. I put this up, which is not a meta-analysis, but a very recent publication from very large prospective study of incident cases of diabetes in people who were free from diabetes at baseline in the EPIC and Interact Study. What they show, I think very persuasively, in a nested case control analysis of nearly 29,000 people with 14,000 or so incident cases of diabetes, that total dairy intake is neutral, cheese seems to be protective again by about 12% or so, and fermented dairy is protective by about 12%, 13%, 14% or so. So I think that's to me very strong evidence that the consumption of dairy products is protective, but again remember of course that we cannot separate out the potential confounding within these studies. Dairy products and hypertension. The meta-analysis shows us that hypertension, the risk of hypertension is reduced in relation to total dairy intake to low-fat dairy and to milk. In terms of cardiovascular disease, again there's been some meta-analysis, somewhat limited data surprisingly, the effects appear to be somewhat neutral and I believe we're going to be hearing more about that later today. And finally, I've looked at the issue of dairy products and cancers. Again, huge literature, particularly in terms of prostate cancer. The World Cancer Research Fund and the American Institute for Cancer Research in 2007 made these statements that there was evidence for a reduced risk of colon cancer, an increased risk of prostate cancer and a decreased risk of bladder cancer, but my reading of the literature is that the association with prostate cancer, and this has also been done in Mendelian randomisation studies again from Epic very recently, is that that association has now been refuted. So the results are mixed and confusing, but I think there's a very important point to make here, which is that there's no sort of elephant in the room. There's certainly no studies that show a very strong association between dairy products and any cancers or any other disease endpoints. In fact, my reading is that on balance it tends to work in a beneficial direction in all cases. And in terms of the cancer risk, there is quite a lot of specific current interest in fermented products. So my conclusions on the basis of this quite brief tour around the global health issues of dairy products. First of all, I think there's very strong evidence, and I'm fascinated by this, that dairy foods confer, or we should be more careful in that statement and say they conferred, we don't know that they would still confer, but over evolutionary time they have conferred an amazingly strong evolutionary benefit. There are huge variations in intake around the world, and those seem to be well correlated with persistence of lactase digestion into adulthood, whether that is cause and effect, I don't know. There are some very important adaptations to low calcium intake. Our physiology is pretty smart. It can cope on lower calcium intakes, and we've got a lot of research to do to understand the implications of playing around with that natural physiological adaptation. There may be some, under some circumstances, some adverse effects. I think that a global recommended nutrient intake specifically for calcium may not be appropriate. Again, we'll be discussing that later. And I think there's very strong both social, ecological, culinary evidence that fermentation of milk confirms many, many advantages. I still remember vividly, actually, the first time that I came across yoghurt, I was probably about five. I was at a friend's house and his mum said to me, come and make this, come and mix this in with the milk, and then we left it out. I was brought up in Uganda, so it was very hot. We just left it out, didn't have to naturally warm it, and then come back later and it's gone, you know, it's turned and it tastes completely different things. I still remember very vividly that amazing transformation. I'll end on what I would suggest as some research needs. Of course, there's a lot to be done in terms of defining the optimal role of dairy foods in a healthy diet. Now, as a consumer and as a cook, and having done this research in preparation for this, I'm now comfortable personally in the position that I love dairy products, I love yoghurt, and I'm going to continue to contribute to put them into my diet because I cannot find any strong evidence of adverse effects. What are the physiological effects of fermented foods? Again, we'll be hearing about this later. I think there is a great deal of evidence to suggest that that process of fermentation interacts with our natural human physiology in some very meaningful ways. Are there specific health attributes of yoghurt? Of course, that's what we're here to discuss. We certainly won't decide it, but to discuss today. Finally, I think it's interesting to ask this question. You see, I haven't mentioned the microbiome yet. No decent speaker these days gets away without mentioning the microbiome. Perhaps I end on this question of would yoghurt be an appropriate vehicle for next generation personalised probiotics? I think, of course, it could well be a decent vehicle for that. So, with that, I'd like to thank my sponsors, the UK Medical Research Council, the Wellcome Trust, NIH, and the Bill and Melinda Gates Foundation, and thank you for your attention. Thank you, Andy, for that really beautiful provocative talk. We have plenty of time for discussion. And, as Sharon said, we're really here to try to discuss what some of the issues are. So, if any of you would like to come up to the microphone, ask some specific questions, but also make some suggestions for what you think it would be interesting to know more about relation to yoghurt. Yes, and please say your name and affiliation before you... Francisco Gwander in Barcelona in Spain, gastroenterologist. So, thank you very much. It was a very interesting presentation. And concerning the data on bone fracture in different populations, I wonder whether the H effect was corrected because probably life expectancy is different from Africa to Sweden. And this would have an impact in the data. Yes, so I'll invite Connie to answer this as well. But my first answer would be a very specific one in relation to the work we've done in the Gambia, which is that Anne Prentis has done longitudinal analyses of bone density and shown that although bone density is a little bit lower in Gambian women, so this isn't just, the fracture correlation is not just due to bone density, as is well known, there's issues around the geometry of the bone formation as well. Although the bone density is a little bit lower, it seems to decrease at a very similar rate. And we actually have a study ongoing at the moment called Gambas, the Gambian Bone Aging Study, which is specifically making measurements of the acute rate of bone loss. So, we'll have a better answer on that as well. But it's certainly not just an issue that we have lost the potentially affected individuals. I don't know if Connie would like to add anything to that. Ellen Walker, Harvard Medical School, and also a gastroenterologist. Andrew, we had a major debate recently at the Harvard School of Public Health. You probably know that Walter Willett feels low calcium is appropriate, and I brought in as a pediatrician a representative from the committee of nutrition of the American Academy of Pediatrics who disagreed with him. The point I'm making is I think the low versus high calcium is a long term adaptation, as I think you showed in your studies. If you're taking in low quantities of calcium, you somehow adjust so you don't get fractures. So I think sudden changes in a population that has a different intake may be detrimental both ways. I'd like you to comment on that. I absolutely endorse that view. I mean what we see with the evolution of the lactase persistence is in evolutionary time an amazingly fast adaptation, but actually it's very slow. It's happened over perhaps several hundred generations. So we come along as health interventionists and want to do that in half a generation, and clearly that may not be wise. I absolutely agree with you and all the work that Anne has done would suggest that there are adaptations set up. We don't know when they're set up. Are they set up early in fetal life? Is it epigenetics? I'm glad I've had the opportunity to add epigenetics to microbiome in my talk. Is it an epigenetic processing? Another speakers are going to address that. So maybe I could park that response and wait until we hear from people who know what they're talking about. So I just want to ask you, when you were looking around in the literature, how hard was it to find specific mention of yogurt rather than dairy? I mean quite hard, quite hard. Again, the issue of separating those effects out, specifically in epidemiological studies. This is obvious. It's very difficult to find a person who eats a whole load of yogurt and no other dairy produce, who doesn't eat cheese, doesn't drink milk, but has a whole bucket full of yogurt every day. So that phenotype, that exposure doesn't really exist. They tend to be very mixed. So I think it is always going to be difficult. We had a little conversation yesterday and someone said, you know, you can't do long term randomised trials. I think you could do pretty long term randomised trials on yogurt if you put your mind to it. So that maybe is another research need. Well, another fascinating topic you raised was the cultural differences in the history of yogurt consumption. Did you come across any good reviews on that or would that be a good idea for us to consider? I think that would be fascinating. Maybe that probably there are books on this, but in the time I had available to prepare myself, I didn't come across them. I wonder maybe we could throw that open to the audience. Does anyone here know of a sort of fantastic ethnographic review of the consumption of fermented products around the world? Interesting. So there we go. That sounds like a research need.