Select Committee on Science and Technology Minutes of Evidence

Examination of Witnesses (Questions 100-119)

Professor Peter Burney, Professor Aziz Sheikh, Dr Graham Devereux and Professor John Warner


  Q100  Lord Rea: I wonder if you can bring us up to date about recent trends in the UK in the incidence and prevalence of allergic diseases.

  Professor Burney: Amongst children the main evidence comes from a study called the ISAAC study, which is an international study that has studied children about 10 years apart and looked at the difference between not individual children but schools of children, so they are looking at the same age groups, 6 to 7 year olds and 13 to 14 year olds on each occasion. In the UK among younger age groups there were insignificant increases in asthma and rhinitis during that period, and there was a significant increase in eczema, but in the older children there were significant reductions in the prevalence of asthma and rhinitis and a non-significant reduction in eczema. That is important in a way and it has got a lot of attention because up until those studies and studies published over the last few years almost all the studies that had looked at trends in disease had shown upward for asthma, eczema and rhinitis, so it looks as if there is a flattening off of this trend, though this is not seen in all countries. In some places the prevalence is still going up, in some places it seems to be going down, so for the first time there is a rather mixed set of evidence. That is for children. The other issue is the issue that I referred to a little bit earlier, and that is what is happening in adults. In adults, as I say, the picture is dominated by what happens from cohort to cohort, in other words, from one birth generation to the next, and what we know about is people born up to about 1970. We know in this group that the prevalence of people who are sensitised to allergens has increased in each successive generation and that these are therefore likely to be increasingly affected by the kinds of diseases that follow from sensitisation. We do not have that information for children yet, so what is happening there is a bit of a blank and we cannot answer the question of when this increase stopped. The ISAAC study, the study that I talked about earlier, did not measure sensitisation. They looked just at the change in the resulting diseases.

  Q101  Lord Rea: What further work do we need to do in order to explain further these trends in children, adolescents and adults?

  Professor Burney: Explaining why those changes have happened is an even more difficult question. However, just to know how the epidemic is going to evolve there are two big issues that need to be reviewed. I have said that sensitisation is increasing from generation to generation from people born in about the 1930s to the 1970s. I think we do need to know what is happening to sensitisation in birth cohorts since then; that is important. We also need to know more about what happens to people who are sensitised in terms of disease later on in life. There would be a reasonable guess that things will get worse because by and large as you get older you get frailer, you pick up more complications, you pick up more problems, so we suspect that they would have more disease as well as this added burden of sensitisation, but we do not really know that and I think that will be important to know at the end of life. At the beginning of life, as I say, I think it will be interesting and important to know what is really happening to sensitisation rates.

  Q102  Lord Taverne: In your paper you mention that there is this unsolved puzzle of why only a proportion of individuals who test positive for specific IgE develop atopic forms of allergy. Is this becoming more common or is this something which is a complete mystery? Does this have an effect on the development of allergies?

  Professor Burney: As I say, it is a mystery. I do not think that we know. Someone else on the panel may know more. There are some hypotheses about why some people seem to be able to carry an allergy and not have a disease as a result, but most of them are really quite speculative at the moment and people have been looking at why these things may happen. I suppose the interesting results, which are not really relevant to the UK, are from places from where there are a lot of parasites and in some of those studies there is a suggestion that the parasites will provide some chemical signal for the body to dampen down inflammation and that might be an explanation in some places. Another reason might be, for instance, the diet, so that again we do not have a very specific answer to that question but it may be that diet is important in your response to allergen if you have an allergy. The short answer is that we do not know but I think it is a very important area to investigate.

  Q103  Lord Taverne: Is it a high proportion or a very low proportion of those who do not develop as you would expect?

  Professor Burney: Quite a high proportion. I guess 50 per cent of the population probably in some age groups will have sensitisation to a common allergen, but the proportion that will have recognisable asthma will be less than 10 per cent probably, so it is quite a high proportion.

  Q104  Lord Rea: Are we making the best use, in looking into the epidemiology and course of allergic conditions, of the very considerable database that we hold in the NHS, particularly general practice records? Could we do more to make use of this information?

  Professor Burney: People certainly are using the data. The data are becoming more available now and they are more easily accessible. Perhaps Professor Sheikh could say more.

  Professor Sheikh: I will take this with your previous question about disease trends. Building on Professor Burney's response, we know relatively little about disease trends in the more severe systemic allergic problems such as anaphylaxis, food allergy, urticaria, because a lot of the primary work has not been done. When we do begin to exploit these national data sets what we find is that hospital admissions for the systemic allergic disorders have increased very rapidly over the last decade and a half, and these increases seem to be continuing. We are involved in work using primary care databases to look at these trends to see whether similar patterns hold true and these data also show that the systemic allergic disorders are increasing. What we also know is that multiple allergic disorders also seem to be increasing so, whilst currently in the older age spectrum approximately five per cent of people will have multiple allergic problems, if we look in the younger groups it is something like 10 per cent. These are some of the people who are getting into most problems and are running into most difficulties with NHS care provision. In terms of unlocking the potential of the data sets, we still have quite a long way to go and one of the key things is allowing data linkage to take place, because not only do we have healthcare data sets; we also have educational data sets, we have got census records, so we can begin to look at the impact of allergic diseases if we can make progress on data linkage, but that needs some thinking through because of the factor around confidentiality and privacy issues.

  Q105  Chairman: Do you think the UK Biobank will have a role here?

  Professor Sheikh: In terms of data linkage it could possibly because it is going to be a vast resource and the idea of linking with other national data sets would be very interesting.

  Q106  Lord Haskel: When we were discussing the first question we spoke about the importance of defining terms. In oral evidence the Department of Health said that the introduction of the Systemised Nomenclature of Medicine, or SNOMED, as they call it, would help to classify allergic diseases in a much more specific way. Do you think that this will improve the way in which clinicians and researchers exchange clinical knowledge and aid research into national trends?

  Professor Sheikh: Certainly SNOMED has the potential to do that and the advantage of SNOMED is that it has been developed in the US, together with the SNOMED CT version, the clinical terminology version, which has been developed in association with the NHS. It is being used in about 40 countries in the world at the moment and so in terms of international comparisons through routine data sets the potential is phenomenal. It also allows more specific coding because there is far more flexibility within this new coding structure in the way that the system is configured, so in terms of potential, yes. However, what I am slightly concerned about is that at the moment we need to get the allergy community involved in the range of terminology that exists to see whether it is fit for purpose and I do not think that has been done as yet. What also concerns me is that for use in real time settings there needs to be simulation work that as far as I know has not been done yet. Retrospective coding is one thing but, in terms of real term consultation coding, again, that work needs to be done to see whether this system is fit for purpose. We also need to ensure that there is training in place because a particular advantage of this is that SNOMED CT through Connecting for Health would be implemented across the NHS, so in primary care and in hospitals and NHS Direct, for example, and so we have a potential for linked data which is very powerful. What we need to ensure is that there is consistent training across all of those different NHS sectors to ensure that the data we collect are ultimately meaningful because in order for Professor Burney and I to correctly interpret data it is essential that these are collected accurately.

  Q107  Lord Haskel: The problem with classifying data in this way, of course, is that if something new comes along you tend to miss it because it does not fall into any classification. Do you think there is any danger of that?

  Professor Sheikh: I think there is less danger of that with SNOWMED-CT because it does allow you to construct codes using a number of fields. Also, there is a commitment to this being an iterative process, so, as with the current coding system, if there is a need there is the potential to include new codes, although at the moment this is quite a laborious process. With SNOMED-CT it has been thought through a bit better as far as I can see.

  Professor Warner: In a way one of the problems in relation to feeding data into this system goes back to the previous question that was asked about being allergically sensitised and having a raised IgE as distinct from having IgE mediated disease. It is very easy for somebody to misinterpret information based on a history and allergy tests in relation to the patient's individual problem and then the data go in incorrectly and that rather destroys the whole object of the exercise. That is a feature of allergy practice at the moment, not being something that is uniformly included in undergraduate training or in postgraduate education programmes, means that there is not sufficient allergy expertise around to ensure that the data feed-in in the first place is accurate.

  Dr Devereux: As a chest physician I often see people who are sent to my clinic with a diagnosis of asthma and a huge per centage of them do not actually have asthma. This is where the database is going to collapse. It is so easy for a GP to make a diagnosis of asthma because it is the commonest thing that is around, whereas when they come to my clinic I will do the tests and find that some of them do not have asthma; some of them have bronchitis and emphysema, some of them are just overweight, so maybe out of 10 people I see with a GP diagnosis of asthma probably only about three or four of them turn out to have asthma in the end, so you are right.

  Q108  Lord May of Oxford: What do we know about the incidence and prevalence of allergic afflictions compared with other countries, and particularly, in so far as it looks like we are high among the OECD countries, how much of this is real prevalence as distinct from differences in the definition of things we are talking about or, more importantly, the statistics?

  Professor Burney: Again, most of that information does not come from routine statistics; it comes from surveys, so they are relatively well controlled. There is a problem, obviously, in using questionnaires in being very precise about what people have and whether the questionnaires are working exactly the same way in one language as against another, but for the most part we can be fairly confident. If I go back to the ISAAC studies, which are worldwide studies of children, they are based on questionnaires essentially. Some of them are rather elaborate video questionnaires where the children are shown videos of something and asked, "Are you like that?". They took quite a lot of trouble to try and take out the effect of language problems that you might have. For the ISAAC studies it is quite true, certainly in Europe that the UK comes high in the lists. In the six to seven year olds I think there was nowhere that was higher, if we include the Channel Islands and the Isle of Man within the UK. I do not think there was anywhere that was higher in Europe for asthma and rhinitis in the reports that were made from these questionnaires. For eczema I think it was only Sweden that had a higher prevalence. For the 13 to 14 year olds, again, in asthma only Ireland was higher and in rhinitis there were a number of places that were higher, but by and large the answer is that for those questionnaire-based data we can be pretty sure that that is true. For the adults we have less information across the world, but certainly in western Europe, and here we have evidence from serology, from looking at specific IgE to allergens in the sera from people from different parts of Europe, again the UK has, or rather in the centres in the UK—I should be a little bit more precise because these are not universal studies; they do not try and capture the whole population of England—they had amongst the highest prevalence of sensitisation, and that was pretty well true for all of the centres that we looked at, and they had correspondingly higher levels of disease, so I think we can be pretty sure that that is true. The only places in the ECRHS, the European study that looked at adults, which had as high or higher rates of sensitisation were other English speaking places like Australia and New Zealand, which had very high rates as well, so we can be pretty sure about that. In the ISAAC study, which looked at more places worldwide, the other hotspot was Latin America where they seemed to report very high levels, but for those we do not have the equivalent data yet on sensitisation rates, so we do not have an objective measure to measure that against, but that will come.

  Q109  Lord May of Oxford: What are the ideas that are floating round about the underlying reason for our being higher than others?

  Professor Burney: There are a number of reasons. It has been very difficult to demonstrate a reason in a lot of different kinds of studies. When you actually get to the data you say, "Let us adjust for the following things", and then look to see whether that explains the difference, and almost always the answer is that it does not explain it, so explaining the differences has been extremely difficult. There are general thoughts around.

  Q110  Lord May of Oxford: We have, of course, heard some of them. I wondered what your view of their comparative merits was.

  Professor Burney: I would probably at the moment put quite a lot of emphasis on diet. I think diet is probably important, but the detail of what in the diet is much more difficult to disentangle at the moment. There could be genetic differences. We know that there are, but whether they are relevant in this case I am not sure. It has been noted, for instance, that by and large genes follow language and quite a lot of the places with high prevalence are English speaking. Against that the Australian centre in the ECRHS is Melbourne, where the population is very largely of south European origin, so maybe this is not a good explanation.

  Q111  Lord May of Oxford: It is not really part of Australia.

  Professor Burney: The other hypothesis that is causing a lot of interest at the moment is the so-called hygiene hypothesis. Again, we are not entirely sure what it is in the hygiene hypothesis that is explaining the differences. It is still really a hypothesis and, to be truthful, it is probably lots of hypotheses which are adapted to fit different stories and I think that has to mature a bit. We need to know a little bit more about exactly what it is about "hygiene" that is important. I suppose those would be the two main hypotheses, diet and hygiene, that people are most interested in at the moment.

  Q112  Lord May of Oxford: What about the playing in dirt hypothesis?

  Professor Burney: That is the hygiene hypothesis. There are a lot of separate bits of information about that. Some of the early theories on mechanisms really do not seem to stack up so well but the basic original data were that if you have older brothers and sisters you are less likely to be allergic, and that seems to be very consistent in all the data sets. That was the original part of the hypothesis. It has since been expanded in various ways so, for instance, one of the interesting groups that have been looked at, mostly in central Europe, is farmers and it is noted that people who are brought up on farms, particularly farms with animals, are less likely to become allergic, which was a bit of a surprise to some allergists who thought that being in amongst all those allergens might have had the opposite effect, but it does not seem to. That is a strong part of that hypothesis. In other parts of the world the hypothesis does not stack up so well. If you go to Africa where the environment is completely different hygiene does not seem to explain the distribution of disease. People have tried to explain it, for instance, by looking at parasites as part of this general hygiene hypothesis. In fact, parasites do not protect against sensitisation, as predicted by the original hygiene hypothesis, but have a different effect and on a different part of the mechanism, so that is probably a different story.

  Q113  Lord Soulsby of Swaffham Prior: Having started off on the hygiene hypothesis, I have two questions. The first one is what practical advice should the Government be giving to parents regarding hygiene to minimise the risk of their children developing allergies?

  Professor Burney: My view would be none. Again, we have been into this a little bit before. You have to have a very high level of understanding of something before you start giving advice to people. One of the things that people should be a little bit aware of is that there is a hint in some of the studies that the same risk factors, that protect you from allergy are also associated with lower lung function, so it may well be that nanny is right, that playing in the dirt is not such a good thing. It may be good for your allergies but you really need to know across the board what the effects are going to be before you start giving advice.

  Professor Warner: I think the word "hygiene" is now probably the wrong one to use to explain this hypothesis. There was a very nice paper published in the journal that Professor Kay edits very recently where it has been suggested by the authors, and indeed other people, that we should talk about microbial exposure or microbial deprivation hypothesis. Hygiene hypothesis gives out the wrong message to people, and the problem is that it is interpreted as suggesting that you should roll your babies in dirt. The consequence is all kinds of nightmares of infections that we have effectively eliminated by good hygiene, and that is not what we want. It also gives out a very bad message, for instance, in relation to immunisation. This is an area where there is enormous worry that families are not immunising their children because they have got the message that immunisation means less infection, means more allergy. That is an incredibly bad public health message and we have got to counteract that. In fact, all the studies that are being done, and I have been involved with one that we are just putting together a paper on at the moment, show quite the converse, that if you immunise babies effectively there is, if anything, marginally less allergy. This is a complex area and we have to disentangle all the different components. I still think it is an interesting area in relation to potential for intervention but we are going to have to understand what microbial factors are involved and how they can be administered to reduce allergies.

  Q114  Lord Taverne: Is it the new Liu & Leung paper, the paper you mentioned?

  Professor Warner: The paper I mentioned is by Bloomfield et al. I am sure Professor Kay would be able to give you the details. It is Clinical and Experimental Allergy 2006, volume 36, pages 402-425.

  Q115  Lord Soulsby of Swaffham Prior: I just want to come back on the hygiene hypothesis. There has been a suggestion that domestic pets, dogs and cats and maybe other pets, play a role in this. Would you subscribe to that?

  Professor Burney: There is certainly some evidence that children brought up with dogs have less sensitisation. I think we probably need a little bit more evidence to support it, but that seems to be true.

  Q116  Lord Soulsby of Swaffham Prior: Because this country, the United Kingdom, has a high companion animal population but it is not the only country in the European Union that does, of course. France and Germany have high populations too, and then there is the reverse of that: countries that do not keep domestic animals because of religious reasons. What is the issue there? For example, in a Muslim country is there as much sensitisation and rhinitis as in a non-Muslim country?

  Professor Burney: I do not think we know much about that. I should also add another issue over the pets, and that is that, again, there is misunderstanding about this. Pets can be a problem for people who have already been sensitised, so they are not an undiluted joy in this way. I think the answer would be that in Muslim countries by and large they are often exposed to allergens but cats are kept outside the house and therefore they have a much lower exposure. That may not be such a good thing as some people believe that very high doses of cat allergen may be protective and lower levels are more likely to sensitise but that is talking about the allergen, so there is a paradox about this. Cats carry allergen but they also probably carry some of this microbial exposure that might be protective as well, so it is a paradoxical issue. I cannot really answer your question about the countries in the Muslim world. I do not have any data on that.

  Q117  Lord Taverne: Is there anything in the suggestion "dogs good, cats bad"?

  Dr Devereux: The first data were with cats and most of the studies have shown that cats are beneficial; you are less allergic to cats than you should be, but also there is some dog work. On the stuff about Muslim countries, there is a study from Saudi Arabia showing that Saudi Arabians, if they live in the cities, have a very high prevalence equivalent to what we have in the westernised countries, whereas the ones living out in the deserts with a nomadic lifestyle have a very low prevalence of asthma and allergic disease. It is not Muslim; it is where you live. It depends where you developed or whether it was in the traditional lifestyle.

  Q118  Viscount Simon: Is that the same in, let us say, central Africa?

  Dr Devereux: In Africa there are studies showing very similar things, that populations living in the towns and cities have more asthma and allergic disease. People living in the townships out in the bush are less likely to have asthma and allergic disease. However, the first studies in the 1970s showed that there was quite a big gap but now the gap is narrowing and it is thought that people in the townships and the bush are taking on westernised lifestyles and that is maybe why the gap is disappearing now.

  Q119  Viscount Simon: Or maybe second homes.

  Dr Devereux: I suspect that if you are poor in Africa you cannot afford a second home.

  Professor Burney: No, but movement between urban and rural areas does happen, particularly with children. Children will be sent from the towns in some places to the country to be educated because it is a less dangerous environment, so that is a complication.

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2007