TUESDAY 16 FEBRUARY 1999

PROFESSOR JOHN BOURNE, DR CHRISTL DONNELLY AND DR ROSIE WOODROFFE

  160.  Is there more that MAFF can do to overcome the public scepticism of this trial?
  (Professor Bourne) We would advise Ministers on the appropriate release of data from the trial area, but we would certainly wish to be assured that premature release of data would not compromise the trial.  I have suggested to the Minister of State that before any data are released, he does have the agreement of my Group to the release of that data.

  161.  I would suggest that there you are as a committee of scientists or statisticians, with the light of truth shining in your eyes, to define what truth is; and you are trying to impose a scientific experiment, some kind of mechanical grid on the country, which may or may not correspond with the facts of geography; which may or may not be acceptable to the bleeding hearts, who do not want to kill little badgers, so you are having to conciliate them; and which may be screwed-up by human nature.  Just to take one instance.  If I was a farmer and I thought there was a correlation between badgers and TB infection in my herd, I would go down and mow them down.  How are you going to balance that?
  (Professor Bourne) You have put your finger on all the problems facing us.  We accept that.  However, I think what we are putting in place is a pragmatic approach to answer a really serious problem.

  162.  But since it is illegal you would not know if I had mowed them down.
  (Professor Bourne) That is true to a degree.  In the survey-only areas, of course, which we are surveying, we would have some idea if illegal killing was being carried out.  But you are right.  It is very difficult to estimate that and relate that to the discussion we previously had.  I think we do need to satisfy you and satisfy members of the public that what we are taking is a very pragmatic approach to a very difficult problem.  There is no simple answer to this.  There is no quick fix to this.

  163.  But if you are divided between the people who want to kill them and the bleeding hearts who do not want to do anything to them, you cannot satisfy them both.
  (Professor Bourne) We cannot.  There will be extremes of view.  However successful we are in getting the message across, what we are taking is a pragmatic and sensible approach to addressing a very serious problem.

  164.  If I go out and mow them down and you do not know, it is going to make your results wrong.
  (Professor Bourne) It would confuse the results but you are talking here about the power of experiments.  Perhaps we will come back to talk about that a little later with Christl Donnelly, who can provide you with a statistician's viewpoint about this and the strength of what we are doing.  But to come back to the concept of what we do, and whether this will be acceptable by badger groups on the one hand and farmers on the other, I think farmers have to appreciate that we sincerely believe that this is the only legitimate way to find an answer to their problem.  It is not short-term.  Perhaps there should be more discussion between farmers and Ministers about short-term aspects that may be used to control this disease.  That is not our responsibility.  We are there, as you know, to put in place a scientific trial.  We are also there to advise on other data which we may use to provide scientific answers to a whole range of questions which we have identified.  There are problems here.  On the one hand, if farmers do interfere with what we are doing, the evidence for involvement of the badger is being diluted.  On the other hand, if conservationists do not accept what we are doing, the message to the farmer is: "There is no other alternative, chaps.  Go out and kill more badgers," which is exactly what we wish to avoid.  We stated that very clearly—and it is not just a pious wish.  We are not a bunch of blue sky scientists.  We do have our feet on the ground.  Our hope is that we can put in place a policy which does allow the co-existence of badgers with cattle.  If we did not believe we could achieve that we would not be doing it.

  165.  Okay.  How about non-compliance at the other end? The level of non-compliance which would invalidate the experiment.
  (Professor Bourne) Non-compliance has not been a feature with the Cornish triplets.  We have had a very high level of compliance with farmers.

  Chairman: You did suggest that Dr Donnelly might talk about the power of experiment.  This is probably a good moment to do that.

  166.  But may I move this on.  I am sure Dr Donnelly is going to be a brilliant statistician and statistically it will be valid, but what I am worried about are the practical problems which Dr Donnelly cannot actually take account of.  She can tell us how we can take account of non-compliance and homicidal murder—or badgericidal murder, I am not quite sure.
  (Dr Woodroffe) That would be melicidal murder.
  (Dr Donnelly) May I circulate this graph.
  (Professor Bourne) These are diagrams from Krebs' report.
  (Dr Donnelly) They are page 93 but just so that everyone has them to hand.  The issue of non-compliance will be in either side—whether or not it is coming from badger activists who are releasing badgers which would have been culled otherwise, or whether or not it is farmers who are doing illegal culling—this will be to make the incidence in the treatment arms more similar.  The greater the extent to which the treatments become more similar, it takes longer to find that difference.  So the experiment has been designed to detect over a five-year period differences as low as a 20 per cent reduction in TB incidence, but that does depend on the underlying rate of TB breakdowns in the control or survey-only area.  If we assume that there will be 100 breakdowns per year—this is in the control area—this gives us the green line.  The green line tells you how many years will be required to detect a difference of a certain level.  So, for example, a difference of a 20 per cent reduction in TB incidence could be detected in the five years of the experiment.  Now, the effect of non-compliance will depend on unknowns, one of which is: what would be the per cent reduction if there was 100 per cent compliance with the trial? If everyone co-operated fully and everything went to plan.  We do not know where that will be but that number is then reduced by any level of non-compliance.  So obviously if there was complete non-compliance and, as you said, all badgers were killed equally in all of these areas, that would lead to them being exactly the same and we would never detect any difference.  But the circumstantial evidence, as evaluated by Krebs, did suggest that there would be an effect on breakdowns.  The extent to which that is affected by non-compliance will affect the number of years required until we can find that difference.  It is really a level of magnitude.  But being able to detect as low as 20 per cent—possibly lower, as the underlying breakdown rate is increasing—that may increase to 150 breakdowns a year and then you would be on the blue line.  You can see that we are able to detect relatively low reductions in breakdowns.  Of course, some people believe the actual number is zero and some people believe it is 90 per cent.  This really gives us the one opportunity to tell that difference, to tell where we are.

  167.  So don't mess with me because it will be take longer.
  (Dr Woodroffe) Yes, exactly that.

  168.  It is a model for people who might want to mess with it.
  (Dr Woodroffe) Absolutely.

  Mr Mitchell: Could I ask about the implementation of these experiments.

  169.  First, could we just ask about the question of power, if we can move on to that, because it is not just human intervention, but there could be other differences within the triplets.  There might be climatic differences or soil might make a big difference.  Are you satisfied with what you have already seen in Devon, the non-compliance that you may have experienced, that you are going to have enough power in this experiment at the end of the day to produce a valid result?
  (Dr Donnelly) We will be looking for the level of a reduction but also the stability of that.  That is going to be critical.  For example, if in some areas we see a 60 per cent reduction and in other areas we hardly see any, it will be very important to analyse the secondary outcome variables and find out why that would be.  It might be due to climate.  It might be due to soil type.  Then, as well, we have a questionnaire being given out to farmers who experience breakdowns, as well as farms that do not.  Those comparisons will allow us then to distinguish why it is integral in some areas compared with others.  That will obviously be important but there will be some fluctuation, even if everything were equal, due to statistical fluctuation.

  170.  Even if the worse case scenario was likely to be—probably on the basis you have already learnt in Devon—do you believe you will actually come up with a result at the end of this experiment?
  (Professor Bourne) Yes, we do.  I am not sure we have experienced a worst case scenario.

  171.  Make a better guess.
  (Professor Bourne) We have been reassured by the level of interference.  We have no doubt we will come up with the answers.  The question of timing is important.

  172.  I have a question on the variables.  How many variables do you have? What are you listing? The Chairman mentioned the weather and soil.  There are other creatures within the test zones.  Are you studying those variations within populations? How many variables are there which you are comparing?
  (Professor Bourne) There are a number of questions there.  I will try to winkle out some of these and then Dr Woodroffe can contribute.  Questions with respect to the source of infection.  We have identified it may be cattle, it may be the badger, it may be other wildlife, there may be interaction between wildlife and badger, wildlife and cattle.  This is an important component of the study.  Again, research proposals have been received by MAFF.  They have been considered by the Group.  They are still being considered by the Group to ensure that the work, with respect to wildlife, is done appropriately.  It will be done appropriately within the trial areas.  With respect to the whole range of other risk factors you mentioned, we believe that the legitimate way to identify research and to put some quantitative assessment on the risk, is through the epidemiological questionnaire which we are putting in place.  MAFF have applied a questionnaire over the last many years to TB outbreaks.  This is TB 49.  It was criticised by Krebs as being, perhaps, too weighted towards the badger.  I think that was correct.  What we have done is to remodel that questionnaire, extensively change it.  We have adopted what we hope to be an open-minded approach to addressing a whole range of risk factors.  I do not know whether you want to comment further on that, Rosie, and then it might be appropriate for Christl to comment on the analysis of the multiple questionnaire.
  (Dr Woodroffe) I should add perhaps, to both of you raising these questions about other variables which might be important, that there are other aspects of contracted-out research going out to independent researchers to look at the environmental correlates—

  173.  Have you considered the extent to which these might compromise the trial?
  (Dr Woodroffe) I was leading to that.  What this additional research will do will be to try to identify the environmental correlates of high TB risk to cattle.  The things like soil type and so forth will, when the data come in—and they are operating on a slightly shorter timescale—inform the analysis of the trial data.  It is something that will be possible: to use those data to analyse the trial results more carefully.

  174.  Just one point on that.  I am surprised at the difficulty in estimating badger numbers.
  (Dr Woodroffe) Yes, it is surprising.

  175.  Because the badger is a fairly large creature.  Lives in a lair which is relatively well established.  I imagine wildlife groups have affectionately studied it for many years, so I am surprised that it is so difficult to analyse the numbers.  If you then come to comparators—weasels stoats, foxes, other creatures which may or may not be carriers of TB—how are you going to have reliable variable numbers on these other wildlife species, which may be affecting the end result?
  (Dr Woodroffe) You can count badgers if you are willing to go out and catch them.  It is just that it is very difficult to do over large areas.  Likewise, there are a variety of techniques which have not been calibrated because it has not been so important up to now to have an indirect way of measuring their densities over large areas.  I think that they will use live trapping techniques: there are very, very well established techniques for counting small mammals, weasels, stoats and things.  There are standard ecological techniques for estimating animals in the wild.
  (Professor Bourne) You are right to ask these questions.  They are the very questions we have asked ourselves.  How does one really understand the level of TB prevalence in wildlife populations other than badgers? How does one understand the dynamics of that infection? As I indicated, this work has been based upon scientific proposals, which have been submitted to MAFF and which have been considered by the Group.  We are in the process of considering these.  It has been suggested that various approaches may be made—for example, by blood sampling wildlife—which frankly is a view we do not accept.  There is no way from a blood sample that you can determine prevalence of TB.  It has to be by individual examination by an appropriately equipped laboratory.  So there are big questions here.  The next question is: what is the quantitative number which you wish to sacrifice in order to find this data? Those are the very issues we are addressing ourselves.  As I say, we are still asking for these proposals to be revised and come back to us.  I am expecting some of these proposals to come back to us tomorrow so that we can finalise them, to get them on the road by the beginning of the next financial year.

  176.  It is vital that the proper resources in terms of finance, people, staff, be allocated to it.  The Country Landowners' Association have said MAFF is not allocated enough.  You, yourself, have inferred something of the same when you have said that only a rolling programme is possible because of limited resources.  Has there been any disagreement between the Scientific Group and Government on the resources to be allocated to the culling trial?
  (Professor Bourne) No, there has not.  The only problem, which was quickly resolved by the Minister, was this question of weekend working by the wildlife field staff.  That was very rapidly resolved.  You must appreciate that throwing money at a problem is not necessarily the answer.

  177.  That is a stock comment.  Everyone says that money is not the answer to the problem.  But you want more money, do you not?
  (Professor Bourne) We do not.  It is not as simple as that.  We have come up against a problem of logistics with respect to getting wildlife staff in place and giving them the appropriate training.  It was conceived by the Krebs Committee that these ten triplets would be in place within the ten-month period.  Also, in our first report we suggested we would put two in place in the first year, that is, by January of this year.  Then eight in place during 1999.  It has turned out to be hopelessly impossible to do that because of the problems we have had with training the appropriate number of field staff to ensure that the job is done properly.  It is one thing having money to hire staff.  It is another thing to have the resource to train these people.

  178.  That is a practical problem, not financial.
  (Professor Bourne) That is a practical problem, correct.  I am reassured that the necessary amount of cash is there to carry out the fieldwork.  What we still have to put to the test—and again I am reassured by veterinary services that they will cope with this—is the amount of resource needed for post morteming badgers as they come in from the culling areas, and also adequately from the post morteming badgers as the Road Traffic Accident Survey gets into place.  The assurance we have is that the resource is there.  I must say that MAFF have met every challenge thus far that we have put to them.

  179.  I like their advert for caring killers.  I thought that was a master stroke of advertisement.  What are the latest plans on these targets? How many triplets will be working by the end of 1999 and how many by the end of 2000?
  (Professor Bourne) By 1999 we would hope to put in place another four triplets, and to complete the ten triplets by the end of next year.