III. THE KREBS REPORT
33. On 23 July 1996 Douglas Hogg MP, then Minister
of Agriculture, Fisheries and Food, responded to growing concern
over Government policy on bovine tuberculosis by announcing an
independent scientific review under the chairmanship of Professor
John Krebs, Chief Executive of the Natural Environment Research
Council. Five reasons were later given for the initiation of the
review: (a) the link between badgers and bovine TB remained unproven;
(b) the only reliable method for showing that badgers have bovine
TB involved killing them first as the blood test on live badgers
was insufficiently sensitive; (c) there was still little sign
of a successful vaccine against bovine TB in badgers which had
long been seen as the solution; (d) the effectiveness of badger
culling remained doubtful; and (e) bovine TB was spreading.
These reasons more than justified the setting up of an independent
inquiry on this scale. As the Minister currently in charge of
the policy commented to us: "it is a tribute to the previous
Government they had set up the Krebs inquiry, because if they
had not we would have had to do something similar."
34. The terms of reference of the Krebs review were:
"To review the incidence of tuberculosis in cattle and badgers
and assess the scientific evidence of links between them; to take
account of EU policies on reducing and eliminating the incidence
of tuberculosis in cattle; to take account of any risk to the
human population; and accordingly to review, in the light of the
scientific evidence, present Government policy on badgers and
tuberculosis and to make recommendations".
The NFBG and others suggested to us that the focus on badgers
coloured the inquiry and prevented the team addressing other issues.
However, Professor Krebs told us that the terms of reference had
been discussed before they were finally agreed by Ministers and
that all members of the group were satisfied with them. The group
regarded the terms of reference as "enabling rather than
proscriptive ... we felt free to range quite widely in looking
at the scientific issues related to the bovine tuberculosis problem".
We agree with Professor Krebs that the terms of reference were
appropriate and not restrictive, but we understand the concern
of the conservationists about the fact that the title of the review
was "Bovine tuberculosis in Cattle and Badgers".
35. Five other experts were appointed to the Review
Group, all academics. They were assisted by Dr Simon Frost and
Dr Rosie Woodroffe, then from the Departments of Zoology at the
Universities of Oxford and Cambridge, respectively. None of the
group had direct veterinary experience. A call for written evidence
resulted in 68 submissions from organisations, academic institutions
and individuals. Six presentations were heard from other interested
parties, including MAFF and academic experts, and meetings were
held with representatives from the farming industry, veterinary
interests and wildlife organisations. The group also made site
visits to farms and to the MAFF Wildlife Unit and the Woodchester
Park Badger Research Station and held discussions with representatives
of the appropriate government departments from Northern Ireland,
the Republic of Ireland and New Zealand, countries with a similar
problem with bovine TB in cattle. This extensive programme of
consultation and research may have contributed to the delay in
the timetable for the report. When the review was announced, it
was expected that the group would report in early summer 1997.
In the event, they held a meeting to discuss the main questions
to be addressed in the final report on 25 June 1997 and the report
itself was published on 16 December 1997.
36. The Krebs report is an impressive document. Well-written
and researched, it covers a wide range of issues and provides
much useful analysis of the history of Government policy on the
control of bovine tuberculosis and of the current scientific understanding
of the epidemiology of the disease. It concentrates on evidence
of TB in badgers and the transmission of M. bovis from
badgers to cattle but, as the NFBG acknowledge, it goes "beyond
[the] terms of reference" to "highlight the other possible
contributory factors to the TB problem - including other wildlife,
climate and poor animal husbandry".
Of its many conclusions, the one which was frequently repeated
in evidence to us was that: "The sum of evidence strongly
supports the view that, in Britain, badgers are a significant
source of infection in cattle".
Krebs went on to state that "Most of this evidence is indirect,
consisting of correlations rather than demonstrations of cause
and effect; but in total the available evidence, including the
effects of completely removing badgers from certain areas, is
compelling". He added two caveats: "it is not, however,
possible to state quantitatively what contribution badgers make
to cattle infection" and "it is not possible to compare
the effectiveness of [previous policies for killing badgers];
nor is it possible to compare any of them with the impact of not
killing badgers at all, because there have been no proper experiments".
His primary conclusion was that "The control of TB in cattle
is a complex problem and there is no single solution. We recommend
a combination of approaches on different timescales".
37. The Krebs report contains 16 pages of conclusions
and recommendations. The latter were grouped into four categories:
those aiming to understand the causes of herd breakdown, to evaluate
the effectiveness of currently available strategies to reduce
herd breakdowns, and to develop improved strategies to reduce
herd breakdown, and other recommendations which were mainly concerned
with the Government's approach towards the commissioning and use
MAFF summarised the main recommendations as:
1. The development of a cattle
vaccine, considered to be the best long term option to control
2. Major research initiatives aimed at achieving
a better understanding of the causes of TB, and developing improved
strategies to reduce outbreaks;
3. A randomised culling trial to test the effectiveness
of different strategies and to provide unambiguous evidence of
the role of badgers in cattle TB, the trial to be overseen by
an expert group; and
4. That the Government should work with the farming
industry to improve husbandry methods to minimise contact between
cattle and badgers. 
The research proposals included epidemiological work
to look at the local correlates of risk to cattle, more research
to establish transmission routes, investigation of other potential
wildlife reservoir species, spoligotyping as a possible way of
identifying a link between TB in badgers and in cattle, and the
reinstatement of the survey of badgers killed in road traffic
accidents. All the measures were presented as a package but the
culling trial has naturally drawn most attention and criticism,
and it is this part of the report which is often meant when reference
is made to the "Krebs experiment".
38. For the culling trial, Krebs recommended that
ten groups of three areas in TB hotspots, each measuring 10km
by 10km, should be the basis of a field trial. At the time this
covered most of the main hotspot areas. In each triplet one area
would be subject to proactive culling of badgers, another to reactive
culling following the identification of TB in cattle, and the
third to survey work only. The Krebs group undertook a statistical
power analysis (see Glossary) on the rate of breakdown and concluded
that the proactive strategy could provide clear results within
one year but a full quantitative assessment which could provide
a sound basis for future policy would take longer.
In evidence to us Professor Krebs described his design as set
out in the report as "sketching out a concept rather than
writing an implementation plan".
For this reason, we can divide the factors that made many witnesses
claim that the trial was "fundamentally flawed" or "significantly
flawed, and unnecessary"
into objections of principle, process and practice. In the next
section we discuss the latter two which relate to the implementation
of the trial by the Bourne Group and MAFF. Here, we consider whether
Krebs was right in principle to recommend such an experiment.
39. The aim of the culling trial is "to quantify
the impact of culling badgers".
As Krebs himself pointed out, among the opponents of the recommendation
"there is a great diversity of views ranging from those who
said the trial was a complete waste of time because we already
know that badgers give TB to cattle, to those who said the trial
is a complete waste of time because we know that the badger is
Evidence to us reflected this dichotomy of opinion although significantly
more organisations expressed the first view than the latter. While
it is perhaps to be expected that the RCVS or farmers should judge
that the badger was already proven to be the source of bovine
TB in cattle,
it was less predictable that the Wildlife Trusts should adopt
the same view on the link and that even the NFBG would accept
However, their agreement on the futility of the trial resulted
from very different propositions. The RCVS had wanted Krebs to
recommend "action that would result in the removal of all
badgers in infected areas",
while the Wildlife Trusts believed that "to kill 20,000 badgers
or whatever in order to find out for certain whether there is
a link seems to us a waste of badger lives".
40. This difference illustrates the need to put badger
culling on a scientific footing. Until there is a clear and quantitative
understanding of the link between badgers and cattle TB, it is
impossible to develop an effective and rational policy on its
control. The disappointment and frustration of both farmers and
wildlife groups that Krebs was unable to come up with a definitive
answer to the problem is natural and to some extent can be seen
as the inevitable result of a clash between scientists who want
real scientific proof and practitioners who want action, as most
clearly evidenced by the attitude of the RCVS. However, we
accept Professor Krebs's conclusion that the evidence of a link
between badgers and cattle TB is compelling but not conclusive
and that a field trial is required to test and quantify the link
between badgers and cattle.
41. It has been argued by several witnesses that
the trial is unnecessary as the link has already been established
by previous culls, especially those at Thornbury, Gloucestershire
from 1975 to 1981 and East Offaly in the Irish Republic from 1989
These examples of large scale badger clearances, together with
two further instances at Hartland in Devon and Steeple Leaze in
Dorset, are described in the Krebs report.
In addition, we are grateful for a memorandum from the Irish Government
on the outcome of the East Offaly project. In all four cases,
where badgers were cleared the incidence of TB in cattle fell
significantly. However, the difficulty with these results is that
in none of the cases were adequate experimental controls in place.
As Krebs commented, "badger removal might have caused the
observed fall in herd breakdown rates, but the possibility remains
that some other unidentified factor could have been responsible".
The study of the Irish experiment concluded: "As this study
involved just a single area with high badger numbers and a high
cattle density, it is not possible to directly extrapolate from
the outcome to other areas. Further studies will be required to
confirm the conclusions".
The Irish Republic has instituted a new trial involving four areas,
each consisting of an area designated for complete removal of
badgers and one for more limited removals, the results of which
will be reviewed in December 2002.
Their action in establishing this trial following the East Offaly
experiment underlines Professor Krebs' argument that these previous
clearances provided "pretty strong circumstantial evidence"
but that in order to convince the doubters it was necessary "to
address it more rigorously".
42. One factor in deciding whether the culling trial
is justified in principle is the number of badgers likely to be
killed in order to implement it. Media headlines have repeatedly
cited as fact the figure of 20,000 badgers.
This is based on an estimate made by the NFBG
and is at odds with Krebs' own estimate of 12,500, which has been
accepted by the Government.
The discrepancy is largely accounted for by differing views on
badger densities in the trial areas. The figure given by Krebs
also reflects varying totals of badgers culled during each year
of the trial, ie 7,500 to establish the proactive and reactive
triplets in the first year and around 1,250 for each subsequent
year. To put both estimates into perspective, however, in the
last year of the interim strategy (1997) 2,447 badgers were killed
in official programmes,
and every year some 50,000 badgers are killed on the roads.
Our concern for animal welfare extends not just to wildlife but
to domestic animals as well. We conclude that, seen in context,
the number of badgers likely to be culled in the trial will not
substantially affect the overall UK badger population and is justified
in pursuit of a soundly-based policy which should save unnecessary
slaughter of both badgers and cattle in the future.
43. The emphasis on the culling trial has deflected
attention away from Krebs' other proposals. Professor Krebs told
us that "We clearly said at the beginning that this is a
multifaceted problem; it is complex; there is no simple, quick
and we believe that his report reflects this complexity. It goes
some way towards setting out the holistic approach called for
by the NFBG and others, while reflecting the balance of evidence
that the badger has a part to play in the transmission of bovine
TB to cattle, a fact that justifies the particular attention paid
to the badger. It was suggested to us that the lack of a theory
of causation invalidated the trial and that it would be preferable
to abandon the trial in favour of molecular testing.
Both Professor Krebs and Professor Bourne were adamant that while
molecular testing may have developed sufficiently in a few years'
time to help identify transmission routes and indeed the Krebs
report contained proposals for using this method, it was not feasible
that "we could put in place now molecular epidemiology which
would negate the need of doing the trial".
On balance, therefore, we conclude that Krebs' approach of
combining a culling trial with other research is the correct one
and we see no reason why any of the Krebs proposals should be
abandoned. We agree broadly with Krebs' conclusions and recommendations.
44. Nevertheless, we have reservations about the
priority Professor Krebs awarded to one particular area of interest.
From the evidence we have received, we are convinced that more
attention should have been paid to the role played by husbandry
in preventing TB breakdowns.
The Krebs group did briefly touch on this subject in the report
and suggested that areas outside the trial would be suitable for
an experimental comparison of proactive husbandry methods. They
recommended that the farming industry itself should take the lead
in this "comparison of the impact of simple husbandry techniques",
aided by advice from MAFF. Krebs concluded that "Husbandry
may well play an important role as part of the long-term solution".
However, the weight of this finding is not balanced by analysis
in the report itself which we find disappointing in this respect.
Professor Krebs explained that "we did not write a great
deal about it because we did not have the expertise" and
the group also wished to complete their work in good time.
We understand this explanation and acknowledge that the introduction
of the subject was at the instigation of the group since it did
not fall into the remit given by the Government, but we regret
that Professor Krebs and his colleagues were unable to review
the literature relating to this important area with the thoroughness
of the rest of the report.
45. Our one remaining concern about the Krebs report
is that it failed to consider the situation of the farmers in
areas outside the hotspots chosen for the experiment. As we have
seen, the problem is worsening with the result that, as the NFU
told us, "when Krebs came out the ten triplet areas that
he had proposed would have covered at least 75 per cent of affected
areas and in this period where we have gone on it now covers less
than 50 per cent".
Of course, Professor Krebs is not to blame for the delays in implementing
his recommendations but in any case he offered farmers outside
the triplets only the chance of participating in the limited husbandry
experiments and he strongly recommended that no culling should
be carried out in these areas.
In this way, he took no account of the need for short term containment
measures while the long term strategy is developed. At the moment,
badger-culling is the only measure for which there is any evidence
of short-term effectiveness. In the treatment areas, this is recognised
since the trial itself is a control measure as well as a source
of data but farmers in the other areas face continuing problems
with TB with no means of addressing them.
Conclusions on the Krebs report
46. On the whole, notwithstanding the reservations
expressed above, we congratulate Professor Krebs and his team
on his cogent report. His findings have been accepted by the vast
majority of witnesses to our inquiry, while the NFU told us that
the report "represents the only real method of finding a
lasting answer to this problem in the management of TB".
The TB situation has become even more urgent in the 16 months
since Professor Krebs published his report, with the continuing
rapid increase in the number and spread of herd breakdowns. We
turn now to the progress made by the Government in implementing
80 Government response to the Krebs report, para 6. Back
Ev. p.40. Back
84 Qq 2, 3. Back
p.38, para 20. Back
Executive summary, para 5. Back
paras 6 and 7 Back
para 1. Back
Chapter 7, pp.135-7. Back
Factsheet C4. Back
Ev. pp.175, 189. Back
Executive summary para 9. Back
424; Ev. p.123. Back
352, 292. Back
pp.186, 189-90, 193, 231. Back
pp. 30-1. Back
example, see the Guardian, 12 April 1999. Back
5.6. 29, 5.6.30. Back
Debates, 24 June 1998, col.
111 Q 31. Back
3-4, 203-4. Back
paragraphs 106 to 113 below. Back
Executive summary, para 15. Back