Volume 1: Findings and Conclusions
5. The animal health story
Ruminant feed ban
The first BAB
UKASTA's information about breaches of the ban
Cross-contamination in feedmills
What went wrong?

300 Mr Kevin Taylor ¹ became responsible for providing veterinary advice on all aspects of the control of BSE from the time that it became a notifiable disease in 1988. He told us that there was no practical way in which the ruminant feed ban could be enforced, as there was no test which could identify rendered ruminant protein in animal feed. Effectiveness depended on voluntary compliance with the ban. Because of the long incubation period, years would elapse before it would become apparent whether there had been strict compliance with the ruminant feed ban. We consider that it was reasonable to expect that neither feedmills nor farmers would deliberately incorporate MBM in cattle feed. Other sources of protein were available that were only marginally more expensive.

301 No guidance was given to the County Councils and Unitary Authorities, whose duty it was to enforce the ban. We had evidence which suggested that some local authorities made attempts to check on compliance with the ruminant feed ban by sampling, but found this impossible. It is possible that others may have checked the records of feedmills to ensure that MBM was not a component of cattle feed, although strictly they had no statutory right to demand to see these. ² In general we do not believe that any steps were taken by local authorities to enforce the ban during any part of the period with which we are concerned.

302 Mr MacGregor proposed that the introduction of the ruminant feed ban should be handled in a low-key way on the assumption that MAFF had a system for notifying all those who were affected, and in particular farmers. In the event MAFF officials made no attempt to contact renderers, the feed trade or farmers directly, but relied upon meetings with trade associations, or farmers' unions, together with a press release, in order to publicise the introduction of the ban. ³

303 Representatives of the feed industry told us that when the feed ban was introduced, a number of factors combined to detract from any impression of urgency about its implementation:

• the grant of a period of grace in which to use up current stocks of feed;
• the absence of any feed recall;
• the fact that neither import nor export of MBM was to be prohibited;
• uncertainty as to whether MBM was indeed the vector of BSE and, if it was, as to which rendering systems were unable to inactivate it; and finally,
• the belief that a very large amount of infective feed would have to be consumed to transmit the disease.

304 Farmers who gave evidence told us that they did not appreciate the gravity of the situation at the time. It was only the occasional farmer who had experience of BSE and that experience was normally of no more than a single case. They continued to use up any stocks of cattle feed remaining at the time that the ban came into force. ⁴

305 A relatively relaxed attitude to enforcement of the ban was illustrated by the decision of Mr Meldrum in February 1989 that the development of an ELISA test, to detect the presence of ruminant protein in animal feed, should be carried out 'in house' by a senior scientific officer at Worcester VIC, Mr Mike Ansfield. This course had a number of attractions, not least that MAFF would retain the intellectual property in the test, which might prove commercially valuable. It was estimated, however, that it would take between 12 and 18 months to develop the test. The more costly alternative of seeking external collaboration in producing a test would have been likely to produce swifter results. ⁵ As we have commented above, this attitude was a consequence of a failure to appreciate the need to guard against cross-contamination of cattle feed.

306 Although no anxieties were expressed about the adequacy of the action taken by MAFF to eradicate BSE, there were concerns about the risk that BSE might pose, in the interim, to humans and to non-ruminant animals. The scale of infection of cattle during the period before the ruminant feed ban was introduced proved to have been greatly underestimated. By the end of 1988 cases were being reported and confirmed at a rate of over 100 cases a week. The Southwood Working Party had envisaged cases remaining on a plateau at about that rate, but by the end of April 1989 the rate had increased to about 150 cases reported each week. In June 1989 the Government announced its intention to ban SBO (brain, spinal cord, tonsils, thymus, spleen and intestines) from all human food ('the human SBO ban'). This led a large part of the feed industry to impose a voluntary ban on including those categories of offal in animal feed, a ban that MAFF made statutory in September 1990 ('the animal SBO ban') after a number of cases of Feline Spongiform Encephalopathy (FSE) had been identified and BSE had been experimentally transmitted, by inoculation, to a pig. Those events are dealt with later in this chapter.

307 The first case of FSE in May 1990 led to considerable public concern about its implications for human health and to an Inquiry into BSE by the Agriculture Committee of the House of Commons. The primary concern of the Committee was the implications of BSE for human health. So far as animal health was concerned, the Committee observed that the ruminant feed ban, if strictly applied, should arrest BSE. They recommended, however:

That the Government establish an expert committee to examine the whole range of animal feeds and advise on how the industries that produce these should be regulated.

308 This recommendation was accepted by the Government, which set up the Lamming Committee (the Expert Group on Animal Feedingstuffs) in 1991.

309 By the end of 1990 MAFF officials and Mr Gummer, now the Minister of Agriculture, Fisheries and Food, had no reason to doubt the efficacy of the ruminant feed ban. The rate of reported cases had soared until in some weeks these exceeded 400, but they were all cases of cattle born before the ruminant feed ban came into force. Mr Ansfield appeared to have made substantial progress in the development of an ELISA test. His test could detect both ovine and bovine protein in meat and bone meal (MBM). It remained to test it on compound feed.

310 The development of the ELISA test resulted in some concern on the part of UKASTA. It feared that the test would identify small quantities of ruminant protein in cattle feed resulting from cross-contamination with pig and poultry rations in the feedmills, or from traces in tallow incorporated in cattle feed. Its concern was not that this would be sufficient to infect cattle, but that it might result in prosecution of its members for breach of the Regulations. This concern was conveyed by Dr Danny Matthews ⁶ to Mr Meldrum. He told us that at this point he did not recall any concerns at MAFF that cross-contamination of feed might be taking place on a scale sufficient to undermine the effectiveness of the ruminant feed ban.

The first BAB

311 On 22 March 1991 the first BAB ⁷ was reported to Mr Gummer. This was made public by a news release five days later. It caused considerable excitement within MAFF as urgent consideration was given to whether it was a case of maternal transmission or whether it might have been infected by feed. It was, however, only the first of what was to become first a trickle, then a stream and finally a flood. By the end of the year 300 BABs had been reported, of which only 11 had been confirmed. Investigations by Dr Matthews and his colleagues suggested that at least the majority of these cases were caused by feed containing MBM that was still in the feed chain when the ban came into force on 18 July 1988.

312 The Lamming Committee ⁸ met for the first time on 15 February 1991. At their second meeting on 13 March they heard evidence from Mr Meldrum. He told them that he was not totally content with the current controls, as at

. . . present there was no test for ruminant protein in feed. However, an ELISA method was currently being evaluated for use in the field.

313 He said that he was fairly confident that on-farm mixers would observe the controls, despite the absence of a test.

314 Unfortunately, hopes that the ELISA test was almost ready for use were dashed when it was found that most compound feeds produced a positive result even when they included no MBM.

315 When the Lamming Committee reported in June 1992, they commented about BSE that the evidence suggested that in the majority of cases the controls were working, despite the fact that the ruminant feed ban and the SBO ban were to a considerable extent dependent on self-regulation by the industry. They welcomed the development of the ELISA test.

316 As the number of BABs increased, so did MAFF officials' conviction that feed containing ruminant protein had been fed to cattle for a significant period after the ban came into force. In September 1992 Dr Matthews minuted Mr Meldrum commenting that it was clear that the major compounders had needed at least three months to clear stocks, in some cases longer. He added that smaller compounders, who were disproportionately represented among suppliers to owners of BAB cases, having not been party to discussions prior to the introduction of the ban, might be expected to have taken longer to clear their stocks.

UKASTA's information about breaches of the ban

317 By this time 220 BABs had been confirmed. Mr Meldrum wrote to Mr James Reed, the Director-General of UKASTA, suggesting that there had been a time lag of between three and six months before the ban became fully effective. In response to Mr Meldrum's request for information, UKASTA asked all companies represented on its Executive Committee to answer a questionnaire. At a meeting on 10 November, they gave Mr Meldrum the results of this survey, on condition that the information would be treated with the utmost confidence. The survey showed that most compounders had continued to manufacture cattle feed containing ruminant protein into July 1988 and did not clear stocks from their premises until August or September, or even, in a few instances, October. When giving evidence to us, UKASTA representatives suggested that the stocks of cattle feed may have been cleared by incorporation in feed for non-ruminants, so that the survey may not have disclosed deliberate breach of the ban by UKASTA members. We reject this suggestion, as did Mr Meldrum. The contemporary evidence of the meeting on 10 November is unequivocal. That evidence is reinforced by the fact that over 11,000 cattle born in the last five months of 1988 contracted BSE, as did a further 12,600-odd that were born in 1989.

318 These figures will, of course, reflect the use by farmers after 18 July 1988 of feed purchased before that date, but we are satisfied that they also reflect deliberate breaches of the ban by some compounders and others in the supply chain. As a whole the animal feed industry does not emerge from the BSE story with credit.

319 MAFF officials seized eagerly on evidence of breaches of the ruminant feed ban, for the alternative explanation that maternal transmission was occurring was less palatable. In November Ministers were told that there was clear evidence that ruminant feed containing MBM would have been available for six months after the ban came into force. At the year end a MAFF progress report expressed continued confidence that the ban would bring the epidemic to an end.

320 The first half of 1993 saw MAFF officials frustrated in their desire to start testing feed for the presence of animal protein by continuing difficulties in developing the ELISA test - accentuated by suspension of work on the project while the Worcester VIC was relocated to Luddington. MAFF's difficulties were compounded by the fact that they had no legal power to carry out random sampling. Samples could only be taken when there were reasonable grounds for suspecting that the Regulations were being broken. The lengthy incubation period made it difficult to demonstrate such grounds.

321 In September a briefing paper prepared for Mrs Gillian Shephard, who had succeeded Mr Gummer in July as the MAFF Minister, and Mr Nicholas Soames, the Parliamentary Secretary at MAFF, stated that there had been 4,010 confirmed BABs, the great majority of which had had access to ruminant protein in their feed. The paper went on to make the point that the animal SBO ban introduced in 1990 had had the effect of reinforcing the ruminant feed ban.

322 Problems in relation to sampling continued in the first half of 1994. The ELISA test was ready for field testing, but sampling capacity at Luddington was limited and there was no hope of embarking on large-scale monitoring at feedmills. Furthermore, the lawyers were having difficulty finding a path through the maze of different Regulations relating to animal feed that would enable mandatory sampling to be introduced.

323 Towards the end of 1993 Mr Wilesmith had begun to feel concerned that cross-contamination might be taking place at feedmills manufacturing multi-species rations. This concern was taken up by Mr Bradley early in the following year. In a minute to Mr Kevin Taylor, he commented that they had both believed that the animal SBO ban would have stopped any infected ruminant protein getting through into the animal feed chain, but if the SBO ban was being abused there was a weakness in this argument. It was at this time that concern was growing about reports of non-compliance with the animal SBO ban. ⁹

324 By the middle of 1994 MAFF officials had worked out a sampling procedure which they recommended in a submission to Mr Soames. Sampling of cattle feed should initially be carried out on farms on a voluntary basis. Any positive results would lead to mandatory sampling at the feedmill which had supplied the feed. MAFF would carry out the testing themselves rather than entrusting the ELISA test to the local authorities, which had statutory responsibility for enforcing the ban. There were a number of reasons why MAFF officials wished to keep the testing 'in house' - one being apprehension that some local authorities might prove over-assiduous in enforcing the ban. It seems to us that the test was not sufficiently robust at this stage to be used in statutory enforcement of the ban. The decision of MAFF officials that MAFF should use the test on a voluntary basis under a uniform scheme to operate across the country was reasonable.

Cross-contamination in feedmills

325 In June 1994 the possibility that cross-contamination in feedmills was a cause of some of the BABs was discussed with UKASTA's Scientific Committee. They commented that equipment used in feedmills was being updated 'as and when required'. This was the start of a series of meetings between MAFF officials and UKASTA in which each had a similar hidden agenda. MAFF was concerned not to do anything that would lead UKASTA members to cease using animal protein as an ingredient of feed for non-ruminant animals. UKASTA, for its part, was anxious that its members should be able to continue to do this without incurring risk of prosecution should it result, on occasion, in cross-contamination of ruminant feed. UKASTA was to threaten repeatedly that it might have to advise its members to cease using animal protein, while MAFF officials sought to allay UKASTA's anxieties by reassuring its members that sampling was not being used as a precursor to prosecution. In reality, the limitations of the ELISA test, coupled with the requirement under the Order to prove knowing incorporation of ruminant protein, meant that MAFF officials were in no position to contemplate enforcing the ruminant feed ban by criminal proceedings.

326 In July 1994 Mr William Waldegrave succeeded Mrs Shephard as Minister of Agriculture, Fisheries and Food. The following month he was informed of the first four BABs to be reported that had been born in 1991.

327 By September 1994 a number of factors had combined to indicate that cross-contamination in feedmills was a serious problem:

• Inactivation studies had shown that the three systems which provided most of the UK rendering capacity were not capable of inactivating the BSE agent.
• Epidemiological investigations had revealed a correlation between the incidence of BABs and the ratio of cattle to pigs in the different counties. The incidence of BABs was highest in those counties where mills were producing both pig and cattle feed in large quantities. Mr Wilesmith concluded that cross-contamination was likely to be occurring at the mills rather than on the farms, although he recognised that cross-contamination on the farm was possible.
• There was clear evidence of failures to comply with the requirements of the animal SBO ban.
• Four BABs born in 1991 had been confirmed.
• Voluntary on-farm ELISA testing had produced the first positive result.
• Interim results of MAFF's attack rate experiment had shown that 1 gram of infective material was sufficient to transmit BSE when administered orally.

328 This last factor produced a radical change of attitude on the part of both MAFF and UKASTA to the dangers of cross-contamination of feed. In reporting to Mr Waldegrave on 21 November 1994, Mr Richard Packer, the Permanent Secretary at MAFF, stated:

The trade's protestations that cross-contamination never occurred have been reversed; they are now more or less telling us that where the same mill is used for ruminant and non-ruminant feed, some cross-contamination is inevitable, although this is usually at low levels.

329 Mr Packer had plainly been misinformed. UKASTA had expressed concern about cross-contamination at the outset, but had been led to believe that this would not matter because a large quantity of infective material had to be eaten in order to result in infection.

330 At this point UKASTA appeared to come closest to advising its members to cease using MBM in feed. It attempted to elicit from MAFF an assurance that the rendering processes would produce MBM that was 'safe'. Mr Packer was not prepared to provide this. However, he did produce for UKASTA a statement summarising the steps MAFF had taken to prevent transmission of BSE to cattle. This emphasised that the controls over the implementation of the animal SBO ban were being strengthened and that more effective rendering processes were being adopted. The statement ended:

The Ministry considers there to be no reason in principle why [ruminant protein] should not continue to be used in non-ruminant feed, even in premises preparing feed for ruminant and non-ruminant species, provided that steps are taken to prevent accidental inclusion in ruminant rations.

UKASTA accepted this statement as satisfactory reassurance and the use by its members of MBM in non-ruminant feed continued.

331 We had evidence from some of the major feed compounders that once they had been made aware that cross-contamination was a cause for concern, they took steps to identify the critical control points and to modify their production lines so as to reduce the risk of contamination occurring. UKASTA and MAFF reached agreement under which compounders were permitted, under a quota system, to submit samples to Luddington for ELISA testing in order to check that their production was free of contamination. This sampling was carried out in parallel with sampling by MAFF of feed on farms on a voluntary basis. Mr Meldrum told us of at least one occasion on which this led to the identification of a mill where cross-contamination was occurring, and to the mill in question taking steps to remedy the problem. Problems were, however, still being experienced with the ELISA test and it was apparent to both MAFF and UKASTA that it was capable of giving false positives and false negatives.

332 As at 23 January 1995, the number of confirmed BABS had risen to 15,771, of which 812 had been born in 1990 and 9 in 1991. In the following month it was confirmed that the attack rate study had demonstrated that 1 gram of material was sufficient to produce oral transmission. When this was reported to Mr Waldegrave, he asked whether further steps needed to be taken to ensure that compounders' feedlines were clean. Mr Meldrum replied that the short answer was 'No'. The important thing was to prevent infected material entering the feedlines. As to this, the only action that he could recommend was to continue to intensify controls on the disposal of SBO. Mr Waldegrave accepted this advice.

333 In May 1995 MAFF officials were giving consideration to arranging advisory visits to feedmills in order to give guidance on how to avoid cross-contamination and, at the same time, to replacing voluntary sampling on farms with unannounced sampling visits to mills. Our impression is that UKASTA was less than enthusiastic about these proposals. Its first duty was to protect its members' interests and it showed a continued awareness of the need to protect its members from the risk of prosecution. However, consideration of voluntary visits and sampling was overtaken by a Decision of the European Commission ¹⁰ adopted on 18 July 1995. This required routine monitoring of feedmills, and in particular of mills which produced both ruminant and non-ruminant feed, to include official ELISA tests for the presence of animal protein.

334 Discussions with UKASTA about implementing this Decision did not receive an enthusiastic response. UKASTA did, however, cooperate in the drafting of a letter from MAFF to all manufacturers and mixers of feedstuffs, drawing attention to the need to avoid cross-contamination and giving guidance on how to do so. We found this a bland document. In particular it made no mention of the fact that experiments had demonstrated that as little as 1 gram of infective material could result in oral transmission of BSE. A revised Advisory Note directed specifically to farmers was drafted by MAFF in November 1995. This was an admirable document giving detailed advice on all the different ways in which feed might become contaminated on the farm or in the course of farm mixing. Unfortunately, this draft got bogged down in the course of the consultative process, involving input from the Spongiform Encephalopathy Advisory Committee (SEAC) and the Parliamentary Secretary, and had not been sent out when it was overtaken by events in March 1996. This was one of a number of examples in the BSE story of the best being the enemy of the good.

335 By 24 August 1995 the number of confirmed BABs had risen to 21,475, of which three had been born in 1992. Although it was not initially appreciated, the effect of the Commission Decision requiring mandatory sampling of feed was to give MAFF officials the right to enter mills and carry out the sampling. An Animal Health Circular was drafted instructing State Veterinary Service staff on measures to implement the mandatory sampling regime, which was initiated early in 1996. Although the ELISA test was still not perfected - we understand that it remains imperfect to this day - the first round of tests produced four positive results from 25 mills tested.

336 On 6 July 1995 Mr Douglas Hogg succeeded Mr Waldegrave as Minister of Agriculture, Fisheries and Food. One of his first acts was to introduce the Specified Bovine Offal Order 1995 which, as we explain later in this chapter, dramatically improved the regime for enforcing the animal SBO ban. Later in the year, Mr Hogg discussed with Mr Meldrum whether further measures should be taken in the feedmills to address the risk of cross-contamination. Mr Meldrum explained that mandatory sampling was to be introduced and advised that it would not be practical to require feedmills to set up separate production lines for ruminant and non-ruminant feed. Mr Hogg accepted this advice.

337 SEAC reviewed from time to time the implications of the BABs and the action that MAFF officials were taking to address the cause of infection. ¹¹ The Committee urged the importance of the development of the ELISA test, but in general endorsed the action that MAFF was taking. On the identification of the probable link between BSE and the new variant cases of CJD, SEAC's attitude changed. Members considered that it was of paramount importance to bring the BSE epidemic to a close as swiftly as possible, thereby protecting both animal and human health. To achieve this SEAC proposed a ban on the use of all meat and bone meal of mammalian origin in farm animal feed. This would remove all possibility of the contamination of ruminant feed. The Government accepted this advice and gave effect to it on 29 March 1996. ¹²

338 As at end-June 2000 the number of confirmed BABs stood at 41,538. Of those 179 were born in 1995 and 2 in 1996. For each confirmed case, several will have been slaughtered before developing clinical symptoms. Almost all of these cases will have resulted from eating MBM derived from apparently healthy animals, because animals showing signs of BSE were being slaughtered and destroyed.

What went wrong?

339 When looking back with the benefit of hindsight, we have identified a number of things that went wrong in the history of the ruminant feed ban.

340 At the time that the ban was introduced, it was thought that all that the Regulations needed to do was to prevent the deliberate inclusion of ruminant protein in cattle feed. The Regulations were not designed to make unlawful the accidental contamination of cattle feed with small quantities of feed containing ruminant protein. Nor did they confer adequate powers of entry, inspection of records and sampling.

341 For the same reason, the development of a test to detect the presence of small amounts of ruminant protein in cattle feed was not treated as a matter of high priority. Five years were to elapse before the ELISA test was developed to a point at which some practical use could be made of it.

342 These shortcomings were symptomatic of a lack of rigorous thought about the implementation of the ruminant feed ban and the risk of cross-contamination at the time that it was introduced, which we have discussed in Chapter 3 above.

343 The risk of cross-contamination was then masked by the introduction of the animal SBO ban. We believe that it was because of the second line of protection apparently afforded by this ban that the Lamming Committee had no concerns about the possibility of contamination of ruminant feed. The Committee expressed concerns about the lack of control of on-farm mixing, but not in the context of BSE.

344 No sense of urgency attended the introduction of the ruminant feed ban. This was because of a fundamental misunderstanding of the scale of infection that was taking place. It was believed that infection had probably been occurring at a uniform rate of perhaps no more than 60 cases a month. In fact, the latent snowballing effect of recycling had boosted the rate of infection to 10,000 cases a month or more. ¹³ No one is to be criticised for failure to appreciate the scale of the problem. We do, however, censure (although we do not have the means to identify) those in the feed industry who deliberately breached the ruminant feed ban by continuing to supply ruminant feed that contained animal protein after 18 July 1988, when the ban came into force.

345 For some years MAFF officials proceeded on the basis that all necessary steps had been taken to eradicate BSE. As Mr Thomas Eddy ¹⁴ was to remark to Mr Waldegrave in February 1995, the long incubation period meant that five years had to elapse before it could become apparent whether precautionary arrangements and compliance by the industry were adequate. As the numbers of BABs increased, and their dates of birth grew later and later, MAFF officials progressively extended the period of carry-over of cattle feed containing ruminant protein that they assumed must have occurred. To an extent they were correct and we do not feel that they can be criticised for not appreciating until 1994 that a significant cause of infection of BABs was cross-contamination of cattle feed.

346 At that stage there were a number of alternative options to address the problem. The most radical was to prohibit the use of MBM in all animal feed. As Mr Meldrum remarked to Mr Hogg, the economic consequences of this would be 'devastating' and a serious waste disposal problem would be created. In the absence of evidence that BSE was transmissible to humans, we do not consider that this extreme measure was called for. To have adopted it, simply to prevent cross-contamination of feed in feedmills and on farms, would have been an admission of defeat. Other, less drastic, viable options were open.

347 At one time we were attracted by the view that feedmills should have been required to process feed for ruminants and feed for non-ruminants in separate production lines. We were, however, persuaded that to have insisted on the heavy expenditure necessary to achieve this would also have been disproportionate. MAFF's approach was to concentrate on procuring proper implementation of the SBO ban. This included requiring renderers to process SBO in dedicated plant. We consider that it was reasonable for MAFF officials and Ministers to conclude that it was not necessary to require feedmills to undertake, in parallel with renderers, the expense of installing duplicate lines. Instead MAFF sought to encourage feedmills to take voluntary steps to reduce the potential for cross-contamination.

348 With hindsight, we can deduce that the measures that MAFF had already taken had had a dramatic cumulative effect in reducing infection year on year. Looking back five years from end-June 2000, we see only 232 BABs which were born in 1995, and only 2 born in 1996. But for the events of March 1996 MBM would have remained part of the diet of pigs and poultry and MAFF would have been able to claim that, by a combination of the ruminant feed ban and the animal SBO ban, they had virtually eradicated infection of cattle with BSE.

349 It is this consideration which has led us, at the end of the day, to conclude that no criticism need be made of the somewhat muted attempts by both MAFF officials and UKASTA to get feedmills and farmers to take steps to tackle cross-contamination. When it was appreciated that this was occurring, and that a quantity as small as 1 gram of infective material would suffice to transmit the disease orally, one might have expected UKASTA urgently to draw these facts to the attention of its members and MAFF to do the same in relation to cattle farmers and to feedmills that were not members of UKASTA.

350 We suspect that the more measured approach that was adopted was explained by a shared reluctance on the part of MAFF and UKASTA to adopt a course that might lead to feed compounders ceasing to use animal protein as a feed ingredient. For the reasons that we have given, we do not feel that this was an unreasonable attitude to adopt.