Volume 5: Animal Health, 1989-96
2. The ruminant feed ban, 1989-96
The chronological account
Developments in 1994 - on-farm testing regime is established

Developments in 1994 - on-farm testing regime is established

2.239 In a statement to the Inquiry, Dr Matthews summarised what he perceived to be the situation regarding testing for ruminant protein in feed and the practical difficulties involved in establishing a testing regime:

By the beginning of 1994 it was felt that the test was ready for field testing, but capacity at Luddington VIC was limited. Despite the potential of ELISA tests for testing considerable numbers of samples at speed, as I recall, the major difficulty with this test is the amount of time needed to prepare samples for testing. Because of this, and concerns about the robustness of the test, there was no hope at that point of embarking on large scale monitoring at feedmills. In addition, because source test materials had been supplied by a limited number of companies during the developmental stages it was likely that field testing would present a new range of proteins against which the antibodies had not yet been tested. A pilot study therefore began in June 1994, targeting farms on which BABs born in 1990 or later were being presented . . . The method chosen to select farms was no more than a means of stratifying the sampling regime, and could not attempt to identify the source of infection for the BAB being dealt with at the time. It was also considered possible that a farm that was presenting BABs born in 1990 or later could well prove to be an indicator of a supply mill that had had a long term cross-contamination problem. ¹

Doubts cast on routine sampling powers

2.240 In her minute to Mr Howard dated 17 December 1993, Ms Behzadi advised that section 76 of the Agriculture Act 1970 gave powers to take routine samples, provided that the requirements of section 77 were met. On 24 January 1994 Ms Behzadi told Mr Howard that she was 'concerned as to whether your plans for the testing at Luddington will fulfil the requirements of section 77 of the Act'. ² Section 77 required the division of samples and their analysis by an 'agricultural analyst'. Agricultural analysts were to be appointed by local authorities. ³

2.241 On 31 January 1994 Mr Howard circulated a minute summarising the progress made in establishing routine sampling of ruminant feed. He advised that despite earlier confidence, it was not feasible to use the Agriculture Act 1970 for routine sampling, since it required sampling to be undertaken by an 'agricultural analyst' as defined in the Act. Mr Howard attached a draft minute from Mr Taylor to Mr Meldrum, which suggested ways of overcoming the difficulty, but he said that 'in the short term it looks as though we will have to be content with voluntary sampling'. ⁴

2.242 It was suggested in the draft minute that:

One solution to this problem would be to make the test available for local authorities to carry out. I understand though that the test requires the production of specific antisera to bovine, ovine, and porcine protein, and vigorous quality control tests to ensure specificity and sensitivity of the test. Whilst Luddington have sufficient tested material for conducting several thousand tests, they are not equipped to produce kits for use by other laboratories on a large scale. Whilst the test methods have apparently been published with a view to patenting, the question of mass production has not been addressed.

Another solution would be to ask local authorities to appoint analysts from Luddington as 'agricultural analysts' as required under the Agriculture Act. We are pursuing this possibility with lawyers which if legally acceptable would have to be discussed with the local authority associations. ⁵

2.243 The draft minute stated that samples taken in the immediate future would have to be on a voluntary basis with all testing done at Luddington VIC. This could be done either using veterinary staff carrying out investigations into BSE suspects, or by manufacturers through UKASTA. The first option, which could be pursued along the lines suggested by Dr Matthews in his minute of 1 October 1993 (see paragraphs 2.224-2.228), would provide data quickly and in the long term could operate effectively alongside a statutory sampling regime. For the second option:

UKASTA would probably get support from some of their members for the carrying out of voluntary sampling, by MAFF on a chargeable basis, as they have supported the banning of the incorporation of ruminant protein into feedingstuffs for ruminants. Costs imposed on the compounder could be a determining factor because in addition to the charge for taking the sample, the VIC would be charging for the testing of samples. ⁶

2.244 It was recommended in the draft minute that arrangements should be made to pursue both options for voluntary sampling. It was also recommended that the statutory sampling regime should be pursued and introduced as quickly as possible once the difficulties were overcome and the Ministers' agreement obtained. ⁷

2.245 In his covering minute, Mr Howard sought advice from Ms Behzadi on whether local authorities could legally appoint analysts at Luddington as 'agricultural analysts' under the Act. He also questioned whether feedstuffs from all investigations into BSE suspects could be sampled. Mr Howard disagreed with Dr Matthews's recommendation that samples should be held until BSE was confirmed in the suspect (see paragraph 2.228), and said:

. . . we are trying to find out if the ban is being complied with rather than associate positive test results with cases of BSE. The sooner we collect some meaningful data the sooner we will know if there is a problem. ⁸

2.246 In a manuscript note on the covering minute, Mr Robertson indicated that he was content with the draft minute, but noted:

It must be made clear to Mr Gayford [of the VIS] that this will not be a subject to be raised by him at his meeting with UKASTA and BVA unless they mention it, in which case he can give a low-key answer. He will need a short brief on this subject so that he knows what to say if he is pressed on the matter. ⁹

2.247 On 2 February 1994 Mr Bradley minuted Mr K Taylor on the policing of BSE controls. He believed that 'we are content at present that all the necessary controls to protect public and animal health are in place'. However:

I, amongst others, have stressed the importance of ensuring that these controls are being effectively policed with particular respect to the ruminant feed (RF) ban and the SBO ban. It has always been important to do this but any deficits will be more plainly revealed for all to see in the next few years and there could be serious financial implications re compensation and disposal, not to mention drains on veterinary resources. ¹⁰

2.248 A number of events had focused Mr Bradley's attention on policing. First, the Lamming Report had welcomed the development of tests for detecting ruminant protein in feed and recognised that when they became available they would provide an additional safeguard (see paragraph 2.152). Mr Bradley suggested:

They are available and they have not been used anywhere to my knowledge. The Swiss have been testing feed for well over a year and the Lamming Report was published in 1992. In 1994 we are still discussing legal aspects. I believe there could be substantial criticism of this retrospectively especially when it is formally shown that the RF ban was significantly abused and that continuing abuse may be occurring (ie BSE in animals born in 1991, though it is premature to accept this as a fact or as feed source). Why cannot feed on MAFF farms be checked now or why can it not be done on a voluntary basis and why cannot we move legally at a much faster pace? ¹¹

2.249 During oral evidence, Mr Bradley explained his belief that other countries were already testing feed for the presence of animal protein at this time:

When we had the feeding of ruminant protein ban in originally ruminant protein and subsequently mammalian protein, it was a question of being able to have a check system to see that this was effective. And there had been an ELISA test which, according to the Lamming Committee's report, the working group on feedstuffs had been validated in about 1991 or 1992, whatever date was their report. However, in other countries they were using other methods quite outside of this ELISA. This was a microscopic examination, so they would take a sample of the meat and bonemeal or the cattle food product, examine it under the microscope and I believe by looking for bone structure they could actually identify whether mammalian protein was present. Obviously they could not detect species but could detect whether mammalian protein was present. This had been used in the Republic of Ireland to the best of my knowledge, and in Switzerland. If that is what I said, that was the period over which it was being used. To the best of my knowledge this method has not been used in the UK. ¹²

2.250 In his minute, Mr Bradley also voiced concern about the enforcement of the SBO ban (see Chapter 4). He said:

I believe we are both of the opinion that whilst the RP [ruminant protein] ban was effective, though not completely so, after 18 July 1988 any infected RP getting through would be stopped by the SBO ban 2 and a bit years later. If the SBO ban itself is being abused then there is a weakness in this argument. Furthermore, even if the SBO ban was completely effective and cattle were therefore not being exposed to infection via protein in the feed any abuse of the [RFB] would still be revealed by the test we hope to apply. If a positive test was announced then we would be unable to determine whether the protein contained infectivity and certainly would not be able to convince our extremist critics that it did not. It is therefore absolutely essential that effective audits are carried out and policing is continuous and sound and is seen to be so by publication of the results where this is appropriate or, at the very least, to have them available to counteract criticism. ¹³

2.251 BABs born in 1991 were also a source of concern to Mr Bradley and he thought that if feed was the cause 'we really do have to get to the bottom of it'. He concluded:

For all these reasons I believe we have to quickly and effectively re-assess, and if necessary, improve the policing of controls both via MAFF and the Local Authorities. Any trickle of infected RP into the cattle feed chain could result in an unfortunate plateau of confirmed BSE cases in a couple of years which will be hard to deal with and may even prevent export of live breeding cattle or, re-introducing a ban if we are successful in having it lifted this year. We can afford no trouble at all and must see that it does not happen. ¹⁴

2.252 On 15 February 1994 Mr Howard recirculated a draft minute from Mr Taylor to Mr Meldrum, which included comments from recipients of the first draft on 31 January 1994. His covering minute noted:

. . . this minute concludes that the best option open to us is to carry out voluntary sampling on-farm at the same time that the investigations into reported BSE suspects are done. There are difficulties regarding statutory sampling. In short we either make the 'Luddington test' available to local authorities or we appoint analysts at Luddington as 'agricultural analysts' under the Agriculture Act. ¹⁵

2.253 Following her receipt of this minute on the next day, Ms Behzadi pointed out a further obstacle. Section 77(4) of the Agriculture Act 1970 required that when a method of analysis had been prescribed, it must be followed. Part 4 of schedule 2 of the Feeding Stuffs (Sampling and Analysis) Regulations 1982 prescribed the method for the determination of protein in feedstuffs. ¹⁶ Ms Behzadi therefore believed that 'we are faced with the problem that it seems as though the only method of analysis which could be adopted by an agricultural analyst in the routine sampling of feeding stuffs for ruminant protein . . . is that method laid down in Part 4 of Schedule 2'. The ELISA test would not be useable. ¹⁷

2.254 After consulting with a colleague, Ms Behzadi revised this opinion on 18 March 1994. She advised that an agricultural analyst would not be tied to the method in Part 4 of schedule 2, since it did not deal specifically with the detection of ruminant protein. Therefore, the Agriculture Act 1970 could be used as a basis for statutory sampling, so long as 'agricultural analysts' undertook the testing. ¹⁸

2.255 On 12 March 1994 an article by Miss Linda Hoinville of CVL's Epidemiology Department entitled 'Decline in the incidence of BSE in cattle born after the introduction of the feed ban' appeared in the Veterinary Record. It raised the possibility of accidental cross-contamination, stating:

The gradual decline in the incidence [of the disease] following the introduction of the feed ban is consistent with a decrease in the amount of contaminated feed remaining on farms or in the feed supply chain. This suggests that the feed ban was not immediately fully effective. The continued exposure of animals born in 1989 could be due to this gradual exhaustion of the supply of feedstuffs manufactured before July 1988. The accidental inclusion of meat-and-bone meal in cattle feedstuffs or contamination with ingredients used in pig and poultry rations is also possible. ¹⁹

Study of the effect of rendering on BSE infectivity

2.256 As discussed in paragraph 2.5, when the RFB was introduced in 1988, it was intended to be a temporary measure pending the results of BSE inactivation experiments. A study to determine the effectiveness of various rendering processes on the inactivation of the BSE agent was established in November 1990. On 21 March 1994 Mr Meldrum briefed Mr Soames on the preliminary results of these experiments. The brief was copied to Mrs Shephard and Mr Packer, amongst others. It said:

The first phase of the project studied BSE. The bioassays have not yet reached an end point, so it cannot yet be concluded that any system provides inactivation. It is, however, already clear that the three systems which collectively provide most of the British rendering capacity do not provide effective inactivation - in fact it appears that only two plants are satisfactory and this would represent less than 20% of total red meat rendering capacity. At least one of the systems appears to have no detectable effect on the amount of infective agent present. This is particularly worrying because these three plants process a significant volume of product and are only used in the UK and USA.

The results support the hypothesis that BSE was caused by the presence of the agent in animal protein which was fed to cattle and underline the wisdom of the measures which have been implemented since July 1988 to prevent ruminant derived protein being fed to ruminant animals. ²⁰

2.257 Mr Meldrum advised that existing control policies were 'vindicated' and that no change was needed to respond to the findings. He noted that once the results were publicised, the rendering industry could be affected by buyers unwilling to purchase MBM from plants using processes that did not inactivate the BSE agent. Mr Meldrum continued:

The Government might also be criticised for allowing the continued use of rendering systems which were perceived to be ineffective in activating SE agents, and by that means exposing pigs and poultry to an unnecessarily high dose of the agents. One possible response would be to require an additional pre or post-rendering heat treatment process but the evidence to support any particular process which might be used is not yet conclusive . . . However it is likely that the industry would nonetheless install such systems to protect their sales. ²¹

2.258 Mr Hollis responded to this point the next day:

Adding such a requirement now would look rather odd. If we felt that rendering was ineffective back in 1988, why have we waited until now to protect pigs and poultry? It would also play into the hands of the Germans by apparently demonstrating that we have not been as strict as we should have been. And what would our consumers of pig and poultry meat make of such a change? Finally, I understand that such a requirement would force renderers to install new equipment costing over £200m.

If the CVO is happy to continue the current feeding practices for pigs and poultry, and his minute suggests he is, then I strongly recommend that we defend the status quo rather than introduce new regulations.

To achieve this I suggest we present the news in a positive rather than a negative way, drawing on the arguments in . . . the CVO's minute. These early results demonstrate the soundness of our vets' hypothesis on the reason for the appearance of BSE. They underline the wisdom of the measures implemented since 1988. They present no new threat to animal or public health. ²²

2.259 Following agreement at a meeting with Mr Soames, Mr Thomas Eddy, Head of Animal Health (Disease Control) Division, prepared a note on the handling of the interim results of the rendering study. The final version was submitted to Mrs Shephard on 25 March 1994. Under the heading, 'Discussion', the note stated:

The actual results so far are not, in our view, technically a problem. They confirm our hypothesis that the disease was spread by BSE contaminated material surviving the rendering process and being incorporated into meat and bone meal. Ruminants have been protected from this risk by the ruminant feed ban introduced in July 1988. Pigs and poultry have never been shown to develop a BSE like disease through feeding - either in the field or in experiments - though pigs can be infected by material injected into the brain . . . However, the SBO ban, extended to all animal feed in 1990, means that protein from the only tissues shown to be infective in tests is not used for animal feed, or indeed for any other purpose. In our view therefore there are no new animal or public health measures which need to be introduced as a result of these findings. ²³

2.260 An attached question and answer brief stated:

Q. Is UK meat and bone meal safe now we know rendering does not kill BSE?

A. Yes: do not forget that we believe the disease was spread by contaminated feed so all the controls assume that rendering does not kill the agent. Those controls are that ruminant protein cannot be fed to ruminants and that the only tissues found to be infective (brain and spinal cord which are covered by the specified bovine offals) are destroyed and cannot be used for any purpose; not for animal feed and not even for fertiliser. ²⁴

2.261 The note was discussed at a meeting with Mrs Shephard on 28 March 1994, with discussion focused on how the results should be publicly disseminated. ²⁵

2.262 In a statement to the Inquiry Mr Andrew Fleetwood ²⁶ explained the results' implications:

These results provided the first direct experimental evidence that infectivity may survive some forms of rendering . . .

These results came as no surprise. They corroborated epidemiological evidence which suggested BSE had been present in MBM, having survived the rendering process. The particular concern, shared by all those involved including the industry, was that there was now definite evidence that BSE infectivity would survive certain processes, but no evidence of what processes would eliminate it.

However, to put this into context, it is important to note that eliminating BSE infectivity through the rendering process was a second line of defence - the first being the various legislative measures to cut off recycling of infection. ²⁷

Implementation of a voluntary feed sampling regime

2.263 On 25 March 1994 Mr Howard minuted Mr K Taylor, hoping to 'take forward positively proposals to commence sampling as soon as possible'. He thought it would be helpful to summarise the position on voluntary sampling:

Sampling would be carried out by veterinary/technical staff at the time of the investigations into reported BSE suspect cases; the number of premises from which samples would be taken to be decided. Mr Gayford tells me that although Luddington have capacity to test only 20 samples a week, resources could be made available, within reason, to meet our needs. Mr Gayford would like testing to commence as soon as possible and would prefer it to be on a voluntary basis initially for a run-in period.

Ms Behzadi advises that if a farmer agrees to the sampling and testing of feedstuffs it would be unreasonable on our part not to provide the test results. This would be even more important should positive test results occur when we may consider prosecution action. Positive test results from voluntary sampling could also lead to civil action being taken. Evidence in such instances would be based largely on the sampling and testing techniques in use which would be without statutory backing and possibly open to challenge. ²⁸

2.264 On the statutory routine sampling of feedstuffs Mr Howard surmised:

We have established that routine sampling could be carried out under the provisions of the Agriculture Act 1970, by the local authority, and that the method of testing developed at Luddington, although not included as a recognised method of analysis under the 1982 Feedingstuffs Regulations, could also be used.

Regarding the appointment of an agricultural analyst as defined under the Act, Mr Gayford has said that there is no one in the VI Service possessing the necessary qualifications. This leaves us with the alternative of buying the services of an agricultural analyst who would be responsible for the testing of samples at Luddington in accordance with the statutory requirements. Mr Gayford agrees in principle with this idea. I understand that this function could be carried out in an overseeing capacity with the delegation of duties to staff at Luddington. ²⁹

2.265 Mr Howard felt that MAFF should press ahead with voluntary sampling rather than wait for resolution of the problems associated with statutory sampling. He recommended:

(a) as a priority we discuss our proposals with the local authorities (LACOTS)

(b) we reach agreement on the methodology for the sampling and testing of feedingstuffs with a view to commencing shortly

(c) the necessary arrangements are made for the statutory sampling by local authorities with testing at Luddington under the supervision of an agricultural analyst. ³⁰

2.266 Mr Taylor briefed Mr Meldrum on 8 April 1994 and suggested initial steps that could be taken to allow sampling and testing 'without further delay'. He stated that the ELISA test had been available since the opening of Luddington VIC but action was delayed because it was hoped to include statutory powers to sample in the proposed BSE Amendment Order, which for other reasons was still being discussed. Mr Taylor explained the legal difficulties that prevented sampling unless there was evidence of a contravention of the ban and advised:

Unfortunately, while that situation has still to be resolved, the need to be seen to carry out some policing of the ruminant feed ban is now urgent. ³¹

2.267 He proposed that voluntary on-farm sampling should start 'as soon as possible' with the target date of 1 May 1994. Divisions would be set a target of submissions from two to three farms per month, targeted towards farms with BABs born in 1990 or later. This could mean the submission of 100 to 300 samples per month, as some farms would have more than one feed. A press release would be issued to warn that sampling was about to begin, 'in the hope that any compounder knowingly using ruminant protein will cease forthwith'. ³²

2.268 Mr Taylor hoped that the issues associated with nomination of an 'agricultural analyst' would soon be resolved, so that statutory sampling could be initiated. However, the voluntary 'trial run' would help to identify the pitfalls in the field and in the laboratory before local authorities were briefed. ³³

2.269 On 29 April 1994 Ms Behzadi advised Mr Howard on the relationship between the proposed voluntary sampling regime and article 15 of the Bovine Spongiform Encephalopathy Order 1991. She noted that it was agreed that if positive results were obtained under voluntary sampling, MAFF should then sample under article 15 of the Order, since they would have reasonable grounds for suspecting the RFB was breached. If the intention was to establish the offence of feeding ruminant protein to a ruminant, testing would be on farm. If the intention was to establish the offence of selling or supplying, testing would be at the relevant premises. Under this process, samples acquired during voluntary sampling would not be used for evidential purposes. Ms Behzadi set out procedures that should be followed to help avoid problems at the prosecution stage. ³⁴

2.270 Although Mr Meldrum and Mr Taylor discussed the issue, Mr Meldrum did not respond in writing until after the proposed implementation date of 1 May 1994. He informed Mr Taylor that in the 'absence of comment from others I assume that we are all agreed that we should move into a voluntary sampling arrangement as you propose, to be announced by Press Release and to come into effect, I assume, on 1 June'. Mr Meldrum concluded:

However, I think it essential that not only should we consult Ministers but we should also advise the major trade organisations including BVA, NFU, NCBA, UKASTA, UK Renderers and De Mulders of what we propose. They will need to have some assurances as to what action we would take should we find a positive sample on the basis of voluntary sampling. How would we respond to that question? ³⁵

2.271 On 19 May 1994 Mr Eddy distributed to Mr Meldrum, Mr Martin Haddon, ³⁶ Mr Taylor and Dr Cawthorne a draft submission for Mr Soames on the proposed voluntary testing. In his covering minute, Mr Eddy noted that Mr Meldrum felt that a press release announcing the testing regime should be issued. However, Mr Dugdale thought it would be unnecessary and 'could be counter productive by drawing attention to the weaknesses of our ruminant protein ban enforcement up to now'. Mr Meldrum responded by way of manuscript note:

I have no difficulty with the thrust of [the draft submission]; others may have minor drafting proposals. I support a low key Press Notice; a PQ will only raise questions later. ³⁷

2.272 A final submission was forwarded to Mr Soames on 24 May 1994. In his covering minute, Mr Eddy noted that the debate over the degree of publicity to be afforded the announcement had yet to be resolved and that he would be particularly grateful for Mr Soames' views on that point. ³⁸

2.273 Mr Soames was invited to agree:

(a) that we should introduce the new test to detect ruminant protein in animal feed but that

(b) we should introduce the test only on a voluntary basis under MAFF control and use compulsory powers only where we had detected a positive result from the initial voluntary test;

(c) that we should issue a press notice. ³⁹

2.274 The submission reiterated the information previously provided by MAFF officials on the background to the testing and why it was preferred that samples be collected on a voluntary basis and submitted to Luddington VIC. However, on local authority involvement, it continued:

There is also the problem that some local authorities are unnecessarily assiduous in relation to BSE, for instance some ban the use of beef in school meals, and we cannot guarantee that they will treat the existence of this test in a prudent way. We therefore feel that on balance it would be better at this stage to retain the test capacity in-house so that we can monitor the situation and see whether the ruminant protein ban is being fully effective or whether there is a need to put significant extra resources into much more widespread use of the test which would involve local authorities in routine sampling of feedstuff supplies. ⁴⁰

2.275 The submission concluded by addressing the publicity issue:

There are in effect two main options. One would be simply to introduce the test on a voluntary basis but not to give it any publicity. There are certain attractions to this, not least because any publicity highlights the fact that we have not up to now had this test and that the policing of the ruminant protein ban has not been as tight as it might have been with the test. But against this there is the general presumption that we have always aimed to be open on BSE and that the test does represent a strengthening of the control arrangements which is good news and should therefore be given publicity. ⁴¹

2.276 The Q&A brief attached to the submission said that the test had not been introduced earlier 'because a reliable test was not available'. It also addressed the question how, without such a test, it was possible to know that the ban was complied with:

Enforcement is a matter for local authorities. It can be checked by visiting animal feed suppliers and looking at their records. But this new test will be a supplement. ⁴²

2.277 It also answered the question whether there was any reason to think that the ruminant protein ban was not being enforced:

No, since I am sure that everyone in the industry realises how important it is in the fight against BSE but this test will provide a useful confirmation. ⁴³

2.278 A draft news release was also attached to the submission, together with a draft letter to be sent to relevant industry bodies informing them of MAFF's intentions regarding feed testing. ⁴⁴

2.279 In a statement to the Inquiry, Mr Soames explained that although he responded that he was content to introduce the ELISA test on a voluntary basis under MAFF control, he told Mr Eddy a press release was not necessary given that the capacity of the test was limited. ⁴⁵

2.280 On 29 May 1994 Mr Bradley wrote to Mr Meldrum recommending steps that could be taken to enhance MAFF's actions to eliminate BSE. Whilst he was mostly concerned with improvement of the SBO controls, on the ruminant feed ban he said:

It seems imperative that the RFB is completely enforced by the power of law. If we do not ensure it is watertight we will pay for it with continuing BSE confirmations in 4-5 years time and it may jeopardise live cattle exports if we successfully get these re-instated. Testing feed is one way this could be done and there seems to have been an inordinate delay in getting this established. ⁴⁶

Notification of voluntary testing regime

2.281 On 15 June 1994 Mr Howard informed Mr Hayward of the Press Branch that it had been agreed that voluntary routine sampling would commence on 1 July 1994 and that an Animal Health Circular (AHC) was being distributed within the next few days. He attached a draft letter to relevant organisations informing them of the sampling programme prior to its introduction. ⁴⁷

2.282 AHC 94/93 was issued on the same day. It stated that it was 'essential to ensure that feedingstuffs do not present an on-going risk to cattle through the inclusion of ruminant protein whether by accident or deliberately'. It also explained the voluntary testing regime. A protocol for the collection of samples was included and VOs were advised:

It will be obvious that sampling at the time of examination of a suspect is not intended to quantify the risk via feed to that suspect. The aim is to use BABs born in 1990 or later to indicate premises on which there may be a potential long term feed contamination problem. This is no more than a simple means of targeting limited resources, and must not be misinterpreted as implying that such cases have been infected via feed. ⁴⁸

2.283 DVOs were requested to ensure that samples were submitted from two farms within the Division per calendar month. Ideally these were to be from farms with BABs born in 1990 or later but if none were reported appropriate alternative suspects were to be identified and samples taken. If more than one source of ruminant feed was identified on any given farm samples of each were to be collected and sampled - this specifically included protein supplements used for home-mixing. ⁴⁹

2.284 The circular also advised that where there was evidence that feed containing ruminant protein was knowingly fed to ruminants, the instructions for statutory sampling should be followed. This involved collecting further samples of all feed intended for ruminants on the supplier's premises, which should be unannounced. A protocol for statutory sampling was included. ⁵⁰

2.285 On 22 June 1994 the letter was sent to relevant organisations, notifying them of the general arrangements for the voluntary routine sampling regime, commencing on 1 July. It noted:

It has been our intention to carry out surveillance of this kind for some time but until recently no test has been available to determine the presence of ruminant protein in feedingstuffs. Such a test has been developed and validated at the Ministry's Veterinary Investigation Centre at Luddington and we are now able to proceed with sampling and testing which will provide data about the effectiveness of the ruminant protein ban. ⁵¹

Mr Wilesmith briefs UKASTA on BAB case control study

2.286 On 21 June 1994 Mr Wilesmith and Miss Hoinville presented to UKASTA's Scientific Committee an assessment of the risk factors for BABs. The note of the meeting records that during discussion, it was noted that:

The industry's voluntary ban on the use of specified bovine offals in all feedingstuffs became generally operative from early 1990 and was given statutory backing in September of that year. Prior to the feed bans, the practice in some rendering plants was for specified bovine offals to be collected separately from other material and then put back, en masse, with the other animal by-products. This, in turn, could have resulted in animal protein containing plugs of infected material being produced and subsequently used in ruminant feedingstuffs. ⁵²

2.287 The extent to which the development of BSE was associated with the doses of infected material that cattle might have consumed was discussed:

It was considered that, basically, the larger the dose the shorter the incubation period for BSE. However, a picture was emerging, which was supported by work undertaken in mice, that BSE could develop further to one initial exposure to infected material. ⁵³

2.288 The Scientific Committee advised that equipment used in feedmills was 'being updated as and when required'. The Committee was advised that before undertaking any specific work on collecting information on the mechanical aspects of feedmills, the CVL would review available data to assess whether BSE cases had been supplied from particular mills. ⁵⁴

SEAC discusses cross-contamination

2.289 An emergency meeting of SEAC was held on 25 June 1994, during which the Committee considered a paper prepared by Mr Wilesmith, Miss Hoinville and others on a case control study of BABs. The paper concluded that maternal or horizontal transmission was a risk in only a small number of cases. SEAC agreed that feed contamination was likely to be the major source of infection and expressed its concern that:

. . . the later BAB cases suggested that there could still be some contaminated material slipping through the controls into animal feed through cross-contamination in mills producing ruminant and monogastric feed and through lax compliance with the SBO rules . . . ⁵⁵

2.290 SEAC was advised that Mr Wilesmith had recently met UKASTA and had been told that whilst cross-contamination had been a problem, improvements had been made lately. Mr Eddy also advised that the ELISA test would be introduced to detect ruminant protein in feed and SEAC hoped 'that this could be used pro-actively particularly in the areas where there were suggestions that contaminated feed might still be a problem'. ⁵⁶

Further monitoring is debated in light of cross-contamination concerns

2.291 On 4 July 1994 MAFF met UKASTA representatives to discuss the programme of voluntary routine testing of feedstuffs for ruminants, which had commenced on 1 July 1994. UKASTA opened by expressing its disappointment at not having been informed in advance of MAFF's intentions. This was particularly because UKASTA would have been willing to discuss the possibility of the work being undertaken at compound feed mill level rather than on farms. UKASTA's note of the meeting records that MAFF had considered taking a high profile stance on policing the feed bans, but on reflection, had decided not to do so. ⁵⁷

2.292 MAFF explained that it was important to find an explanation for the occurrence of BSE in animals born in 1989 and later, since the general view was that if maternal or vertical transmission was occurring, it was doing so at a low level. It was noted that the source of infection had to be from something else, such as ruminant protein. Furthermore:

An important point was that the CVL considered that an animal could become infected with BSE following one meal. From (incomplete) rendering experiments it had been demonstrated that BSE could survive the cooking process and, with the removal of water and fat, the degree of concentration of the agent potentially increased in the meat/bonemeal. ⁵⁸

2.293 The ELISA test's detection sensitivity was discussed and the note records that:

UKASTA advised MAFF that the technique could be expected to detect ruminant protein present in a ruminant feedingstuff due to accidental cross contamination if that portion of the feed were sampled. The Association was concerned about the implication this could have on individual member companies' liabilities with regard to farmer customers whose cattle had developed BSE. The only protection available to the industry was, therefore, to withdraw totally from the use of meat and bonemeal as a feed ingredient. ⁵⁹

2.294 UKASTA expressed its concern that pockets of old feed in bulk bins might be sampled, to which MAFF:

. . . advised that officials were looking for fresh feedingstuffs and were being steered away from sampling feed in bulk bins. The Ministry would not be taking any statutory action if ruminant protein was found in a sample of ruminant feedingstuff. The supplier concerned would, however, be contacted with a view to further investigation being carried out. ⁶⁰

2.295 The issue of contamination between rations intended for different species was also raised. A trend had been identified whereby the eastern and northern regions of England were experiencing a disproportionate level of BABs. This raised the unanswered question as to whether these cases were caused by cross-contamination in mills producing for both monogastric and ruminant animals. This was particularly relevant given that MBM was used in higher levels in the eastern and northern regions. ⁶¹

2.296 In a statement to the Inquiry, Dr Matthews explained how the regional distribution of BAB incidence led to suspicions of cross-contamination in feedmills:

As the numbers of BABs accumulated, the incidence of such cases in East Anglia and Northern England was disproportionate to the earlier incidence of BSE. There did however appear to be a statistically significant correlation between the incidence of BABs born in 1990 or later with the size of the local population of pigs and/or poultry. This association did not relate to the farm level, and consequently the most likely common factor was the feed mill that produced feed for pigs and poultry as well as ruminants. ⁶²

2.297 On 11 July 1994 Mr Howard wrote to Miss Nelson with MAFF's comments on UKASTA's note of the meeting. He confirmed that the content of the meeting should be treated as confidential. He also said that Dr Matthews was following up a request for further information about the sensitivity of the ELISA test and that the results of the tests would be discussed with UKASTA at its next meeting. They would also discuss 'the way ahead if we opt for sampling and testing under the Agriculture Act 1970'. ⁶³ In a manuscript note at the foot of the letter, Miss Nelson noted:

Danny Matthews is urging Luddington to release [a] copy of the techniques. This is being resisted as it is to be patented and therefore Luddington would not wish to see copies made until its position is secure. However, details to be published in about 3 months. D Matthews still pressing.

N.B From samples taken so far - results negative. ⁶⁴

2.298 Miss Nelson forwarded a final copy of UKASTA's note of the meeting to Mr Howard on 20 July 1994. She reiterated UKASTA's wish to see details of the ELISA technique at an early opportunity. Miss Nelson sought clarification on whether the test could distinguish between ruminant protein and that of porcine or avian origin, and requested that UKASTA should meet MAFF to discuss the preliminary results of voluntary routine sampling, before they were released to the farmers from whom they were taken. She advised that UKASTA wanted to organise a meeting with Mr Wilesmith to discuss features of CVL's epidemiological work that had come to light including 'the fact that an animal could become infected with BSE following one meal'. ⁶⁵

2.299 Meanwhile, on 14 July 1994, Mr Ansfield minuted Mr Prince about the ELISA technique. He suggested that a commercially available ELISA test, while relatively insensitive to meats or MBM cooked to high temperatures, could be made to work if advanced techniques in MAFF's system were used as a prerequisite to testing the sample on it. The commercial kit was very expensive and Mr Ansfield sought guidance on whether some kits could be purchased for evaluation. ⁶⁶

2.300 Mr Ansfield subsequently advised Mr Gayford that he had subjected the commercial kit to extracts prepared using MAFF's method. He summarised that it had the potential for use with MAFF's extraction procedure for ovine and porcine MBM. However, neither the bovine or avian modules were satisfactory and so they would not be used. ⁶⁷ The testing of the commercial kit against every available feed was not pursued. ⁶⁸

2.301 Mr F Whaley, a Regional Veterinary Officer (RVO), raised cross-contamination concerns earlier in the year, on 27 May 1994. He had briefed Mr Crawford and Mr K Taylor on the recent Eastern Regional meeting of the UKASTA/BVA/MAFF liaison group. He reported that attention had focused on 'the apparent carry over of infection to BABs and a possible explanation of how infective ingredients or feedstuffs could have been and continue to be carried over to livestock'. UKASTA representatives 'quoted their own experience of material being discharged from bulk bins years after entry':

Pig nuts purchased over 3 years earlier were discharged on-farm from a bulk bin which had subsequently been charged with a different formulated and sized nut. Discovery was made during an investigation by the milling company following a complaint.

Barley was discharged from a bulk bin which had been used for several years for wheat at a mill. ⁶⁹

2.302 It was explained that most bulk bins discharged by a vortex process, similar to an hourglass, from the centre and top. Therefore, UKASTA representatives thought it possible that 'some material will remain in situ in bulk tanks indefinitely if they are constantly replenished'. Mr Whaley commented:

It is easy therefore to visualise the intermittent discharge over very long periods of infective ingredients during milling and formulation of feedingstuffs on the farm.

This may of course be 'old hat' to you, but I do not recall it having been actively discussed as a likely cause of BAB infection. ⁷⁰

2.303 Mr Whaley suggested that the new ELISA test provided a 'potentially invaluable tool' for testing the theory, and recommended that consideration should be given to the sampling of both ingredients and feedstuffs stored in the bottom of bulk bins on and off farm. ⁷¹

2.304 On 14 July 1994 Mr Taylor forwarded Mr Whaley's minute to Mr Meldrum. His covering minute said:

In view of the concern about the possibility of cross-contamination, at feedmills and elsewhere, of ruminant feed with ruminant protein intended for rations for other species, the attached minute from Frank Whaley is of considerable interest. We need to consider whether this should lead us to revise our instructions for collecting samples for testing for the presence of ruminant protein. ⁷²

2.305 Dr Matthews advised Mr Taylor, by way of a manuscript note on Mr Taylor's original minute, that sampling from the bottom of bulk bins would be impossible unless they were emptied. However, Dr Matthews believed that Mr Whaley's information warranted a VO inspection of plants and practices to assess the scale of the problem. ⁷³

2.306 On 22 July 1994 Mr Bradley voiced his thoughts to Mr Taylor, Mr Eddy and Dr Matthews on the issue:

If the industry has not emptied out the bins, generally tidied up their quality control and policed the controls in an adequate fashion they are being irresponsible as well as [flouting] the law. Any deficits there are will be revealed in 4-6 years by an occurrence of BSE that could be avoided. I think they should be very severely fined if they are found wanting in any way. We have the knowledge to prevent all feed borne BSE and the law must be rigorously applied to compounders and farmers if ever we are to rid ourselves of the problem.

The preparation of ruminant and non-ruminant feed using the same equipment on the same premises is a weakness and will be compounded by the same situation on farms even though risks are less now the SBO ban has bitten.

Mr Wilesmith tells me that several mills in E. Anglia are wise to the problems and have installed new equipment and cleaned up to avoid cross contamination or contamination of new with old. We really do need to see all the loopholes plugged. ⁷⁴

2.307 On 27 July 1994 Dr Matthews reiterated his opinion that mills should be formally inspected. He pointed out to Mr Taylor that 'we are all now aware' that cross-contamination in the feedmill could occur, particularly if they were producing multi-species feeds. He said that discussion with UKASTA indicated 'that this may be a real problem', and was distressing the industry as they attempted to ensure they were not identified as the ones producing ruminant feed containing ruminant protein. Dr Matthews suggested 'they may go to the extent of excluding meat and bone meal altogether'. He attached a report submitted by a VO, which identified likely problem areas in a mill he had visited:

My reason for copying the report is to suggest that apart from relying on compounders getting their act together from fear of prosecution or litigation, it may be worthwhile formally inspecting plants in order to assess risks. We did just this with respect to the rendering industry in 1988, but with the exception of mills that may have been visited for salmonella or residue sampling purposes, or possibly for factory inspection approval, we have not formally forced compounders to seriously review their procedures. While such visits may in part become advisory, it could give us a better feel for the scale of any potential problem which the small scale on-farm sampling may not. ⁷⁵

2.308 The report identified routes for possible incorporation of ruminant protein into cattle feed:

1. Augers may not be routinely cleaned between being used for pig and/or poultry feed and cattle feed.

2. Screenings from pig and poultry feed are not kept separate and can become incorporated into cattle feed.

3. Sweepings from the factory floor are recycled and may be put into cattle feed.

4. Animal fat is used in cattle feed. This is acquired from renderers who process abattoir waste. This includes bovine and ovine offal, and ovine brains. I understand this material contains some protein, as it is not profitable to extract all the protein from offal when making tallow.

5. Ruined batches of pig and poultry feed and feed returned from customers may be recycled and then be incorporated into cattle feed as a raw ingredient. ⁷⁶

2.309 On 27 July 1994 Commission Decision 94/474/EC was adopted. Article 3 read, in part:

3. The United Kingdom shall take appropriate measures to ensure that:

. . .

(b) ruminant protein destined for use in pig and poultry rations and other uses is not included in ruminant rations;

(c) scientific tests are used to monitor the implementation of . . . (b). ⁷⁷

2.310 On the same day, Mr Eddy sought Mr Dixon's opinion on whether this meant MAFF could overcome the existing legal constraints on testing for ruminant protein by introducing legislation under the European Communities Act. ⁷⁸

2.311 Prompted by Mr Bradley's minute of 22 July 1994, Mr Eddy advised Mr Taylor on 4 August 1994 that he had 're-read all the papers' on carry-over of feed. He said:

It does seem to me that we have a problem which can extend from the farm right back through the feed supply chain wherever bulk bins are used. Mr Whaley's minute is worrying in suggesting that material can lie around in these bins for over 3 years and that if the supply is constantly replenished then some material can remain indefinitely. I would have thought that this was not ideal on organoleptic grounds but it is certainly worrying in relation to the controls we have introduced on BSE and the lag times between the introduction of these controls and their effectiveness. It does seem to me that we need to issue some sort of guidance to farmers and everyone in the supply chain encouraging them to empty out bulk supply bins to ensure that there is no carry-over of old potentially contaminated stock. Until we have cleared this out of the system we are never going to know where we are on the various measures to prevent BSE through contaminated animal feed. ⁷⁹

2.312 He concluded by saying that Mr Howard had an ongoing remit to redraft the advisory leaflet to farmers, first issued in June 1990 (see paragraphs 2.64-2.65), though he suspected it had been held up by other more urgent work. Mr Eddy also asked whether Mr Taylor would like a leaflet for the feed industry to be drafted along similar lines. ⁸⁰

2.313 On 3 August 1994 a meeting was held between UKASTA and MAFF to discuss the ELISA technique. Mr Ansfield presented technical information about the test, and it was reported that Luddington VIC's capacity for testing was 'fully occupied' with the samples taken from farms during routine examinations of suspect BSE cattle. MAFF advised that no samples had proved positive for ruminant protein thus far. ⁸¹

2.314 UKASTA advised that individual member companies would wish to submit samples of compound feedstuffs to check that their manufacturing procedures did not allow carry-over of ruminant protein into ruminant rations. The note of the meeting records:

The interest [of UKASTA members] was for initial screening and to see whether there was any possibility of cross contamination taking place between the production of one brand of feedingstuff and another. It was estimated that approximately 1,000 samples could be submitted for analysis.

MAFF advised that such a requirement would tie up the existing analytical capacity at Luddington for about a year. The question of whether further resources should be made available was for the Ministry officials at Tolworth to give a ruling upon. ⁸²

2.315 On 17 August 1994 a meeting was held between CVL and UKASTA representatives. Following discussion about concerns regarding the animal SBO ban (see Chapter 4), attention turned to exposure and dose. UKASTA's note of the meeting records:

The CVL reiterated the finding known for some time that a single exposure to the BSE agent [was] likely to cause the disease to develop. There was no evidence of the occurrence to BSE resulting from a cumulative exposure.

The BSE epidemic is consistent with a low dose exposure. Studies were being carried out in order to ascertain the magnitude of the dose required. In one study animals had been dosed with 100g of brain from terminal BSE cases. Results of this indicated that the incubation period for BSE in the trials was likely to be the same as that found in the field. In another experiment, single doses of infected material had been given to animals at levels of, respectively, 1g, 10g and 100g.

A discussion took place on the implications to the feed industry if it was subsequently shown that a dose of 1g had caused BSE. Setting aside considerations of the effectiveness of the SBO legislation and even though it was possible to try and minimise the risk of cross contamination between feeds for ruminants and those for monogastrics, the only protection for a feed manufacturer might be to stop using meat and bonemeal in any feedingstuff. ⁸³

2.316 BAB cases were also discussed during the meeting:

Particular attention was drawn to the data indicating that, for the Eastern Region of the country, cattle born in the 1985/1986 cohort, the incidence in homebred cases was 6.86% whereas in 1990/91 the incidence had increased to 15.49%. The CVL considered that it was not possible to explain the majority of the BAB's other than by accidental cross contamination of ruminant meat and bone meal in ruminant feedingstuffs. This was of concern particularly in light of the low exposure/single dose needed for BSE to develop. ⁸⁴

2.317 UKASTA noted that protein concentrates were species specific, as required by legislation, but the manufacture of different type of feedstuff caused the industry concern because of the potential for cross-contamination between it and the subsequent batch of feed. The possibility of species specific mills was discussed, but it was noted that this was impossible for the majority of compounders who only had one mill. Pig-and poultry-only mills could not be justified in the western part of the country. ⁸⁵

2.318 CVL reported that, under the voluntary sampling scheme, 109 samples had been taken and all were negative. Samples had only been taken from farms on which animals born in 1990 or later had developed BSE, and between one to five feeds had been sampled per farm. It was recorded that 'samples were taken both because of fear of cross contamination and the need to be seen to be policing the feed ban'. UKASTA repeated its request for member companies to be able to submit samples to Luddington VIC for analysis to ascertain whether their procedure minimised the potential for cross-contamination. CVL said it would be possible for officials to get the necessary legal powers to carry out surveillance at feedmills, but that before any change in the current sampling procedures was introduced, it was necessary to assess the results obtained from the on-farm sampling scheme. It was agreed that MAFF and UKASTA should discuss a future scheme covering sampling at feedmills at a further meeting, and that MAFF would contact UKASTA in due course. The aim would be to design a scheme under which representative samples were taken without the need for any legislative backing. ⁸⁶

2.319 The CVL reported that it was drafting a questionnaire designed to obtain data on the distribution of mills against a population study of BABs. In response to UKASTA's concern about the possibility that the information gathered might be used in a civil action against an individual compounder, MAFF stated that it would not be possible to identify any particular feed manufacturing plant from the final report. ⁸⁷

The first BABs born in 1991

2.320 On 22 August 1994 Mr Meldrum advised Mr Packer that MAFF was now aware of four cases of BSE in cattle born in January, May, August and 'between June and September' 1991. None of the cases had been formally confirmed, though 'all diagnostic investigations are complete and there is no reason not to do so'. Further epidemiological enquiries were to be made, but detailed investigation into feeding histories could not begin until the cases were officially confirmed. Mr Meldrum advised that Ministers should be informed, but that deliberate publicity was unnecessary and probably unwise. However, he noted that the information should not be hidden, but routinely released in answers and reports without comment. ⁸⁸

2.321 On the following day Mr Packer forwarded the briefing to the new Minister of Agriculture, Mr William Waldegrave. ⁸⁹ A manuscript note on the submission, dated 26 August 1994, records that 'the Minister is content with [Mr Meldrum's] proposals for handling these cases'. ⁹⁰

UKASTA considers excluding all MBM following confirmation that only a 'very low dose' is required to transmit BSE

2.322 On 7 September 1994 Mr Reed provided Miss Nelson with UKASTA's objectives for a meeting with MAFF on the following day. Mr Reed proposed that UKASTA should revisit a number of topics to clarify MAFF's intentions and to allow UKASTA to get certain key points across. His proposed questions were:

- What is the latest thinking on infective doses? If these are now thought to be small and 'one meal', what control measures might be envisaged that are not already in place? . . .

- What are the real objectives of the on-farm tests to detect bovine [and ovine] MBM, given that these should have little or no relevance to the infection of a beast long before? How can [UKASTA] reassure [its] members that a deliberate attempt to blame feed suppliers for the epidemic is not being mounted to distract attention from the inadequacies of the control measures in place?

- If cross-contamination is now suspected as the main cause of BSE in BABs, why is the rendering of SBOs permitted on the same lines as other offals from cattle and sheep? What steps might be envisaged to control the cross-contamination risks in rendering plants, feedmills, transit, and on farms?

- Apart from the wish to assess the level of cross-contamination risk in feedmills and identify where the risk occurs, what is the aim of the proposed questionnaire? Will similar surveys be carried out simultaneously at rendering plants and on farms? If not, why not, since the risks of cross-contamination may be just as great or greater, and the consequences just as deadly? UKASTA members' willingness to co-operate honestly in the survey will depend on their being convinced that an even-handed, open-minded but closed-mouthed approach will be employed . . . ⁹¹

2.323 In turn, Miss Nelson advised Mr Reed that Dr Cooke wished two key points to be made, to reflect discussions between BOCM Pauls and Dalgety Agriculture. They were:

i) Will the Ministry give a written undertaking that, in its opinion, proper heat treatment will inactivate both the BSE and scrapie agents in processed ruminant meat and bone meal such that a 1% cross contamination from one compound feed to another is of no concern;

if not;

ii) should the Ministry not introduce legislation prohibiting the use of meat and bonemeal in animal feedingstuffs. ⁹²

2.324 The meeting was held on 8 September 1994. ⁹³ The infective dose required for transmission of BSE was the first topic discussed. UKASTA's note of the meeting records that:

The CVL had always considered that the infective dose of exposure to BSE was relatively small. Studies were being undertaken on the oral exposure of cattle. These included an experiment whereby single doses of infected material had been given to animals at levels of, respectively, 1g, 10g and 100g. Although the work was not yet complete, the indications were that a very low dose was sufficient to cause BSE. ⁹⁴

2.325 It was recorded in the note of the meeting that following the completion of the case control study on BABs, 'the potential for cross contamination at feedmills should be a cause of concern'. It was reported that there was a correlation between the incidence of BABs and the ratio of cattle to pigs and poultry on a county basis. It was noted:

The objective of the Ministry's animal health control policy, is to prevent or reduce the exposure of animals to the BSE agent to a level that is not significant. The policy was largely dependent on the effectiveness of the ruminant protein feed ban to prevent the infection of cattle, and the specified bovine offal ban to prevent infection of other animal species. The objective had not changed although the legislation had been refined in the light of pertinent information. ⁹⁵

2.326 UKASTA expressed its concern that the feed industry was being singled out as the cause of the BSE epidemic, and pointed out that any cross-contamination at feedmills 'should not be of any consequence if the controls in place further up the supply chain were totally effective'. UKASTA stated that, in the current circumstances, the only way the industry could fully protect itself against possible claims for causing BSE was to stop using MBM in mills making ruminant feed. ⁹⁶

2.327 Discussion turned to the ELISA test. It was recorded that:

MAFF reiterated that, although it was not possible to eliminate maternal/vertical transmission as a cause of BSE, all evidence pointed to the main cause being contaminated feed. The ruminant feed ban was the cornerstone of the Ministry's control policy for BSE and thus it was necessary for the legislation to be enforced. The development, by Luddington VI Centre, of its ELISA technique meant that the sampling and testing of feedingstuffs for the presence of ruminant protein could be carried out. ⁹⁷

2.328 MAFF reported that under the voluntary sampling scheme 154 samples had been taken from 62 farms, and that the first positive result had been obtained from a compound feedstuff sample. It was being assessed whether there was any cross-reaction with individual raw materials that might not have been verified during the technique's development. Enquiries at the compound feedmill would be made. UKASTA reiterated that the only way the industry could fully protect itself would be to stop using meat and bone meal. ⁹⁸

2.329 There was further consideration of CVL's proposal to produce a questionnaire to collect information that might be relevant to 'levels' of potential cross-contamination. UKASTA expanded its concerns:

The fear was that any evidence submitted to the MAFF could be accessed by the prosecution in future possible Court cases from farmers seeking compensation from individual feed compounders. UKASTA had co-operated fully with MAFF throughout the investigations into BSE. Now, however, it was considered essential that the questionnaire should be addressed equally to other sections of the supply chain for processed animal protein in order to put the role of the feed compounder into perspective.

MAFF considered that it was possible that action could be taken now by individual farmers based on published scientific evidence. It was also noted that even if the questionnaire was not circulated evidence still existed that the disease was feed borne. The CVL was, however, to send UKASTA a copy of the proposed questionnaire together with background information on its importance . . . ⁹⁹

2.330 UKASTA also drew attention to the case of an SVS officer who had commented to a farmer that the BSE agent was present in feedmills due to use of contaminated MBM. MAFF undertook to contact the officer concerned and also to try and stop this 'type of disinformation being circulated by MAFF officials'. ¹⁰⁰

2.331 Mr Reed said in a statement to the Inquiry that:

In the context of BSE, cross-contamination was not thought to be a crucial issue before 1994, because our understanding was that the SBO ban had successfully tackled the source of infection, and that a relatively high level of exposure to infective agent was required for transmission to occur. During that year, however, more cases of infected cattle born after the ruminant feed ban occurred, research results showed that lower doses of infected brain tissue were capable of transmitting the disease. ¹⁰¹

SEAC publishes its report on BSE knowledge and research

2.332 In September 1994 SEAC published a report entitled 'Transmissible Spongiform Encephalopathies: A Summary of Present Knowledge and Research'. On BABs, it stated:

There was surprise in some quarters that despite the imposition of the animal health controls in 1988 (ruminant feed ban) and 1990 (SBO ban for feeding to other species) 12,807 cases of BSE have been confirmed (to mid September 1994) in animals born after the 1988 ban, and this has given rise to comment. This was because it was expected that the ban could be implemented immediately, but this view did not take account of human nature and the practical difficulties involved. As nothing was done to eliminate the large amount of feed already in the distribution 'pipeline', the benefits of the ban were not instantaneous and complete. ¹⁰²

2.333 The report noted a number of factors that suggested BSE was a food-borne epidemic:

1. The number of confirmed BABs was much lower than the expected number of confirmed BSE cases had the epidemic run its natural course;
2. The majority of BABs were born in the calving season following the introduction of the RFB; and
3. Enquiries had shown that BAB cases had a greater association with feed history than possible maternal or horizontal transmission. ¹⁰³

The first positive ELISA test result

2.334 On 9 September 1994 Mr Robertson briefed Mr Meldrum on the first positive test from on-farm surveillance of ruminant feeds. He noted that there was uncertainty in the result as the sample had been taken from the bottom of a hopper, which made it difficult to determine the supplier and delivery date. Luddington VIC reported that the sample contained bovine, ovine and porcine protein. Mr Robertson suggested this could be the result of 'legitimate inclusion of porcine protein with bovine and ovine contamination'. However, the possibility of a non-specific result could not be excluded, since no feed made by the compounder in question had been tested during validation of the test. ¹⁰⁴

2.335 In a statement to the Inquiry, Dr Matthews mentioned the significance of the ELISA test for establishing cross-contamination in mills as a continuing source of BSE:

[I]t was inevitable that concerns about infection via cross-contamination of feed grew progressively as more BABs were confirmed and other sources of infection were not identified. The introduction of the ELISA test in June 1994 provided confirmatory evidence soon after that cross-contamination was a real issue that had to be dealt with. In other words hard evidence was found as opposed to anecdotal evidence and interpretation of epidemiological data. ¹⁰⁵

Development of the CVL questionnaire

2.336 On 19 September 1994 Mr Wilesmith wrote to Miss Nelson with information on the proposed survey of feedmill practices. He summarised the further epidemiological studies involving feed compounders that could be undertaken and the reasons for them. Mr Wilesmith also wished to 'allay the fears of UKASTA members of incriminating themselves as a result of participation in the proposed survey'. ¹⁰⁶

2.337 Mr Wilesmith highlighted the importance of explaining the occurrence of BSE in animals born in 1989 and 1990 for 're-establishing the international trade in animals and their products'. He explained that the first phase of the proposed epidemiological studies involved a questionnaire to obtain data and information on 'practices within feedmills which have some relevance to the possibilities of cross-contamination during 1989/90'. The next phase was 'a between herd case-control approach' using herds which had a homebred 1989/90 born BSE case and those that did not. A part of the comparison to investigate risk factors would involve management and husbandry practices and feeding history including risk factors obtained from the questionnaire. Mr Wilesmith assured Miss Nelson that the results of the analyses could not be used to incriminate any particular mill as the cause of BSE in the cases being studied. Mr Wilesmith explained that the final 'optimistic' stage was to use the questionnaire information to assess the degree to which practices had changed over time. ¹⁰⁷

Ministers are updated on the BAB situation

2.338 On 21 September 1994 Mr Eddy informed Ministers that MAFF would now have to confirm the first four cases of BSE in animals born in 1991. Mr Eddy did not recommend that MAFF issue a press notice as the cases would become public knowledge via the regular published statistics, for which a question and answer brief was attached. ¹⁰⁸

2.339 The current number of confirmed BABs was 12,860. Of those, 358 had been born after the end of 1989. Mr Eddy advised that infected feed was the likely explanation for BAB cases born in 1988 and 1989, and probably for those born in 1990 and 1991 too, though investigations were continuing. He stated:

It is certainly true that feed can remain on farm for a long time, particularly when farms and suppliers use bulk feeding hoppers which may not be completely cleaned out before refilling. But there may be cases due to maternal . . . and horizontal . . . transmission although the Ministry's attempts to clarify this through statistical analysis have failed to show any firm evidence of maternal transmission and only limited evidence that horizontal transmission is occurring. ¹⁰⁹

2.340 Mr Eddy explained that there was 'nothing particularly significant in 1991 as opposed to 1990 in epidemiological terms', though 'the media may well find it more difficult to accept carry over of contaminated feed as a cause'. Mr Eddy advised that the BSE Sub Group of the EU Scientific Veterinary Committee was due to meet the following week and said, 'We do not believe that there is anything to be gained by not revealing these cases'. ¹¹⁰

Compounders consider excluding MBM from ruminant rations

2.341 On 23 September 1994 Mr Fleetwood informed Mr Meldrum and others that one feed compounder was 'close to taking the decision to stop purchasing meat and bone meal for any of their rations'. He noted that if one major compounder made this decision, 'the rest are likely to follow with significant consequences for the disposal of ruminant by-products'. ¹¹¹

2.342 Mr Robertson, to whom the minute was copied, pointed out that action by compounders based on positive results from on-farm sampling would be premature, since the ELISA test could be over-sensitive. He noted that further samples were being taken at the mills involved to clarify the situation. ¹¹²

2.343 During a UKASTA Scientific Committee meeting on 29 September 1994, it was noted that:

A number of sites where cross contamination between animal proteins and other types of raw materials might occur were identified. These included not only on-farm but in-store, in the country of origin, in boats, in transport as well as different points within the feed mill. It was noted, however, that it might be counter productive to stress these varying numbers and sites. ¹¹³

2.344 The Committee also expressed concern that:

. . . MAFF had commenced on-farm testing without necessarily thinking through the consequences for the whole of the agricultural industry. Officials were aware that one course of action open to feed compounders was to stop using meat and bone meal in the manufacture of any feedingstuff. An alternative for the industry was the establishment of ruminant feed only mills. Such a step would only be open to those companies with more than one manufacturing site. ¹¹⁴

2.345 MAFF's proposed survey of feedmills was also discussed. The Committee acknowledged the survey's aims (see paragraphs 2.336-2.337), but noted that there were still difficulties with companies incriminating themselves if they completed the questionnaires. It was decided that whilst the CVL was finalising the survey, UKASTA should cooperate and members should send their comments on the proposed survey to the Secretariat. ¹¹⁵

2.346 On 3 October 1994 Mr Robertson updated Mr Meldrum and others on results from the voluntary routine sampling regime. He reported that 482 samples collected from farms had given negative results, with only two farms returning positive results. Statutory sampling was undertaken on 19 September 1994, pursuant to article 15 of the Bovine Spongiform Encephalopathy Order 1991, at the mill of origin relating to the first farm previously reported. Of the ten samples collected, two were 'weakly positive' for bovine, ovine and porcine antisera. Further samples of the ingredients of the two positive rations had been requested. Statutory sampling at the mill in question relating to the second farm had also been carried out and the results of the tests were pending. Mr Robertson advised that:

Both the renderers and the compounders are closely watching the developments in the matter of the testing of feedstuffs, and there have been indications from the compounders that, if too many problems arise as a consequence of positive results from the tests