Opening Session

Chaired by the Chairman of the Select Committee, Lord Smith of Clifton.

Lord Smith welcomed the delegates and thanked them for attending.

He said that the conference was intended to provide the best possible information for the Committee. It was unlikely that consensus would be reached within the working groups, but it would nonetheless be useful for the Committee to be able to take a helicopter view of all shades of opinion before they began to draft their final report.

Five speakers were invited to give short presentations.

Dr James Anderson, Home Office Inspectorate

Dr Anderson said that the Animals (Scientific Procedures) Act 1986 was "enabling legislation" which was the UK implementation of Directive EC 86/609. The core of the Act was a principle — the cost/benefit analysis. This meant that the Act was perpetually applicable and constantly contemporary.

Over the 15 years of the operation of the Act there had been advances in animal facilities and accommodation and, principally, the establishment of a "culture of care". An important feature of the Act is that all the key questions on scientific validity and alternatives were matters of judgement concerning the available information. Confidence in the Act came down to both scientists and anti-vivisection organisations having confidence in the Inspectorate.

Dr Anderson said that compulsory training modules had been introduced for personal licence holders (1994), project licence applicants (1995) and Named Veterinary Surgeons (1996). Training for NACWOs was in prospect. He also discussed the Ethical Review Process. This had had a beneficial effect on the consideration of animal welfare, although it also had the potential to become bureaucratic. The Inspectorate's review gave examples of best practice to help ERPs operate more efficiently.

Dr Pete Clifton, British Psychological Society

Dr Clifton said that the 1986 Act was a great improvement over the previous legislation: accommodation standards had improved and more veterinarians and animal welfare experts were involved. He considered that the Act struck an excellent balance.

The cost/benefit analysis was hampered by the need for more research into animal behaviour to assess animal welfare better, and the fact that the benefits of "blue skies" research were intrinsically unpredictable. Replacement alternatives were often not possible, so emphasis should be placed on reduction and refinement.

Bureaucracy in obtaining licences meant that some experiments were simply not done in the UK. There had also been difficulties with permissions for visiting scientists. Finally, he said that the inclusion in the Statistics of GM animals used solely for breeding was misleading.

Dr Maggy Jennings, RSPCA

Dr Jennings said that the RSPCA had a goal to end all animal experiments, but worked in the short term to minimise animal suffering. Animals lives also had intrinsic worth and should not be wasted. The RSPCA had four specific concerns.

One, alternatives and toxicology. A fundamental change was required: instead of justifying animal use, thought needed to be given to how the need for animals could be removed. In vivo and in vitro testing should be integrated, rather than attempting to replace each individual animal test with a non-animal alternative. This shift in attitude required changes in tertiary science education.

Two, freedom of information. There was no information on how costs and benefits were determined by the Inspectorate, and the Statistics gave no account of animal suffering. Confidentiality hampered the development of alternatives and limited input into the regulatory process. It also prevented effective communication between Government bodies such as the APC, UKXIRA and the Home Office.

Three, regulation. The life experience of experimental animals should be taken into account as "cost". The "acquisition of knowledge" was too broad a category for benefit, as it could justify almost anything.

Four, the ERP was the key to improving local implementation of the Act. There should be increased lay and animal welfare membership.

Dr David Smith, AstraZeneca

Dr Smith said that animal testing removed severely toxic compounds, identified likely adverse effects, and established the safe starting dose and dose response: "The right dose differentiates a poison from a remedy" (Paracelsus). Risk assessment was not absolute safety. Acceptable levels of risk were determined by society, which is increasing its demand for reassurance. Identifying low-incidence effects required many more animals.

He said that the easy gains in replacement alternatives had already been made. There had been limited progress with the use of whole organs. In particular the liver, where human toxicity was most frequently found, was not well understood. Emphasis should be placed on improving the "failure rate" of compounds at the molecular level before they were ever tested on animals. Regulatory "box-ticking" should be removed: he cited a joint industry/animal welfare initiative to dissuade the EU from any possible increase in group size. The numbers of animals used were not, however, as important as the welfare of the animals that were used. A holistic approach to the Three Rs was advocated.

Toxicologists had problems with: the supply of animals (the closing of breeding farms in the UK simply meant that more animals were transported from abroad); intimidation by Animal Rights groups in the UK, which affected the ability to recruit and retain staff and over-optimistic assessments of alternative technologies. Toxicologists were making efforts to engage with animal rights groups, but disliked change for change sake if clear animal welfare benefits were not forthcoming.

Michelle Thew, BUAV

Michelle Thew said that she shared Dr Smith's conviction that there was a need for new medicines and a need for them to be as safe as possible. However, researchers would not experiment on humans without obtaining their consent, whatever the need. She considered, similarly, that the deliberate infliction of suffering on another animal, if not for its benefit, was immoral.

It was generally accepted that animals did indeed suffer, but over 60% of procedures were carried out without anaesthesia. The Act should operate with clear and transparent policy guidelines and reflect informed public opinion — currently it was operated by ex-animal researchers for animal researchers. She quoted Dr Mark Matfield from the Research Defence Society to the effect that security concerns were often over-stated. These concerns were nonetheless used to prevent transparency and accountability.

Duplicate testing was a problem, as the Home Office only encourages, but does not require, the sharing of data. Improvements such as the banning of research on Great Apes, or for cosmetics, were a start, but why was research not also banned on all primates, or for household products? Transparency was central for a proper debate. The BUAV could only advocate for animals if it had access to adequate information.

Working Groups

Report of Working Group 1: Centre(s) for Alternatives

Chair/Rapporteur: Dr Kenneth Boyd

1. It was generally agreed that targeted resources would help to promote and develop alternatives, though more in fundamental research than in toxicology.

2. Many of the best 'alternatives' had been developed by animal scientists in the course of their work, as they had the necessary expertise in a particular scientific field. Scientists must be encouraged to continue the development of alternatives.

3. This was not sufficient by itself, however, particularly as scientists tended to concentrate on reduction and refinement rather than on replacement. The MRC's Centre for Best Practice, for example, was focused on welfare rather than replacement.

4. There was already an enormous amount of information available, but it was on many different databases and there was little quality control. An overview of alternatives was also needed as even when individual replacement tests were developed, they often did not fit into the regulatory toxicology system.

5. Alternatives needed to be developed in context. Ring-fenced funding in other areas had not improved the quality of work done. Free-standing centres for alternatives would not be able to draw on the immensely wide range of necessary expertise. The onus would also then be taken off animal scientists to consider alternatives themselves.

6. There was very little enthusiasm for a free-standing 'wet lab', an independent monolithic research facility, or another validating body.

7. What was needed was a centre for strategic planning to co-ordinate funding, give alternatives science academic status, provide reliable high quality information, and encourage scientists themselves to develop alternatives.

Possible Structure of an Alternatives Centre

8. There was a remarkable degree of consensus in the group of the type of alternatives centre that would be most constructive. A hub and spokes model was proposed.

9. There should be a virtual centre at the hub. This would be a portal to relevant databases for alternatives (possibly including its own database), and provide a forum for sharing information (possibly including databases of 'negative results'). It could co-ordinate existing 'alternatives' funding by charities, industry and the Home Office.

10. The centre should be independent and funded by a wide range of interested parties, not just by industry. It should have close links to the Home Office and their internal databases. The Home Office could require project licence applicants to consult the centre as proof of their search for alternatives.

11. In addition, the centre should co-ordinate 'spokes'. These would consist of small research groups with different specialisations incorporated into existing research centres at Universities and medical schools. These small units would draw on the existing expertise in research centres, and act as drivers to incorporate alternatives research into the every day business of research science.

12. Some people were unhappy with the word 'alternatives'. It was suggested that the national hub should be called the 'Centre for the 3Rs'.

Other possible actions

13. Improving an awareness of and need for alternatives could also be achieved in other ways. One, three Rs terms should be added as search keywords to existing databases.

14. Two, Funding bodies should require applicants to demonstrate their search for alternatives and state what innovative techniques relating to the 3Rs their projects involve. Journals should require articles based on animal experimentation to incorporate a short note on the use or development of the 3Rs. These two steps would keep the issue of the 3Rs at the forefront of scientists minds.

15. Three, there was a need for better training in statistics and experimental design.

Plenary

16. In plenary, it was commented that "the world is full of partly developed databases". The difficulties in maintaining even the virtual centre should not be underestimated. The centre would require a commitment to long-term funding.

17. Such a centre should act as a "champion" for alternatives, even if it did not conduct any actual research itself.

Report of Working Group 2: Toxicology

Chair/Rapporteur: Dr Barry Phillips (RSPCA)

How effective and reliable are in vivo toxicity tests?

1. In toxicology, in vivo toxicity tests (those using live animals) form only part of a complex process of risk assessment that takes account of other sources of information, including analysis of chemical structure and in vitro tests, and is also subject to expert interpretation and judgement. The purpose of toxicological risk assessment is to evaluate the likelihood of harmful effects on health from exposure to specific doses (or concentrations) of particular chemicals, and thus to inform the risk management process (actions such as prohibition of the use of a chemical, the specification of maximum recommended doses, or the provision of advice on safety precautions).

2. The effectiveness of in vivo toxicity tests per se is difficult to assess and cannot simply be deduced from the effectiveness of the entire risk assessment process. Thus, although there was agreement that human health and safety are, in the main, well protected by the current process, there was some disagreement as to whether this was due to the use of animal tests, or in spite of their use.

3. There is little hard scientific evidence upon which to base a judgement of the reliability of in vivo toxicity tests. This is due to the ethical constraints on testing chemicals in human subjects which preclude the confirmation in humans of adverse findings in animals. Only in the case of pharmaceutical substances, administered in doses judged to be safe, or after accidental exposure to chemicals, can the results of animal tests be checked against actual human experience.

4. Recently, the International Life Sciences Institute (ILSI) undertook a study of a series of pharmaceutical compounds that had shown toxicity in human clinical trials.[221] This study found that in 71% of cases, the effects seen in people were foreshadowed in the animal tests carried out prior to the clinical trials. When the results of tests on rodents were considered in isolation, only 43% of the human toxicities could have been predicted. Tests on dogs or primates were necessary to increase the prediction of toxicity to 71%. Of the 29% of effects not detected in animal tests, the majority were of a type that the animal tests were not designed to detect, or were intrinsically undetectable in this type of test e.g. headache and dizziness, and certain skin reactions.

5. It was agreed that the ILSI study has limitations, and that the general reliability of animal tests is very difficult to judge, as some animal tests may be more reliable than others. In broad terms, those tests designed to detect severe or rapidly developing effects, such as acute oral toxicity or irritancy, appear to be more predictive of human health effects than those attempting to detect more subtle, long-term effects. Deficiencies in the design of certain tests may also compromise their reliability. For example, the use of very high doses of chemicals in rodent cancer bioassays has given rise to a high incidence of findings that are unlikely to be of significance to human health.

6. It was noted that rare effects, in a small sub-group of the human population, are very difficult, if not impossible, to predict using laboratory tests or even in clinical trials.

7. It was pointed out that the validation requirements for non-animal tests is in marked contrast to the almost total lack of formal validation of animal test methods. Despite uncertainty about the reliability of animal tests, alternative methods are expected to give the same results. A figure of greater than 80% concordance with animal test results was quoted as a requirement for validation of an alternative method in the USA. A better method might actually be rejected because it gave correct results (correct prediction of human health effects) more often than the existing test.

8. It was concluded that the effectiveness and reliability of animal tests is unproven. It was recommended that the reliability and relevance of all existing animal tests should be reviewed as a matter of urgency.

9. It was suggested that the long-term solution to the problem was to work for a better understanding of the basis of disease and toxic effects, so that tests based on knowledge of the fundamental mechanisms involved could be designed rationally.

What is the scientific justification for (a) the use of a second, non-rodent species in chronic toxicity testing, and (b) the seemingly formulaic use of certain species (such as the beagle)?

10. The justification generally given for the use of a second, non-rodent, species is that rodent studies do not detect all toxic effects because of species differences. However, the scientific basis for the use of two species is questionable; tests could be conducted in any number of species and the relevance of the findings for man would be equally uncertain for all the species used. The use of two species is a compromise which provides some reassurance that important toxic effects have not been missed.

11. On a case-by-case basis, an appropriate test species can be chosen rationally if existing knowledge about specific chemicals indicates that one species is more likely than another to respond to a chemical in a similar way to man. In some cases, studies of human tissue in vitro can guide species choice. A number of organisations have published guidelines on the choice of appropriate test species. However, for practical reasons, the choice of species is very limited; a major consideration is experience in using a particular species, and the existence of a body of historical data on that species.

12. A study of the role of dog studies in pharmaceutical testing[222] found that for 115 pharmaceuticals, dog studies confirmed the findings from rat studies in 63% of cases but provided additional information in 37% of cases. New findings in the dog resulted in termination of drug development in 11% of all studies. Although the relevance of the findings for human safety could not be ascertained, the study recorded a 'consensus among toxicologists that the dog plays an essential role in the safety evaluation of pharmaceuticals'.

13. The formulaic use of two species in safety testing was not considered to be a scientifically justifiable practice, but rather an acknowledgement of the problem of species differences in extrapolating the results of animal tests to predict effects in humans.

14. The routine use of the dog in addition to the rat is a compromise which is largely dictated by practical considerations. Nevertheless, studies in dogs have provided important information.

Are new compounds ever tested on humans when they have already showed an adverse reaction in animals? If so, what is the purpose of the animal experiments?

15. All chemicals have some degree of toxicity. In the case of pharmaceuticals, compounds are intended to have beneficial effects on the human body; their pharmacological activity. If this activity is excessive, it becomes detrimental. The purpose of animal tests is to define the dosage which will yield a beneficial rather than a detrimental effect. Inevitably, the animal tests will be designed to include doses which have adverse effects.

16. In addition to the pharmacological effects of medicinal chemicals, they and all other chemicals have adverse side effects at certain doses. The purpose of animal tests is to define the nature of these effects and the likelihood of their occurrence at particular doses.

17. The process of risk assessment, based on animal test data, followed by risk management measures, is intended to ensure that human exposure is restricted to levels that do not result in an unacceptable risk of adverse effects. It is to be expected that chemicals will be used for various purposes despite the fact that adverse toxicological effects have been found in animals.

18. One specific example was discussed—the fact that many licensed medicines have been shown to induce cancer in rodents. It was acknowledged that rodent cancer tests can give very misleading results, due to factors such as hormonal disturbance, chronic irritation, and stimulation of rodent-specific receptor molecules. It is also widely accepted that studies are poorly designed, using unrealistically high doses of chemicals. For this reason, the results are often considered irrelevant to human risk assessment. Nevertheless, the tests are useful for identifying potent carcinogens. Another factor mentioned was the clinical risk/benefit calculation, where a clinician must decide whether the risks of adverse effects from a particular treatment outweigh the risks of withholding treatment. The use of medicines with known adverse effects in animals may or may not be justifiable, but the information on risk is essential.

19. The number of people who suffer from 'Adverse Drug Reactions' is often cited as evidence that animal tests fail to protect people from toxic medicines. The point was made that the majority of such cases were due to poor practice and overdosing. Others argued that the figures often cited for Adverse Drug Reactions had already taken such factors into account.[223] The risks associated with potent medicines may be well defined, but inappropriate use cannot always be prevented.

20. It was concluded that all chemicals have toxic effects at high enough doses. The purpose of the animal experiments is to estimate the potency of the chemical so that margins of safety can be established.

Can there be absolute and objective measures of toxicity?

21. It was considered that toxicity is a deviation from normal which is not absolute, short of death, but that many forms of toxicity can be measured objectively.

What is a realistic timeframe for the reduction of animal use in toxicology by (a) 20%, (b) 50% and (c) 90%? Can the complete elimination of animal use in toxicology ever be envisaged?

22. There was general agreement that a 20% reduction could be achieved very quickly by means of harmonisation of test guidelines, reductions in the number of animals used in each test, and greater use of available alternative methods. A 50% reduction would require considerably more development and scientific research but was feasible within 10 years. A 90% reduction would probably take at least 20 years, and would need a major breakthrough in modelling of biokinetics (mathematical methods for predicting uptake, metabolism and excretion of chemicals in the body) and/or a rapid development of toxicogenomics (molecular biological techniques).

23. Funding of research on alternatives was considered to be poor. At the EU level, research funded by DG Research has not led to the development of any new tests. In the UK, there is some support provided by the APC and potentially by the MRC, but there appears to be a lack of high quality applications for work in this field. A new approach was considered essential to identify promising areas of research, raise the status of alternatives-related research, and provide funds on a large scale.

24. For chemicals testing, a forum for discussion is needed, analogous to the International Conference on Harmonisation (ICH) for pharmaceuticals.

25. Some members of the group believed that complete replacement of animals in toxicology was possible as long as there was sufficient will and determination. A minority view was that all animal testing could be replaced immediately, although it was acknowledged that this was not the accepted view of most scientists and regulators, and that changing these attitudes was probably a long-term objective.

Which research areas are most likely to lead to the replacement of animals in toxicology (e.g. computer modelling, in vitro human tissue testing, or in vitro animal tissue testing)?

26. The group agreed that the three examples quoted are important areas but that progress could also be made by making greater use of human volunteer studies and that the development of computer based biokinetic/pharmacokinetic modelling was fundamental to the successful application of in vitro tests.

27. The use of human tissue suffers from problems of supply which need to be addressed urgently. The use of 'humanised' cells (mammalian cells containing inserted human genes) could also be helpful.

28. New molecular techniques such as toxicogenomics and metabonomics were considered unlikely to be useful in replacing animal tests but could be effective in reducing animal use or in developing refined test designs.

29. Ultimately, an increased understanding of the mechanisms of toxicity, e.g. sensitisation, irritancy and carcinogenicity, will lead to the rational design of in vitro tests.

Does the search for alternatives focus too much on replacement? Is sufficient attention given to reduction and refinement?

30. Considerable advances have been made in reduction and refinement, for example in the acceptance of alternatives to the oral LD50 test. Harmonisation of guidelines is also a very effective reduction strategy.

31. The development of new alternative methods as replacements usually requires more financial support because it involves a great deal of laboratory work, and extensive inter-laboratory validation studies. In comparison, the development of reduction and refinement alternatives is generally less costly and their introduction less contentious; extensive validation is usually not required. Education and training is a particularly important and cost-effective strategy for promoting the use of alternatives, good study design and improved animal welfare.

32. The group considered that the three Rs are inter-linked and should all the subject of research.

What are the obstacles to promoting alternatives in toxicology?

33. The demand in the USA for testing chemicals for endocrine disrupting properties was cited as an example of testing driven by legal requirements rather than sound science. This and several other examples, such as testing for effects in juvenile animals and for behavioural endpoints, were considered to relate to a general problem of regulators in the USA exercising a high level of individual power.

34. In Europe, the current legislation on chemicals requires the submission of the results of prescribed sets of tests; the 'checklist' approach. Unnecessary or irrelevant testing may be done simply to complete a required data set. It is hoped that the proposed new EU chemicals strategy will be based on a sensible, structured approach to testing.

35. The group considered that some animal testing was done for 'administrative', rather than scientific reasons. More flexibility and intelligence is needed in chemical testing.

36. The main obstacle to the validation of alternative methods is the poor quality of the data from the corresponding animal test with which the new method is compared. Even when sound data exist, obtaining test results from chemical companies is often a problem. A way forward must be found to enable validation to proceed without the need for animal test results.

37. Acceptance of alternative methods by the OECD is crucial to their widespread use in toxicology. This has proved to be an extremely slow process.

Report of Working Group 3: Freedom of Information

Chair/Rapporteur: Dr Jon Turney (UCL)

Section 24 of the 1986 Act

1. Section 24 of the Act does not, in theory, work as a blanket ban on the distribution of any information held by the Home Office relating to animal experiments. The Home Office does not accept that all information is confidential and it considers all requests for information. Animal welfare groups asserted that, in practice, the Home Office has cited section 24 as a reason for withholding information.

2. Section 24 should be repealed because, in practice, little information is considered by the Home Office to be non-confidential and little is therefore released to the public on request. Repealing section 24 would help to change the culture of secrecy at the Home Office in line with the principles of openness outlined in the Freedom of Information Act 2000.

3. By itself, however, repealing section 24 would not be sufficient to enable the public to access information. Under the Freedom of Information Act, section 38 (relating to putting individuals at risk) and section 41 (relating to confidential information) could still be used to prevent the release of information about animal experiments into the public domain.

Duplication

4. Some felt that duplication was a real problem. Others argued that it was an essential part of the scientific process. As in other areas, more transparency could prevent unnecessary duplication and reassure the public when duplication was necessary.

Involvement of Animal Welfare Groups in Cost/Benefit Assessment

5. Cost/benefit assessment is a moral and subjective judgement and should reflect public opinion rather than the opinions of a narrow group of scientists, as at present.

6. Animal welfare groups do not wish to be involved in licensing because it would conflict with their remit, involving them in the regulation of animal experiments. Ethical committees do not have the resources to do a rigorous scientific review of the need for each animal experiment.

7. Welfare groups could be involved prior to licensing, on some sort of board for assessing the need for animal experiments, to which scientists could be brought to make suggestions for alternative experiments.

8. Animal welfare groups need to feel as though they could have an impact. For this to happen, and to allow for meaningful dialogue, there needs to be greater transparency and public access to information.

Information for the general public

9. Relevant government publications are currently difficult to access. More data need to be placed in the public arena, using the world wide web.

Role of Scientists

10. Scientists could help to increase understanding by avoiding the use of euphemistic language when referring to animal experiments, and by providing frank descriptions of what was done to the animals.

11. Failed experiments, not just those that were successful, should be written up and published.

12. Scientists do write up experiments that disprove hypotheses, because this is as scientifically important as experiments that proved hypotheses.

13. Scientists need to communicate about their experiments in a precise manner so that others can repeat the experiments. Lay summaries could be included in addition to the usual article written for other scientists.

14. Scientists should criticise other scientists in the glare of the media, not just in scientific journals.

The media

15. The media present the issues in a polarised manner and has treated all positions unfairly.

Education

16. Very few science graduates have participated in experiments on animals throughout school and a decreasing number during university education.

17. Issues about animals should be taught in schools on a philosophical basis in order to encourage children to think through the issues carefully. The debate should not refer just to animal experiments but how society uses animals in general, for example in food production.

Statistics

18. The published Statistics are not intended to enable public understanding, but are for statisticians. They are particularly unhelpful as they show only aggregated figures; however, even disaggregated figures would not be much better.

19. Suffering can not easily be recorded in a meaningful way. It might help to increase the number of severity categories and give some general descriptions about the type of suffering, in addition to cross-references to the Statistics.

20. Actual levels of suffering could be retrospectively reported, and related to broad categories of procedures or common experiments.

Training for scientists and laboratory workers

21. Scientists and laboratory-workers, particularly senior personnel, should be re-trained every few years.

22. The focus should not be on re-training but on continuing professional development.

Points made in Plenary Session

23. There is a need for more collaborative discussion groups.

24. The media is not one entity: some journalists and broadcasters are becoming more nuanced and responsible in their reporting of the issues.

Report of Working Group 4: Regulation and the Ethical Review Process

Chair/Rapporteur: Dr Jane Smith (Boyd Group)

Should the role of the Inspectorate be reviewed?

1. There was agreement that 21 inspectors had not been sufficient. Some members felt that the new target number of 33 would be sufficient; others that numbers are not the crucial issue. Whether numbers are sufficient depends on what precisely the role of the Inspectorate is to be.

2. It was noted that currently over half of all inspections are unannounced, reducing the validity of accusations of 'cosiness', and more such unannounced inspections would be welcomed by all.

3. There was disagreement about the extent to which the present inspection system is balanced. Certain participants argued that inspectors are not sufficiently disinterested, because nearly all have previously been licensed to work experimentally on animals. They therefore come from within the system rather than sitting outside of it. For this reason, some maintained that there should be inspectors charged solely with animal care and welfare, who had never held licences, and could act as the animals' advocates. Others argued that this role was already provided for by the NVS/NACWO team.

4. It was agreed that there is currently a lack of transparency in the workings of the Inspectorate. For example, the proportion of project licence applications that are turned down both (i) initially; and (ii) even after final resubmission, is neither published nor known to welfare and other campaigning organisations. Moreover, it is unclear how severity limits and bands are defined and how the cost benefit assessment is applied in practice. There was consensus that more openness is needed.

5. It was noted that separating the roles of the Inspectorate into (i) inspecting; and (ii) providing expert advice would increase the demands on the Inspectorate as the 'inspecting' Inspector would not already have the detailed, contextualised knowledge of the project licence. With more inspectors, a greater degree of team working and specialisation within the Inspectorate would be feasible. It was agreed that there was a need for more independent enquiry in cases of criticism of the Inspectorate or alleged wrong-doings under the Act.

How could the Ethical Review Process be improved?

6. It was agreed that ERPs should have true external members. There was debate about whether these should be strictly 'lay'; who should be classified as a lay person; and who would command the confidence of the wider community. It was agreed that there was a need for feedback on the contribution that lay members have made, and could make, within the ERP.

7. Members disagreed as to whether an anti-vivisectionist could be a member of an ERP and whether anti-vivisectionists counted as 'lay people'. Some anti-vivisectionist participants indicated that they would, in any case, not wish to be involved in an ERP.

8. It was suggested that, in order to represent the interests of animals, there was a need, not only for 'lay' people, but also for independent external experts in animal welfare.

9. It was observed that it was difficult for lay people to challenge the claim that knowledge per se is a benefit. Those in favour of animal experimentation could always argue that there might be instrumental (e.g. medical) benefits from almost any scientific procedure at some point in the future.

10. Some participants felt that ERPs duplicate the work of the Inspectorate; others that ERPs and the Inspectorate complement one another. At present ERPs technically only advise certificate holders and the role of ERPs would therefore change if they were to have the power to approve minor amendments. Although there was some support for moving to a system in which ERPs could approve very minor or welfare-friendly amendments to project licences, there was also some reluctance on the part of participants who were members of ERPs to move too far into a judgmental, rather than an advisory role.

Are levels of bureaucracy having an adverse impact on (a) animal welfare, and (b) scientific research in the UK?

11. It was observed that it is difficult to comment on the impact of bureaucracy because every case is different. It could take considerable time to gain approval for a PPL, but it was agreed that this is often for good reasons and is not always the 'fault' of the ERP and/or Inspectorate. Project licences last for five years, and it can take time to get such wide ranging documents right. There was disagreement as to the extent to which such delays really hold back UK science.

12. There was more agreement, however, that the process of approval for applications for minor amendments to projects can be unnecessarily cumbersome. If an ERP approval process were to be available for minor and welfare-friendly amendments, there would be a need to define 'minor'. It was noted that this question already occurs in the context of fast tracking minor amendments by ERPs. Although many ERPs have their own guidelines on what counts as "minor", it would be impossible to draw up a definitive list, even within a single establishment. A case-by-case approach would be more appropriate and trust in the NVS and NAWCO is crucial.

13. There was disagreement about the extent to which members of the Inspectorate are consistent with regard to the detail that they require on project licences. One person maintained that such differences are not great and that the Inspectorate goes to great lengths to minimise them; another person maintained that some Inspectors actively help project licence applicants to write their applications so as to allow for their flexible interpretation whilst other inspectors require great specificity.

222   Broadhead, C. L. et al., "Prospects for reducing and refining the use of dogs in regulatory toxicity testing of pharmaceuticals", Human and Experimental Toxicology, vol. 19 (2000), 440-47. Back

223   See, for example, Lazarou, J., Pomeranz, B.H., Corey, P.N. Incidence of Adverse Drug Reactions in hospitalised patients: a meta-analysis of prospective studies. Journal of the American Medical Association, Volume 279 (15), 1200-05 (1998). Back