Volume 8: Variant CJD
3. Establishment of the CJD Surveillance Unit
Public Health Laboratory Service involvement
PHLS and CJD surveillance
Second request for PHLS involvement in CJD surveillance
What could the PHLS have contributed to CJD surveillance?

3.20 The CJDSU was established expressly to undertake CJD surveillance rather than following the usual route of disease surveillance and control by the Public Health Laboratory Service (PHLS), which did not become involved in CJD surveillance until after 20 March 1996.

3.21 The PHLS is an executive non-departmental public body with responsibility for providing a microbiological and epidemiological service to health authorities and local authorities for the diagnosis, control and prevention of infection and communicable disease. ¹ In 1986, it comprised a network of 52 diagnostic laboratories throughout England and Wales with headquarters at Colindale in London. It was, and continues to be, supported by two surveillance centres, the Communicable Diseases Surveillance Centres, or CDSC, which are based in London and in Wales and by a specialist research establishment, the Centre for Applied Microbiology and Research, at Porton Down. Since the PHLS operated only in England and Wales, it was not considered to have a role in UK-wide CJD surveillance, ² although precedents for UK-wide surveillance by PHLS do exist. ³ Details of the Welsh PHLS and the mechanisms of disease surveillance in Northern Ireland and Scotland can be found in vol. 9: Wales, Scotland and Northern Ireland.

3.22 The CDSC served as the epidemiological arm of the PHLS by keeping human infectious diseases under surveillance. The Centres worked with other PHLS units to provide expert epidemiological support for the study of infectious diseases including the investigation of outbreaks. Their surveillance function was based upon regular returns of diagnostic data from the PHLS laboratories, supported by other information, including reports from clinicians and others. ⁴

3.23 Strong links existed between PHLS and its sponsor Department, DH. Members of the PHLS board which determined policy included a Deputy Chief Medical Officer (DCMO) and a DCMO from the Welsh Office (until July 1989), and other DH staff attended as observers. ⁵ In addition, the PHLS was required to produce an annual Accountability Review for discussion at a meeting with Health Ministers. The Review included a Corporate Plan produced in the light of regular dialogue between DH and Welsh Office officials and the PHLS.

3.24 From 1988 onwards, annual 'Customer Liaison' meetings were held between the PHLS, DH and Welsh Office officials and served to discuss progress on the PHLS's ongoing work and concerns. In addition, there was day-to-day contact between PHLS and DH scientific and medical staff, particularly in relation to current outbreaks of infectious disease.

PHLS and CJD surveillance

3.25 Before 20 March 1996, several discussions had taken place between PHLS and DH officials about the possible assistance the PHLS might be able to provide in CJD surveillance. One of the surveillance options initially proposed by Dr Pickles to the CMO on 2 March 1989 in response to the recommendations of the Southwood Report, was for the CDSC to undertake surveillance of CJD (see paragraph 2.31). However, Dr Pickles did not approach the CDSC for help and this option was not pursued. ⁶ Nevertheless, the PHLS was concerned about the risks to humans from BSE and their 1989 Corporate Plan, which was prepared in mid-1989, highlighted the need for continued surveillance and research to evaluate these risks. ⁷

3.26 However, on 12 December 1989, 'a formal decision not to involve CDSC was taken at the PHLS Accountability Review meeting'. ⁸ The CMO's speaking notes for that meeting indicate DH's reasons for not involving the CDSC at that time:

Although some of the animal disorders may in the end turn out to be communicable, I would not myself have classed most cases of Creutzfeldt-Jakob Disease this way. We have spent much time with expert advice considering how to monitor cases of CJD and concluded it would not be appropriate for CDSC to be involved at this stage. In general, I would not encourage PHLS/CDSC to become involved in monitoring diseases unless they are known to be communicable. For the moment, that includes CJD. ⁹

3.27 In February 1990, Dr Pickles wrote to the Director of the PHLS, Dr (now Sir) Joseph Smith, to update him with the work funded by DH being carried out on spongiform encephalopathies. ¹⁰ The letter was produced following discussions between Dr Pickles and Dr Smith about possible involvement of the PHLS and CDSC. In evidence to the Inquiry, Dr Pickles said that she wrote the letter at Dr Smith's request to assist him in persuading doubting colleagues that PHLS involvement was unnecessary. ¹¹

3.28 In her letter, Dr Pickles detailed the reasons why DH was reluctant for PHLS to become involved. These included:

1. The role of the PHLS was limited in practice to the study of the human health aspects of infections that were known to transmit from animals to man. This had not been shown for BSE.
2. The Tyrrell Committee had concluded that the most effective way to monitor CJD was to embark on a study similar to that conducted by Professor Matthews in the late 1970s and early 1980s. Experiences in the early study suggested that direct verification of cases would be essential and therefore that clinical neurologists/neuropathologists would be indispensable. The PHLS and CDSC did not have this expertise available.
3. It was generally accepted that CJD was not communicable in the normal sense (ie, transmission had only been shown iatrogenically and experimentally). It did not therefore seem necessary to involve the CDSC in monitoring, not least because it might have given the impression that CJD could spread from person to person. Moreover, CDSC involvement could have been misinterpreted to suggest that a rival study had been set up. Similarly it was not desirable for local groups to report on CJD cases without consideration of the complete and accurate set of data collected by Dr Will.
4. It was not considered part of the mainstream business of the PHLS to be involved in laboratory studies of CJD, investigating the neuropathology or nature of the agent. In addition, it was considered desirable from a health and safety aspect to keep this type of work confined to as few laboratories as possible.

3.29 Dr Pickles's letter was submitted by Dr Smith to the PHLS Board for their consideration in April 1990. ¹² Dr Smith provided the Board with background information on CJD surveillance and noted that the PHLS was not undertaking significant microbiological or epidemiological work on slow viruses, something for which it might later be criticised. However, he also stated that:

On the other hand, comprehensive surveillance and research is in progress in the UK and unnecessary duplication of research effort is to be avoided. The PHLS has to prioritise the use of its resources and currently savings in the order of £1.0 million have to be found.

3.30 The PHLS Board concluded that there was not a need to undertake studies in view of the extensive work in hand by DH which it would not be cost-effective to duplicate. ¹³ Indeed, the memorandum submitted by the PHLS on 11 June 1990, in response to an invitation from the Agriculture Select Committee for its inquiry into BSE, reflected the conclusions of the PHLS Board:

. . . the service has no body of data to justify a memorandum to the Agriculture Committee, but we are of course keeping a close watching brief on the situation. If appropriate areas of work were to be identified which were not thoroughly covered by other groups, or for which a duplication of effort by PHLS staff appeared to be justified, this work would be seriously considered. As yet, however, such needs have not been identified and it is believed that the necessary areas of investigation are already being very adequately addressed by various expert groups. ¹⁴

3.31 Dr Pickles has pointed out in evidence to the Inquiry that these comments suggest Sir Joseph was not supportive of PHLS involvement. ¹⁵

3.32 However, in his evidence to the Inquiry, Sir Joseph said that he felt at the time that the PHLS should have played a role in the investigations, given the expertise and experience within the organisation. He said:

From, I believe, early in 1990, however, it was made increasingly clear to me that DoH and Ministers did not wish the PHLS to work upon BSE/CJD, nor to be seen to work or comment upon the subject, and especially that CDSC should not be involved. This caused me much concern. I thought that the PHLS should be involved in the critically necessary human epidemiological studies of BSE/CJD, and that the PHLS could make a valuable contribution to their planning and operation. ¹⁶

3.33 Sir Joseph felt that their involvement would have had a positive effect in reassuring the public, ¹⁷ and that the public was not sufficiently aware of the work of the PHLS to associate their involvement with the possibility that BSE might be communicable to humans. ¹⁸

3.34 Similar views were held by Professor Stephen Palmer and Dr Roland Salmon, consultant epidemiologists from the Welsh Unit of the CDSC. They felt that there was insufficient evidence to support the pronouncement by the CMO that beef could safely be eaten by everyone. ¹⁹ Indeed, Dr Salmon felt that a number of features of the biology of BSE and CJD suggested that a small risk of transmission to humans could not be excluded. ²⁰ In addition, both were surprised that it appeared that SEAC had not considered certain scientific publications which reported associations between CJD, food consumption and animal contact. Moreover, they felt that SEAC was not completely on top of the epidemiology and that there was a preoccupation with biological rather than statistical issues. ²¹

3.35 Both Professor Palmer and Dr Salmon were therefore concerned about the lack of PHLS involvement. They had reservations about the CJDSU epidemiology study and considered that 'such an important and long term study (20 years +) might be better directed by either MRC or PHLS or both'. ²² The contribution that members of the PHLS felt could have been made by the PHLS/CDSC is discussed later in this chapter. ²³

Second request for PHLS involvement in CJD surveillance

3.36 In January 1993, Dr Diana Walford was appointed Director of the PHLS. Dr Walford reopened the question of PHLS involvement in BSE/CJD. She believed that the argument that BSE was not a human pathogen could not be sustained, because this was not known at the time. Her opinion was that the PHLS should have been engaged in work in order to ascertain whether or not it was a human pathogen. ²⁴

3.37 Furthermore, Dr Walford also believed that the PHLS's unique expertise in communicable disease epidemiology, coupled with its experience in the investigation and handling of major national incidents, should have been fully utilised. ²⁵

3.38 However, in evidence to the Inquiry, Professor Will has stated that the work of the CJDSU was supervised by the MRC Allen Committee, whose membership included epidemiologists. ²⁶ Further, he has pointed out that the CJDSU received expert epidemiological and statistical advice from the London School of Hygiene and Tropical Medicine (LSHTM). ²⁷

3.39 Dr Walford with the support of other PHLS staff, ²⁸ discussed this topic with Sir Kenneth Calman (CMO) several times between March and September 1994. ²⁹ Sir Kenneth continued to express the view that the PHLS should not be involved in this area of work. ³⁰ He believed that there was no gap in surveillance that the PHLS could readily fill and that there was no room for PHLS involvement while a diagnostic test was still unavailable. ³¹

3.40 In November 1995, when the report of a small number of cases of CJD in young people became known, Dr Walford again became concerned that the PHLS was not involved in human TSE work. She took the initative of contacting Dr Will directly to ask him if he would welcome any epidemiological assistance from the PHLS. ³² Dr Will asked to see several of the PHLS surveillance databases to try to identify possible misdiagnosed cases of CJD and suggested that Dr Walford should contact Professor Smith (LSHTM), who had been providing statistical input to the CJDSU. ³³

3.41 Dr Walford kept Sir Kenneth fully informed of her discussions with both Dr Will and Professor Smith. ³⁴ But when she informed him, on 22 December 1995, of a proposed meeting between CJDSU and PHLS, he made it clear that he did not wish the meeting to go ahead. Sir Kenneth felt it might compromise the position that SEAC, whose epidemiological expertise had been strengthened by recent appointments, ³⁵ should be the single source of scientific expertise on prion disease. ³⁶ The proposed meeting was cancelled. ³⁷

3.42 Since the announcement on 20 March 1996 of the possible link between vCJD and BSE, the PHLS has been involved in a number of aspects of work on CJD, including:

1. a project to detect, retrospectively, under-ascertainment of vCJD in Wales;
2. establishing active surveillance, through the British Paediatric Surveillance Unit, of progressive intellectual and neurological deterioration in children under 16 years of age in the UK, to determine whether cases of CJD were occurring in that population;
3. a project to set up a panel of clinical samples from patients with neurological disorders, for the evaluation of candidate tests for CJD;
4. a project to develop a diagnostic test for CJD;
5. regular monitoring of the trend in incidence of vCJD, in collaboration with the CJD Surveillance Unit; and
6. various reviews of the methodology and work of the CJDSU.

Furthermore, a PHLS statistician was appointed as a member of SEAC's Epidemiology Subcommittee and Dr Walford was herself appointed a member of the CMO's Committee on the Human Aspects of Spongiform Encephalopathies. ³⁸

What could the PHLS have contributed to CJD surveillance?

3.43 In evidence to the Inquiry, Sir Joseph Smith and Dr Walford identified several areas in which they felt that PHLS could have contributed to CJD surveillance. These include paediatric surveillance, development and administration of the questionnaire and field work, dissemination of information to health professionals and facilitating links with clinicians other than neurologists. In the following sections we describe these points and the contrary arguments put forward, especially by Dr Pickles. In the discussion section at the end of Chapter 5 we return to the role of the PHLS in CJD surveillance.

Paediatric surveillance and access to medical registers

3.44 A close link existed between the British Paediatric Surveillance Unit (BPSU) and the PHLS, who supplied a consultant epidemiologist as an advisor to the BPSU. The association between the two organisations established a system of 'active case searching' whereby paediatricians reported monthly on cases of specific diseases. Where these specific diseases were not reported, a system of 'follow-up' checks ensured that no cases were missed. ³⁹ Sir Joseph noted that this surveillance had been very successful in the surveillance of Reye's syndrome ⁴⁰ and SSPE. ^{41 42}

3.45 Dr Walford also felt that the paediatric surveillance was a good model for how the PHLS might have approached surveillance of a rare condition like CJD. The PHLS could have worked with partners such as BPSU, British Paediatric Association, British Neurological Surveillance Unit and with neurologists to cover both the UK as a whole and all age groups.

3.46 However, in evidence to the Inquiry, Dr Pickles pointed out that at the time the CJD monitoring study was initiated, only one case of CJD in a person under the age of 16 had been reported worldwide. It did therefore not seem appropriate to involve the BPSU at the time. About access to medical registers, the CJDSU could have had access to the registers of neurological disease held by the PHLS (eg, SSPE, hospital discharge statistics, death certificates) on an informal basis as envisaged by the PHLS Board in their April 1990 meeting. ⁴³ Dr Will had in fact examined the PHLS's SSPE register in 1995, since it was possible that CJD cases had been misdiagnosed as SSPE (he did not, however, find any evidence of misdiagnosis).

Development and administration of the questionnaire

3.47 Dr Walford noted in her evidence to the Inquiry that food questionnaires, such as the CJDSU questionnaire, entail a high possibility of ascertainment bias and that the design and administration of such questionnaires is a skilled job. ⁴⁴ She considered that however skilled the research neurology registrars available to the CJDSU may have been, they did not have the required training in the field, nor the skills of PHLS epidemiologists, to ensure sufficient accuracy in data collection. A partnership between the PHLS and the CJDSU would have allowed a PHLS epidemiologist to accompany the neurologist to interview the relatives of CJD victims.

3.48 In her evidence to the Inquiry, Dr Pickles recognised the experience and expertise of the PHLS epidemiologists, but questioned whether their experience of relatively acute events such as food poisoning was pertinent to the investigation of causative factors which were widely used and where no information on the necessary dose for exposure or duration of exposure was available. She also doubted whether anyone, regardless of experience, would be able to obtain reliable or accurate dietary history in CJD cases, where information had to be obtained from family members. Dr Pickles noted the methodological advantages of having a single interviewer who needed to be a neurologist.

Dissemination of information

3.49 The PHLS had and has a variety of mechanisms for disseminating information. A weekly publication, Communicable Diseases Report, provides information about infectious diseases to public health doctors, microbiologists and the Environmental Health Departments on current infection statistics and issues. The consultant in communicable disease control in each district would also receive the publication and disseminate the information to those who needed it. Electronic means for the dissemination of information also existed. The PHLS 'Epinet' was used to alert all public health professionals, departments of public health and district directors of public health to specific matters of concern. Latterly, the internet has provided an opportunity for more widespread dissemination in the form of the PHLS website, on which the Communicable Diseases Report is also available. ⁴⁵

3.50 Both Sir Joseph and Dr Walford felt that these vehicles could have been used to alert general practitioners and psychiatrists to CJD through public health professionals, though they appreciated that direct information to relevant professional associations needed to be supplied by DH. However, DH had its own networks for disseminating information and advice to the medical profession through letters from the CMO and the publication Health Trends. ⁴⁶ With regard to CJD, Dr Will arranged for the inclusion of relevant enclosures circulated by the Association of British Neurologists to its members, communications were relayed directly to neurologists and relevant information was provided directly to professionals in public health if required or requested. ⁴⁷

Facilitating links with clinicians other than neurologists

3.51 It has been suggested that the PHLS might have been able to facilitate links with clinicians other than neurologists, eg, psychiatric geriatricians, to whom CJD might present. ⁴⁸ However, DH expected that given the neurological nature of CJD, patients would be referred to a neurologist even if they had been seen by other specialists first. ⁴⁹ It was considered in this case that Dr Will would have had closer links to neurologists than the CDSC could have developed.