Social Audit Ltd
P O Box 111 London NW1 8XG
Telephone/Fax 44 (0)171 586 7771


The Rt Hon Frank Dobson PC MP
Secretary of State for Health
Richmond House, Whitehall,
London SW1A 2NS 17 June 1999

Dear Mr Dobson,

I recently read a guest editorial in a professional journal (Clinical Research Focus, November 1998) in which the writer drew attention to significant risks for patients entering clinical trials. The author is Director of a commercial agency specialising in training and auditing for compliance with Good Clinical Practice requirements and her opinion is based on extensive experience. Over the past 10 years, her company has conducted some 800 GCP audits of investigator sites, clinical reports, clinical laboratories and Standard Operating Procedures, involving over 200 companies in over 25 countries - but mainly the UK. Detailed findings from a representative sample of 226 audits involving 31 companies have been published in Applied Clinical Trials (February 1998 to June 1999), based on the company's computerised records of compliance for some 1,500 checklist items per audit. The writer, Wendy Bohaychuk PhD, concluded thus:

"Our database of GCP compliance assessment basically shows that new GCP guidelines and regulations have not yet improved the situation for patients over the last few years and frankly, after 10 years of detailed auditing, I would never go into a clinical study myself and I would certainly try to discourage anyone in my family from doing so."

That seems to me a matter of real concern, as I'm sure it will to you - but the question, as always, is what might be done about it? In the first instance, this should of course be a question for the Medicines Control Agency and Committee on Safety of Medicines, rather than your Department - but I am extremely reluctant to turn to them, for reasons that will be obvious to your Permanent Secretary, at least. (The Office of the Parliamentary Commissioner for Administration recently asked him to intervene personally, to encourage the MCA/CSM to provide information I first requested in March of last year.)

Should you request it, I would be happy to provide further evidence of what I see as the extensive abuse of secrecy in this field - and you will see from the enclosed letter, sent in response to the consultation on the draft Freedom of Information Bill, that I do not expect things to improve. Nevertheless, it seems to me that Tony Blair came exactly to the point, when he spoke at the 1996 awards ceremony of the Campaign for Freedom of Information:

" … when a health scare like BSE occurs, the people want to know the facts, people want to know what the scientific advice is in full, and they need to be sure that the public interest always comes first. They want to know if there was any relaxation of regulations which resulted in public safety being compromised … The only way to begin to restore peoples' trust is therefore to be completely open about what the risks are …."

The case for openness about risks to patients in clinical trials seems equally compelling. Inevitably, some additional risks are involved in clinical trials - though of course there could be advantages for some individual patients who enroll, just as there may be benefits to society at large. But how can one reasonably expect patients to agree to participate, giving truly informed consent, if so many people are routinely exposed to avoidable risks of the kinds described - especially without evidence of effective regulatory control?

The evidence from this series of audits is not reassuring. Overall, no company achieved compliance of 80% or more of checklist items. The range was from 43% to 79%, with an average compliance level of 64% in the UK - the same average level for all countries combined. Here are a few examples of non-compliance findings from these audits. They suggest patients need far better protection than they get now, and/or raise serious doubts about the validity of reported results:

"Ethics committees were not informed of all serious safety events occurring in the study or elsewhere during the study" (81% of audits)

"There was significant under-reporting of safety information arising during the study" (31%)

"General practitioners were not informed that their patients were entered into clinical studies before the study began" (71%)

"Study subject was not clearly informed of instructions for using the medicine/device" (43%)

"There was inadequate (or incorrect) supporting evidence that suitable patients were entered into the study" (29%)

"There was inadequate evidence to show that study medications were stored safely and securely during the study" (55%)

"Informed consent was obtained from study subjects after the start of the study or after the start of any study procedures" (37%)

"The consent documents submitted to the ethics committee/IRB differed from the documents actually used at the study site" (26%)

"There were unexplained or unacceptable discrepancies in study medication accountability" (36%)

This evidence suggests it would most helpful if the MCA/CSM were to pass on their recommendations to patients invited to participate in clinical trials, explaining also in broad terms the basis on which their judgement is made. (By this I refer to their evidence, rather than authority - and to hard data relating to risk, rather than lack of evidence of harm). It would be especially useful to have answers to the following questions:

1. Whether or not the evidence available to the MCA broadly supports or refutes these published audit findings?

2. How many GCP audits has the MCA itself carried out to date and what plans, if any, the Agency has for expanding its work in the field?

3. Whether and to what extent independent evidence of levels of non-compliance with GCP is at present routinely scrutinised by the MCA (or CSM), as part of the process of examining Product Licensing applications?

4. Whether the MCA has ever conducted any enquiry into general standards of compliance with GCP, and/or the need for auditing by government regulators and, if so, when and what conclusions were reached?

5. Whether any sort of disciplinary action had ever been taken against any company or companies arising from non-compliance with GCP, and whether any record of such actions is publicly available?

6. Whether companies are formally required to demonstrate a satisfactory level of compliance with GCP as a condition of granting a Product License?

I have no doubt you will wish to satisfy yourself on these points, but I would appreciate any assistance you might also feel able to give in securing publicly available answers to these questions. Thank you for your interest and attention. Along with some of our website visitors, I shall look forward to hearing from you.

Yours sincerely,
Charles Medawar


Contents page
Correspondence with government