Intended for healthcare professionals


Clinical research under the cosh again

BMJ 2004; 329 doi: (Published 29 July 2004) Cite this as: BMJ 2004;329:241
  1. Charles Warlow, professor (charles.warlow{at}
  1. Department of Clinical Neurosciences, Western General Hospital, Crewe Road, Edinburgh EH4 2XU

    This time it is ethics committees

    Today the BMJ publishes a clutch of papers on the regulation of clinical research by ethics committees.15 All describe, in one way or another, how ethics committee review may impede and delay research, sometimes even to distort the methods so much that the conclusions are flawed and patients damaged—an unintended unethical consequence.

    Although this is not just a problem in the United Kingdom, to know exactly what is going on in other countries is difficult. If the situation is anything like that in the United Kingdom it will be confusing, changing all the time (just last month the UK government notably altered its human tissue bill), and made even more confusing by varying guidance from official bodies such as the General Medical Council and the BMA. Clearly, international differences are a particular problem for multicentre research across national boundaries. For example, unlike in the United Kingdom, the United States has provision for waiver of consent for research including patients with sudden mental incapacity, such as those with cardiac arrest.6

    But even within the United Kingdom there is confusing variation. In Scotland, but not quite yet in England, legislation exists to protect the rights of incapacitated adults that made unbiased research into sudden brain injury impossible until trumped on 1 May 2004 by the implementation of the European Clinical Trials Directive, which makes it (more or less) possible again, although problems remain.79 In England, but not in Scotland, the research use of routinely collected health data is regulated by legislation under section 60 of the Health and Social Care Act,10 but, illogically, audit escapes almost any regulation at all. Not surprisingly, much research is now conveniently rebadged as audit.

    So what is worrying the researchers this week? Hester Ward and colleagues describe how their case-control study in Creutzfeldt-Jakob disease has been made extremely complicated and expensive by the demands of modern data protection.1 Worse, their methods are now so severely compromised that the response rate among the controls is only a meagre 16%, and so the results are likely to be unreliable. Konrad Jamrozik bemoans the length and complexity of research ethics committee review and makes suggestions for improvement.2 As of 1 March 2004 the new arrangements are meant to be an improvement on the old, but with a 68 page form, with one extra for student projects, Wald remains unconvinced.4 The questions are very little to do with ethics, and for his own simple project the form required 44 hours to complete, at a cost of about £850. Jones and Bamford describe their dismay when they discovered that apparently trivial (to the investigators) changes in their protocol had to go back for a lengthy approval process by the ethics committee and R&D department, and every other project in their unit was scrutinised as well.3 And finally, Michael Parker and colleagues try to sort out just what is research and so requiring ethics committee approval, and what is clinical practice when it comes to investigating rare genetic disorders—another minefield for the unwitting researcher.5

    These are increasingly familiar problems and a threat to the future of clinical research.11 Rightly or wrongly, clinical researchers are exhausted by the demands of ethics committees that seem more concerned with the science (which they cannot necessarily judge) and editorial control of patient information sheets than with ethics. But what can be done? The epidemiologist Sir Richard Doll has been quoted as saying that no one with the power to do anything will take any notice until a bona fide researcher is jailed for transgressing some trivial regulation, and he sportingly volunteered to be the test case himself.12 Others have suggested that we must involve patients and patients' organisations. After all government is more likely to listen to consumers, in this case of research findings, than researchers. In the meantime, when discussing any flaws in their study, researchers should make plain which scientifically inappropriate aspects of the methods were forced on them by ethics committees and how the results may be biased as a consequence. If nothing is done, clinical research will wither, and that is not to anyone's advantage.


    • Education and debate pp 277-89

    • Competing interests CW is a member of the Royal College of Physicians (London) committee on ethical issues in medicine and the MRC committee on the ethics of research involving human participants or tissues and personal information, but the views in this paper are his own. His recent application to a research ethics committee for a student project was turned down and is currently subject to appeal.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract