Investigating allegations of research misconduct: the vital need for due process
Commentary: Response from members of the Griffiths inquiry
Editorial: Inquiring into inquiries
"Almost every statement made about the design, conduct, and reporting of the neonatal CNEP trial in the Griffiths report was ill informed, misguided, or factually wrong," concludes a review of its findings in this week's BMJ.
The internal review chaired by Professor Rod Griffiths was set up in February 1999 after several parents alleged that their premature babies had been entered into a controlled trial of a breathing support technique, known as continuous negative extrathoracic pressure (CNEP), without their consent. The resulting report was published in May this year.
Dr Edmund Hey, a retired paediatrician from Newcastle upon Tyne, and Dr Iain Chalmers, director of the UK Cochrane Centre in Oxford, conducted their review on behalf of the Medical Defence Union. In their swingeing critique the authors raise serious doubts about the reliability of the report, the review, and the appropriateness of the ensuing recommendations.
The authors defend the trial's design, which, they claim, the panel had failed to understand. They refute the allegation that no external peer review had been conducted and they accuse the panel of making "an unwarranted slur on the professionalism and skill of all the nurses concerned." They say that a consent document existed among the research records for every child in the study. Assertions made by a small number of parents about events up to 10 years ago went unchecked, say the authors, yet now form the basis for recommending major changes to the governance of clinical research in the NHS.
"We can still agree that there has been a scandal, but suspect that the scandal is what has been done to, not what was done by, the medical and nursing staff in Stoke on Trent."
Responding to the criticisms, the members of the panel say that Hey and Chalmers have completely misunderstood the terms of reference of the investigation and have ignored the main substance of the report. Constrained by the fact that some of the issues and individuals involved are still being legally investigated, they nevertheless claim that the consent forms did not contain all the necessary information and that there were discrepancies between the stated ages at which babies were entered into the trial. Parental recall may indeed be faulty, they agree, but add: "this does not mean that all parents are wrong and all researchers are right."
An accompanying editorial by Richard Smith says that the whole system of such inquiries needs to be re-evaluated, and that their process, including the appointment of the members and the purpose of the investigation, should be made clear from the outset. "Inquiries should publish their material and methods, check oral allegations against documentary evidence, and send drafts of evidence accusing individuals to those individuals, so that errors of fact can be corrected."
Contacts:
Dr Iain Chalmers, UK Cochrane Centre, Oxford
Email: ichalmers@cochrane.co.uk
Professor Rod Griffiths, NHS Executive, West Midlands Regional Office, Birmingham
Dr Richard Smith, Editor BMJ, London
Email: pressoffice@bma.org.uk