Skip to main content
NCBI home page
The BMJ logoLink to The BMJ
. 2000 Nov 25;321(7272):1345.

Investigating allegations of research misconduct

Worst outcome of Griffiths report would be that research becomes increasingly difficult

Terence Stephenson 1
PMCID: PMC1119076  PMID: 11090523

Editor—The Griffiths report contained a critique of two specific research projects and proposals on the regulation of research generally.

Hey and Chalmers express no view on these proposals.1 However, the accompanying editorial by Smith agrees with the report's main recommendations on the need for better research governance in the NHS without citing supporting evidence.

The Griffiths report made three main recommendations which could inhibit clinical research.

Firstly, concern about consent for research at times of stress. Most intensive care research requires consent at times of stress and if the researchers waited this research would be impossible. The financial cost of the suggested consent from a third party would inhibit research.

Secondly, the inquiry stated that brain damage could not be reliably assessed at a short period of follow up and was critical that follow up was determined by the amount of funding available. To place a requirement that all research in children involves long term follow up studies would abolish much research. The resources required are not addressed.

Thirdly, trusts should ensure that arrangements are in place for active monitoring of the progress of research. There is no indication that new funding will be available for this. Furthermore, “research involving vulnerable groups should be subject to an even greater degree of independent supervision than clinical research in general.”

Not for the first time will there be the paradox of well-meaning discrimination against children. This has been the case for trials of new drugs, in which the perceived obstacles deter research in children.2 The Department of Health's draft research governance “to protect participants, improve quality and stop research fraud” gives no evidence of the magnitude of these problems and does not discuss the resource implications of implementation.3 “There are already powerful incentives to adhere to many of the principles” in the framework, including the law, duty of care, and high professional standards of researchers. No evidence is quoted to assess how often these safeguards fail. Unreferenced “recent enquiries into adverse incidents relating to research” are inadequate justification.

Patients must be protected from inappropriate research, and informed consent must be obtained. However, extrapolating from the Griffiths report to clinical research throughout the United Kingdom must be thought through. Currently, 40% of drugs prescribed for children are not licensed,4 partly because of the existing deterrents to research in children. More hurdles are likely to inhibit research further,5 and ultimately it is children particularly who will suffer.

Footnotes

Professor Stephenson undertakes research in an NHS trust and a university, some funded in the past by industry. He was a member of the trust's local research ethics committee for five years and is a lead clinician for the directorate's research and development. He is a member of the Committee on Safety of Medicines and chair of the Joint Standing Committee on Medicines of the Royal College of Paediatrics and Child Health and the Neonatal and Paediatric Pharmacists Group, both of which have advised on drug trials in children. However, the views expressed are his alone.

References

  • 1.Hey E, Chalmers I. Investigating allegations of research misconduct: the vital need for due process (with commentary by R Griffiths, TE Stacey, J Struthers) BMJ. 2000;321:752–756. doi: 10.1136/bmj.321.7263.752. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Aynsley-Green A, Barker M, Burr S, Macfarlane A, Morgan J, Sibert J, et al. Who is speaking for children and adolescents and for their health at the policy level? BMJ. 2000;321:229–232. doi: 10.1136/bmj.321.7255.229. . (22 July.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Department of Health. Research and development. Research governance framework for health and social care (draft) consultation paper. www.doh.gov.uk/research/announcements/researchgovernanceconsult.htm; accessed 15 November 2000.
  • 4.Conroy S, McIntyre J, Choonara I, Stephenson TJ. Drug trials in children: problems and the way forward. Br J Clin Pharmacol. 2000;49:93–97. doi: 10.1046/j.1365-2125.2000.00125.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Anonymous. UK paediatric clinical research under threat. Arch Dis Child. 1997;76:1–3. doi: 10.1136/adc.76.1.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Nov 25;321(7272):1345.

CNEP trial was greatly flawed

Carl Henshall 1, Deborah Henshall 1

Editor—Hey and Chalmers are quick to discredit the Griffiths inquiry for not pursuing issues beyond its remit, but they themselves have conducted a biased review.1-1

The scoring system has been proved to allow bias in favour of continuous negative extrathoracic pressure (CNEP) ventilation over intermittent positive pressure ventilation. A baby who died receiving CNEP could score higher than its surviving matched pair receiving conventional treatment. If this scoring system and the statistical approach were supposed to enable termination of the trial if the subjects suffered harm why did the study continue after the 10th infant was strangled to death by the neck seal?

The trial did not run as it was designed to. Protocols were ignored and criteria overlooked, rendering the results invalid. The pilot study on the neck seal and whether it affected cerebral blood flow was not done until half way through the main study and used children who were already part of the main study.1-2,1-3 This suggests that the study was carried out without the researchers knowing the safety and efficacy of CNEP, rendering any consent form invalid.

Peer review was conducted by the National Perinatal Epidemiology Unit, Medical Research Council, and National Heart and Chest Hospital in 1990. How is this peer review when the trial began in 1989?1-2

The “bonding questionnaire” asks parents how they bonded with their children while in neonatal intensive care. It implies that CNEP was standard practice and does not mention the consent process.

We were constantly discouraged from handling our children because of the loss of negative pressure through the portholes. No research has been conducted on the effect of opening a porthole on an infant's cerebral circulation, despite the researchers' concerns.1-4 However, nurses mentioned that transferring a sick infant into CNEP compromised the rules on minimal handling.1-2,1-4 It could take up to one hour and at least two nurses to establish CNEP. One baby died during transfer.

Hey and Chalmers disagree with the Griffiths inquiry that “it is not possible to be sure who completed some of the consent forms.” Perhaps the inquiry could not believe that any parent would misspell their name or produce a perfect signature when unconscious.

The worst implication in this article is that CNEP could be beneficial neurologically. Some of the children who scored maximum points for normal cranial ultrasound findings had severe brain damage. In one such case an independent autopsy ascertained the cause of death as cerebral haemorrhage.

The retrospective audit being set up will definitively find the true number of children neurologically compromised by the use of CNEP; Hey and Chalmers should not pre-empt this.

Footnotes

Mr and Mrs Henshall are the parents of a child who was neurologically damaged by the CNEP trial and of another who died.

References

  • 1-1.Hey E, Chalmers I. Investigating allegations of research misconduct: the vital need for due process (with commentary by R Griffiths, TE Stacey, J Struthers) BMJ. 2000;321:752–756. doi: 10.1136/bmj.321.7263.752. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-2.Samuels MP, Raine J, Wright T, Alexander JA, Lockyer K, Spencer SA, et al. Continuous negative extrathoracic pressure in neonatal respiratory failure. Pediatrics. 1996;98:1154–1160. [PubMed] [Google Scholar]
  • 1-3.Palmer KS, Spencer SA, Wickramasinghe YA, Wright T, Southall DP, Rolfe P. Effects of positive and negative pressure ventilation on cerebral blood volume of newborn infants. Acta Paediatr. 1995;84:132–139. doi: 10.1111/j.1651-2227.1995.tb13595.x. [DOI] [PubMed] [Google Scholar]
  • 1-4.Raine J. Manchester: John Rowlands University; 1993. Continuous negative extrathoracic pressure in the neonatal respiratory distress syndrome [MD thesis] [Google Scholar]
BMJ. 2000 Nov 25;321(7272):1345.

British Association of Perinatal Medicine welcomes analysis of Griffiths report

Andrew Wilkinson 1,2, Janet Rennie 1,2

Editor—Incompetent would not be too harsh an adjective to describe the Griffiths report,2-1 particularly considering that the authors use equally strong language following their cursory indictment of the randomised controlled trial of continuous negative airway pressure (CNEP) in North Staffordshire in general and of Professor David Southall in particular. Many professionals and parents have suspected that the conclusions of the report were wrong. Now, the analysis by Hey and Chalmers shows that the inquiry team's report was indeed deeply flawed.2-2

Hey and Chalmers considered in detail documents that the inquiry team could have consulted but did not. Their assessment reveals an alarming degree of incompetence by the inquiry panel, which seems to have been content to rely on allegations made in oral evidence offered years after the events in question.

In the same issue Smith acknowledges that his previous editorial was based on a false assumption that the inquiry had been well conducted.2-3,2-4 Hey and Chalmers have done patients, clinicians, and researchers a great service in showing that what has been suspected by many of them is true. After external scientific and ethical review the CNEP trial in Stoke was carried out to well established standards. Indeed in many ways it was ahead of its time, a fact reflected in its publication in the world's leading paediatric journal.2-5

Neonatal research in the United Kingdom has always set high standards and has an excellent record of collaborative trials to evaluate and establish more effective and safer treatments. As in all fields of medicine, good research is essential to serve patients well and the understanding and willing involvement of parents will always be required.

The Griffiths panel failed to appreciate this, possibly because the distance of the panel members from the clinical arena clouded their interpretation of the facts. The response of the Griffiths panel members to the Hey and Chalmers paper ends with a statement that they consider their “job done” and will not take part in further debate. It is disingenuous of them not to defend their report, considering the extent of the damage it has done, not just to the doctors and nurses in Stoke on Trent but to all those caring for newborn babies and their parents.

We trust that Hey and Chalmers's painstaking analysis of this affair will set the record straight about the quality of the research and care in Stoke on Trent in the early1990s and re-establish confidence in the governance of British neonatal research after the damage done by the Griffiths inquiry.

Footnotes

Professor Wilkinson and Dr Rennie are practising clinical and research neonatologists.

References

  • 2-1.NHS Executive West Midlands Regional Office. Leeds: NHS Executive; 2000. Report of a review of the research framework in North Staffordshire Hospital NHS Trust (Griffiths report)www.doh.gov.uk/wmro/northstaffs.htm . ( www.doh.gov.uk/wmro/northstaffs.htm, updated 8 May 2000.) , updated 8 May 2000.) [Google Scholar]
  • 2-2.Hey E, Chalmers I. Investigating allegations of research misconduct: the vital need for due process (with commentary by R Griffiths, TE Stacey, J Struthers) BMJ. 2000;321:752–756. doi: 10.1136/bmj.321.7263.752. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-3.Smith R. Inquiring into inquiries. BMJ. 2000;321:715–716. doi: 10.1136/bmj.321.7263.715. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-4.Smith R. Babies and consent: yet another NHS scandal. BMJ. 2000;320:1285–1286. [PMC free article] [PubMed] [Google Scholar]
  • 2-5.Samuels MP, Raine J, Wright T, Alexander JA, Lockyer K, Spencer SA, et al. Continuous negative extrathoracic pressure in neonatal respiratory failure. Pediatrics. 1996;98:1154–1160. [PubMed] [Google Scholar]
BMJ. 2000 Nov 25;321(7272):1345.

Nurses undertaking the CNEP trial give their support and describe their experiences

Teresa Wright 1,2, Kate Lucking 1,2

Editor—The Griffiths panel wrote the following:

“9.3.5 The Review Panel is in no doubt that the nursing sister assigned to the project worked many more hours than she was contracted for, but she did not appear to have been provided with a protocol or system of documentation which made sure that everything was complete for all patients.

“Nursing staff and the sister in particular had not been trained or had adequate research experience for the job that they were being asked to do.

“Supervision from the researchers was inadequate, and the staff were poorly supported by trust nursing management.”3-1

As the clinical nurse specialist employed for the project, one of us (TW) took on a support nursing sister (KL) shortly after the trial had started. KL covered periods of leave and helped in the running of the trial while also working as a sister on the neonatal unit. She then took on the role of clinical nurse specialist on a full time basis to supervise the last six months of the trial. Thus we were the two senior nurses running the project in Stoke.

We strongly refute the suggestion that there was an inadequate protocol or means of documenting all of the material relating to each of the patients in the trial. We kept each patient's data in a dedicated folder, and we understand that all of this material is available at the trust for relevant staff, the quality of our work being there to see.

The panel did not ask to see either of our curricula vitae, which would have shown the previous research experience of TW, including involvement in West Midland's perinatal audit, a small part in a research study comparing two preterm formula milks, and the development and organisation of a small randomised controlled trial of continuous negative extrathoracic pressure (CNEP) in chronic lung disease of prematurity.

It is also a matter of record that the panel interviewed only one of us (TW).

It was intimated in the Griffiths report that we did not receive support from Professor Southall during the period of the trial. Although Professor Southall was based in London for a large part of this time, we always found him to be available to talk to and extremely approachable if we had any queries or concerns, however minor, about the trial. He provided us with continuous support, advice, and reassurance, being available seven days a week day and night on a rotational basis with Dr Martin Samuels, who was also extremely supportive.

References

BMJ. 2000 Nov 25;321(7272):1345.

It is time to learn the lessons from Stoke

David Hall 1

Editor—The careful and scholarly analysis of the Griffiths report by Hey and Chalmers is most welcome.4-1 The multiple inquiries into the research and the child protection work of Professor Southall and his colleagues at the North Staffordshire Hospital have unavoidably been very damaging   —   not only to the doctors and other professionals caught up in this but to the clinical investigation and management of child abuse,4-2 the readiness of academics to embark on innovative and therefore high risk research, and the reputation of the NHS for impartial handling and investigation of complaints.

It is easy with the wisdom of hindsight to be critical of government, the management of the trust concerned, and the panels of investigators, but the serious allegations which had been made could not be ignored, and doubtless everyone involved acted in good faith. Unfortunately, those ordering and carrying out the investigations lacked adequate guidelines to inform the proceedings, and this must now be rectified urgently. Unless inquires are conducted with the same rigour as the research that they purport to investigate, their findings will always be subject to criticism, but to do this adequately will need substantial funding. Furthermore, such inquiries will be subject to the same variability of opinion as the refereeing of research papers, so panels must be chosen with great care. No research team is perfect, and any clinical or academic unit subjected to the scrutiny which Professor Southall has undergone4-3 would surely reveal imperfections, errors, and omissions. But I doubt whether many of us would come out of it as well as he has, especially some 10 years after the events that were the subject of the original complaint.

The original charges against Professor Southall were to do with his work in the child protection field. One disturbing result of the whole saga is an increasing reluctance on the part of paediatricians to take part in child protection work. This has always been a difficult area of practice because clinical decisions can easily be challenged, uncertainty is the rule rather than the exception, and giving evidence in court is often intimidating. As a result of the attacks on Professor Southall's work on how parents fabricate illness, this reluctance has escalated. The complaints procedure must now be revisited and restructured in cases of child abuse.

If the lessons are learnt and acted on swiftly, this unhappy series of events may yet have some benefits for research, clinical practice, and children.

References

  • 4-1.Hey E, Chalmers I. Investigating allegations of research misconduct: the vital need for due process (with commentary by R Griffiths, TE Stacey, J Struthers) BMJ. 2000;321:752–756. doi: 10.1136/bmj.321.7263.752. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4-2.Marcovitch H. Diagnose and be damned. BMJ. 1999;319:1376. doi: 10.1136/bmj.319.7221.1376. . (20 November.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4-3.Royal College of Paediatrics and Child Health. www.rcpch.ac.uk; accessed 15 November 2000.
BMJ. 2000 Nov 25;321(7272):1345.

Nothing but the truth must emerge from these investigations

Penny Mellor 1

Editor—Perhaps Professor Southall would like to explain how he had copies of the consent forms he sent to Hey and Chalmers5-1 and whether he had parental consent to release part of a child's medical records to a third party.

Professor Southall states on a public website (www.baspcan.org.uk) that an investigation into his child protection work at North Staffordshire Hospital was undertaken because of two ridiculous statements attributed to me. In fact, I went to North Staffordshire with a parent and presented documentation to support allegations; the parent accompanying me laid several of her own complaints about the way in which her case had been handled (she has now been fully vindicated and cleared). A preliminary NHS inquiry found the concerns so serious that two independent expert panels were convened to take a further and more in-depth look at Professor Southall's work.

The Griffiths inquiry was set up to look at the research undertaken at North Staffordshire. Covert video surveillance was research; it has been published as research and therefore did fall within the remit of the inquiry. Ms Nash and Dr Harrison visited my home just as the inquiry was being set up. I presented them with a police statement written by Professor Southall in 1992 in relation to such a case. The content of this police statement showed a total disregard for the health and safety of a child by the doctors viewing the events. It was also pointed out that while this had happened, there had been absolutely no approval from either Staffordshire police or social services because of their concerns for the welfare of the child.

Professor Southall's suggestion that those campaigning about the continuous negative extrathoracic pressure (CNEP) trial are trying to sabotage his child protection work is outrageous and shows a distinct lack of compassion for the parents who have lost their children during a clinical trial. I am appalled that anyone would stoop this low, let alone somebody who claims his life's work is for children.

I stand by every word that I have ever had published about Professor Southall. There has never been a concerted effort to stop the protection of children. The exact opposite is true: it is for the sake of children that I have campaigned so hard for nothing but the truth to emerge.

References

  • 5-1.Hey E, Chalmers I. Investigating allegations of research misconduct: the vital need for due process (with commentary by R Griffiths, TE Stacey, J Struthers) BMJ. 2000;321:752–756. doi: 10.1136/bmj.321.7263.752. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Nov 25;321(7272):1345.

Some questions still need answers

Brian Morgan 1

Editor—Hey and Chalmers's clarification of their position on the continuous negative extrathoracic pressure (CNEP) trial in their rapid response to their article is welcome.6-1,6-2 They say: “We have not concluded that there were no irregularities in the way the CNEP study was undertaken.”6-1

However, there was another clarification which Chalmers could have offered, indeed might have thought helpful to declare as a conflict of interest, even have drawn to the attention of the Medical Defence Union when reviewing this sad business, which is that international collaborators with the Cochrane Centre in Oxford, of which he is director, will very likely review the CNEP study as part of its ongoing work   —   unless they decide to exclude the study altogether.

For example, a recent Cochrane Library review of continuous distending pressure for respiratory distress syndrome in preterm infants omitted the Queen Charlotte's-North Staffordshire CNEP study, despite a report having been available since 1996, saying that this study was “awaiting assessment.”6-3 Hey and Chalmers's lively criticism of the Griffiths inquiry will I am sure not influence the neonatal editorial group responsible for this aspect of Cochrane reviews. I hope it doesn't prejudice them against CNEP.

I am, however, confused by another electronic response in which Rowlands says: “I became interested in the human rights and welfare of the NHS aspects of this issue purely as a reader of the Independent whose coverage has been particularly deplorable   —   purely sensational and not at all concerned with the damage it has wrought.”6-4

Hey and Chalmers, however, rely on unattributed hearsay evidence from the health editor of the same newspaper: “Jeremy Laurance's article in the 22 September issue of the Independent states that ‘Health department officials are known to have had strong reservations about the quality of the (Griffiths) report.’”6-1 Is the Independent “deplorable,” “sensational,” “not at all concerned with the damage it has wrought,” or is it instead a reliable source of information?

And if I might add, the 18 000 word rebuttal to the Griffiths report, written by Professor Southall, was withdrawn by BASPCAN (the British Association for the Study of Child Abuse and Neglect) from its website (www.baspcan.org.uk) for a time after representations. Another (3500 word) dossier, written by Professor Southall and formerly lodged with the BMA,6-5 was also removed, not yet to be restored. The rebuttal has been reinstated, but given a statement from BASPCAN that such postings are time limited and that the Charity Commission is looking into an alleged misuse of the BASPCAN site, there is at least the possiblity that the rebuttal may yet again be removed.

Should, or would, the removal of these documents also lead to the withdrawal of articles which rely on them as sources?6-2,6-5

Footnotes

Mr Morgan as a freelance journalist may earn income from writing or broadcasting about these events.

References

  • 6-1.Hey E, Chalmers I. The Griffiths inquiry into research in Stoke on Trent failed to use due process. Electronic response to: Investigating allegations of research misconduct. bmj.com 2000;321 (www.bmj.com/cgi/eletters/321/7263/752#EL58; accessed 15 November 2000).
  • 6-2.Hey E, Chalmers I. Investigating allegations of research misconduct: the vital need for due process (with commentary by R Griffiths, TE Stacey, J Struthers) BMJ. 2000;321:752–756. doi: 10.1136/bmj.321.7263.752. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6-3.Ho JJ, Subramaniam P, Henderson-Smart DJ, Davis PG. Cochrane Library. Issue 4. Oxford: Update Software; 2000. Continuous distending pressure for respiratory distress syndrome in preterm infants (Cochrane review) [DOI] [PubMed] [Google Scholar]
  • 6-4.Rowlands I. Competence of the Griffiths review. Electronic response to: Investigating allegations of research misconduct. bmj.com 2000;321 (www.bmj.com/cgi/eletters/321/7263/752#EL38; accessed 15 November 2000).
  • 6-5.Marcovitch H. Diagnose and be damned. BMJ. 1999;319:1376. doi: 10.1136/bmj.319.7221.1376. . (20 November.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Nov 25;321(7272):1345.

Response to concerns raised by Mr and Mrs Henshall

Edmund Hey 1

Editor—Mr and Mrs Henshall voice important concerns in their rapid response and the letter above.7-1 Professor Southall is genuinely distressed that NHS managers did not let him discuss these directly with the family when they were first raised and equally frustrated that legal advice restrains him from responding personally to this correspondence even now. Because the family deserve some answers I will address as many of these concerns as my knowledge of the documents allows.

They say that the 10th baby recruited to the study was “strangled to death,” but this is to misunderstand the autopsy information. There was trauma to the skin round the neck, which triggered increased nursing vigilance and further (very effective) improvements in the design of the neck seal, but no evidence that this contributed to the baby's death. Rather the reverse. Skin blood flow was so poor, because the baby was already dying, that a pressure sore developed.

Clear rules for monitoring the progress of this study were in place and in the hands of an independent statistical adviser, who would have stopped the study at once had any evidence of excess harm come to light. Such rules have increasingly become a component of all good trials in the past few years, but the continuous negative extrathoracic pressure (CNEP) trial was ahead of its time in having these in place in 1989.

The scoring system designed to make this possible has been widely misunderstood. It was only possible for a baby who died to be awarded a better (higher) score than one who survived intact by treating those who survived with severe lung damage (the main problem the trial was trying to address) as “intact” survivors. Scores were not the only way progress was monitored. Nor were they used to evaluate the final outcome. An audit of all deaths, and all survivors with cerebral untrasound abnormality, was also maintained. Perhaps a belief that everyone would assume this was undertaken explains its omission from the final trial report.

The Henshalls rightly question the statement in the parents' information sheet about the technique having been “shown to be safe and effective” and wonder why, if that were true, further studies were undertaken in parallel with the CNEP trial to look at what such treatment did to cerebral blood flow. All ethics committees require evidence of safety, but, in the nature of things, such statements have to be more provisional than is generally acknowledged. As the recent spotlight on what the last chief medical officer had to say about British beef reminds us, all that can ever be said is that there is, as yet, no evidence that something is unsafe.

Early work using CNEP had been promising,7-2 and no evidence had emerged to suggest such treatment was unsafe. The CNEP trial was undertaken specifically to confirm that it was indeed safe when offered to very small babies and to determine whether the benefits outweighed any possible hazard. The advent of infra red technology while the trial was in progress made it perfectly proper to do further studies to confirm that cerebral blood flow was unaffected. Indeed, it would have been unethical not to do this.

The Henshalls say that they have evidence that the CNEP trial did not run as it was designed to and that protocols were ignored. The Stoke clinicians deny this, and Dr Chalmers and I uncovered no significant protocol violations during our examination of the records. We would accept, however, that the trust never gave us access to any of the clinical case notes. Since these concerns were first raised four years ago, it is clearly time that they were addressed publically. The families concerned deserve this, and so do the clinicians in Stoke. The editor of this journal clearly concurs.7-3

The family question the timing of peer review. The study was peer reviewed by experts in the National Heart and Chest Hospitals before it opened in London in October 1989. In the context of my review of the Griffiths inquiry I merely referred to the additional review it underwent before it opened in Stoke in April 1990.

The family are right to emphasise that the nurses were concerned from the outset that handling and frequent opening of the chamber portholes could be detrimental. The trial was designed to try to test this but was only able to look at outcome to discharge and not long term outcome for funding reasons. The Henshalls rightly emphasise that there is only a limited correlation between the appearance of the brain on early ultrasonography and long term (⩾2 years) prognosis and that all major neonatal trials should address long term prognosis. Even the published trials funded by the Medical Research Council in the United Kingdom have not, as yet, done this. It is time public pressure made all funding bodies do this because some treatments of short term value may, in the long term, do more harm than good.7-4 It remains scandalously difficult to get money to research this.

Finally, although the Henshalls's own child is tragically disabled, it remains to be shown that this is a consequence of her CNEP care rather than her preterm birth. While they can name individuals in this regard, they also know that the General Medical Council bars any doctor from making clinical information public by way of rebuttal without their consent.

References

  • 7-1.Henshall C, Henshall D. Investigating allegations of research misconduct. Electronic response to: Investigating allegations of research misconduct. bmj.com 2000;321 (www.bmj.com/cgi/eletters/321/7263/752#EL1; accessed 15 November).
  • 7-2.Samuels MP, Southall DP. Negative extrathoracic pressure in treatment of respiratory failure in infants and young children. BMJ. 1989;299:1253–1257. doi: 10.1136/bmj.299.6710.1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7-3.Smith R. Inquiring into inquiries. BMJ. 2000;321:715–716. doi: 10.1136/bmj.321.7263.715. . (23 September.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7-4.Tarnow-Mordi W, Mitra A. Postnatal dexamethasone in preterm infants. BMJ. 1999;319:1385–1386. doi: 10.1136/bmj.319.7222.1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Nov 25;321(7272):1345.

Authors' reply

Edmund Hey 1, Iain Chalmers 1

Editor—Our article presented the results of our investigation into the report of the Griffiths inquiry and showed that it had not been well conducted. Had it been, the inquiry report would not contain so many factual errors. Others, apparently including civil servants, share our concerns: on 22 September the Independent reported, “Health department officials were known to have strong reservations about the quality of the report.”8-1

It is disingenuous of the panel members8-2 and ministers8-3 to ignore our criticisms of the conduct of the inquiry by trying to divert attention instead to the series of recommendations made about research governance. These were not a focus of our article, although we could perhaps have emphasised more strongly that national recommendations should not be based on an incompetent inquiry into just two studies conducted nearly 10 years ago in one NHS hospital.

One of us (EH) has responded separately to the issues raised about the continuous negative extrathoracic pressure (CNEP) study by Mr and Mrs Henshall (above). Although it was not the purpose of our investigation to judge the quality of this study, we stand by our suggestion that its general conduct was “exemplary.” However, without knowing what specific criticisms some families raised with the panel or examining relevant papers to which we have not had access, we could not possibly conclude that there were no irregularities. We have seen some documents that others have not seen; others have seen documents we have not seen. What remains unacceptable is that the Griffiths panel apparently never sought or examined many key documents. To insinuate now that these were “owned” by Professor Southall (rather than by the whole study team) and state that “he did not provide” these documents to them8-2 is a further distortion: the panel's own records make clear that they knew that NHS managers had told Professor Southall not to hand over the crucially important study log book.

Like us, the clinicians in Stoke still do not know, except in the most general terms, what concerns parents have raised about the conduct of the CNEP study. People cannot be expected to defend themselves, or provide relevant documentation, until they know what they have been accused of. Many of the factual errors in the Griffiths report could have been avoided had the panel been more specific about the allegations, examined relevant documentation, and invited those they criticised to comment on a draft of the report.

It is easy to understand the frustration of families who still feel they have not been told what went on in the CNEP study, four years after some of them raised concerns about its conduct. Many of your correspondents on bmj.com (www.bmj.com/cgi/content/full/321/7263/752) have chided the medical profession for arrogance and secrecy in these matters. However, we have seen correspondence showing that, years ago, Stoke clinicians wanted to meet the parents who first raised these concerns but were prevented from doing so by NHS managers. Perhaps it is managers who need to heed the words of the under secretary of state for health, who, in his first public comment on our article, is reported to have said “The NHS belongs to the public, who have a right to know what goes on in it.”8-4

Until there are proper ground rules for the conduct of inquiries of this sort in the NHS nobody is going to receive much justice. Some parents clearly feel that justice has still not been done to them or their children. Equally, many of the doctors and nurses in Stoke who tried to provide care for these children and to find ways of improving the care of all vulnerable preterm babies feel that their efforts have been criticised unjustly. Others, not as directly involved, are confused by partial information and continued lack of transparency. It is a serious indictment of the way the NHS is currently managed that the president of the Royal College of Paediatrics and Child Health needs to ask for such inquiries to be “conducted with the same rigour as the research they purport to investigate.”

Consensus has been sought and reached on the elements needed for the conduct and reporting of some other forms of research.8-5 The obvious inadequacies of the Griffiths inquiry and report make it abundantly clear that consensus is urgently required on how official inquiries in the NHS should be conducted and reported. Until that happens, they are unlikely to serve the interests of those using the NHS, let alone those working in the service. Many will be alarmed to learn that ministers still “do not believe that the [Griffiths] review was out of order or kilter with others which have taken place, or are taking place, within the NHS.”8-3

Finally, we confirm that Mr Morgan is right to conclude that the content of the Cochrane review he mentions is the sole responsibility of its authors and the editors of the Cochrane Neonatal Group in North America.

References

  • 8-1.Laurance J. Top doctors hit back at “witch hunt” inquiries. Independent 2000 Sept 22:10.
  • 8-2.Griffiths R, Stacey T, Struthers J. Commentary: Response from members of the Griffiths inquiry. BMJ. 2000;321:755–756. . (23 September.) [PubMed] [Google Scholar]
  • 8-3.North Staffordshire NHS Trust. Oral questions and answers. Hansard. 10 October 2000. Column 150.
  • 8-4.Blackhurst D. Southall criticised by CNEP report author. Sentinel 2000 23 Sep:9.
  • 8-5.Begg C, Cho M, Eastwood C, Horton R, Moher D, Olkin I, et al. Improving the quality of reporting of randomised controlled trials. The CONSORT statement. JAMA. 1996;276:637–639. doi: 10.1001/jama.276.8.637. [DOI] [PubMed] [Google Scholar]

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES