Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
When you are in a hole stop digging
- John A M Hemming (2 December 2006)
-
Was the doctor really an expert witness?
- Raymond G Holder (2 December 2006)
-
Facts and opinions
- david leopold (4 December 2006)
-
Author's reply
- Jonathan Gornall (4 December 2006)
-
Re: Was the doctor really an expert witness?
- Jayaprakash Ayillath Gosalakkal (4 December 2006)
-
Problems estimating the proportion of repeat sudden infant deaths that are unnatural
- Martin P Samuels (7 December 2006)
-
Jonathan Gornall is to be congratulated
- A Nigel P Speight (8 December 2006)
-
Increased suspicion
- Hugh Robertson-Ritchie (8 December 2006)
-
The Answer to Crib Death
- Michael D Innis (8 December 2006)
-
Why ?
- Richard G Wilson (8 December 2006)
-
A significant omission
- Richard Webster (9 December 2006)
-
SIDS and the law
- Catherine Williams (11 December 2006)
-
Re:, Was message of sudden infant death study misleading? BMJ 2006, 2 Dec.
- Robert G Carpenter, Alison Waite, Robert Coombs, Charlotte Daman-Willems, Angela McKenzie, Jonne Huber (12 December 2006)
-
Vaccine deaths not considered amongst 'judge and jury' researchers
- JohnP Heptonstall (13 December 2006)
-
Ellipsis ... To mislead or clarify?
- David JR Hutchon (13 December 2006)
-
The standard of proof matters
- John A M Hemming MP (13 December 2006)
-
Re: Why ?
- Penny Mellor (14 December 2006)
-
Serious concerns
- Brian Morgan (15 December 2006)
-
Should these articles have been published?
- John Stone (15 December 2006)
-
Exactly who should get it right?
- Hilary Butler (15 December 2006)
-
Author's reply
- Jonathan Gornall (18 December 2006)
-
Re: Problems estimating the proportion of repeat sudden infant deaths that are unnatural
- Alison J Waite, Robert Carpenter, Robert Coombs, Charlotte Daman-Willems and Angela McKenzie (19 December 2006)
-
Silence in court
- Penny Mellor (19 December 2006)
-
False Positives Do Matter
- John A M Hemming MP (19 December 2006)
-
The Emperors have no clothes
- Robert Wheatley (25 December 2006)
-
If only...
- Brian Morgan (25 December 2006)
-
SIDS is a residual diagnosis - sometimes
- John A M Hemming MP (26 December 2006)
-
Convicted by suspicion, insinuation, and supposition
- Charles Pragnell (27 January 2007)
-
The presumption of murder
- Richard Webster (7 February 2007)
-
Re: The presumption of murder - author's reply
- Jonathan Gornall (9 February 2007)
-
Presumption of Murder: From What
- L. Travis Haws (10 February 2007)
-
Re: Re: The presumption of murder - author's reply
- Michael D Innis (10 February 2007)
-
Suspicion or vigilance?
- Richard Webster (26 February 2007)
|
|
|||
|
John A M Hemming, Member of Parliament SW1 0AA
Send response to journal:
|
Jonathan Gornall seems to be arguing that it is reasonable to convict people of murder if 60% of children who are dying are dying from natural causes, 27% from indeterminate causes and 13% from unnatural causes. He accepts that it is unreasonable to convict people of murder if 87% of deaths are from natural causes. The fact is that "Meadow's law" was always rubbish. Too many people have been convicted and or/had their children taken away from them on the basis of expensive 4th rate science dressed up as medical opinion. SIDS has always been known for years as having a genetic link. Recent work in Boston [1] has come closer to identifying the differences that result in a pre-disposition to SIDS. Research in China [2] has demonstrated how mice develop a tolerance to hypoxia through exposure to hypoxic conditions. The Boston study strengthens the hypothesis that medullary 5-HT dysfunction is associated with SIDS and may lead to death by a failure of respiratory and autonomic responses to homeostatic stressors during sleep. It may be that this route does give a better mechanism to distinguish between SIDS and smothering. I would not wish, however, to use this mechanism alone as a basis for prosecution or child protection proceedings. The Boston study also hypotheses that that sudden death results when 3 factors impinge on the infant simultaneously: (1) an underlying vulnerability; (2) an exogenous stressor (eg, prone sleep position, bed sharing); and (3) the critical developmental period, ie, the first 6 months of postnatal life, when the infant is at greatest risk for SIDS. That is also a good argument, taking into account the Chinese research, for not exposing infants vulnerable to SIDS to hypoxic conditions. In the mean time Mr Gornall should throw away his spade. [1] Multiple Serotonergic Brainstem Abnormalities in Sudden Infant Death Syndrome http://jama.ama-assn.org/cgi/content/abstract/296/17/2124 [2] Acute Adaptation of Mice to Hypoxic Hypoxia Guo-Wei Lua, Ding-Wu Dingb, Mei-Tang Shia Capital University of Medical Sciences, Beijing, China Biological Signals and Receptors 1999;8:247-255 (DOI: 10.1159/000014594) Competing interests: Chairman - Justice for Families |
|||
|
|
|||
|
Raymond G Holder, retiredengineer BH9 3NF
Send response to journal:
|
I followed the Cot Death court cases with great interest, as I saw a possible connection with one of my own problems, not generally recognised by UK medicine. Dr Meadows based his case on a sort of "probality squared" argument, with no apparent knowledge of scientific developments in this area. I had earlier been reading Fernando Scaglia's paper on Carnitine Deficiency, as I suffer from this, he is a a paediatrician in USA, and relates how genetic faults can cause lack of carnitine in the newborn, occuring at of the order of 1/40,000 births, possibly resulting in Sudden Infant Death Syndrome, as fat cannot be metabolised to provide energy for heart operation, with consequent failure unless urgent supplementation is given. He details a long list of lab etc tests to be carried out to establish the existence of the condition. The Polio Network of W Australia have also found that a carnitine deficient Post Polio mother, possibly after a difficult pregnancy, can give birth to a carnitine deficient daughter, whose offspring may also be carnitine deficient. Supplementary carnitine during pregnancy may alleviate this problem. The doctor had obviously never heard of this problem, and from snippets of news on SIDS, it seems to be not at all well known. I Emailed the defence lawyers for Trudi Pattel with the reference when her trial was in progress on a similar charge. It raises the question, who is to justify a witness being declared "expert", as time goes on, the most expert among us in all walks of life can become experts in outdated technology, as I know only too well from my own experience. Competing interests: None declared |
|||
|
|
|||
|
david leopold, Physician Swansea sa34je
Send response to journal:
|
Dear Sir. Jonathan Gornall has surely rendered a service to science in the cause of Child Protection in so painstakingly and neutrally presenting some little-known facts about this fatefully influencial study. It would appear that the import of the study should have been entirely converse. Who can say whether some prosecutions which have failed might have been proven? And there must surely be some cases where prosecutions which have been abandoned, might now be reopened. This is medical Journalism at its very best. Yours faithfully
Competing interests: None declared |
|||
|
|
|||
|
Jonathan Gornall, Freelance journalist London E1
Send response to journal:
|
What exactly is the basis for Mr Hemming’s apparent characterisation of Professor Roy Meadow as a purveyor of “4th rate science dressed up as medical opinion”? Those who continue to campaign against Professor Meadow and the very existence of MSBP/factitious or induced illness have had great success in trading up the issue of the Sally Clark statistic into a widespread belief that Professor Meadow was wrong on all fronts, that MSBP has never existed and that every case with which Professor Meadow was involved was, by default, flawed. I am prepared to be corrected but, to the best of my knowledge, none of this has ever been shown to be true, despite the wishful thinking of various advocates, defence lawyers and their clients and much widely disseminated disinformation to the contrary. For his part, Mr Hemming seems to be unaware that Professor Meadow was found guilty by the GMC only of misusing (unintentionally) one statistic in one trial, and that both the High Court and the Court of Appeal have ruled that this was an honest mistake that did not amount to serious professional misconduct [1]. I would draw the attention of the MP and his fellow campaigners to a report published at the weekend [2] of a court judgment handed down on Friday – coincidentally, the same day as my article appeared in the BMJ. It is possible that they missed it, as it appeared only in the Guardian. It began: “The paediatrician Sir Roy Meadow was yesterday vindicated by a high court judge over his evidence in a care case six years ago, when he told the court a four-month-old boy probably died because he was intentionally smothered by his mother. In a 70-page judgment following an exhaustive and highly unusual review of the issues surrounding P's death in January 1999, Mr Justice McFarlane said: ‘I am driven to the firm conclusion that no criticism of Professor Meadow's role in this case can be sustained’.” The child’s parents had won the right to a review of their case on the ground that Professor Meadow had been involved in the original hearing. The judge, however, rejected their criticism of Professor Meadow’s evidence. “Indeed,” he added, “the passage of time and the exhaustive additional investigations have proved that, on the medical issues that were before the court in 2000, he was correct.” Hardly a judgment that chimes with Mr Hemming’s assessment of Professor Meadow’s work as “4th rate science dressed up as medical opinion”. 1 GMC and Professor Sir Roy Meadow with Her Majesty’s Attorney General, Case No: CO/5763/2005, Court of Appeal (Civil Division), October 10, 2006. 2 Dyer, C. Paediatrician in baby death case vindicated after six years. Guardian, December 2, 2006. Competing interests: Author of article |
|||
|
|
|||
|
Jayaprakash Ayillath Gosalakkal, Consultant Paediatric neurologist Le1 5WW
Send response to journal:
|
Carnitine defeciency is indeed a real condition and there are many causes for it. Screening for such conditions are often not done due to lack of awareness or costs. I was reminded of the nature of expertise when I read the Judges opinion on the diagnosis and treatment of epilepsy by experts He said
"The Assessors considered three standards that were possibly applicable to
Dr Holton's work, namely that of a consultant paediatrician with a special
interest in neurology; that of a consultant paediatric neurologist; and
that of a consult paediatrician. They rejected the first of these
standards on the basis that Dr Holton had never trained as a paediatric
neurologist or as a consultant paediatrician with an interest in
neurology. They added:
I think this should give us all pause to consider what a modern medical system should be doing? Should people be automatically considered experts because of the position they are holding? I personally feel nobody should be considered a sub specialist unless they have undertaken an accredited training program or equivalent in that speciality. I have always believed in the dictum ,don’t trust something based on “who is saying it rather than what is being said”. The recent reports on Prof Emery (2) just reinforces my belief (1) MR JUSTICE STANLEY BURNTON Neutral Citation Number: [2006] EWHC 2960 (Admin) Case No: CO/1764/2006IN THE HIGH COURT OF JUSTICE QUEEN'S BENCH DIVISION ADMINISTRATIVE COURT (2)Jonathan Gornall Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168 (2 December) Competing interests: None declared |
|||
|
|
|||
|
Martin P Samuels, Consultant Paediatrician University Hospital of North Staffordshire
Send response to journal:
|
In his critique of the paper by Carpenter and colleagues (1), Jonathan Gornall has shown how the categorisation of second sudden unexpected infant deaths was changed with the result that the data from the Care of Next Infant (CONI) programme produced a ratio of natural:unnatural deaths of initially 1.5:1 (2, 3) and later 6.7:1 (4). This increase in the likelihood of second sudden infant deaths being natural would be welcomed if it was based on new information about the deaths. Unfortunately, there were 13 cases with inadequate information and placing these into the category of natural deaths, as the authors did in their more recent analysis, is as mistaken as placing them into a category of unnatural death. Amongst these 13 cases, there was one with rib and clavicular fractures at post-mortem, one where a mother was murdered, 2 with extensive legal investigations, and 4 where older children were taken into care. It would have been more appropriate for Carpenter and colleagues to have divided these cases as 27 natural, 6 unnatural and 13 undetermined. If this latter proportion were all allocated to either natural or unnatural deaths, the proportion of unnatural deaths varies between 13% (6/46) and 41% (19/46). This range provides a more objective estimate of the uncertainty surrounding the categorisation of death, rather than the analysis by Carpenter and colleagues, which involves a dichotomous categorisation of infant deaths, with probably erroneous perceptions about the likelihood of abuse in a second infant (1). However, even this range is probably inaccurate given that an unknown proportion of the 18 deaths categorised as SIDS may have been covert homicides, as there is no definitive test to distinguish between natural and unnatural mechanisms (5). Jonathan Gornall omitted the other major problem with the Carpenter study; it had concluded with the interpretation that most repeated unexpected infant deaths were probably natural, even though their study was not a population based study. This would be acceptable if they knew they had good case ascertainment of all subsequent siblings, but the CONI programme is voluntary and families are not tracked after a sudden infant death to know whether they are referred to the CONI programme. Even when referred, the authors stated that only 4% of subsequent siblings were missed (6). Our data from the CONI programme run in North Staffordshire between 2003 and 2005 show 16% (11/70) of families known to have previous SIDS and referred to CONI do not take up the scheme, a fourfold difference. We have no idea of rates of unnatural deaths in those families that do not get referred or take up CONI. This further impairs any attempt to identify a true population estimate for the ratio of natural versus unnatural death in repeat siblings. References 1. Gornall J. Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168. 2. Foundation for the Study of Infant Deaths. Report on 5000 babies using CONI (care of next infant programme). London: FSID, 1998. 3. Emery JL, Waite AJ. These deaths must be prevented without victimising parents. BMJ 2000;320:310. 4. Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35. 5. Di Maio VJM. Repeat sudden unexpected infant deaths. Lancet 2005;365:1137-1138. 6. Carpenter RG, Waite A, R C Coombs RC, Damen-Willems C, Huber J. Repeat sudden unexpected infant deaths. Lancet 2005;365:1138. Competing interests: None declared |
|||
|
|
|||
|
A Nigel P Speight, Consultant Paediatrician University Hospital of North Durham DH1 5TW
Send response to journal:
|
It is alarming that the inherently biased methods in the Lancet article were not detected by the Lancet reviewers, and it is a reproach to our profession that it has taken a freelance journalist to expose this. Jonathan Gornall has performed an important public service in this difficult area and it is sad to have to point out that in the current climate this has actually required a degree of courage. The recent reporting of the GMC case against Professor David Southall has been utterly biased in terms of the tabloids and not much better from the broadsheets and the BMJ. Professor Southall has given evidence in his defence for six days last week and it has been virtually unreported. The current climate of one-sided reporting is creating a syndrome which could be called "Abuse Denial" in which we are in danger of turning the clock back to the days before Maria Colwell, and our painstakingly constructed machinery for protecting children is in danger of being rendered impotent. Long ago Henry Kempe stated that the first stage of development that every individual, profession and society must pass through in their approach to child abuse is Denial, and his words seem sadly apt today. Unfortunately, it is easy for people who have never actually witnessed child abuse to disbelieve in its reality. The media, editors of journals and Members of Parliament bear a heavy responsibility to "get it right". Paediatricians have had to overcome their own feelings of disbelief and would be only to happy to help educate those who still have trouble believing in Child Abuse. Competing interests: None declared |
|||
|
|
|||
|
Hugh Robertson-Ritchie, general practitioner 128 Canterbury Road, Folkestone, CT20 2JT
Send response to journal:
|
John Hemming MP starts his response in a very unconvincing way, saying, “Jonathan Gornall seems to be arguing …”. Well, Jonathan Gornall does not argue what John Hemmings says; instead, he is writing about the evidence for suspicion of foul play. The statement in Gornall’s article, "... suspicion of foul play should increase with each subsequent sudden infant death within a family" [1] (presumably ‘Meadow's Law’) is vindicated in the Lancet paper by Carpenter et al. [2] even using their own interpretation of their figures. Out of the 6373 first sudden infant deaths, it seems that __none__ were considered at the time to have sufficient evidence to be classified as suspicious, whilst out of the 46 families where a second sudden infant death occurred, __13%__ were considered to be unnatural. That is an increase from 0% to 13%, a very big "increase in suspicion". [1] BMJ 2006;333:1165-1168 [2] Lancet 2005;365:29-35. Competing interests: None declared |
|||
|
|
|||
|
Michael D Innis, Retired Home
Send response to journal:
|
Editor, The Lancet article concerning the "sudden unexplained infant deaths" in 20 regions in Europe [1] included immunization/vaccination (I/V) in the risk factors but limited it to the previous 7 days in contrast to the American Vaccine Adverse Event Reporting System which extends to reactions occurring within 4 weeks of I/V. Had the Europeans used the American reporting system they may have come up with the same conclusion as Joseph G Hattersley [2]who concluded it was an adverse reaction to vaccination. "The Answer to Crib Death Sudden Infant Death Syndrome” (SIDS) Joseph G. Hattersley Abstract “(1) Two doctors on opposite sides of the globe eliminated crib death among their patient populations for 40 years using ascorbate supplementation. Unknown to each other they arrived at the same regimen. (2) Crib deaths nearly disappeared in Japan in 1975 when first inoculations were postponed until the 24th month of life. These findings and their explanation are explored. SID is traced to a nonspecific or general adaptation stress syndrome defined by Hans Selye. It is precipitated by a deficiency of ascorbate and also of vitamin B6 and zinc.” So said Hattersley. Dr Archie Kalokerinos [3] and Dr Frederick R. Klenner [4] are the "two doctors on opposite sides of the globe" mentioned by Hattersley. I would have added the name of Professor Alan Clemetson [5] to the pioneers using ascorbate supplementation to eliminate Sudden Infant Death Syndrome in their practise Michael Innis Reference: 1. Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35 2.The Answer to Crib Death (yahoo) 3. Kalokerinos A Every Second Child Keats Publishing Inc New Canaan Connecticut 1981 4. Klenner, F.R. (1971), “Observations on the dose and administration of ascorbic acid when employed beyond the range of a vitamin in human pathology,” J. App. Nutr. 111: 210-214 5. Clemetson CAB Vaccinations, Inoculations and Ascorbic Acid. J Ortho Mol Med(1999) vol 14 No 3 137 – 142 Competing interests: I have given evidence for the Defense in the UK, USA and Australia and I have been paid for my services. |
|||
|
|
|||
|
Richard G Wilson, Honorary Consultant Paediatrician Kingston NHST. KT2 7QB
Send response to journal:
|
Mr. Gornall seems at first to be just another journalist trying to make his name by fabricating a scandal where none exists. He presents no new facts but a series of opinions , suggestions and insinuations. He edits the comments from those he criticises. The paper by Prof. Carpenter was clear, honest , peer reviewed and debated at the time.This and Dr. Stanton's paper provide facts which show that the DiMaios and Prof. Meadow were wrong. The only "serious concern" is about the origin and purpose of what is stated to be "an investigation by the BMJ ". His reply shows that he confuses the subject of Fabricated or Induced Illness by Carers with the problem of unexpected death in infancy . There is a small overlap but they are quite different issues. It also suggests that he was heavily briefed by unnamed individuals who in the normal open way would have been identified as co-authors. These two serious topics should be addressed by a reasoned and free debate not by tabloid journalism. (This is a personal view and not made on behalf of any organisation with which I am associated). Competing interests: Chair RCPCH Working Party on FII 2002. Trustee F.S.I.D. |
|||
|
|
|||
|
Richard Webster, Author Oxford OX2 6TX
Send response to journal:
|
Jonathan Gornall’s article raises a number of important questions about the 2005 Lancet paper on recurrent sudden infant deaths which is its main focus. On the evidence he presents it would seem that the paper’s conclusion - that 87% of the unexpected deaths of a second child were due to natural causes - is overstated, and that the paper’s authors have presented as clear-cut a situation where many uncertainties remain. Since any medical paper should categorise and report its results with the utmost scrupulousness, the questions which Gornall raises are important and his article is helpful in a number of respects. The central problem with his article, however, is that the conclusions he appears to draw from his investigation simply do not follow from the evidence he presents in it. It is difficult to read what he has written without forming the impression that Gornall believes the appeals of Angela Cannings and Donna Anthony, in both of which the Lancet paper was invoked by the judges, should not have been granted. Yet even if we accept that John Emery, the pathologist who initiated the Lancet study but who died before it was submitted for publication, would have disagreed with all the statistical revisions Gornall documents, the position which Emery outlined before his death provides significant grounds for rejecting as unsafe the convictions not only of Angela Cannings and Donna Anthony, but also of Sally Clark. Gornall himself quotes – or partly quotes – the key passages from the report which Emery produced for Mrs Clark’s defence team before her trial in November 1999, six months before his death. These passages were, as Gornall notes, included in the judgment given in Mrs Clark’s first, unsuccessful appeal in October 2000. Emery wrote as follows: “The occurrence of repeat unexpected deaths is a topic on which I have been doing research over the last few years and have been able to do a confidential study involving approximately 100 such deaths. “These have all been families who have presented as having two ‘cot deaths’. My findings have been that approximately a third of the deaths have been due to a whole series of rare natural causes which had been missed by those doing the necropsy. A third of the deaths were associated with different forms of child abuse and having a whole number of features which indicated that they were unnatural deaths. In the final third no suspicion of unnatural death was found and no natural cause was found and these children constitute what may be termed at the moment as ‘true idiopathic unexpected child deaths’.” To this quotation, Gornall appends his own brief note in which he suggests that “His [Emery’s] conclusion was stark.” He then gives this conclusion as a further quotation from Emery: “The occurrence of two unexpected deaths in a family thus raises a definite suspicion of unnatural death which in my experience is confirmed ... in a third of such cases.” Gornall goes on to imply that Emery’s stated position actually supports, or is compatible with, ‘Meadow’s law’, namely that “One sudden infant death is a tragedy, two is suspicious and three is murder, unless proven otherwise.” The first point which might be made here is that the figures Emery himself gives would not seem to justify any conclusion which is ‘stark’ in the sense that Gornall intends. What Emery’s figures suggest is that, of the second infant deaths he studied, at least a third could be definitely categorised as natural deaths and, in that no suspicious signs were found in another third, the proportion of natural deaths might be as high as two thirds. Even if we take the smaller of these two figures, it would utterly confound Meadow's notorious 1 in 73 million calculation. It is notable that Gornall's article studiously avoids mention of this discredited piece of evidence. The second point is that it seems rather odd that Gornall’s quotation of a passage which is clearly crucial should contain an ellipsis. What does this mysterious dot-dot-dot hide? What was it that Jonathan Gornall chose to leave out? If we check back to the text of the judgment in Sally Clark’s first appeal we find, interestingly, that it was only a single short word, so economy of space can scarcely have been his motive. This is what the sentence said before Gornall edited it: “The occurrence of two unexpected deaths in a family thus raises a definite suspicion of unnatural death which in my experience is confirmed only in a third of such cases” (paragraph 116). By leaving out the word ‘only’ Gornall actually changes the emphasis of the entire quotation. Whereas the sense of the original would tend to imply that Emery had serious reservations about ‘Meadow’s law’ and felt it needed qualification at the very least, Gornall has changed the wording so that he can the more easily recruit Emery’s statement to support this law. It is indeed important that medical papers should categorise and report their results with the utmost scrupulousness. But it is also important that those who seek to criticise them should show an equal degree of scrupulousness. richardwebster@ntlworld.com - www.richardwebster.net Competing interests: None declared |
|||
|
|
|||
|
Catherine Williams, Reader in Law, Sheffield University Law School, Sheffield University, S10 1FL
Send response to journal:
|
In his important article Jonathan Gornall asks, was the message of the sudden infant death study by Carpenter et al misleading? If it was this has implications for those working in the field of child protection, including lawyers acting for the various parties and the courts. Experts are expected to provide well researched, unbiased evidence to a court.[1] If experts are criticised for presenting misleading evidence to the courts, quite apart from the damage this does to their personal reputation, this has the knock on effect of damaging the work being done to protect abused children. In investigating possible child abuse and mounting care proceedings, although understandably parents may view this as an accusatory system with a focus on their ‘guilt’, the purpose is not to establish ‘who did it’ but rather to protect the child. This is quite different from the criminal law system where, as Jonathan Gornall rightly points out, guilt must be proved beyond reasonable doubt. Carpenter et al in their opening summary state: ‘We aimed to estimate the probability that a second infant death is natural versus unnatural.’ This, of course, is a false dichotomy. It leaves no room for the ‘it could be, it couldn’t be’ abuse scenario, the routine situation that courts find themselves in when faced with a disputed case of child abuse. The Carpenter article has been criticised particularly by Chris Bacon [2] and David Hall wrote: ‘In the 18 cases fully investigated and thought to be true cases of a second SIDS, the authors stated that all the families had a ‘high frequency of SIDS risk factors’ and in this series of repeat deaths the authors did not describe any unequivocal example of two SIDS in one low risk family.’[3] Thus it would appear that the conclusion of Carpenter et al that second SIDS deaths are not rare and that the majority, 80-90%, are ‘natural’ simply cannot be supported. Jonathan Gornall has added further confirmation for viewing the article with considerable caution. In a contested child abuse hearing it is not the job of doctors to decide whether or not a child has been abused, it is for the court to decide. Courts should be given the whole picture, but doctors supply only part of the jigsaw. What a doctor must do is to present his or her findings, with some indication, where possible, as to the likelihood or otherwise of it being abuse. So, for example, the court should be told that there have been previous Apparent Life Threatening Events (ALTEs), a child has current or old fractures, blood has been found in the mouth or the child has had nosebleeds, etc. These factors are not diagnostic of child abuse or that a child has been killed, but they are relevant factors for the court to consider when deciding whether or not a child is at risk of abuse. Saying that SIDS has occurred is not a diagnosis of the cause of death, but a failure to diagnose the cause. There is a cause for the death, but it is not known. It will either be medical or non-accidental. That this is so is amply demonstrated by the Carpenter et al article itself, as the authors comment on two of the deaths originally classified as SIDS. Subsequently one of those deaths was found to be due to Long QT syndrome and the other due to the father having murdered the child. Lawyers and courts must be able to rely on medical evidence presented to them. Just because a court makes a finding that the threshold for making a care order has been passed this does not mean that a care order has to be made. There are a host of options open to a court ranging from making a care order and removing the child, through to making no order at all, with various other possibilities in between. However, the decision as to what order to make must be based on the best evidence available to the court. If courts are misled in to thinking that two or more deaths in a family are extremely likely to be ‘natural’ this puts other children in the family at very high risk, if not of death certainly of possible abuse. It is not the prime responsibility of courts and lawyers to have regard to the reputation or well being of adults, but to consider the welfare of the child and to do what is best for that child. If medical practitioners do not present their work carefully and in an unbiased fashion mistakes will be made, and it will be children who will suffer. [1]Re J (Child Abuse: Experts Evidence) [1991] FCR 192. Also reported as Re R (A Minor)(Experts’ Evidence) [1991] 1 FLR 291. [2]Bacon C. Repeat sudden unexpected infant deaths. Lancet 2005; 365:1137. [3]Hall DMB. The future of child protection. J R Soc Med 2006; 99: 6. Catherine Williams, Reader in Law, Sheffield University Competing interests: None declared |
|||
|
|
|||
|
Robert G Carpenter, Hon. Professor, Medical Statistics Unit, London School of Hygiene & Tropical Medicine, London 19 Lansdowne Avenue, Orpington, Kent BR6 8JT, Alison Waite, Robert Coombs, Charlotte Daman-Willems, Angela McKenzie, Jonne Huber
Send response to journal:
|
Sir, We were surprised and disappointed that you did not invite a reply to Gornall’s allegations(1) relating to our paper(2) to appear with his article. We presume that you will print our response. Sudden infant deaths are always tragedies that deserve full and detailed investigation. Up until recently this has not always been the case. The CONI (Care Of Next Infant) scheme supports families who have had a previous cot death or sudden unexpected infant death. Since 1988 approximately 10,000 children have been supported by the scheme of whom some have died whilst in the programme. As far as possible we have tried to fully investigate these deaths with detailed visits to the families in their homes and careful review of the pathology followed by a case discussion. There had also been a coroner’s enquiry, (as should happen with all sudden and unexpected deaths), and sometimes a police investigation. In a few cases there were criminal prosecutions. Our report in the Lancet was a review of the first 57 deaths. We stand by our findings that of these deaths, 9 were inevitable (died of known cause) and of the 48 unexpected deaths 7 were probable homicide. Professor Emery led the CONI steering group (of which we are all members) up till his death, when Professor Huber joined our team. Professor Huber is a paediatric pathologist who was a close friend and long time colleague of Professor Emery. We are unable to answer for Professor Emery (he died in 2000) but as we were all close colleagues we have not materially changed the cause of death for any of the cases he knew about before his death. There are differences between our earlier work and that reported in the Lancet. In the latter, we reported but excluded from the analysis the sudden deaths over one year of age and have clarified “non natural” deaths. In previous unpublished reports to FSID, The Foundation for the Study of Infant Deaths, we had placed in the “non natural” category four infants who died in bed with their parents as they were possibly accidental and avoidable. In our Lancet paper(2) we classified these cases as SIDS, in line with the CESDI SUDI study(3), as stated in the text. Finally the non-natural group included two cases for which an open verdict had been recorded. The histories of these and other cases for which an open verdict was recorded led us to classify them on a case by case basis which we described in the paper. These revisions account for the differences between our Lancet publication 7/57 (probable homicide) and our earlier report to FSID 14/44 (non natural). We did not alter the classification of any case in which death had previously been attributed to either NAI or suspected or proven filicide. The assertion that we gave “no explanation for the decision to categorise as natural all 13 deaths for which there was insufficient evidence to reach a conclusion” is a misrepresentation of our text. The text makes clear that in the 13 cases that we were unable to review, we relied on the findings of coroners’ and higher courts in categorising them as natural. The text of our paper states “Eight of these CONI deaths were certified as SIDS. None were attributed to homicide or non-accidental injury.” The assertion that “In five of these cases a police investigation was under way” is not correct. We could only make the statement that we did, after all police enquiries had been completed. Risk factors such as smoking and bed sharing do not provide a cause of death. We reported that in many families with repeated Sudden Infant Deaths there are many risk factors. To understand the significance of risk factors requires knowledge of their prevalence in both the community and in cases of natural and unnatural infant death. Without such information, the implication that any sudden unexpected death with significant risk factors is unnatural or homicide, is wholly unfounded. We conclude that Gornall’s article(1) raised no issue of substance that was not explicitly dealt with either in the original article(2) or the ensuing correspondence(4). R.G. Carpenter, A. Waite, R..C. Coombs, C. Daman Willems A. McKenzie, J. Huber. References 1. Gornall J. Was message of sudden infant death study misleading? BMJ 2006 2 Dec. 333: 1165–8. 2. Carpenter RG., Waite WJ., Coombs RC., Daman-Willems C., McKenzie A., Huber J., Emery J. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural. Lancet 2005, Jan. 1; 365:29-35. 3. Fleming P., Blair P., Bacon C., Berry J. Sudden unexpected deaths in infancy. The CESDI SUDI Studies 1993-1996. 2000. The Stationery Office. London. 4. Repeat sudden unexpected and unexplained infant deaths: Correspondence. . Lancet 2005, March 26; 365: 1137–8. Competing interests: None declared |
|||
|
|
|||
|
JohnP Heptonstall, Director of the Morley Acupuncture Clinic Leeds LS27 8EG
Send response to journal:
|
Sir/Madam Perhaps what suprises me most about the references provided by the author, which inform of various medics and researchers assessing infant deaths, are the fact that nowhere can I find any reference to probable vaccine-induced death (though these are known to occur and are so referenced in ARVI minutes through the 1980s when yellow card reports were discussed at those meetings, 1, and Sally Clarks' second child died within hours of vaccination); and, perhaps even more startling, the fact that said research teams presumed to have the facts and expertise required to presume about murders/filicides that had no benefit of properly conducted criminal investigations. It is usual that a proper criminal investigation is carried out, potential lawbreakers are identified, a prima facie case is suggested and placed in a court of law where experts in law, witness and other evidence and careful deliberation of same, are required before anyone can presume on murder. Regards John H. References 1. eg. ARVI 86/3rd meeting 3rd October 1986, ARVI 87/ 1st meeting 6th February 1987. Competing interests: None declared |
|||
|
|
|||
|
David JR Hutchon, Consultant Obstetrician and Gynaecologist Memorial Hospital, Darlington, DL3 6HX
Send response to journal:
|
Richard Webster's rapid response is most interesting and shows how subtle chnages can change the emphasis. I was not aware of the language rules regarding ellipsis until I looked it up as a result of this response. Previously I would have expected a large section of irrelevant text to have been replaced by the elipsis. However I note that the English dictionary states that "The use of ellipses can either mislead or clarify, and the reader must rely on the good intentions of the writer who uses it." There is no doubt in my mind that leaving out the word "only" changes the emphasis of the sentence. Clearly tha vast majority of us would not have appreciated the possible implication of this subtle change. Although I appreciate it is difficult, the editorial staff should try to stop this sort of misuse of the English language. Garnall should also be asked to respond and justify his use if the ellipsis in this case. Competing interests: None declared |
|||
|
|
|||
|
John A M Hemming MP, Member of Parliament SW1 0AA
Send response to journal:
|
Catherine Williams makes a good start at looking at the issues that drive law in relation to children. The standard of proof does matter. If we make the assumption that a particular diagnosis implies homicide in say 75% of cases then is it really appropriate to convict a parent of murder when one knows that 25% of the people are innocent. 25% is quite clearly a reasonable level of doubt. The figures cited above indicate that the level of doubt is well over 50% although there is some doubt as to how far. It is a more complex case when it comes to child protection proceedings. When a child is taken into care it is bad news for the child. The question that really needs to be considered is whether it is worse news not to be taken into care. This is done on a different standard of proof. That is reasonable for an emergency protection order, but is not reasonable for a contested adoption. There is also a question of a propensity to intervene. No-one wishes to be the whipping boy (or girl) of the next Climbie Inquiry. This has the effect of driving intervention where it is not necessary. The reason why our child protection system is such a mess is in part the complexity of the issues. Articles like those of Mr Gornall do not help. Competing interests: Chairman - Justice for Families |
|||
|
|
|||
|
Penny Mellor, Campaigner/researcher Home WV9 5HX
Send response to journal:
|
Dr Speight needs to understand that the crime of fillicide/infanticide is not deemed to be child abuse per se-it is deemed to be an act carried out by a mother whilst not in her right mind having been affected by the massive hormone imbalances that child birth induces in the body or through extreme tiredness-Smothering a baby in these circumstances is nothing like a parent/carer repeatedly battering or torturing a child until the child is killed. We all recognise child abuse exists, contrary to the persistent publicity to the contrary. On a final note Dr Samuels responds that in North Staffordshire's Coni programme there is not a 100% uptake-to be frank, given the reputation of his colleague for falsely accusing a parent of murder based on a TV programme-that's hardly surprising is it? Competing interests: Campaigner against false allegations of MSbP etc |
|||
|
|
|||
|
Brian Morgan, Freelance Journalist Cardiff CF11 6LF
Send response to journal:
|
Two at least of the numerous respondents have raised very serious issues concerning the substantive BMJ article. These are (a) that there may have been unidentified collaborators. Further to this if there were undisclosed individuals involved in preparing this article, their interests have not been declared. And (b) the removal of one word from a quote changed the relevance of the passage quoted to the points being made in the article. Is it the intention of the editor of the BMJ to investigate, and then to publish a statement? Should the substantive article be withdrawn pending any investigation? Brian Morgan brianmorgan@ntlworld.com Competing interests: None declared |
|||
|
|
|||
|
John Stone, none London N22
Send response to journal:
|
The editor needs to consider whether the series of articles by Jonathan Gornall in British Medical Journal this year should have been published, and most particularly the present piece [1] and its immediate predecessor [2]. Before making specific points about these two articles I believe it is fair to say that they represent a departure in the style of journalism for BMJ, being akin to magazine articles, or even gossip. It may be that Gornall, who is an experienced newspaper journalist, may have been insufficiently aware of the traditional constraints of writing for a scientific/medical journal, which should not have caused BMJ to suspend traditional publication criteria. In the case of the present article [1] the central claim that the views of a dead person had been misrepresented by his colleagues, turns out to have been false [3], and Gornall had adapted Professor Emery's report for the Sally Clark defence team to make it sound as if his concern was exactly opposite to what he was saying. If BMJ was concerned about the ethics of Carpenter et al, they now need to consider even more urgently their own. In the case of the previous article [2] there was serious concern about substance. Twice challenged by myself Gornall offered no defence of expert anonymity or court secrecy, so whether certain persons may have misbehaved was of no consequence to the fundamental issue [4, 5, 6]. The article, in effect, constituted nothing more than an ad hominem attack, and an attempt to defend the indefensible [5]. I do not suggest we get into the Stalinist practice of withdrawing articles, but I do think BMJ ought to address the fact that serious errors of judgement have been made, and give some explanation to readers as to why this occured. [1] Gornall, J, 'Was message of sudden infant death study misleading?', BMJ 2 December 2006, http://www.bmj.com/cgi/content/full/333/7579/1165 [2] Gornall, J, 'How doctors' anonymity in family courts is under threat', BMJ 11 November 2006, http://www.bmj.com/cgi/content/full/333/7576/1024 [3] Webster, R, 'A significant omission' BMJ Rapid Responses 9 December 2006, http://www.bmj.com/cgi/eletters/333/7579/1165#150643 [4] Stone, J, 'What is Gornall's argument?' BMJ Rapid Responses, 16 November 2006, http://www.bmj.com/cgi/eletters/333/7576/1024#148953 [5] Gornall, 'Author's Reply', BMJ Rapid Responses, BMJ Rapid Responses, 17 November 2006, http://www.bmj.com/cgi/eletters/333/7576/1024#149367 [6] Stone, J, 'Re: Author's Reply', BMJ Rapid Responses, 21 November 2006, http://www.bmj.com/cgi/eletters/333/7576/1024#149522 Competing interests: None declared |
|||
|
|
|||
|
Hilary Butler, freelance journalist home, NZ 1892.
Send response to journal:
|
Jonathan Gornall's article misses the crucial point.
Dr Jayaprakash Ayillath Gosalakkal says: "I think this should give us all pause to consider what a modern medical system should be doing? Should people be automatically considered experts because of the position they are holding?" Dr Martin P Samuels says: "...there is no definitive test to distinguish between natural and unnatural mechanisms" Catherine Williams, reader in Law, says: "Lawyers and courts must be able to rely on medical evidence presented to them...If medical practitioners do not present their work carefully and in an unbiased fashion mistakes will be made, and it will be children who will suffer." The authors of the Lancet paper reply: "we relied on the findings of coroners’ and higher courts in categorising them as natural." In a medical article in 2002 (1) The following comment was made: "Virtually all deaths certified by coroners undergo postmortem examination, which is generally assumed to provide a more accurate cause of death. However, most coroners are not medically qualified and their role is primarily to exclude foul play rather than to provide accurate epidemiological data. Comparatively few coroners' postmortem examinations include bacteriology, and coroners may not always obtain pathology reports from specimens taken before death by clinicians. ...In the eight deaths certified by coroners, the postmortem examinations did not include bacteriological examination. In seven of these, bacteriology showed pertussis. However, the coroners and their pathologists apparently did not have access to this information, either at the time of certification or later." The truth of this paragraph is what should be focused on. If a pathologist's mind-set is solely on ONE purpose (to exclude foul play) then they're less likely to provide the most scientifically accurate possible explanation for the death- as shown in the study above where seven deaths catagorised as SIDS, were during the process of the study, found from bacteriology, to be from pertussis. The real issue here is the adequacy of the qualifications of pathlogists and coroners. What is their "expertise"? What training do they have? Are they given sufficient information from the start of investigations, to provide an open-minded analysis of all possible causes of death? Who assesses whether the pathologist has all the relevant information, has done all the relevant tests, or even whether they have done the autopsy correctly in the first place? If pathologists don't get it right in simple situations, where there is a test available, how might that play out in more complex situations where other fixed mindsets come into play? How many other "Sally Clarks" have been sent to jail in such circumstances? Contrary to Dr Nigel Speight's assertion that: "The media, editors of journals and Members of Parliament bear a heavy responsibility to 'get it right'." ... it is my opinion, that if the medical profession would ensure that coroners get it right in the first place, then the role of the media, journals and members of Parliament would be greatly clarified, and lightened. Parents of babies who die would welcome greater reassurance that doctors, paediatricians, "expert" witnesses, and pathologists; whom coroners or courts depend on for valid causation (or explanation) of death, are worthy of their trust. (1) Crowcroft, N.S., et al 2002 "Deaths from pertussis are underestimated in England" Arch Dis Child 2002;86:336-338 Available at http://adc.bmj.com/cgi/content/full/86/5/336 Competing interests: I was involved peripherally on the Sally Clark case. |
|||
|
|
|||
|
Jonathan Gornall, Freelance journalist London, E1
Send response to journal:
|
IT IS a measure of the nature of the increasingly polarised debate over child protection that one cannot contribute to it without being characterised as a supporter of one “side” or the other. One consequence of this polarisation is that issues vital not only to the protection of children, but to the provision of support to often-chaotic families, are not debated, but merely shouted down. It is therefore disappointing, but not surprising, that while various voices have been raised in an attempt to drown out the facts in my article [1], not one of these critics has sought to address the issues. The central issue here is whether the index of “suspicion” should rise with each repeat unexplained infant death within a family. Clearly, judging by the CONI data even as analysed by Carpenter et al [2], it should. Equally clearly, that is not a generally popular proposition but, in the light of the evidence that supports it, nevertheless one that merits serious consideration. How that raised suspicion should be acted upon, by doctors and by society as a whole, is another issue. My view – and I record it only because it seems to chime with that of many doctors with whom I have spoken – is that any mother who harms or kills her own child is acting most likely under the influence of some temporary mental imbalance, and as such deserves not prosecution and imprisonment, but care and compassion. It is indeed difficult to understand why some such cases find their way to the criminal courts, while others go no further than the family courts. This, however, is the system we have and, until it is changed, paediatricians charged with the protection of children have no choice but to operate within its boundaries. That is, of course, a debate well worth having – but how can it be had when any attempt to assemble the facts is met with only partisan hostility? The responses of some of the respondents are as predictable as their agendas are transparent. However, I expected Carpenter et al at least to take seriously the concerns of the former President of the RCPCH. In my article I conveyed some of Professor Hall’s concerns. For emphasis, here in full is the last paragraph of the letter he wrote to the editor of the Lancet on November 11, 2005: “Recently I was revisiting the literature on sudden infant death in preparation for a review of child protection in general, a matter which as you know is of great concern to paediatricians at the present time. I was alarmed by the conclusions reached by Carpenter and colleagues and when I discussed this with Dr Bacon [former medical adviser to the Foundation for the Study of Infant Death, the paper’s sponsor] I found that he agreed with me - the conclusion reached was seriously misleading and could have serious consequences.” In his response, Carpenter fails to address Professor Hall’s concerns. However, he does make a number of points in response to my article, one of the most telling of which is this: “We stand by our findings that of these [57] deaths, 9 were inevitable (died of known cause) and of the 48 unexpected deaths 7 were probable homicide.” Carpenter’s determination to stand and fight on this point would, of course, be admirable – were anyone challenging it. No-one disputes that the seven deaths were unnatural. The point - that he continues to fail to address - is whether it is justifiable to classify all of the remaining 41 deaths unequivocally as natural. Another bold defence of an unassailed position follows: “We did not alter the classification of any case in which death had previously been attributed to either NAI OR SUSPECTED OR PROVEN FILICIDE” (my emphasis). Again, no-one is suggesting that they did. The point I made in my article was that the authors “did not make clear that after Professor Emery's death they recategorised deaths that he had classed as UNNATURAL OR OF INDETERMINATE CAUSE as natural deaths”. In the 1998 paper, Emery had grouped together as “non-natural deaths” the following cases: 5 proven filicides, 2 non-accidental injuries (NAI), 2 open verdicts, 4 overlying and one case in which, although there was insufficient evidence, filicide could not be excluded. Of these 14 cases, the seven in the first two categories (filicides and NAIs) did indeed, as Carpenter states, remain classified as unnatural in the Lancet paper. They are the only ones that did. The point – which Carpenter again fails to address – is that after Emery’s death, without his blessing and yet under the imprimatur of his apparent authorship, the remaining seven cases that Emery had classified as “non-natural” were allocated to the “natural” category. Carpenter says my assertion that the authors gave “no explanation for the decision to categorise as natural all 13 deaths for which there was insufficient evidence to reach a conclusion” misrepresents their text. I beg to differ. Although this paper purported to be the result of a thorough review of all the unexpected CONI deaths, based in each case upon detailed inquiries, “including a family interview, a review of autopsies and case discussion”, when faced with no fewer than 13 cases about which they were, for various reasons, able to discover very little, the authors were content to abandon the advertised rigours of their methodology and fall back upon “the findings of coroners’ and higher courts in categorising them as natural”. This compromise alone, I would suggest, invalidates Carpenter’s findings. There was only one intellectually honest place for these deaths - the "grey area" category that Carpenter refused to countenance. The authors’ own words say it all: “We cannot exclude the possibility in our study that some of the 13 cases in which enquiries were not possible were cases of covert homicide.” If that is so, then how on earth can Carpenter justify classifying as “natural” all 13 of those cases – including seven in which parents or their representatives flatly declined to engage with the study? He can’t - which is, presumably, why he doesn’t try. Carpenter is absolutely right that “Risk factors such as smoking and bed sharing do not provide a cause of death”. Neither, however, do they justify the default conclusion that deaths featuring such factors can be classified safely as natural. It is precisely the decision by Carpenter et al to dispense with the “grey area” category, recognized by Emery and others working in this difficult field, which invalidates the conclusions of this study. To describe such worrying deaths unequivocally as natural is, as Professor Hall says, “seriously misleading”. Richard Webster chooses to ignore the overwhelming evidence of Emery’s position (and I listed all the references in my article and gave copies of every document for the benefit of the BMJ’s editors and lawyers), preferring instead to seize upon my omission of the word “only” from Emery’s report on behalf of the Sally Clark defence team. He suggests that “Gornall has changed the wording so that he can more easily recruit Emery’s statement to support this [Meadow’s] law”. This is an exercise in sophistry. The “only” – whether in or out – changes nothing. Neither Emery’s statement in this case nor his much-recorded views in the other documents I referenced require any “recruiting”. It is a cold fact that Emery believed that the occurrence of two unexpected deaths in a family raised a definite suspicion of unnatural death – a suspicion that was confirmed, he felt in this instance, in "only" a third of cases. In other studies Emery found 55%, 42% [3], 34.5% [4] and 40% of such deaths [5] to be unnatural. Webster and others might be better employed addressing the reality of these sobering percentages rather than attempting to obscure them behind a smokescreen, but even (or “only”!) “a third” is a proportion both shocking and sufficiently substantial to suggest that Professor Meadow was rather closer to the truth with his so-called “law” than many would care to admit. Richard Wilson, who appears less concerned with any message within my article than with whence it might have come, makes the suggestion that I was “heavily briefed by unnamed individuals who in the normal open way would have been identified as co-authors”. But in the “normal open way” of which profession? Setting aside the apparent distaste for journalists so evident in Wilson’s intemperate response, I invite him to consider that I am not a scientist who submitted a paper, but a journalist who wrote an article. (What’s more, I wrote it all by myself, although once or twice, I confess, I was obliged to resort to a medical dictionary.) It is true that I had heard a number of doctors expressing disquiet about the Carpenter paper and it was this that persuaded me to take a look for myself, but this, generally, is what journalists do. They follow leads picked up from a variety of sources. In researching the article I did, of course, talk to several people - frankly, I would have expected criticism had I not - including three of the authors of the paper. This particular interview lasted more than two hours and took place in the presence of the FSID’s director, scientific adviser and vice-chair (as I suspect Wilson, a trustee of the foundation, knows). The only key player I was unable to interview was Professor Emery, but fortunately the papers and other documents he left behind speak for themselves. Wilson professes a concern for “reasoned and free debate”, but his response appears to betray an enthusiasm if not for censorship, than at least for a debate that takes place on his terms. I am surprised to see Brian Morgan taking up Wilson’s point. As a journalist himself, I’d have expected him to defend my right to protect my sources. Perhaps on this occasion Morgan is writing not as an independent journalist, but in his other, polarised role as a “supporter” of MP John Hemming’s campaigning organisation Justice for Families [6]. In which case, I think we should be told. 1. Gornall J. Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168. 2 Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35. 3 S Wolkind, EM Taylor, AJ Waite, M Dalton, JL Emery. Recurrence of unexpected infant death. Acta Paediatr 1993; 82: 873-6 4 Waite A, Emery J. Repeat cot deaths in families enrolled onto a support programme. Abstract, 4th SIDS International Conference, Bethesda, 1996 5 Waite A, McKenzie A, Carpenter R, Coombs R, Daman-Willems C, Emery J. Report on 5000 babies using the CONI (Care Of Next Infant) programme. Foundation for the Study of Infant Death October 1998 6 http://john.hemming.name/national/familylaw/justiceforfamilies.html Competing interests: Author of article |
|||
|
|
|||
|
Alison J Waite, National CONI Co-ordinator Room C1, Stephenson Wing, Academic Unit of Child Health, Sheffield Children's Hospital, S10 2TH, Robert Carpenter, Robert Coombs, Charlotte Daman-Willems and Angela McKenzie
Send response to journal:
|
Martin Samuels addresses the issue of ascertainment of the population of siblings born following a sudden unexpected death in infancy (SUDI) which we also discussed in our paper (1) and the ensuing correspondence (2). However, the figures Samuels uses to support his argument are incorrect. He quotes the total number of referrals to both the Care of Next Infant, CONI programme (eligible to parents who have had a child die suddenly and unexpectedly) and to the CONI PLUS programme (eligible to families enrolled for other reasons). The numbers of referrals to CONI in North Staffordshire were 32 of whom 2 declined. Thus the correct incidence of non-take up of the CONI programme in North Staffordshire is 2/32 or 6%, which is close to our previously estimated figure of 4% nationally. CONI centres are encouraged to pro-actively offer the programme to all families with a parent history of SUDI and to inform us of the number of families who decline and for what reason. In 2003 we introduced a reporting form for this purpose and this is included as a reminder in every CONI stationery pack. 1. Carpenter RG., Waite WJ., Coombs RC., Daman-Willems C., McKenzie A., Huber J., Emery J. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural. Lancet 2005, Jan. 1; 365:29-35. 2. Repeat sudden unexpected and unexplained infant deaths: Correspondence. . Lancet 2005, March 26; 365: 1137–8. Competing interests: Members of CONI steering group |
|||
|
|
|||
|
Penny Mellor, Campaigner/researcher Home WV9 5HX
Send response to journal:
|
Jonathan Gornall writes "How that raised suspicion should be acted upon, by doctors and by society as a whole, is another issue. My view – and I record it only because it seems to chime with that of many doctors with whom I have spoken – is that any mother who harms or kills her own child is acting most likely under the influence of some temporary mental imbalance, and as such deserves not prosecution and imprisonment, but care and compassion. It is indeed difficult to understand why some such cases find their way to the criminal courts, while others go no further than the family courts." [1] If the paediatricians are so compassionate then they themselves could have stopped criminal proceedings. all they had to do was refuse to testify for the prosecution, stating that criminal persecution of mentally ill mothers is not in accordance with their beliefs-or they could have stood in court and made the same plea to the judge and jury-they didn't-it is the experts that drive any prosecution-without their input there is no case to prosecute-so how does Mr Gornall and his anonymous sources reconcile those comments with the facts? At least I stand up to be counted, which is more than can be said for the anonymous doctors who allegedly state one thing privately and quite another to their paymasters. http://www.bmj.com/cgi/eletters/333/7579/1165#151673 [1] Competing interests: Campaigner against false allegations of MSbP and accused of being "polarised" |
|||
|
|
|||
|
John A M Hemming MP, Member of Parliament SW1 0AA
Send response to journal:
|
The polarisation of the debate in respect the law relating to Children and Families referred to by Jonathan Gornall is along the dimension of whether or not false positives matter. I am quite clear that they do. It is not unreasonable to say that the suspicion relating to multiple unexplained deaths increases. To argue, however, that this should be used as a cause for action requires substantially more evidence. The consequences of a false positive whether it be in a court or a diagnosis are important. Driving the argument, therefore, to increase the number of false positives as Jonathan Gornall does is going to raise concerns. He says "How that raised suspicion should be acted upon, by doctors and by society as a whole, is another issue." In this he confuses suspicion and determination. The erroneous equation of suspicion and adequate proof is what results in the "partisan hostility" to Mr Gornall's thesis. Standards of proof start with reasonable suspicion, move through various civil standards and go through to beyond reasonable doubt. Suspicion itself is not necessarily reasonable. For example for an arrest a PC needs "reasonable suspicion". In essence reasonable suspicion is based upon there being good reasons - which may over time be found not to be true, but are sufficient at the time. Mr Gornall also says "It is indeed difficult to understand why some such cases find their way to the criminal courts, while others go no further than the family courts." It is not difficult to understand because the answer is different standards of proof. Criminal Courts are on the basis of "beyond reasonable doubt" and Family Courts are supposed to be on the "balance of probabilities" - although I would argue as to whether actually the burden of proof has shifted. (A different argument) Logical argument, a proper understanding of statistics and an understanding of legal procedures are needed for a proper debate. Mr Gornall should spend the Christmas break reading up on standards of proof. Competing interests: Chairman - Justice for Families |
|||
|
|
|||
|
Robert Wheatley, Consultant Paediatrician Central School Clinic, Blackpool FY3 9HG
Send response to journal:
|
Gornall is to be congratulated for a very readable and well reasoned riposte [1] to the views put forward by Carpenter et al.[2] That it takes an outsider to point out to us what should be blindingly obvious to us all is our shame, not any failing on his part. The very simple and critical point is this: we do not know why infants die of SIDS. None of us do. If we did, they would not be SIDS, by definition. It follows, therefore, as night follows day, that if we do not know why these infants have died, we cannot possibly know whether their deaths were "natural" or "unnatural". The approach taken by Carpenter et al [2] appears to have been that in the absence of any clear evidence of homicide, the deaths were assumed to have been "natural" - whatever they mean by that loaded word. This manipulation of language throws up some other interesting insights into the mindset employed when their paper was written. For instance, several infants clearly had factors indicative of possible overlying - including nine with PM findings consistent with asphyxiation [2]. Whilst these may well have been accidental, I can think of no other circumstances in which death by externally imposed asphyxiation would be considered "natural". This is very bad science indeed, though I will resist trying to decide whether it is fourth, or merely second rate. That it comes from such eminent practitioners only serves to increase the sense of disbelief. What is worse though, is Dr Wilson's intemperate and ill considered response. For somebody who has held such important and responsible positions in our college to fail to understand such a simple and highly relevant point is worrying indeed. The Emperor truly has no clothes - no wonder he's so cross! Sadly, however, it is clear that this is worse than just bad science - though that is bad enough. The stated aim of the study is to estimate the probability that a second infant death is natural versus unnatural[2]. This was published at a time, in early 2005 when these issues were very "live" in the public imagination, and a highly public vendetta was being pursued against certain prominent paediatricians. Such was the weight of media attention and pressure, that later that year, the GMC was moved to deliver a verdict in one such case that was described by a high court judge as "approach[ing] the irrational"[3]. The authors gave no thought as to the near impossibility of their task, but produced an extremely tendentious interpretation of the data available to them, which, as it happened, chimed in with the lynch mob mentality prevalent at the time. This is not science; this is politics dressed up as science; and very destructive politics at that. 1. Gornall J. Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168. 2. Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35. 3. Appellate Court Decision. In: The High Court of Justice, Queen’s Bench Division, Administrative Court, Mr Justice Collins. Meadow, appellant; General Medical Council, respondent. 2006 Competing interests: None declared |
|||
|
|
|||
|
Brian Morgan, Freelance Journalist Cardiff CF11 6LF
Send response to journal:
|
... Professor Emery were still alive he could tell us exactly what he meant by the word 'only' in the context of that sentence. Maybe his meaning is what comes up in my on-line English dictionary - which is "no more than, (implying that more was hoped for or expected) merely". It is not for someone else to edit the text of what the professor wrote, clearly without his knowledge or consent. If this was what Professor Emery intended his sentence to mean, then 'stark' it may have been, but not for everyone and certainly not for Foundation for the Study of Infant Death director, Joyce Epstein. Nor must I add, as someone who has reported on this topic for fourteen years, does it differ from what Professor Emery told me personally - that he disagreed profoundly with the accusatory approach of certain experts, whom he named. I had previously interviewed him and quoted him (1) and afterwards kept in touch. If only the author of the substantive article had been in a similar fortunate position then I doubt if we we would be having this debate. And if removing 'only' and replacing with '...' makes no difference, perhaps the author would tell us why he did this? Sophistry? Then, when it comes to a journalist's duty to protect sources, of course I would defend that right. I was prepared to face criminal prosecution on a conspiracy charge with a possible seven year prison sentence rather than disclose to the police whose case I had been working on. It's by no means clear that the substantive article had anything to do with protecting sources. The concern was that there might have been unidentified co-authors - I was concerned their conflicts of interest were not disclosed. The named author denies there were any and assures us he was the only author. He therefore is responsible for altering Professor Emery's text. As for reference to myself as a 'supporter' on John Hemming's "Justice for Families" website - which of the three words in the title does the author disagree with? This is some of what is said about me on that site - they are essentially my words - I stand by them - perhaps the author should interview some of the falsely accused parents? Brian Morgan - Investigative journalist who has been working to highlight the injustices suffered by families wrongly accused of child abuse since 1991, particularly the way the secrecy of the family courts has allowed unreliable and unscientific medical evidence from a very small group of paediatricians and psychiatrists to be relied upon by another very small group of judges. 1 The Independent, 21st August 1992: page 1 - Parents of sick children 'being falsely accused'; page 3 - Parents accused as their baby lay dying. Competing interests: None declared |
|||
|
|
|||
|
John A M Hemming MP, Member of Parliament SW1 0AA
Send response to journal:
|
The reason why this area of medicine and law elicits so much passioned debate is that it is an intellectually complex area where the outcomes are actually very important to the people concerned. Robert Wheatley points out one aspect of SIDS which is that SIDS relates to undiagnosed deaths. Hence the Boston work stops a number of deaths being SIDS as it becomes possible to diagnose with certainty the cause of a number of deaths previously considered to be SIDS. I would still be nervous about using the residual SIDS diagnoses as proof of murder. This logical confusion exist in much of the debate, however. Hence SIDS would remain as a diagnosis for the cases identified by the work in Boston. It remains, however, important to decide how certain one should be to prosecute a mother for smothering her baby when there is reasonable doubt. There are at least three medical areas that have substantial debates about the use of a probabilistic diagnosis, frequently without proper evidence to justify it, to determine legal action. These are SIDS, SBS and Metaphyseal Fractures. The College of Radiologists, to their credit, have done some work on reporting what proper evidence exist relating to Metaphyseal Fractures. Sadly the RCPCH in their Child Protection Companion have not done the same work, but skirted around the issue. One hopes that they will take this issue up in the future. Competing interests: Chairman - Justice for Families |
|||
|
|
|||
|
Charles Pragnell, Expert Witness - Child Protection and Child/Family Advocate UK/Australasia
Send response to journal:
|
I would suggest that the terms `natural’ and `unnatural’ are complete misnomers when applied to the deaths of infants as in this context. The most correct terms should be `medically explained’ and medically unexplained’. Where the deaths of infants are medically explained then they should not be a consideration in this kind of research and to imply some relationship proportionately with those which are medically unexplained has no relevance and informative value. Where the deaths of children are medically unexplained I concur with the correspondent who argues that such deaths should be subjected to a rigorous examination by a Coroner’s Court, and especially I would suggest, those child deaths which occur whilst children are under the care of medical practitioners in hospitals and are being subjected to medical experimentation for research purposes. Such deaths and their causes should be investigated by independent medical practitioners. Where a death of an infant is medically unexplained then the first consideration and presumption must be the limitations of medical science, in that the child may have died of a yet unknown cause. The second presumption must be the limitations in the knowledge and experience of the investigating medical practitioner – at least three possible causes of infant’s deaths which appear to be rarely, if ever, considered have been suggested by correspondents to this column. i.e. genetic, vaccine related, and carnitine deficiency and there are others such as serotonin deficiency. Although there are recorded child deaths from vaccine related causes this seems not to be a consideration by medical examiners into child deaths because of the closed minds of many in the medical profession to the fact that vaccines can be extremely dangerous to some children. Probably the most telling comment made by Professor Emery is that “…approximately a third of deaths have been due to a whole series of rare natural causes which had been missed by those doing the necropsy.” It is alarming that in effect at least a third of child deaths are not being competently investigated which completely undermines any suspicions are being held that those deaths may have been caused by an adult carer. The `research’ which is being discussed is entirely without validity or utility as it is based on a false premise which is prejudicial from both a professional and legal perspective. Only after an exhaustive investigation has been made of the two criteria set out above, should the possibility of the death of a child having been caused by an adult be considered. In circumstances where an adult is suspected of killing another adult then factual evidence is usually required of a)the cause of death; b) the motives of the accused person; c) the methods/means by which the death was caused; and d)the opportunity to cause the death by placing the accused person at the scene. Such requirements do not appear to be being applied in cases of prosecution for child deaths and it was apparent in the Clark, Cannings, and Patel cases in particular that there was a notable dirth of such factual evidence in those cases and they were convicted largely on the basis of suspicion, insinuation, and supposition, and mainly those proffered by some of the expert medical witnesses. This is confirmed by the comments of the Learned Judges in the Donna Anthony case, as reported by Gornall, that “the consequent convictions depended substantially on the fact that there were two such deaths in the same family.” . i.e. merely on propensity in the complete absence of factual evidence. The suspicions of any person engaged in child protective services are of an extremely low standard of evidential proof, especially when such suspicions are based on personal prejudices, favoured but unsound theories, and dogma, rather than on factual evidence. Meadow’s `Law’ is in fact a total denial of rational thinking on this matter and of the legal presumption of innocence. To state that the deaths of two infants in one family is suspicious is mere fanciful speculation and to suggest that three deaths is murder (no doubt beyond reasonable doubt in Professor Meadow’s mind) is highly presumptive in the absence of any facts to support such a contention. Dr. Nigel Speight alleges that there is denial by some people that child abuse occurs when in fact the denial is abundantly evident in those people in child protection services who cannot accept that the vast majority of reports of child abuse have no substantive basis and a high proportion of these are false positives indicating that child abuse is grossly over-reported, drawing large numbers of children and their families into procedures which leave them severely traumatised and grossly abused by the system which is intended to protect them but which is currently causing far more harm than good. Repeated errors in the detection and prevention of child deaths also indicates that there are many in child protection services who are in denial that any wrongful beliefs or actions have occurred and they are therefore incapable of change and improvement of their practice or their methods of operation. Competing interests: Concern to Reform the UK Child Protection Services |
|||
|
|
|||
|
Richard Webster, Author Oxford OX2 6TX
Send response to journal:
|
Jonathan Gornall’s various attempts to meet the criticisms which have been made of his article serve at least one useful purpose.[1] They make abundantly clear what was implicit in the article itself – that Gornall has confused two different issues. Some at least of Gornall’s criticisms of the 2005 paper by Carpenter and colleagues are well-founded.[2] By translating uncertain data into precise-seeming statistics, Carpenter et al were inviting just the kind of criticism which Gornall made in his article and which Chris Bacon made at the time. As Bacon wrote in the Lancet in March 2005: ‘This paper will be influential, but those quoting it should be aware that its data do not support such clear-cut conclusions. Until we have more reliable means of identifying covert homicide, we cannot be certain of the proportion of repeat sudden infant deaths that is unnatural. It may well be rather higher than this analysis suggests.’[3] Carpenter and his colleagues were justified in replying that they had ‘described the cases in substantial detail so that the nature of the data and problems of interpretation were completely transparent.’[4]. However, to imply that this was a satisfactory answer to Bacon’s criticisms was naïve. In the end the commendable transparency of the paper makes it easier to see that the criticisms were, in some respects, well-founded. But the debate over whether it is possible to determine precisely the proportion of second cot deaths due to natural causes has practically no bearing on whether ‘Meadow’s’ law’ is valid. Gornall’s attempt to recruit John Emery in support of this law is a non sequitur. The great danger of Gornall’s article is that it may have succeeded in persuading the world of what is false by urging upon it what is true. That a freelance journalist who has no medical or scientific background should have so muddled the issues is perhaps not surprising. What is more disturbing is that Gornall’s article should have been published by a reputable medical journal and hailed as a contribution to scientific understanding by a number of trained clinicians who ought to know better. What this suggests is that a significant part of the medical profession has still not understood the multiple statistical fallacies which underpin both ‘Meadow’s law’ and the miscarriages of justice which Professor Meadow and like-minded physicians have helped to bring about. In criticising the paper by Carpenter and his colleagues, Gornall complains that whereas Emery found that there were no suspicious signs or circumstances in approximately two thirds (or 66%) of second unexpected infant deaths, Carpenter’s paper suggested that 87% of such deaths were due to natural causes. The difference is clearly significant. But it is also, from at least one perspective, marginal. Both Emery and Carpenter were agreed about one crucial point – that second unexplained infant deaths are frequently due to natural causes. The figures produced by Carpenter suggest that there is almost a 9 in 10 chance that a second unexpected infant death will be due to natural causes. Emery’s estimate was lower but still suggests that the chances of a second sudden infant death being natural may be as high as 2 in 3. The best-known instance was, of course, the evidence he gave in 1999 in the trial of Sally Clark that, in a family like hers, the chances of two unexplained infant deaths being due to natural causes were 1 in 73 million. It may well be that this ultra-extreme calculation was influenced by Meadow’s disturbingly casual reading of an unhelpful passage in the latest CESDI report (Confidential Enquiry into Stillbirths and Deaths in Infancy).[6] But even if we discount the 1-in-73-million figure entirely we are still left with Meadow’s claim, made in his original statement for the trial, that the chances of two natural cot deaths occurring in the same family were 1 in 1,000,000. Yet more importantly we have Meadow’s own contemporaneous pronouncement on this subject in his 1999 paper, ‘Unnatural Sudden Infant Deaths’. This was the paper in which Meadow set out to give an account of the ‘clinical features of 81 children judged by criminal and family courts to have been killed by their parents’. Meadow had himself been involved in these cases and his explanation of how it was that 81 cases of murder came to be drawn from only 50 families is revealing:
The clear import of these words is that the courts had themselves construed the supposed statistical improbability of recurrent natural cot deaths as a pointer to guilt. What is not so clear, but may be inferred, is that one of the reasons they did so was that this fallacious argument had, in some cases at least, been urged upon them by Meadow himself.[8] Because of the attention which has been given to Meadow’s now notorious 1-in-73-million figure, the statistical basis of such probability calculations has been much discussed. Indeed it has been examined and commented on by at least eight judges, and by many more journalists. However, it is far from clear that any of these examinations has succeeded in conveying the sheer magnitude of the statistical errors which Meadow perpetrated. Even more importantly, these examinations have generally failed to grasp the extent to which the same kind of statistical fallacies appear to have informed Meadow’s habit of mind in relation to all unexpected infant deaths.[9] The appearance of Gornall’s article in the BMJ indicates that a great deal of darkness still surrounds this subject. That any article should seek, on the basis of statistics suggesting that the chances of a second unexpected infant death being due to natural causes are in the region of 2 in 3, to restore the reputation of a paediatrician who gave a jury the impression that the chances this happening were either 1 in a million or 1 in 73 million, is bizarre. What such an article betrays is basic innumeracy, combined with a general lack of understanding and judgment on a scale that is disturbing. That a reputable medical journal should have chosen to publish such an article (and to promote it by way of a press release) is regrettable to say the least. The fact that the author of the article in question should have doctored the meaning of one of his quotations by replacing the word ‘only’ with an ellipsis, and that the BMJ should have published without comment his self-serving defence of this act of intellectual dishonesty, merely compounds the original editorial misjudgment.[10] In view of the widespread misunderstanding of the statistical issues with which the BMJ article deals it should be more widely appreciated that, although none of the judges or journalists who have grappled with Meadow’s abuse of statistics appears fully to have grasped the nature of the problem, the subject has been addressed comprehensively elsewhere. I refer to a series of essays by Ray Hill, Professor of Mathematics at the University of Salford, the clarity and sensitivity of which has been matched only by the general neglect they have suffered. Of these four essays, one remains unpublished, one was delivered as a paper to a conference on developmental physiology, one was published in the journal Pediatric and Perinatal Epidemiology, and one in Significance, the journal of the Royal Statistical Society. All four papers are available for downloading here (or from the author’s university webpage) and all four merit careful reading.[11] It is only necessary to add that perhaps the most disturbing parts of Jonathan Gornall’s article were those sections in which he appeared to be casting doubt on the soundness of the judgment of the Court of Appeal in overturning the convictions of Donna Anthony and Angela Cannings. Gornall’s misconceived comments on these cases border on the defamatory; they should, like the article in which they appeared, never have been published. Notes Further discussion of this topic will be be found on my website at www.richardwebster.net/cotdeaths.html Email: richardwebster{at}ntlworld.com 1. Gornall J. Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168 [free full text]; Gornall J. Author's Reply [1], BMJ Rapid Responses, 17 November 2006 [free full text]; Author's Reply [2], BMJ Rapid Responses, 18 December 2006 [free full text] ; Gornall J, Author's Reply [3] BMJ Letters, BMJ 2007; 334: 7-7. 2. Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35. [PubMed] 3. Bacon C. Repeat sudden unexpected infant deaths. Lancet 2005; 365:1137.[PubMed] 4. Carpenter RG, Waite A, Coombs RC, Daman-Willems C, Huber J. Correspondence: Repeat sudden unexpected infant deaths. Lancet 2005; 365:1138. [PubMed] 5. Meadow R. Fatal abuse and smothering, in Meadow R, ed ABC of Child Abuse, 3rd edition London: BMJ, 1997: 29. 6. Fleming P, Blair P, Bacon C, Berry J. Sudden unexpected deaths in infancy. The CESDI SUDI Studies 1993-1996. 2000. The Stationery Office. London. 7. Meadow R, Unnatural Sudden Infant Deaths. Archives of Disease in Childhood, 1999, 80: 7-14. [free full text]. The paragraph quoted here was cited by Mr Justice Collins in his judgment on Meadow's appeal against the GMC's finding of serious professional misconduct. Mr Justice Collins makes the following observation about it: 'The passage in quotation marks has no attribution. In his evidence, the appellant [Meadow] said he could not recall where those figures came from and he recalled writing them on a blackboard in a lecture and reference from a member of the audience [sic]. But he said that the figure was, as he put it, a ball-park figure. In reality, it seems that it was based on his general experience and was used as an average.' (Meadow v GMC, 17 February 2006. Neutral Citation Number: [2006] EWHC 146, paragraph 47 [free full text]).) This is not the only occasion when Meadow has dignified with quotation marks a passage which is not in fact an exact quotation from anywhere. The passage in which he set out 'Meadow's law' was also placed in quotation marks without being given any attribution. This formulation, which Meadow described as 'a crude aphorism', appears to have taken the form it did because Meadow was quoting inaccurately from memory (or deliberately rephrasing) a maxim first proposed in 1989 by the American forensic pathologists Di Maio and Di Maio: 'While a second SIDS death from a mother is improbable, it is possible and she should be given the benefit of the doubt. A third case, in our opinion, is not possible and is a case of homicide.' (Quoted in Bacon CJ and Levene S, Sudden unexpected death and covert homicide in infancy, Archives of Disease in Childhood 2004; 89: 443-4470 [free full text]). In citing the same passage Carpenter et al [see above 2] point out that in the second edition of di Maio and di Maio a second SIDS death is referred to as being only 'remotely' possible. It should be noted, however, that the maxim as put forward by di Maio and di Maio contains, in the suggestion that mothers who have lost only two children should be given the benefit of the doubt, at least a residue of the presumption of innocence. In Meadow's version this residue has, either accidentally or deliberately, been removed. 8. cf Le Fanu J, Wrongful diagnosis of child abuse—a master theory. Journal of the Royal Society of Medicine 2005; 98: 249-254; [free text at Spiked online]. Le Fanu suggests that Meadow’s argument was in fact circular and that he invoked the authority of the courts ‘without making clear that it was his expert testimony that repetitive SIDS was “murder unless proved otherwise” that had been a major factor in securing those convictions’. It should be noted perhaps that even when Le Fanu is a little cavalier - as he is here in his treatment of Meadow's law - he is frequently perceptive (and very often right). 9. By far the best treatment of this topic is, in my view, that of Professor Ray Hill, see below, note [11]. Perhaps the most illuminating of the discussions which appeared elsewhere was the BMJ editorial by Stephen J Watkins which was published in January 2000. Watkins, S J, Conviction by mathematical error: doctors and lawyers should get probability theory right. BMJ 2000; 320: 2-3 [free full text] 10. Webster R, A significant omission. BMJ Rapid Responses 9 December 2006, [free full text] ; Webster, R. Ellipsis marks an important omission. BMJ Letters, BMJ 2007; 334: 7-7 extract; Gornall J, Author's Reply [2], BMJ Rapid Responses, 18 December 2006; [free full text] Gornall J, Author's Reply [3] BMJ Letters, BMJ 2007; 334: 7-7 extract. 11. Hill R, Cot death or murder – weighing the probabilities, Developmental Physiology Conference, June 2002. [Download Word document – includes unpublished essay: Why Sally Clark is, probably, innocent]; Hill R, Multiple sudden infant deaths – coincidence or beyond coincidence?, Paediatric and Perinatal Epidemiology, 2004 18, 320-326. [Download as PDF]; Hill R, Reflections on the cot death cases, Significance, 2005 2, 13-15. [Download as PDF]. See Professor Hill’s webpage http://www.cse.salford.ac.uk/profiles/profile.php?profile=R.Hill. |
|||
|
|
|||
|
Jonathan Gornall, Freelance journalist www.gornall.info
Send response to journal:
|
I welcome Richard Webster’s generous concession that “Some at least of Gornall’s criticisms of the 2005 paper by Carpenter and colleagues are well- founded” but reject his suggestion that in my article [1] I “confused two separate issues”. The confusion is entirely his. Webster writes: “The debate over whether it is possible to determine precisely the proportion of second cot deaths due to natural causes has practically no bearing on whether ‘Meadow’s’ law’ is valid.” This is simply illogical. What is meant in essence by this so-called “law” is that for each repeat sudden unexpected infant death suffered by one family there is, statistically speaking, a corresponding increase in the possibility that the cause might not be natural. This fact is irrefutable and even Carpenter et al’s contested analysis of the CONI figures supports this proposition [2]. How awareness of that increased possibility should be acted upon in practice is, of course, an entirely different matter, but it is unrealistic to suggest that doctors faced with cases of repeat sudden unexpected death in infancy should simply put it out of their minds when considering a differential diagnosis. Whether or not parents in such cases should end up in front of a judge and jury is an entirely different issue. This remains a question for society as a whole to resolve and it is as unfair as it is facile to saddle with responsibility for that state of affairs the doctors whose key concern is the welfare of any surviving or future children and who are obliged to operate within the parameters of the existing system. Webster’s own confusion extends to reopening the “one in 73 million” debate, which amounts to bowling at drawn stumps. I believe all parties now accept that the statistic was wrongly applied in the Sally Clark case. However, in suggesting that Meadow was responsible for the “ultra-extreme calculation” that led to the figure of one in 73 million, Webster repeats an error made by many commentators before him. The calculation was made by the authors of the CESDI study, in which the passage in question reads: “For a family with none of these three [risk] factors, the risk of two infants dying as SIDS by chance alone will thus be one in (8,543 x 8,543), i.e., approximately one in 73 million” [3]. There is another passage from the same section of the CESDI study that, to my knowledge, has never been quoted in the many attacks on Meadow: “When a second SIDS death occurs in the same family, in addition to a careful search for an inherited disorder there must always be a very thorough investigation of the circumstances – although it would be inappropriate to assume maltreatment was always the case.” Or, put another way, it would be inappropriate in these circumstances to assume that maltreatment was always not the case. This, I would suggest, is an observation whose meaning is not substantially different from Meadow’s “crude aphorism” in the ABC of Child Abuse, “One sudden infant death is a tragedy, two is suspicious and three is murder until proved otherwise” [4]. The representation of this “law” – put forward by Meadow only as “a sensible working rule for anyone encountering these tragedies” - as a “Presumption of guilt” is a misrepresentation. It seems fairly obvious that the phrase “until proved otherwise” is a reference not to legal proof of innocence in a court of law, but to the need in these cases for thorough medical investigation, including the “careful search for an inherited disorder” prescribed in the CESDI study. There is another, also never-quoted, passage in the same chapter of the ABC that places the much-quoted “law” in context and helps to demonstrate that Meadow was as sensitive to the possibility of innocent explanation as he was alert to the sinister: “A much more difficult subject is the covert killing in which a young child dies suddenly and there is uncertainty about the cause. In recent years there has been greater recognition of this occurrence and a willingness to confront the issues. The difficulty has been to do this in a way that is sensitive to the feelings of the parents who suffer the tragedy of a young child dying, suddenly and unexpectedly, from natural causes.” As Webster takes the trouble to refer to Professor Ray Hill and his essays on the subject of the one-in-73 million statistic (“the clarity and sensitivity of which has been matched only by the general neglect they have suffered”), I feel I should reassure readers that Hill has not been quite so neglected as Webster fears. I communicated with Hill [5] while preparing my article because I wanted to find out how a reference to a 2002 draft version [6] of the 2005 Carpenter paper [2] had come to be included in a paper published by Hill in 2004 [7]. Hill, a mathematician, makes no secret of the fact that he was involved as an expert with the defence teams in the cases of Sally Clark and Trupti Patel. Much of the detail of his involvement with these cases can be gleaned from the papers posted on his website [8]. He became involved in the Clark case after her first, failed, appeal in October 2000, and two papers he had written about the case were submitted by the defence team at Clark’s second appeal, in January 2003. Before Patel’s trial began in June 2003 her solicitor asked Hill to write a report and sent him a draft version of what was to become the Carpenter paper. The solicitor had come into possession of the paper because Carpenter was also an expert on the Patel defence team, although in the event he was not called to give evidence [9]. The fact that even Hill’s own conclusions can be said to reinforce Meadow’s point about increased suspicion serves only to emphasise Webster’s confusion. In an article published in 2005 in Significance, the journal of the Royal Statistical Society, Hill wrote: “My calculations gave the following rough estimates. Single cot deaths outnumber single murders by about 17 to 1, double cot deaths outnumber double murders by about 9 to 1 and triple cot deaths outnumber triple murders by about 2 to 1. So each successive death does give rise to some slightly increased suspicion, but to nothing like the extent that Meadow’s law would imply. In particular, where multiple sudden infant deaths have occurred in a family, there is no initial reason to suppose that they are more likely to be homicide than natural.” (This paper can be found on Hill’s website[8].) The phrase in the extract from Hill crucial to this debate is this: “So each successive death does give rise to some slightly increased suspicion” (begging the question – even if one accepts Hill’s “rough estimates” - of whether a reduction in odds from 17-1 to 9-1, or 9-1 to 2- 1, can fairly be described as “slight”). It is vital, of course, to have a realistic understanding of the extent of that “slightly increased suspicion” and, put simply, my criticism of the Carpenter paper was that, by fudging the distinction between natural and unnatural deaths, it hampered such an understanding. I was struck by Webster’s comment, “That a freelance journalist who has no medical or scientific background should have so muddled the issues is perhaps not surprising”. Webster’s website leads me to understand that he “studied English literature at the University of East Anglia”, but makes no mention of medical or scientific training. Nevertheless, he obviously considers himself sufficiently well qualified to criticise not only a mere (unqualified) journalist, but also the staff of the BMJ and “a number of trained clinicians who ought to know better”. I have no idea whether or not Webster is involved with the campaigners who have been working so hard to vilify medical professionals working in the area of child protection, but he does appear to share with them a trait that has been witnessed before in these columns: complete self-confidence in their own ability to grasp the intricacies of this difficult subject far better than the experts they disparage. Webster is, however, not entirely without experience of tackling the subject of false allegations of child abuse. The BMJ generously allowed him to include in his response the link to his website, where he offers for sale various books he has written, including The Great Children’s Home Panic and The Secret of Bryn Estyn – The Making of a Modern Witch Hunt. These are impressively substantial tomes and none the less so for having been published by The Isis-Orwell Press Ltd, a company that appears to publish books only by Webster, and of which he is sole director and shareholder [10]. Like Webster, I too have a website, but mine serves only as a repository of articles and links to documents, including papers, judgments and protocols, posted free of comment as a resource for anyone interested in the frequently colliding topics of cot death, infanticide and child protection. It can be found at www.gornall.info 1 Gornall J. Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168 2 Fleming P, Blair P, Bacon C, Berry J. Sudden unexpected deaths in infancy. The CESDI SUDI Studies 1993-1996. The Stationery Office. London, 2000;91-92 3 Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35. 4 Meadow R. Fatal abuse and smothering, ABC of Child Abuse. BMJ Publishing Group, 1997: 29. 5 Personal emails, September 9 to November 20 2006 6 Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Carpenter RG, et al. Repeat sudden infant deaths in families enrolled on to a support programme. Draft manuscript, November 2002. 7 Hill, R. Multiple sudden infant deaths – coincidence or beyond coincidence? Paediatric and Perinatal Epidemiology 2004, 18, 320–326 8 http://www.cse.salford.ac.uk/profiles/profile.php?profile=R.Hill 9 Author interview with Professor Robert Carpenter, August 29, 2006 10 Companies House WebCHeck, www.companieshouse.gov.uk, accessed February 9, 2007 Competing interests: Author of article |
|||
|
|
|||
|
L. Travis Haws, Dentist Lakewood CO 80228
Send response to journal:
|
Jonathan Gornall states that: "What is meant in essence by this so-called “law” is that for each repeat sudden unexpected infant death suffered by one family there is, statistically speaking, a corresponding increase in the possibility that the cause might not be natural." Gornall then claims that "This fact is irrefutable". I am utterly perplexed at how he believes this is irrefutable. If one can't even demonstrate a cause and effect on a death, how does one even attempt to postulate a statistical method. It's almost like saying that two deaths in a family were from two boulders plummeting down a cliff at seperate times. No one knows exactly why each boulder fell at that specific time and targeted two members, or more, of a family, but since it did, those boulders had conspiracy of murder on the mind. Back in the day when medics likely didn't know about meningitis, they probably attributed it to spirits and went henceforth to trephine a hole in the skull to release the spirit. Two members in one family with holes in their head equates to a family of satan worshipers. When all along, there was something in the household or genes that made them susceptible to meningitis. Unfortunately today, alot of SIDS can be explained, many just refuse to accept the data. The facets are too many to elaborate here, but I submit that the converse of Gornall's "fact" and "statistics" is true. So what we end up with here is that medicine can't explain everything, but then claims to know everything, and even put forth statistics, about what they don't know or can't explain. Boy, that's a shocker. Of course we don't know everything in medicine. It is very likely the surface has only been scratched. And to claim one knows what goes on at the bottom of the ocean by view from a surface crack is preposterous and arrogant. Competing interests: Know many falsely accused |
|||
|
|
|||
|
Michael D Innis, Director Medisets International Home 4575
Send response to journal:
|
Editor, In response to Richard Webster assertion that “The debate over whether it is possible to determine precisely the proportion of second cot deaths due to natural causes has practically no bearing on whether ‘Meadow’s’ law’ is valid” Gornall says “What is meant in essence by this so-called “law” is that for each repeat sudden unexpected infant death suffered by one family there is, statistically speaking, a corresponding increase in the possibility that the cause might not be natural. This fact is irrefutable and even Carpenter et al’s contested analysis of the CONI figures supports this proposition.” What should be pointed out is the Scientific meaning of a “Law” in this context is a ‘Regularity in Nature’ from which true predictions of future events can be deduced. The most charitable designation of Professor Meadow’s observation is it is not a Law and should not be called Meadow’s Law but Meadow’s Hypothesis. Descartes also classified ideas, less charitably , and he would have categorized it as Meadow’s Factitious – being “the Sirens and Hippogryphs you construct imaginatively.”[1] What is important is the accuracy of the diagnoses in the CONI deaths. Knowledge of the cause of SIDS has advanced considerably since the collection of that data and it is fair to say no account was taken of the role of vaccines and nutritional defects [2,3,4,5,6] in the deaths of the children. Discussion of the merits or demerits of Carpenter’s report would appear to be unproductive and of no significance without accurate diagnoses. References: 1.The Philosophical Works of DESCARTES . Rendered Into English by Haldane ES and Ross GRT. Vol ii 1934:pp152 -3. Dover Publications. 2.CAB. Child Abuse or Barlow’s Disease Medical Hypotheses 2002;59(1): 52-56 3.Clemetson CAB Vaccinations, Inoculations and Ascorbic Acid. J Ortho Mol Med(1999) vol 14 No 3 137 – 142 4.Kalokerinos A. Every Second Child Thomas Nelson (Australia) Limited 1974. 5.Scheibner V. Karlsson L COT DEATHS LINKED TO VACCINATIONS Nexus, Oct-Nov 1991 6. Buttram, HE Shaken Baby / Impact Syndrome:Flawed Concepts and Misdiagnoses (8-18-03) Competing interests: I have submitted a paper on the SIDS which is under review. |
|||
|
|
|||
|
Richard Webster, Author Oxford OX2 6TX
Send response to journal:
|
In his attempt to rehabilitate the views of Professor Roy Meadow, Jonathan Gornall chooses to spend most of his latest response avoiding the substantive issues and much of the rest of it diverting attention from the fact he has done so. The question at the heart of the argument is not whether those dealing with sudden infant deaths should remain vigilantly aware of the possibility of infanticide. This is accepted by all responsible parties. So too is the view that the probability of infanticide rises with each successive infant death. The debate is not about whether the probability rises; it is about what that probability is and how steeply it rises.1 Meadow believed that the probability of infanticide in recurrent infant deaths was high and that it rose very steeply. It is this belief which is expressed in ‘Meadow’s law’. Because this ‘law’ is indefensible in its original form Gornall is obliged to reformulate it. ‘What is meant in essence by this so-called “law”’ he writes, ‘is that for each repeat sudden unexpected infant death suffered by one family there is, statistically speaking, a corresponding increase in the possibility that the cause might not be natural.’ But this is not what Meadow’s law says. It does not say ‘two sudden infant deaths is suspicious and three is a bit more suspicious’; it says, quite plainly and quite clearly, that ‘three is murder unless proved otherwise’.2 The question Gornall fails to ask is why Meadow suggested that we should move from a position where we merely suspect murder in the case of two deaths to one where, in the case of three, it is deemed reasonable to presume it. Fortunately Meadow himself has provided an answer. For if we ask why two unexplained infant deaths should give rise to suspicion, his clearest answer was in his 1999 paper ‘Unnatural Sudden Infant Deaths’. Here he explained the high number of convictions in families where there had been double infant deaths by writing that ‘the courts were impressed by evidence that it was highly improbable for two or more children to die in infancy of undiagnosable natural causes’. He went on to explain that: ‘if there is a 1/1000 chance of a child dying suddenly and unexpectedly of natural causes in the first year of life, the chance of two children within a family so dying is 1/1,000,000’.3 Meadow arrived at this figure by assuming that the chance of a second sudden infant death in one family was the same as the chance of the first. This led him to square the initial figure of 1 in 1000 in order to reach 1 in 1 million. We know that he held this position with some consistency because he set out the same calculation in his original statement for the Sally Clark trial. We also know that if the logic of Meadow’s calculation is maintained then it follows, as night follows day, that the chances of three infants dying of unexplained natural causes will be dramatically higher. The figure will be reached by multiplying the supposed probability of two natural deaths by a factor of 1,000 all over again – in other words by cubing the original odds. The probability of three natural deaths will therefore be calculated as 1 in 1,000 million – a remote enough chance to make a presumption of murder seem reasonable. There is nothing new about this understanding of Meadow’s law. Professor Robert Carpenter and his colleagues take the same view in the Lancet paper which gave rise to Gornall’s article. They point out that the method of calculating the probability of recurrent natural infant deaths by squaring or cubing the original odds is invalid because siblings share 25% of their genes and are generally exposed to the same environmental influences. In a crucial passage, they write that ‘isolated estimates of the probability of more than one SIDS case in a family, however derived, are invalid and easily misinterpreted’. In a passage which is equally important they note that: ‘Multiple infanticide is also very rare. What is important is the relative likelihood that the deaths are natural versus unnatural, given that two or more deaths have occurred.’4 This is why the study undertaken by Carpenter and his colleagues was so important. By comparing second infant deaths judged to be natural with those judged to be cases of infanticide, they were in a position to arrive at meaningful figures about the probability of recurrent infant deaths being due to natural causes as opposed to the meaningless figures on which Meadow relied. Given the importance of the study, it is right that the findings presented by Carpenter and his colleagues should have been subjected to rigorous criticism. I have already made clear that I believe some of the points Gornall has made are well founded. But, in seeking to rehabilitate Meadow’s law on the basis of these criticisms, he has failed utterly to recognise the extreme nature of the statistical fallacies Meadow embraced. According to Meadow’s ‘invalid’ and misleading figures, the probability of two infants dying of unexplained natural causes was 1 in a million, which, as already noted, means the probability of triple natural deaths would be 1 in 1000 million. The valid figures, based on the relative incidence of two rare events - multiple SIDS and multiple infanticide - are very different. We can now say that they probably lie somewhere between 2 in 3 (Emery) and 9 in 10 (Carpenter) for double deaths, and are in the region of 1 in 2 for triple deaths. To argue about whether Emery’s estimate or Carpenter’s comes closer to the truth is entirely legitimate. To suggest that either figure does anything to vindicate Meadow’s position is, as I suggested in my last response, bizarre. There is, when all said and done, rather a large difference between 1 in a million and 2 in 3; or between 1 in 1000 million and 1 in 2. Gornall, who wrote to me at the end of last year saying he had read my book about the North Wales child abuse scandal and was ‘in awe at the breadth and depth of your research’ chooses to end his latest response by questioning my qualifications to write about another complex area involving arguments about child protection. He goes on to make reference to his internet research in the Companies House archives through which he has established my ownership of the imprint under which some (though by no means all) of my books have been published. Since this information is freely available on my website, where I make it clear that my most recent book was published under my own imprint for legal reasons, it is difficult to discern what purpose has been served by this research.5 Its main function would appear to be to distract attention from the central fact that Gornall finds so difficult to acknowledge, namely that the probability of recurrent infant deaths being due to natural causes is vastly greater than Meadow’s figures – and Meadow’s law – suggest. Given this, the question which must be asked is whether we should still regard families who suffer a repeat cot death with automatic suspicion. As Gornall observed in his original article, even John Emery once wrote that ‘The occurrence of two unexpected deaths in a family thus raises a definite suspicion of unnatural death’.6 The problem here, as Emery himself immediately went on to point out, was that this suspicion was ‘confirmed only in a third of such cases’. The question this inevitably prompts is whether it is either wise or safe to encourage such a level of unsubstantiated suspicion.7 It is the opinion of many observers, both inside and outside the medical profession, that the views of Professor Meadow, and of like-minded physicians who have encouraged an ultra-suspicious approach to recurrent infant deaths, have given rise to many miscarriages of justice and incalculable human misery in the criminal and in the family courts. It is my own belief that it would be a tragedy if, on the basis of entirely misconceived arguments, the views of Professor Meadow were to be rehabilitated along with the climate of corrosive and unwarranted suspicion he helped to create. Notes Email: richardwebster@ntlworld.com Further discussion of this topic will be be found on my website at www.richardwebster.net/cotdeaths.html 1. Gornall J. Re: The presumption of murder - author's reply, BMJ Rapid Responses, 9 February 2007 [free full text]. 2. Meadow R. Fatal abuse and smothering, in Meadow R, ed ABC of Child Abuse, 3rd edition London: BMJ, 1997: 29. 3. Meadow R, Unnatural Sudden Infant Deaths. Archives of Disease in Childhood, 1999, 80: 7-14. [free full text]. 4. Carpenter RG, Waite A, Coombs RC, Daman-Willems C, McKenzie A, Huber J, Emery JL. Repeat sudden unexpected and unexplained infant deaths: natural or unnatural? Lancet 2005;365:29-35 [PubMed]. 6. Gornall J. Was message of sudden infant death study misleading? BMJ 2006;333:1165-1168 [free full text]. 7. See Webster R. A significant omission. BMJ Rapid Responses 9 December 2006, [free full text]. Competing interests: None declared |
|||