The cot death controversy

Until last year Professor Sir Roy Meadow was one the most respected doctors in Britain.

A former president of the British Paediatric Association and the Royal College of Paediatrics and Child Health, he was widely regarded as a leading authority in the sphere of child abuse. After the release on appeal of two mothers – Sally Clark and Angela Cannings – convicted, with the support of expert evidence from Meadow, on charges of killing their infant children, and after the collapse of murder charges against Trupti Patel, Meadow has been reviled by campaigners and condemned in the media. While hundreds of similar cases – both criminal and civil – are being reviewed, in which parents have been convicted or children taken into care, Meadow now faces charges of serious professional misconduct before the General Medical Council (GMC).

Whatever the outcome of these proceedings, two of Meadow’s most controversial contributions to the study of child abuse – sometimes dubbed Meadow’s syndrome and Meadow’s law – have been widely disparaged in the media as pseudo-scientific prejudices. The syndrome, also known as Munchausen’s Syndrome By Proxy, results when a parent or other carer (usually the mother) either invents or intentionally produces physical or psychological symptoms in a child. The law – ‘one sudden infant death is a tragedy, two is suspicious, three is murder until proved otherwise’ – has featured prominently in the vilification of Meadow following the recent court cases.

It is not surprising that filicide, the killing of a child by a parent, provokes strong feelings. It provokes a range of responses based on two radically different perspectives on the issue of child abuse. On the one hand, some believe that diverse forms of child abuse are commonplace within families, even ubiquitous. This view has been promoted by child protection campaigners, and is used to justify the systematic regulation of contact between adults and children, as well as other surveillance measures. On the other hand, others believe that child abuse by parents is rare (and for practical purposes, non-existent), at least in middle-class families; they believe that illness and death in children is invariably the result of natural causes, even if these have not yet been discovered. From this perspective, campaigners supporting parents who have been convicted of injuring, even killing, their own children have set out to discredit the medical and legal processes of child protection – and the individuals involved in them.

Though both positions – the suspicion that all parents may be child killers, and the conviction that all parents accused of child abuse are innocent – are irrational, the former is more dangerous. This is because it currently enjoys the backing of powerful forces in society, including the government, the professions of medicine, law and social work, and influential voluntary organisations, such as the NSPCC and ChildLine. As it has gathered momentum over the past two decades, the campaign against child abuse has resulted in a more intrusive and coercive state policy in relation to family life and all activities involving adults and children. The most striking defect of the ‘all parents are innocent’ position is that, in its transparent absurdity, it does not provide the basis for effective resistance to the dominant trend for exaggerating the scale of human depravity in intimate relationships to justify further professional intervention.

The fact that Meadow has made such a rapid transition from being a leading figure in the crusade against child abuse, to being depicted as a rogue doctor responsible for destroying thousands of families, reflects the febrile climate in which discussion of these issues takes place in Britain today. At a time when claims to victimhood appear to confer social status, the controversy over infant deaths has produced a grisly competition between, on the one side, champions of murdered children, and on the other, campaigners on behalf of unjustly convicted parents. Each side seems ready to deploy any theory or argument in support of its position.

Let’s look more closely at the current controversy, which arises from the convergence of two categories of medical and sociological analysis – sudden infant death syndrome (SIDS) and Munchausen’s Syndrome By Proxy (MSBP). These categories are of fairly recent origin: both only became established in the 1970s.

• From infant mortality to SIDS

Though ‘cot death’ began to be recognised as a distinct concept in the 1950s it was not until 1969 that a definition of ‘sudden infant death syndrome’ became widely accepted (1). Before 1971, SIDS wasn’t accepted as a cause of death for the purposes of British death certificates. In his classic study of ‘the invention of infant mortality’, medical sociologist David Armstrong describes the process through which SIDS emerged as a distinctively modern preoccupation (2). Tracing the history of death certification, Armstrong notes that it was not until the 1870s, and the ‘social recognition of the infant as a distinct entity’, that infant mortality rates began to be published.

He further observes that ‘until the end of the nineteenth century, infant mortality was a problem of the biological realm; in the early twentieth century it became a problem of society’. At a time of national anxiety about the fitness of the population to sustain Britain’s imperial ambitions, antenatal care and infant welfare schemes emerged in response to the new recognition of infant mortality rates as ‘the most sensitive test of the health of communities’.

It was not until the mid-1950s that ‘the subdivision of infancy took on its modern form’. One area of official concern was ‘perinatal’ mortality (babies who were stillborn or died in the first week): this was linked to measures to improve antenatal and obstetric care and the care of the newborn, especially premature babies. Another was ‘neonatal’ mortality (babies dying in the first month) and ‘post-neonatal’ mortality (babies dying in the next 11 months): this group included a substantial number of deaths for which there was no apparent explanation, and which became defined as the problem of ‘cot deaths’, or SIDS.

Many factors contributed to the growing social concern about SIDS. These include the postwar ascendancy of a more prosperous but more privatised nuclear family with a smaller number of children, on whom parental affections and aspirations were more intensively focussed. The dramatic decline in infant mortality in the first half of the twentieth century – and its continuing decline in the second half – were also important. In 1900 in Britain around 150 babies died in the first year of life for every thousand born alive. At the start of the Second World War, the rate was still more than 50, falling to less than 20 by the 1960s. By 1990 it was eight, and in 2000 six. In 1900, some 140,000 babies died in Britain before their first birthday; a century later the figure was around 5000.

For people born in Britain before the Second World War, child deaths – mainly from infectious diseases – were a familiar experience. The fact that today such deaths are a rare occurrence means that each one provokes a much greater private – and public – reaction. When a baby who previously appeared normal dies suddenly without any apparent explanation, the impact is even greater. In the 1970s and 1980s, the number of babies whose deaths were attributed to SIDS remained fairly steady at around 1000 each year. In the 1990s the total fell sharply, to around 400 a year, a decline widely attributed to the impact of the ‘Back to Sleep’ campaign, which emphasised the importance of putting babies to sleep on their backs (and protecting them from cigarette smoke) (3).

• Child abuse and MSBP

Concerns about SIDS have converged with concerns about child abuse, mediated through growing recognition of the concept of Munchausen’s Syndrome By Proxy (MSBP). Though there have no doubt always been cases of children being beaten by their parents and other adults, the modern preoccupation with such abuse is commonly dated to the publication in 1962 of a paper entitled ‘The battered child syndrome’ by the American radiologist Henry Kempe (4). The public inquiry into the death of Maria Colwell in 1973 at the hands of her stepfather, while under local authority care, first attracted major national publicity to the issue of child abuse in Britain. A series of public inquiries into cases of children – Jasmine Beckford, Tyra Henry, Kimberly Carlisle – who had been killed by their parents or carers, continued through the 1980s (5).

Allegations of sexual abuse on a large scale by parents in Cleveland in 1988 provoked another high-profile public inquiry. Though social workers and other professionals have been criticised both for their failure to intervene to protect children and for their excessive zeal in removing children from families accused of abuse, the current consensus promoted by the government and concerned professional agencies favours intervention in the interests of the child. This emphasis was reinforced by the 2002 inquiry into the case of Victoria Climbie, who died in Haringey, north London, at the hands of her great aunt and her partner following the failure of a number of agencies to take appropriate action.

In 1951 the psychiatrist and popular medical writer Richard Asher coined the term ‘Munchausen’s syndrome’ to describe patients who were, like the German baron of the eighteenth century, inclined to travel widely and tell false stories (6). Simulating medical conditions, often skilfully and dramatically, these patients invite intensive investigation and treatment.

In 1977, after encountering a number of cases in which parents presented their children to doctors and hospitals with false claims of illness, Meadow adapted Asher’s term to describe this syndrome (7). The key features are that symptoms or signs are intentionally produced by a parent or carer; that the perpetrator, at least initially, denies inventing or causing these clinical features; that the features diminish or cease when the child is separated from the perpetrator. There are different degrees of the syndrome. The parent may simply present a false story (which may lead to unnecessary investigations and treatments). The false story may be compounded by the fabrication of signs (including tampering with samples or medical records). The parent may further seek to substantiate the false story by poisoning, smothering or otherwise injuring the child.

Meadow observed that the main victims of this syndrome were pre-school children and that in more than 90 percent of cases the perpetrator was the victim’s mother. In less than five percent of cases it was the father. In another five percent of cases it was another female carer. The most notorious case in which MSBP was diagnosed in a carer was that of Beverly Allitt, the nurse whose conviction in 1994 for killing four children and gravely injuring nine more under her care at Grantham Hospital, did much to raise public awareness of this syndrome.

Meadow’s concept of MSBP has not won universal approval. The colourful term has guaranteed public impact but it has also caused confusion. Does the syndrome only apply to cases in which a parent with Munchausen’s Syndrome manifests this through their child (or children)? Does the diagnosis require a deliberate intention to deceive? How, in any particular case, is it possible to distinguish between fabrication and exaggeration?

Some features identified in MSBP – the perpetrator’s denial of the cause of the child’s illness, the disappearance of symptoms and signs when the child is removed from the perpetrator – can be found in other forms of child abuse. In 1995 the American system of psychiatric diagnosis (DSM IV) introduced the category ‘factitious disorder by proxy’ in place of MSBP. In 2002, a working party of the Royal College of Paediatrics and Child Health proposed replacing MSBP with the more prosaic ‘fabricated or induced illness’ (FII) (8). However, a much fiercer controversy erupted when paediatricians, newly alert to the scale of human perfidy revealed by studies of MSBP, turned their attention to cot deaths.

• ‘Unnatural’ SIDS

In the early 1980s, John Emery, a paediatric pathologist in Sheffield, stated publicly what many students of cot death had long whispered in private: that a significant proportion of deaths attributed to SIDS – he estimated 10 percent – were in fact cases of filicide (9). This provoked a controversy that has been simmering ever since. Emery’s view appeared to be confirmed by the findings of official ‘confidential inquiries’ into infant deaths: one study of deaths registered as SIDS between 1993 and 1996 concluded that parental maltreatment was the primary factor in six percent and a contributory factor in a further eight percent. As cases of SIDS resulting from natural causes declined in the 1990s, the proportion attributable to filicide would be expected to rise.

Reviving this controversy in a debate in the British Medical Journal in 1999, Michael Green, Emery’s successor in Sheffield, indicated that ‘in private conversation’ paediatric and forensic pathologists estimated that between 20 percent and 40 percent of SIDS cases were in fact homicides (10). He proposed abandoning the labels ‘cot death’ and ‘SIDS’ in favour of openly declaring that the cause of death was ‘not ascertained’. Putting the case for keeping these labels, Lady Sylvia Limerick – a prominent campaigner on the cot death issue – emphasised (on the basis of the ‘confidential inquiries’ figures) that in 86 percent of cases there was no suspicion of parental responsibility (implicitly accepting that in 14 percent of cases, the SIDS diagnosis may have concealed murder) (11).

The great difficulty lies in distinguishing between cases of SIDS that may have resulted from parental action and those in which death has occurred from natural causes. The difficulty is compounded by the intense distress that inevitably surrounds any infant death, by the pressure to complete formal investigations as quickly as possible to allow funeral arrangements to proceed, by the inadequacies of specialist child protection and paediatric and forensic pathology services (as a result of recent controversies, these are now even more short-staffed than ever).

The stakes are high for all concerned. As recent cases have forcefully revealed, the consequences of a diagnosis of death by parental action are public humiliation, imprisonment and family disruption. If this diagnosis turns out to be incorrect, this constitutes a grave miscarriage of justice. On the other hand, an incorrect diagnosis of SIDS may also have catastrophic consequences. As Green observes, ‘it is a chastening experience to perform a frank homicide necropsy on a child whose elder sibling was signed off as a cot death a few years previously. I should know – I have done it’ (10).

In 1999, pursuing his studies of MSBP, Meadow published a survey of 81 children who had been killed by their parents, under the title ‘Unnatural sudden infant death’ (12). In this paper, Meadow attempted to identify features that distinguished between natural and unnatural deaths. He outlined two distinctive scenarios. In cases of ‘natural’ sudden infant deaths (SIDs), a previously healthy baby is found dead in its cot early in the morning by parents who had put a baby who appeared quite well, or merely snuffly, to bed some eight or 10 hours earlier. In cases of ‘unnatural’ SIDs, a baby who appeared to be normal only a few hours earlier, is found dead – or moribund – during the day or early evening, usually when it was in the sole care of the mother.

In half the cases of unnatural SIDs studied by Meadow, the baby had been discharged from hospital less than a week before death (in 15 cases less than 24 hours before). Some 70 percent had a history of unexplained illness (including apnoeic attacks, or episodes of ‘near-miss cot death’). Most of these babies were less than eight months old. Their mothers tended to be relatively poor (living on benefits and smoking), to have a record of psychological or social problems and to have had difficulties during pregnancy. The conclusion most commonly drawn in these cases was that the baby had been smothered by its mother.

Meadow also drew attention to features of smothering that might be identified on post-mortem examination. These include signs of recent bruising around the face, bleeding in the mouth, nose and throat, and microscopic features of haemorrhage in the tissue of the lungs. Other signs suggestive of abuse may be found in the eye and brain (such as retinal or subdural haemorrhages resulting from shaking), or in the bones (such as fractures at different stages of healing). However, in more than half of the cases in Meadow’s series there were no pathological features to confirm the cause of death: the corpse appeared normal.

The scientific status of Meadow’s work remains contentious. The 81 cases (from 50 families) included in his 1999 paper were collected on an ad hoc basis over an 18-year period. Proof of parental responsibility was based on criminal convictions alone in nine families, on family court verdicts alone in 31, and on both in 10 cases. Though Meadow insists that the likelihood that court verdicts were correct was ‘very high indeed’, recent events make this judgement questionable – especially regarding the family courts where the standard of proof is lower (and where the majority of cases were heard). In 19 families, parents confessed to killing their babies.

Meadow’s extraordinary claim that he is now unable to produce background data from this study, because it was inadvertently shredded by his secretary on his retirement, raises further doubts about this work. However, notwithstanding the controversies over Meadow’s personal definition of MSBP and his research into SIDS, there is a broad consensus among paediatricians internationally that parents do sometimes fabricate and induce illness in their children, occasionally to the point of killing them. Indeed the Department of Health published a consultation document on this matter in July 2001 (13).

Though the features described by Meadow and others may help to clarify the diagnosis of unnatural SID, a degree of uncertainty remains. The clinical features may be suggestive but they are largely circumstantial; the pathological features – if present – may appear categorical, but they have all been contested – and alternative explanations have been put forward to explain them. In his challenge to the concept of ‘cot death’, based on his concern about filicide, Green referred to the ‘”bad old days” of the 1960s, when infant deaths were wrongly attributed to unsuitable bedding or inappropriate care’ and paediatricians ‘used to invent meaningless diagnoses such as “acute interstitial pneumonitis” and “viral bronchiolitis”‘ (10).

It is striking that these same ‘meaningless diagnoses’ – together with some novel theories – have been invoked by campaigners against recent convictions to explain why these babies died. Supporters of Angela Cannings have attributed the deaths of her three children to environmental toxins, such as organophosphates, from the nearby Porton Down biological warfare research centre. Some campaigners have produced family histories showing numerous cases of infant deaths and have promoted one researcher’s claim to have discovered a ‘gene for cot death’, a discovery likely to be discredited as rapidly as those of genes ‘for’ homosexuality, criminality and alcoholism. Still more improbably, others have cited recent immunisations as a possible cause of sudden infant death, and ‘temporary brittle bone disease’ as a cause of unexplained fractures.

It is true that if a series of cot deaths occurs in the same family, this raises the question of whether there may be some inherited predisposition to such events. Yet much research has failed to reveal any plausible mechanism for this. The estimate that the risk of a second cot death in a (middle-class) family where one such event has already occurred is one in 73 million – cited by Meadow in one of the recent cases – has been widely criticised (14). Meadow has argued convincingly that – at least in the Clark case – this estimate is irrelevant because, given the injuries identified in both her children, none of the pathologists involved suggested a diagnosis of SIDS (15).

Yet there can be no doubt that there have been cases in which serial filicide has taken place – and has been wrongly attributed to SID (indeed, one perpetrator’s account has recently been published) (16). Whatever the statistical calculation of risk, any responsible doctor – even a casual observer – would be justified in being suspicious when such an event took place in the same family on more than one occasion, all the more so on a third or subsequent occasion. ‘Meadow’s law’ has been widely quoted, usually without acknowledging that Meadow was himself quoting what he described as a ‘crude aphorism’ (17). His suggestion that this aphorism is not so much a law as ‘a sensible working rule for anyone encountering these tragedies’ seems little more than common sense.

‘Think dirty’ is another crude aphorism that has provoked much popular condemnation. It was proposed in a Canadian protocol for the investigation of sudden infant deaths in 1995 and quoted by Green in his 1999 article (more recently, it was recommended to coroners by Dame Janet Smith in the course of the Shipman Inquiry) (10). Though Green considers it ‘a little graphically phrased’, he believes that it sums up the suspicious, ‘albeit cautious’, approach that is required in dealing with these cases. Indeed, this approach has been supported by research conducted by Professor David Southall and others, using covert video surveillance to investigate children in whom life-threatening abuse was suspected (18). The results were deeply shocking: in 33 of the 39 children filmed, parental abuse was revealed.

Parents were seen attempting to suffocate their children in 30 cases, in two poisoning them and in one inflicting a fractured arm. Though there are proper concerns about the civil liberties implications of such covert surveillance, the grim reality exposed by Southall’s video cameras seems indisputable.

• Diagnosis, crime and punishment

‘To state, as some doctors do, that to make a false positive diagnosis of child abuse is the biggest crime of all is facile. All doctors make errors in their diagnoses, and if one never made a false positive diagnosis of a common condition one would certainly be failing on many occasions to diagnose it. Secondly, it is necessary to separate the diagnostic process from the subsequent management.’ (19)

Meadow here distinguishes between the process of medical diagnosis and the legal process that may follow when a suspected diagnosis of fatal child abuse is reported to the coroner. In recent years, as we have seen, medical recognition and diagnosis of different forms of child abuse have improved. At the same time, in part as a result of the wider cultural concern about child abuse, there have been significant changes in the management of this problem – both by doctors and by the courts.

In a thoughtful contribution to the discussion prompted by Meadow’s 1999 paper on ‘unnatural SID’, senior paediatrician John A Davies put forward a defence of the traditional approach that had the virtue of making it explicit (20). He first indicated that paediatricians had long been ‘well aware that a “not negligible” proportion of SIDS are brought about by gentle smothering’. However, he continued, ‘few paediatric pathologists would claim to be able to distinguish these on post mortem findings’. In these circumstances, doctors usually followed the ethical principle ‘first do no harm’, in the belief that ‘more harm is done by wrecking the lives of innocent parents than by failing to track down the guilty’. He warned that the outcome of concern to prevent later children from suffering the same fate might be ‘to land them in “care” with all that it implies’. His conclusion was that ‘there is surely something to be said for letting ill alone lest worse befall’.

In recent years, the traditional approach described by Davies has been superseded by the more interventionist strategy, a process encouraged by Meadow and a group of like-minded specialists. Meadow’s status as the leading expert witness in this field has provided him with a public platform from which to promote his cause. It has allowed him to bring his experience as a clinician and researcher to bear on particular cases, and to use his status to develop a wider political role as advocate for the redefinition of SIDS and the greater recognition of filicide. One result of the greater public recognition of issues such as MSBP and child killing, partly as a result of Meadow’s work, has been to make convictions in these cases more likely – even though there has not been any great advance in the precision of medical diagnosis.

It is one thing for paediatric pathologists to indulge in private speculation about the proportion of SIDS cases that may be murder, quite another to present such estimates in public as though they were facts. The effect of such statements has been to normalise the idea of parental killing. Given the prevailing mistrust of parents, shared by doctors, lawyers and reflected in the media, the public has become predisposed to the idea that all forms of child abuse – including murder – are much more common than was previously believed.

Even though the evidence for the diagnosis of child abuse in any particular case may be weak – and that for estimates of the prevalence of filicide even weaker – Meadow’s advocacy has tilted the scales of justice firmly against the accused parent. At every stage of the process, from first suspicion, through medical diagnosis and the decision to prosecute, to conviction and sentencing, the chances of a parent ending up in prison (and children in care) have increased. Inevitably, given the continuing uncertainties over the clinical and pathological features of child abuse, the chances of a miscarriage of justice in such cases have also increased.

Is imprisonment a humane response to filicide? In the past, mothers who were accused of smothering their babies tended to have the charge reduced from murder to manslaughter – and they were usually given a suspended sentence and recommended to have psychiatric treatment. Now, the courts have shifted to take a much harder line: in recent cases, mothers have been convicted of murder and have received mandatory life sentences. (The courts have commonly imposed such harsh terms on fathers convicted of killing their – usually slightly older – children.)

It is worth noting that not all those commonly designated as ‘hawks’ in relation to the diagnosis of parental killing are in favour of more punitive treatment of parents. Indicating that ‘the most usual scenario for filicide is for the baby to have been suffocated by an exhausted parent (usually the mother) while trying to quieten his or her crying’, Emery believes that ‘these parents usually barely knew what they were doing and did not intend or want to kill their child’ (21). Emery’s conclusion from a lifetime’s experience of investigating cot deaths was that ‘we need to prevent these deaths, not victimise the parents’. Recognising that such killings do (mercifully, rarely) happen, there must be a better way of dealing with their perpetrators.

Meadow may legitimately point out that, in recent cases, his evidence only played a small part in the convictions, which were the result of a full legal process. However, he carries a wider responsibility for helping to forge a climate of opinion around such cases that made such convictions more likely. Meadow’s penchant for naming syndromes and promoting punchy aphorisms has raised both his own profile and that of child abuse. It has also made him a target for those who believe that they have been unjustly treated as a result of the courts’ reliance on his testimony.

It is true that in the recent appeals, Meadow’s evidence was called into question. It is also the case that, just as the convictions did not rest exclusively on Meadow’s evidence, neither did the appeals. In the case of Sally Clark, for example, the conviction fell largely on the fact that the prosecution had failed to disclose a pathology report showing a bacterial growth in the baby’s cerebrospinal fluid. (Though this provided legitimate legal grounds to uphold the appeal, it still did not provide a plausible explanation for the baby’s sudden death, which remains ‘unexplained’.)

• Conclusion

A special inquiry is now investigating whether the courts, both criminal and family, placed too much reliance on contentious medical evidence in convicting parents and carers on charges of child abuse and child killing. So far little clarity has emerged out of this murky and highly charged controversy.

Meadow’s syndrome remains ill-defined and his law is little more than a rule of thumb. Yet his work represents an attempt to reach a deeper understanding of the disturbing reality that a small number of parents, usually mothers, every year smother their babies. The denial of this reality by those campaigning against Meadow means turning a blind eye to infanticide.

The demonisation of Meadow has the characteristics of a witch-hunt. It is understandable that parents whose convictions have been supported by his expert evidence should be angry and aggrieved. Yet it is clear that the campaign against Meadow and others who take a similar view draws on wider currents of middle-class resentment. Hence it has rapidly achieved a consensus of support as wide as that opposed to the Blair government’s line on student fees or the war in Iraq.

The campaign has won the backing of a range of activists, journalists and lawyers who share a hostility towards science and expertise in general, and towards the medical profession in particular. There is a striking affinity between websites campaigning against Meadow and MSBP and those opposed to immunisation and animal experimentation and those dedicated to sufferers from Gulf War Syndrome, ME and similar syndromes. One common feature of these campaigns is their vituperative personal attacks on prominent adversaries, which have extended in some cases to harassment, assault and referrals to disciplinary bodies (22).

The current cot death controversy is taking place in a climate of increased hostility to scientific and medical expertise. Already, one consequence of this climate has been the reluctance of doctors to work in the field of child protection. In a letter to The Times (London) on 2 February, Professor Alan Craft, president of the Royal College of Paediatrics and Child Health, described child protection work as being in a state of ‘crisis’. Because of increasing lack of trust between public and the medical establishment in this area, ‘many medical posts in the field of child protection remain unfilled’, he wrote. Paediatricians are ‘not surprisingly, increasingly reluctant to act as expert witnesses in these complex cases’ (23). If every expert decision is greeted with automatic distrust, and if experts vacate the field, this will not help to protect children or exonerate innocent parents.

As the campaign against Meadow has gathered momentum, he has appeared as an increasingly isolated figure. Apart from Alan Craft, few colleagues have been prepared to support him in defiance of the mob currently in full pursuit (24). Meadow has himself remained silent, pending his appearance before the GMC, a body with a marked weakness for capitulating to popular prejudice (25). But while the pursuit of Meadow will not necessarily help to roll back the pervasive influence of the child protection industry, any concessions to his detractors are likely to deter further research into this difficult problem and appropriate measures to deal with it.

Dr Michael Fitzpatrick is the author of MMR and Autism, Routledge, 2004 (buy this book from Amazon (UK) or Amazon (USA)); and The Tyranny of Health: Doctors and the Regulation of Lifestyle, Routledge, 2000 (buy this book from Amazon UK or Amazon USA). He is also a contributor to Alternative Medicine: Should We Swallow It? Hodder Murray, 2002 (buy this book from Amazon (UK) or Amazon (USA)).

(1) Beckwith, JB quoted in Green, MA ‘Time to put “cot death” to bed’, British Medical Journal 1999; 319: 697-700

(2) Armstrong, D, ‘The invention of infant mortality’, Sociology of Health and Illness 1986; 8 (3): 211-232

(3) Green, MA, ‘Time to put “cot death” to bed’, British Medical Journal 1999; 319: 697-700

(4) Kempe, CH, Silverman, FN, Steele, BF, Droegemuelleur, W, Silver, HK. ‘The battered child syndrome’, Journal of the American Medical Association 1962; 181: 17-24

(5) Parton, N., Governing the family: child care, child protection and the state, 1991 and Parton, N., (ed), Child protection and family support: tensions, contradictions and possibilities, 1996

(6) Asher, R, ‘Munchausen syndrome’, Lancet 1951; 1: 339-41

(7) Meadow, R. ‘Munchausen syndrome by proxy: the hinterland of child abuse’, Lancet 1977; ii: 343-5. See also Meadow, R. ‘What is, and what is not, “‘Munchausen syndrome by proxy”?’, Archives of Diseases in Childhood 1995; 6: 534-8, and Meadow, R. (ed) ABC of Child Abuse, third edition, 1997, BMJ

(8) Royal College of Paediatrics and Child Health, Fabricated or induced illness by carers, Report of the Working Party of the RCPCH, February 2002

(9) Taylor, EM, Emery, JL ‘Two year study of the causes of post-perinatal deaths classified in terms of preventability’, Archives of Diseases in Childhood 1982; 57: 668-73; see also Emery, JL. ‘Infanticide, filicide and cot death’, Archives of Diseases in Childhood 1985; 60: 505-7

(10) Green, MA ‘Time to put “cot death” to bed’, British Medical Journal 1999; 319: 697-700

(11) Limerick, S. ‘Not time to put “cot death” to bed’, British Medical Journal 1999; 319: 697-700

(12) Meadow, R. ‘Unnatural sudden infant death’, Archives of Diseases in Childhood 1999; 80: 7-14

(13) Department of Health, Safeguarding children in whom illness is induced or fabricated by carers with parenting responsibilities, Consultation Document, July 2001

(14) If the risk of one such event in a family in which the mother is over 27, does not smoke and earns a wage is estimated as 1: 8543, the risk of a second such event is 1: 8543squared, ie, 1:73m

(15) Meadow, R. ‘A case of murder and the BMJ’, British Medical Journal 2002; 324: 41-43

(16) J Stanton, A Simpson. ‘Murder disdiagnosed as SIDS: a perpetrator’s perspective’, Archives of Diseases in Childhood 2001; 85: 454-459

(17) Meadow, R. (ed) ABC of Child Abuse, third edition, 1997, BMJ: p29

(18) Southall, DP, Plunkett, MCB, Banks MW, Falkov, AF, Samuels, MP. ‘Covert video recordings of life-threatening child abuse: lessons for child protection’, Pediatrics 1997; 100 (5): 735-760

(19) Meadow, R. (ed), ABC of Child Abuse, third edition, 1997, BMJ, p51

(20) Davies, JA, ‘Unexplained deaths in infancy’, Letters, Lancet 1999; 353:9154

(21) Emery, JL, Waite, AJ, ‘These deaths must be prevented without victimising parents’, letter, British Medical Journal 2000: 320: 310

(22) Marcovitch, H., ‘Diagnose and be damned’, British Medical Journal 1999; 319: 1376

(23) ‘Need to review child protection’, The Times, 2 February 2004

(24) Craft, AW., ‘Roy Meadow’s Legacy’, Daily Telegraph, 13 December 2003

(25) After Bristol: the humbling of the medical profession, by Dr Michael Fitzpatrick