In November 1999, a solicitor named Sally Clark was found guilty of the murder of her two infant sons and sentenced to life imprisonment. Three years later, her conviction was overturned on her second appeal, but it was too late for Mrs Clark. She died of alcoholic poisoning in 2007, four years after her release. In response to this miscarriage of justice, the Attorney General ordered a review of similar cases, including those of Donna Anthony, Angela Cannings and Trupti Patel.
Two key factors in overturning Sally Clark’s conviction were that evidence of a bacterial infection had been known to the prosecution’s pathologist, but not disclosed to the defence, and the nature of the statistical evidence presented by an expert witness, Professor Sir Roy Meadow, a paediatrician and former Professor of Paediatrics. Meadow claimed that the chances of two children in a family like Clark’s dying of Sudden Infant Death Syndrome (SIDS) were 1 in 73 million. Meadow had calculated these odds by squaring the chances of one child dying (1 in 8,500). Statisticians have pointed out that since SIDS deaths are likely to have genetic and/or environmental factors in common, the probability of a second child in the same family dying of SIDS is increased, not decreased. They have also pointed out double murder is even less prevalent than double SIDS. After the case, the prosecution pathologist was found guilty of serious professional misconduct and Roy Meadow was struck off the medical register, only to be reinstated on appeal a year later.
Statistical anomalies aren’t Roy Meadow’s only claim to fame. He’s also responsible for describing Munchausen’s Syndrome by Proxy (MBP or MSbP). Munchausen’s Syndrome (that is, without the proxy) had first been described in 1951 by Richard Asher, an endocrinologist and haematologist (and the father of Peter, Jane and Clare Asher). It’s a supposed psychiatric disorder in which the patient fabricates the symptoms of an illness in order to get attention and/or sympathy. In MSbP (that is, with the proxy) a parent or carer exaggerates or fabricates symptoms in someone else – MsbP usually involving mothers fabricating symptoms in their children.
Those who have read my blog on autism will be familiar with the some of the theoretical problems involved in defining syndromes and psychiatric disorders. Syndromes often vary considerably between individuals and frequently it’s not clear whether or not a psychiatric disorder actually exists as a distinct disorder – in other words, patients might be experiencing similar symptoms, but it doesn’t follow that the same thing causes those symptoms in every case. Not surprisingly, the term Munchausen’s Syndrome by Proxy has largely been replaced in the UK by the concept of Fabricated or Induced Illness by carers (FII). In the USA, the term ‘factitious disorder by proxy’ is often used.
In order to explore the idea of FII I’ve focused on an NSPCC Research Briefing by Anne Lazenbatt, NSPCC Reader in Childhood Studies at Queen’s University, Belfast and Julie Taylor, Professor of Family Health at the University of Dundee, and also Head of Strategy and Development with the NSPCC. [The briefing no longer appears to be available on the NSPCC website but can be downloaded here.]§
What is FII?
The briefing describes FII like this;
FII occurs when a caregiver…misrepresents the child as ill either by fabricating, or much more rarely, producing symptoms and then presenting the child for medical care, disclaiming knowledge of the cause of the problem. (p.1)
The authors make it clear that there’s been a good deal of debate about the nature of FII, and they point out that it isn’t a clear-cut construct. FII is seen as a ‘spectrum’ (p.4) and The Royal College of Paediatrics and Child Health (RCPCH) introduced the term FII in 2001, and according to the briefing ‘helpfully offer five examples of behaviour across the FII spectrum’. They are (p.5);
1. Simple anxiety or over-interpretation of trivial symptoms.
2. Child’s symptoms are misperceived, perpetuated or reinforced.
3. Carer actively promotes sick role by exaggeration, fabrication or falsification.
4. Carer suffers from psychiatric illness.
5. Child has a genuine and unrecognised medical problem.
Note that according to the briefing, the RCPCH includes ‘simple anxiety or over-interpretation of trivial symptoms’ and ‘a genuine and unrecognised medical problem’ as part of the spectrum of FII. In fact the RCPCH Guide describes the spectrum as a ‘spectrum of cases where FII concerns may arise’ (p.8), which puts a rather different slant on things. The authors of the briefing do at least comment that ‘the extremes are useful to note and should be borne in mind throughout’. Indeed.
Prevalence of FII
The briefing then goes on to look at the prevalence of FII. Estimating the prevalence is challenging, given the width of the FII spectrum and the heated debate about its definition, or even whether it exists at all. In these circumstances, the authors’ comment that ‘epidemiological studies used to demonstrate prevalence rates are fraught with methodological difficulties’ is something of an understatement, but they have a go. Here are some of the estimates reported (pp.4-5):
• 0.5 per 100,000 children
• 2.8 cases per 100,000 children per year
• 89 per 100,000 children over two years
• 97 new cases of FII in two years
• 451 cases from many different countries
The definition of what constituted a ‘case’ varied widely.
Despite the huge variations in estimates, the briefing emphasizes the view of some professionals that FII is more common than might first appear. I thought it might be instructive to compare the prevalence of FII with that of disorders that might fall into the category of ‘genuine and unrecognised medical problem’.
• Ehlers-Danlos syndrome: (all types) 20 per 100,000*
• Fragile X: 25 per 100,000
• Coeliac disease: 1,000 per 100,000
• Autism: 1,000 per 100,000
• Lactose intolerance: (N. Europe) 5,000 per 100,000
In other words, even the highest estimates for FII are lower than for several medical conditions practitioners would have heard of, but which frequently go undiagnosed for years. But my search of the NHS Choices website could find no results for ‘under-diagnosed’ or terms similar to it, despite each of the conditions listed above being reported by researchers as likely to be under-diagnosed. The NSPCC briefing doesn’t expand on this issue, although it does warn that:
It is important to recognise that children may have genuine significant illnesses or medical conditions in addition to ones that are fabricated and/or induced. (p.10)
How is FII diagnosed?
So how is a clinician supposed to distinguish between a genuine and fabricated or induced illness? With difficulty, it seems. Here are examples of warning signs of FII (p.9):
• The child has repeated and unexplained illnesses or symptoms.
• The child has unexplained multiple illnesses or symptoms.
• The child’s supposed symptoms only occur when the mother is present.
• The mother appears to know a lot about medicine.
• Although the mother stays with the child all the time while he/she is in hospital and attends to him/her well, she may not appear as concerned about the child’s well being as the health care professionals who are providing treatment; in contrast she may appear overly concerned.
• The father is not involved in the care of the child, or his involvement is minimal. Note however that fathers are sometimes involved in FII.
One can understand why the NSPCC wants clinicians to be alert to the possibility that symptoms might be being fabricated or induced, and to protect children from further harm from unnecessary medical tests and procedures. On the other hand, it’s difficult to see how mothers who are single parents or whose partners are working, and whose children have genuine multiple, repeated, unexplained, intermittent symptoms are expected to approach health professionals without exhibiting the ‘warning signs’ of FII listed above.
And here are some words of advice for social workers:
It is not only health professionals who have a role in the detection of FII. Social workers play an important part and may struggle because they have little knowledge about FII, or when they suspect FII, they may not be able to convince the GP (Griffiths, 2010).
Of course that could also be because the GP has medical knowledge, which the social worker might also lack. Griffiths suggests specific points for social workers to bear in mind:
• Being honest about suspicions from the start may scare off the parent (making it difficult to gain evidence), attract undue media attention, or worse, can lead to an increase in harmful behaviour in an attempt to be more convincing.
• Consider motivation. For example, the family might be having financial trouble and fabricating or inducing an illness may entitle them to extra welfare benefits.
• Verify the personal histories of family members, as lies may have been told (for example, that one of the parents has a medical background).
• Remember that some parents may be extremely manipulative and convincing. They may be middle class and they will know how to invoke complaints procedures. (pp.10-11)
The last sentence is especially ironic. Interestingly, the Griffiths reference isn’t, as you might expect, to a research paper but to an article in Community Care magazine. I couldn’t find out who Julie Griffiths is, but she’s written a number of articles on the Community Care website. Her advice is summed up by the authors of the briefing as follows;
Although Griffiths offers cautionary advice, the point for social workers is ultimately the same as for health professionals: it is crucial to do the detective work. (p.11)
Yes it is, absolutely, but the detective work is required to find out the underlying cause of the child’s presenting symptoms, not to prove suspicions of FII correct.
Tragically, there’s no doubt that some parents do harm their children and present the consequences of that harm as a medical condition. Controversial covert video surveillance at the Royal Brompton Hospital, carried out by another doctor with expertise in FII, Professor David Southall (also struck off the medical register and then reinstated) and numerous cases have made that clear. However, the content of the NSPCC research briefing implies that some serious questions need to be asked about how FII is being approached.
How valid is FII?
1. The Royal College of Paediatrics and Child Health’s definition of FII is so broad that accurate determination of the prevalence of FII is impossible.
2. Having said that, even the highest estimates for the prevalence of FII are much lower than the prevalence of genuine medical disorders that are frequently under-diagnosed. It might have been helpful for the research briefing to have provided a checklist of commonly under-diagnosed disorders for medical practitioners to rule out, prior to making a default assumption that FII is involved.
3. I don’t have figures for the prevalence of under-diagnosis of genuine medical conditions, although one study of coeliac disease showed an under-diagnosis rate in four European countries between 76% and 94%. Given the range of disorders involved and their estimated prevalence, it’s possible that children might be at significantly greater risk from undiagnosed medical conditions than from abuse or neglect from parents and carers. For obvious reasons the NSPCC sees its task as drawing attention to active abuse or neglect, but if our focus of concern is the health and wellbeing of children, systems neglect is an important issue that shouldn’t be overlooked.
4. The FII coin has two sides; one is the ability of carers to fabricate and induce illness in children without this being detected, and the other is the under-diagnosis of medical conditions. Both sides of the coin call into question the diagnostic capacity of health services. Clearly, it’s impossible for GPs to know about all illnesses, but rather than raise the profile of FII, it might make more sense to improve access to specialist medical knowledge, in order to rule out genuine medical conditions and increase detection of fabricated symptoms.
What’s FII got to do with child development theory?
FII is assumed to be both a medical disorder and a form of child abuse. However, the NSPCC briefing presents FII within a theoretical framework that will be familiar to readers of previous posts. Firstly, although there’s some discussion about its definitions and prevalence and the briefing does allude to FII being manifest at the systems level (p.1) the FII construct is framed in terms of individual carers’ behaviour towards individual children, even though children can come to harm in a variety of ways that don’t involve carers. Also, carers’ behaviour can be influenced by many factors, including economic policies and flaws in the design of the health, education and social care systems. In other words the FII model itself is framed in terms of the behaviour of individuals, rather than being set in the broader ecological context described, for example, by Uri Bronfenbrenner.
Secondly, even in cases where carers are guilty of abusing children, attempts to explain FII appear to have focused on psychoanalytic, attachment and ‘cycle of abuse’ models – an implicit assumption that the carers themselves might have been abused (p.7). What receives less attention is the possibility that the carers might have a neurological problem that results in delusions, lack of impulse control or fabrication that itself might require medical investigation. Alternatively, some carers might have a vested interest in harming their children – qualification for additional benefit payments is mentioned in the briefing. Although this motivation might be perverse, it isn’t necessarily the result of a delusional illness or of childhood abuse.
The briefing cites, uncritically, a definition of FII that includes over-anxiety and undiagnosed medical conditions, a list of warning signs for FII that include responses that would be common in non-abusive parents and carers, and frames FII predominantly in terms of a ‘cycle of abuse’ model involving only the behaviour of individuals. It overlooks systems factors such as the training of health professionals and the demise of the family doctor who would be familiar with a patients’ family history.
I contacted one of the authors of the briefing for her comments. It was clear from her response that the remit of the briefing and the limit on length made a complete analysis of FII impossible. In addition her area of expertise didn’t qualify her to comment on medical conditions that might be mistaken for FII and that the primary purpose of the briefing was to raise awareness of FII amongst health professionals. An analysis like the NSPCC one doesn’t provide an exhaustive overview of the FII research literature, obviously. However, this briefing is written for professionals and, because it’s endorsed by the NSPCC it is, worryingly, likely to be taken as authoritative despite its shortcomings.
Lazenbatt, A. & Taylor, J. (2011). Fabricated or induced illness in children: A rare form of child abuse?, NSPCC.
Royal College of Paediatrics and Child Health (2009). Fabricated or Induced Illness
by Carers (FII): A Practical Guide for Paediatricians.
§ A set of slides on FII by the authors of the briefing is available online at http://www.baspcan.org.uk/files/Lazenbatt%20Ann%20F2.1%20Mon%2010.30.pdf
It highlights many of the problems with the FII concept.