First Published at openinquiry.nz August 16, 2023

Felicity Goodyear-Smith

Child abuse paediatricians are doctors trained in diagnosing child abuse. They advocate for abused and neglected children and for programmes to prevent child mistreatment, and they consider whether conditions bringing children into hospital might have been caused by abuse. However there are situations where over-zealous paediatricians, in the desire to protect children, wrongly equate specific physical findings as evidence of abuse, with devastating consequences to children and their families.

Physical indicators

In 1983 a hymenal opening greater than 4 mm in pre-pubertal girls was said to be ‘indicative of sexual abuse’ (Cantwell 1983). This led to the routine examination of the hymens of young New Zealand girls in children’s camps and hospital wards, resulting in children being removed from their families and bewildered fathers being charged with rape when this ‘sign’ was found positive. Eventually the claim was shown to be flawed. A number of studies demonstrated that the hymenal diameter ranged from 1 mm to over 1 cm in pre-pubertal girls (for example see McCann Table 7). Diagnosing sexual abuse based on hymenal opening size quietly ceased.

In 1986 the sign of reflex anal dilatation as an indicator that a child had been victim of buggery led to over many children being removed from their parents across the world, including over 120 children in Cleveland, the United Kingdom. In 1988 an inquiry by Lord Justice Butler-Sloss discredited the use of this test to diagnose sexual abuse. By then many children had been unnecessarily placed in foster homes and fathers wrongly charged with rape.

In 2007 10-year-old Charlene Makaza, a recent immigrant from Zimbabwe, was admitted to Christchurch Hospital fatally ill. Her uncle George Gwaze was twice charged and twice acquitted of her rape and murder. Doctors misinterpreted Charlenes’s anal findings as trauma, where in fact these were the result of the florid, previously undiagnosed HIV infection and overwhelming sepsis from which she suffered, and which led to her death.

While child abuse is no longer diagnosed on the basis of the physical signs outlined above, I am unaware of any paediatricians involved who have ever admitted their errors, or apologised for laying unfounded criminal charges or removing children from their parents. Further, there are two contexts where medical findings are still wrongfully assumed to be caused by child abuse.

Gonorrhoea

International guidelines assert that gonorrhoea in children is definitive, or nearly always definitive, evidence of sexual abuse, leading to children being placed in care and alleged perpetrators prosecuted. Child abuse paediatricians serving as expert witnesses for the prosecution regularly state that after the neonatal period, this infection can only be contracted from infected mucous membrane to mucous membrane. However there is overwhelming evidence that gonorrhoea may be acquired either sexually or from non-sexual transmission, for example from contaminated fingers, bathing in contaminated water or via fomites such as wash cloths and towels.

Although it is vulnerable to drying, Neisseria gonorrhoeae thrives in moist warm conditions and can be cultured from damp cloths after two or more hours (eg see Alausa et al). A number of accidental cases of infection are described in the literature, such as a throat infection in a young boy who inadvertently ate chocolate agar from a culture plate positive for gonorrhoea, or gonococcal conjunctivitis in a laboratory worker after his contaminated facial mask strap struck his eye.

Pre-pubertal girls are very susceptible to gonorrhoeal infection of their vulva and vagina. However the effect of oestrogen after puberty means that in adolescents and women the vagina does not get infected, but only the internal genital organs such as the cervix and uterus (hence not likely to get infected from non-sexual transmission). The evidence is overwhelming that vulvovaginal gonorrhoea in pre-pubertal girls can result from non-sexual transmission. Prior to the advent of antibiotics, epidemics of vaginal gonococcal infection would sweep through children’s wards and orphanages. Thermometers, enema nozzles, examination gloves, nurses’ aprons, nappies, towels and bedding have all been implicated as agents of transmission, as has communal bathing of children. Epidemics were only curbed by the introduction of strict infection control and quarantine measures.

There are also many reported cases of childhood gonorrhoeal infections in households, especially in circumstances of poor hygiene, shared towels and warm damp bathrooms. While some of these cases may have acquired the infection sexually, in some circumstances the infection in different children may be in the eye, throat or vagina, depending on which tissue happened to come in contact with an infected towel or flannel. There are also epidemics of non-sexually transmitted gonococcal conjunctivitis in children in outback Australia, and in rural communities in Africa, in communities where there is insufficient water supply and poor hygiene. These outbreaks are likely to be transmitted by contaminated fingers and wipe cloths, and by flies.

Despite this large and diverse body of literature, prosecution and family court experts still insist that gonorrhoea can only be transmitted sexually by mucous membrane to mucous membrane.

Shaken babies?

Another example is the shaken baby issue. Young children presenting with bleeding on the brain are assumed to have been abused by their caregivers. The ‘triad’ of brain swelling (encephalopathy). bleeding in the eyes (retinal haemorrhage) ,and bleeding under the lining of the brain (subdural haemorrhage), is considered diagnostic of shaken baby syndrome (SBS) – now called abusive head trauma (AHT). This has led to many cases of criminal convictions (including death penalty sentences in the United States) and children being removed from their families. A recent multidisciplinary academic textbook (see https://shakenbaby.science/) challenges the scientific reliability of these medical determinations of abusive head injuries in young children.

There are no documented, independently witnessed shaking events that have resulted in the findings associated with AHT. Documented cases of abusive shaking of infants have shown no signs of retinal or subdural haemorrhage. A large-scale systematic review in 2017 concluded that there is insufficient evidence for ‘shaken baby syndrome’ and therefore AHT should not be diagnosed on the basis of the presence of the triad (subdural and retinal haemorrhage and encephalopathy).

Biomechanical analysis of infant shaking shows that shaking alone cannot produce the triad. Babies have relatively large heads, weak muscles and elastic ligaments. Shaking that is sufficiently violent to cause intracranial haemorrhage would be expected to first cause damage to the neck or spine, but no neck injury is seen in cases defined as AHT. Conversely, infants who suffer severe whiplash in road traffic accidents demonstrate fractures or dislocations and nerve root injuries in the high cervical spine, but subdural and retinal bleeding and cerebral swelling are not described in these babies. Biomechanical testing further demonstrates that shaking poses a much greater risk of injury to the neck than the head.

Brain swelling from any cause increases pressure within the skull, which can lead to bleeding into the dura. As the eye is essentially an extension of the brain, increased intracranial pressure it can also lead to retinal bleeding. Hypoxia (lack of oxygen) as well as trauma can cause brain swelling and the resulting haemorrhages. Many natural conditions, including infections, metabolic disturbances, immunological diseases, skeletal diseases and vascular malformations, can lead to these hypoxic findings. Detailed microscopic studies of the brains of infants diagnosed with AHT find that the majority do not have torn nerve fibres (the assumed mechanism of brain damage in these cases), but predominantly have hypoxia – a failure of oxygen supply. Non-abusive accidents such as short falls from a couch or bed can also result in the ‘triad’ signs, despite abuse paediatricians’ claims that these findings only result from intentional trauma.

Radiological imaging can identify bleeding in the brain and give rise to a differential diagnosis, but cannot determine whether the cause was trauma. Radiological findings need careful correlation with pathology to reach an accurate diagnosis. The lack of a feedback loop in cases of suspected abuse means that false positives are not identified. When doctors make the error of assuming that a child has been abused based on the presence of an unexplained subdural haemorrhage, they are unable to learn from their mistakes. When findings wrongly assumed to represent abuse are then repeated in the literature as ‘specific’ for abuse, mistakes are promulgated.

Closed inquiry

Speaking out about misinterpretation of physical signs indicating child abuse is a very unpopular stance, and leads to attempts to silence or cancel the speaker. This includes threats to professional registration and employment, media attacks, and publication censorship. The attacks are often ad hominem, rather than any serious critique of the evidence. Those raising these issues are accused of making the world safe for rapists, or being an apologist for paedophiles. Scientific challenges in court may defined as deliberately misleading and dishonest, and may even lead to perjury charges. This has a chilling effect on other expert witnesses wishing to act for the defence. The consequence is that the assumption that gonorrhoea or the finding of the ’triad’ in a child is definitive evidence of child abuse, is seldom ever challenged in our court rooms. Of course, critics challenging the orthodoxy are not alleging that true sexual and physical assaults do not happen, nor are they denying that these actions are unacceptable and cause immense harm. However false allegations are also very damaging, and can destroy both the accused and their families, including the accuser.

In the attempt to charge and convict sexual offenders at all cost, the principle of innocent until proven guilty has been eroded. The basic forensic evidence-gathering principles of objectivity and neutrality have been seriously undermined by advocacy and the therapeutic practices of validating and supporting people deemed to be victims. The irreconcilable conflict between evidence-gathering and therapy is not understood – it is not possible to serve Hippocrates, the Healer, at the same time as Hammurabi, the Law-giver. Advocacy is not justice.