Will Bordell
by Will Bordell


This piece was originally published by The Justice Gap.


A major review of shaken baby syndrome has called into question the science that links the three symptoms to deliberate harming. The study, conducted by an independent Swedish agency that assesses healthcare interventions, has found ‘insufficient scientific evidence’ to assess the accuracy of the so-called ‘triad’ of symptoms – that is, brain swelling, intracranial bleeding, and bleeding in the retina – in identifying traumatic shaking; and that there was ‘limited scientific evidence’ that the ‘triad and therefore its com­ponents’ could be associated with such shaking. You can read the report here.

Despite long-standing concerns about the validity of the science behind the shaken baby syndrome, it has become increasingly difficult for paediatric experts to challenge a diagnosis in courts. Last month, the Justice Gap reported on how British defendants were struggling to find medical experts in the UK prepared to testify on their behalf after the world-renowned neuropathologist, Dr Waney Squier was struck off by the disciplinary arm of the General Medical Council in March last year.

In their report, the Swedish agency for health technology assessment and assessment of social services (SBU) looked at studies of isolated traumatic shaking in children aged 12 months and younger. Assessing evidence from 30 scientific papers, it claimed that only two of them had ‘moderate quality’. All 30 papers included cases where traumatic shaking had either been witnessed or confessed.

Underwhelmed by the vast majority of studies they reviewed, the ‘methodological flaws which characterise this field of research’ proved especially troubling for the report’s authors. ‘The research field is complex,’ the report suggests, ‘but this does not excuse, for example, circular reasoning and inadequate presentation of data collection.’ Even in the two French studies that were of ‘moderate quality,’ the researchers expressed concerns pertaining to poor documentation of precisely what the suspects had confessed to and under what circumstances.

The report goes on to warn scientists that they must conduct studies in this field with an open mind. It argues that preconceptions often lead to circular reasoning and ‘a high risk of bias’. Scrutinising seven other systematic literature reviews that are often cited as authoritative, the researchers conclude that all seven are of ‘low quality’ due to circular reasoning: each one is ‘based on studies in which a team considered that a child had been shaken if it presented with the triad’.

Reports in the Swedish press have suggested that a number of international agencies, including the American Academy of Pediatrics, sent a joint letter demanding that the SBU allow them to review its findings before publication. The report, first published in October last year, is now available in English.