top of page
  • Writer's pictureChris Brook

Independent Science vs Motivated Advocacy: why can’t medical doctors spot the difference?

Updated: Nov 16, 2019

The Swedish Agency for Health Technology Assessment and Assessment of Social Services, called the SBU, was established in 1987 and became a governmental agency in 1992. This makes it one of the oldest medical assessment organizations in the world. The SBU is headed by a 15 person Board of Directors, representing key organizations, both in the Swedish healthcare system and in social services. Scientific Advisory Committees, each comprising up to 10 persons, provides specialist expertise. SBU’s task is to critically review the scientific basis of methods used in health care and to evaluate their benefits, risks and costs.

In 2014, the SBU appointed a panel of experts to review the scientific quality of the evidence base for Shaken Baby Syndrome, to advise whether SBS is a reliable diagnosis. In other words, they were tasked with assessing whether the existence of certain brain injuries could be used to conclude that a baby must have been the victim of abuse, and in particular of having been shaken. This is the Shaken Baby Syndrome hypothesis, that shaking can be “diagnosed” by reference to particular brain injuries. Over more than two years, the expert group formulated their study and systematically reviewed the literature; the group’s findings were then reviewed by three scientific boards within the SBU and assessed by external scientists before being published in 2016. The authors also published their findings in a peer-reviewed medical journal in 2017. The SBU report found that no high-quality scientific studies support the Shaken Baby Syndrome hypothesis. They found that all studies supporting the Shaken Baby Syndrome hypothesis are of “very low quality”, carrying a high risk of bias due to circular reasoning.

The authors of the SBU had no incentive to find this way. Their job was to assess the science and they got paid the same amount, regardless of their conclusions. Indeed, it would have been far easier for them to support the Shaken Baby Syndrome hypothesis, considering the significant backlash they have had to endure since publishing their results, including personal attacks.

By contrast, the information linking shaking to certain brain injuries is primarily spread from the National Centre for Shaken Baby Syndrome. This is not a scientific institution, it is an advocacy group. They train doctors and police in identifying medical conditions that they claim are indicative of child abuse. They receive significant funding to do this work, as well as charging significant money for their training programs. They hold annual conferences that gain great amounts of sponsorship and charge significant money for attendance. They train more people at the conferences, and primarily provide pseudo-scientific information, non scientific information wrapped up as science (see my previous blog post). They train medical doctors how to testify in order to gain convictions, how to respond to common arguments raised by the defense.

One of the reasons that the National Centre for Shaken Baby Syndrome receives so much money, and has its services called upon so commonly, is because of the high number of convictions that occur for abusive head trauma. The more convictions, the more funding, the more funding the more medical doctors they can train in how testify in order to gain convictions.

Now to be clear here child abuse does occur, no one is saying it does not. And it is important work to give parents and communities skills to prevent child abuse. And prevention programs are to be encouraged, as should convictions for child abuse when it occurs. But to be most effective and efficient, this work should be guided by evidence, by research, by science.

The National Centre for Shaken Baby Syndrome refused to accept the findings of Swedish study, presumably because their whole institute was under existential threat. I am sure they justify it in their own minds another way. It is well established in psychology that people are not so good at identifying their motives or understanding the reasons for decisions. Largely, it is to do with the very well established psychological notion that people are extremely bad at changing their minds or admitting they are wrong, particularly when their life is invested in a particular viewpoint. People then succumb to confirmation bias in order to maintain their belief.

Is it really this simple? Can so many medical doctors really be just following along with the crowd, and the teachings of an advocacy group? And ignoring the science? Aren’t doctors really smart? Surely they cannot be persuaded in this way?

Of course doctors are very smart. But cognitive biases, and modes of forming beliefs that are not based on rational evidence, do not discriminate by intelligence. We all have them. As Tolsty said “The most difficult subjects can be explained to the most slow-witted man if he has not formed any idea of them already; but the simplest thing cannot be made clear to the most intelligent man if he is firmly persuaded that he knows already, without a shadow of doubt, what is laid before him.”

And further, medical doctors are not generally trained in science. Sure they take some science classes, but that is different to working as a scientific researcher, and doing a PhD in science which best described as an “apprenticeship”. Doing a medical degree, or even a science degree, or a masters in forensic science, does not capture that training. So I am not surprised by the unscientific evidence given by medical doctors that is leadings to severe over-diagnosis of child abuse.

What really concerns me is the unscientific evidence given by members of the Victorian Institute of Forensic Medicine. The VIFM should be prioritizing the use of science, and should be able to distinguish science from advocacy. I mean, they may have some reason for questioning the findings of the Swedish study. If so, those should be outlined clearly. But as it stands they are going much further than raising questions about the findings. They are dismissing the study as not even constituting a reasonable doubt as to the link between certain brain injuries and shaking. As though the study and it’s authors were unreasonable. Alarm bells should be ringing at the upper echelons of the VIFM and if they are not, questions need to be asked regarding the scientific credibility and judgement of the whole institute.

Further, the Victorian Forensic Pediatric Medical Service also refuses to accept the findings of the Swedish study. This service is part of the The Royal Children's Hospital, Melbourne. It provides “advice and secondary consultations for Victorian Health professionals, Police and Child Protection Practitioners”, “has a leadership role in establishing standards, protocols and guidelines in relation to forensic paediatric medicine in Victoria” and “is a leading Victorian centre for teaching and research in the medical aspects of child maltreatment.” The Victorian Forensic Pediatric Medical Service has state-wide responsibilities in relation to forensic medical assessments and care for abused, assaulted and neglected children.

The VFPMS certainly does a lot of important work. Unfortunately, however, a lot of their advice and the standards they advocate are simply not based on science. This is hugely concerning as their teachings are widespread and engrained in our system. The result is the over-diagnosis of abuse, and the conviction of innocent people. Expect to see more cases of wrongfully convicted appear on this website in the coming months. People convicted on the basis of junk science, just like Joby Rowe. I have already been contacted by three more people claiming they were wrongly convicted of shaken baby syndrome, and I am assessing the evidence.

229 views0 comments

Recent Posts

See All

An Anti-Science conference

The National Centre on Shaken Baby Syndrome conference is being held in Sydney this week. The National (U.S.) Centre on SBS is a well organised advocacy group that can take responsibility for thousand


bottom of page