Top scientists question basis for Victoria’s baby shaking prosecutions

Top Victorian pathologists have raised doubts about the scientific evidence relied on to jail a number of men for shaking babies to death, raising serious questions about whether their convictions were sound.

Three men jailed in Victoria since 2018 for child homicide or recklessly causing serious injury have launched or are about to launch appeals that challenge the basis of shaken baby syndrome, a collection of three injuries said to point to abuse as the cause of harm to infants in their care. The existence of the syndrome has been called into question globally but remains widely accepted by Victorian police, forensic specialists and child abuse experts.

Joby Rowe, Jesse Vinaccia and Jesse Harvey have launched or are planning appeals against their convictions.

Two specialists from the Victorian Institute of Forensic Medicine – which has given prosecution evidence in a number of cases – have raised doubts about the reliability of the syndrome diagnoses, saying the same internal injuries can be caused by other medical conditions. The institute’s deputy director, David Ranson, conceded that in the absence of any other signs of trauma or corroboration of abuse, it was “incredibly dangerous” to prosecute or convict.

The controversy centres on the finding that an infant must have died as a result of violent shaking by a parent or guardian when an autopsy reveals a “triad” of internal head injuries – bleeding in the brain, retinal haemorrhage and swelling of the brain.

Scientists say the same symptoms can be a result of non-violent medical causes, including complications of otherwise silent birth-related head injuries, genetic conditions, infections and bleeding disorders.

The issue raises broader questions about the way scientific evidence is used in courts, with former prosecutors, judges and legal experts saying they are concerned that the use of disputed or untested forensic and medical opinions risks causing miscarriages of justice because of the way juries understand scientific evidence.

The Age and The Sydney Morning Herald have identified at least five recent Victorian cases of child homicide or serious assault where police or prosecutors have relied on the triad to run prosecutions. Three of the young men convicted, Joby Rowe, Jesse Vinaccia and Jesse Harvey, each of whom has been sentenced to eight or nine years in prison, are intending to appeal against their convictions.

But the challenge to shaken baby syndrome has been rejected by police, child abuse experts and medical and forensic specialists, who insist there remains strong evidence to substantiate convictions beyond a reasonable doubt.

The controversy has polarised the forensic, medical and legal communities in the emotive context of babies who have died, amid fears that casting doubt on past convictions could re-traumatise families and potentially allow some abusers to escape detection.

‘Incredibly dangerous’

Last year, Vinaccia became the first person to file an appeal challenging his shaken baby syndrome conviction.

In 2019 he was found guilty of child homicide over the death of his girlfriend’s 3½-month-old son. Under questioning, Vinaccia told police he might have put the boy, Kaleb, down into his cot “pretty rough” and his treatment could have been “a bit bouncy and stuff”. The appeal claims the confession was unreliable, the conviction unsupported by medical evidence, and that police and experts failed to consider that the injuries could be accounted for by Kaleb’s pre-existing medical condition, which had led to him being hospitalised with swelling in his brain.

Rowe was found guilty of child homicide in 2018 over the death of his three-month-old daughter, Alanah. “Violent shaking with or without impact on a soft surface” was found to be the cause of death based on her internal injuries, with experts testifying there was no other “reasonable explanation”. Rowe denied mistreating the child.

Harvey was convicted of recklessly causing serious injury to his seven-week-old son Casey in 2019. He claimed he did not shake or hit Casey, but said the baby bumped his head on the edge of a couch as Harvey sat down. The medical evidence held the child’s internal injuries were equivalent to a 10-metre fall or high-velocity motor vehicle collision.

The Victorian Institute of Forensic Medicine is the independent agency whose expert testimony has been used by law enforcement authorities to launch a spate of recent prosecutions. Its deputy director, Professor Ranson, told The Age and the Herald that the science underpinning the shaken baby diagnosis was now more contested than it was a decade ago, raising questions about the reliability of prosecuting “triad-only” cases.

“That reliance on a single piece of evidence is always dangerous. Whether you like the triad issues or don’t like the triad issues, it’s not the be-all and the end-all. It behoves the entire legal system to say giving weight and relying solely on a single piece of evidence is incredibly dangerous,” Professor Ranson said.

Professor David Ranson of the Victorian Institute of Forensic Medicine.Credit:Eddie Jim

Dr Linda Iles, head of the institute’s forensic pathology services department, is one of two key experts in Victoria who have been called by the prosecution to testify in numerous “baby shaking” trials. Her evidence has helped secure a number of convictions.

However, she told The Age and the Herald that triad-only cases were weak because the evidence is only “indirect”. Despite this, the three symptoms together were the accepted medical and forensic basis for concluding that a baby had been abused and were being used increasingly to secure convictions.

“For a long time, fatal triad-only cases weren’t taken to criminal prosecution. I don’t know what has changed for that to start up again. But I’m only privy to the medical evidence in these cases,” Dr Iles said.

While the validity of shaken baby syndrome has been contested in the medical, scientific and forensic communities internationally for decades, this is the first time the highly regarded Victorian forensic medicine institute has publicly called into question the way the diagnosis is being used in criminal cases in Victoria.

Dr Linda Iles says triad-only cases are weak because the evidence is “indirect”.Credit:Eddie Jim

Another expert who has given evidence in criminal trials, Joanne Tully, the deputy director of the Victorian Forensic Paediatric Medical Service at the Royal Children’s and Monash Children’s hospitals, declined to be interviewed.

Dr Tully’s evidence helped convict Vinaccia and Rowe, and outside court she has been critical of those who question the diagnosis, saying in a 2019 presentation that “high-quality evidence exists but the area is also plagued by non-believers”.

The ‘consensus’

Most cases of child abuse where the “triad” appears are obvious – examinations by paediatricians and pathologists find other indicators of abusive handling such as bruises, lacerations or broken bones.

These cases are often corroborated by a history of abuse in the family, testimony or confessions by offenders.

The controversy arises in cases where triad injuries are found without any other evidence of trauma. In these cases the main evidence base is a series of international studies drawn from the confessions of convicted abusers describing how they injured or killed a child.

In criminal prosecutions, Dr Tully has described this “confessional data” as the “gold standard”.

”One particular [study] questioned obviously a number of individuals who had said they’d done this and who’d been convicted of doing this, and all of those said that the forces applied would be described as violent,” she testified in a 2019 trial.

The American Academy of Paediatrics said there was “no legitimate medical debate among the majority of practising physicians” as to the validity of shaken baby syndrome, also known as abusive head trauma. The controversy “appears to be in the legal system and the media”.

Dr Joanne Tully, deputy director of the Victorian Forensic Paediatric Medical Service at the Royal Children’s and Monash Children’s hospitals.

In Australia, many forensic paediatricians and pathologists broadly support the 2018 “consensus statement” of an international consortium of experts that said there was “no controversy concerning the medical validity of the existence of [abusive head trauma].”

Theory of the crime

But a growing number of scientific and legal critics argue this confidence is misplaced. They say shaken baby syndrome has been effectively reverse engineered from confessions to injuries in a way that cannot be tested or assessed for accuracy.

Gary Edmond, director of the University of NSW’s expertise, evidence and law program, said the fact there was no requirement in the criminal law system that expert opinion be valid or reliable meant juries were routinely exposed to speculative, unproven opinions that had the veneer of science but are not actually scientific. Professor Edmond’s comment did not refer to any specific baby shaking prosecutions.

“People shouldn’t be able to express opinions in criminal cases, especially experts testifying on behalf of the state, unless the underlying techniques have been demonstrated through rigorous evaluation to actually work,” he said.

“You need to have validation studies, you need to have indications of error or statistics to support the claims. How often do these experts make mistakes? Is it one in five or one in 100? That matters. If you can’t tell us that, we can’t evaluate your evidence rationally.

A highly influential 2016 study by the Swedish Agency for Health Technology Assessment found there was “limited scientific evidence that the triad and therefore its components can be associated with traumatic shaking”.

Asked about the lack of reliable validation studies in this area, Victorian Institute of Forensic Medicine pathologist Dr Iles conceded to The Age and the Herald: “There is a problem here.

“Because those things [shaking baby experiments] can’t be done, the evidence base is going to be indirect. All I can say is it has to be something in excess of normal handling because we don’t see children with this catastrophic presentation very often. And that’s really the limit of what we can say because we don’t have any other modelling to go by.”

Dr Iles, who for legal reasons has refused to discuss individual cases in which she has been involved, acknowledges these limitations make triad-only cases weak.

“But it is my responsibility that if I’ve seen this pattern of injuries as inflicted head trauma, and I can find no other cause for this child’s brain injury, then I have to call it based on what I have. I can’t say that it’s undetermined.

“That’s essentially how the diagnostic paradigm works. If you know that this is a pattern of traumatic brain injury and you cannot find an alternative cause, then you’re obliged to say what the diagnostic paradigm allows you to say.

“In terms of the weakness behind the evidence base, I’m very clear about that. Ultimately it’s for
the court to decide.”

However, this raises serious concerns among some in the justice system who fear it is not clear that juries understand the distinction between reasonable medical certainty and “beyond reasonable doubt”. One senior Victorian judge, who spoke on condition of anonymity because they were not authorised to speak publicly, said using the triad to infer guilt beyond reasonable doubt was a “really dangerous exercise”.

Solicitor Michael Nott said it would be “extremely difficult” for the average juror or member of the public to assess complex medical evidence, especially for a highly emotive crime such as the murder of a child.

His client, Nicholas Baxter, was convicted of manslaughter in a shaken baby case by a Queensland jury in 2017 – a decision that was overturned on appeal in 2019. In April Mr Baxter was acquitted in a judge-only trial where the medical evidence was found to be circumstantial and based on a “chain of inferences” that could not be proven beyond reasonable doubt.

“As soon as the community, or a jury, sees a baby has died, there could be a thought that someone has to be held responsible … You’ve got to have cogent, proper evidence. If you’ve got no more than a basic theory about the triad, then what have you really got?” Mr Nott said.

Beyond reasonable doubt

Victoria Police’s homicide squad declined to comment on how it investigates suspected shaken baby cases or decides which are strong enough to refer to the Office of Public Prosecutions. But disquiet has been growing inside the Victorian Institute of Forensic Medicine over recent decisions by police to run criminal cases that rely heavily on triad-only medical findings.

Professor Ranson, who declined to speak about individual cases, said the institute provided information and analysis but it was “not the decision-maker” when it comes to whether to prosecute. He acknowledged there had been tension between its experts and the police.

“There are heaps of cases where we stand up and say to the police, ‘I’m sorry, we’re never going to be able to say that. I know you’d like us to, but we’re not going to do it,’ ” he said.

Dr Lydia Garside, a child protection paediatrician at Sydney Children’s Hospital, Randwick, said communicating the diagnosis in legal settings could be “tricky” but there was no doubt about its validity from a medical perspective.

“We do have to make assumptions and presumptions in clinical medicine. You can apply pure science principles to clinical medicine but you’re always going to be disappointed because there are so many variables. But there is a lot of clinical medicine that can’t be pure science. If there was doubt we wouldn’t call it abusive head trauma.”

Veteran prosecutors caution that the decision to bring cases to trial involves more than a medical or forensic opinion.

“Don’t forget that if any pathologist or expert gives an opinion, they are subject to cross-examination. It’s open to scrutiny, it’s open to be contradicted by another expert giving a contrary opinion. If we as the prosecution can’t exclude that then the prosecution fails,” senior crown prosecutor Raymond Gibson, QC, said.

“We have an interest in running trials that are scrupulously fair. We’re not interested in running cases that rely on dodgy science.”

Start your day informed

Our Morning Edition newsletter is a curated guide to the most important and interesting stories, analysis and insights. Get it delivered to your inbox.

Most Viewed in National

From our partners

Source: Read Full Article