Diagnosing Murder | Ep 2: The Science

Just a warning, this episode contains content that listeners may find distressing. I've learned some surprising things making this podcast, and one of the most surprising is that nobody really acknowledged child abuse until the 1960s. In the words of sociologist Ron Westrom, until then, the fact that parents and caregivers sometimes abused their children was a hidden event. Nobody wants to say anything until somebody else has said something. This is the nature of hidden events. Nobody wants to be the first one. Ron was a professor at East Michigan University. He's retired now, but he built his career on studying things that, for whatever reason, people close their eyes to. The truth of the matter is, none of us tend to be terribly willing to step forward and say something which is going to cause a lot of problems for somebody. And maybe for us, heroes are rare, and especially in this kind of field. Then, one day in 1961, a hero took that step. Out of his work grew medical practices, laws, and institutions that have saved perhaps millions of children from harm. But it also gave rise to an argument, a scientific and medical dispute so heated that some people now describe it as a war. The argument is over one branch of child abuse medicine, what's known as shaken baby syndrome. And while there's no doubt you should never shake a baby, the scientific question is over whether shaking causes the medical findings known as the triad, the kind of brain damage that the theory suggests it does. Decades after it was first discovered, the theory of shaken baby syndrome was the first proposed as a theory, then found its way into medical orthodoxy. That question remains the focus of bitter, angry dispute. In this episode, I'm going to tell you the story of how we got here, what the science says, and how people motivated by the best of intentions came up with the theory that seeing babies being taken from their parents, and parents being sent to jail. And it all starts with a bright young adult. American paediatrician. His name was Henry Kemp. From the newsrooms of The Age and the Sydney Morning Herald, this is Diagnosing Murder. I'm Michael Bachelard and this is episode two, The Science. Henry Kemp was a prodigy. At 34, he became the head of paediatrics at the University of Colorado. And what he saw there was plenty of evidence that some adults were deliberately hurting children. And he said, you know, there were times when you would see the outline of a paediatrician and you would see the outline of a paediatrician and you would see the outline of a paediatrician and you would see the outline of a paediatrician were caused by whiplash after their mothers shook them. And he said, oh my goodness, this could be a very bad thing and we should discourage people from doing it. And so in 1971, ten years after Kemp's big conference revelation, Norman wrote his own paper. It was just two pages long. For the first time, he proposed that there might be a relationship between shaking a baby and subdural injuries. The language of Norman's paper is thoroughly hedged. He wrote, one must keep in mind the possibility of assault in considering any case of infantile subdural hematoma. Keep in mind the possibility. It's hardly a call to arms. Ultimately, though, that's just what it became. And it was only quite recently that I realised the enormity of what has happened. What happened is that Norman's paper was the first paper to be published in the United States as the seed out of which shaken baby syndrome grew. It started when he posted his paper to his friend, a fellow paediatrician in the United States called John Caffey. And from what I can make out, Caffey was a man beset by professional jealousy. Caffey definitely felt that he'd been left out. That's Ron Westrom again. Fate gives you a chance, and if you don't take that chance, then you're going to regret it forever. And so I think this is really what happened to Caffey. Caffey had written about his life as a paediatrician, and about child abuse as far back as the 1940s, well before Kemp. But it seems he hadn't wanted to upset people too much, so he kept it low-key. That meant his paper was overlooked. And 20 years later, Kemp got all the attention. Now, with Norman Guthculture's paper in his hands, Caffey wasn't going to make the same mistake again. He wrote his own paper, which took Norman's theory and elaborated. And Caffey wrote his own paper, but Caffey didn't hold back in his advice about what parents should avoid doing. He had a long list of things, including putting babies in a bicycle or a rocking horse. He warned about bouncers, swings, seesaws, slides, trampolines, powered cradles, and something called powered vibratory training equipment in gymnasiums, whatever those are. He really didn't like anything that looked like shaking a baby. And Caffey pushed him. He pushed his theory hard. He even tried to coin a name for it, Caffey Syndrome. It never caught on. Instead, it got the name we're all now familiar with, Shaken Baby Syndrome. And under that name, it really took off. Out of Caffey and Guthculture's theory, two approaches grew. The first is educational. The developing brain of a baby is so fragile. This Australian TV ad, someone puts an egg inside the hollow skull of a doll, then shakes it. That shaking them, even for a few seconds, can cause serious brain damage. The mess it causes is unambiguous and incontestable. Babies break if you shake them. If you can't stop your baby crying, cry for help. This advertisement and public service announcements like it all over the world, were consistent with what Caffey and Guthculture had envisaged. They simply wanted to put out a cautionary note that said, parents, beware, you may be harming your children without knowing it by doing this, so let's be cautious. Don't shake your children. That's Keith Findlay. He's a retired professor of law at the University of Wisconsin. And Keith says that because they initially saw education as the point of their papers, Caffey and Guthculture were happy to just put out a hypothesis. They didn't really do much research. So it made sense. It had very little downside to embracing it because what was the remedy? Education. Playing it safe. So it was quickly embraced in medical school. But Keith's not an educationist. He's a lawyer. And as a former co-director of the Wisconsin Innocence Project, he watched as this educational approach branched out into something else in society. Entirely. By the early 1990s, the legal community, prosecutors began looking at that and recognizing this was a tool that they could use to prosecute people they believed were involved in child abuse. The way Keith tells it, as the years passed, the doctors diagnosing shaken baby syndrome and the lawyers citing it to bring prosecutions sort of amped each other up. So it's a synergistic thing where the medicine provided this mechanism that the law saw as a tool for prosecuting people and the pressures and powers of the law then put pressure on the medicine to change so that by the early 1990s they were no longer saying this was unknowing infliction of harm to saying that in the 1993 official position paper of the American Academy of Pediatrics, any parent who shook the child this hard would know it would cause harm and perhaps kill the child. That American Academy of Pediatrics paper made this hypothesis in essence settled science. By 1993, it was medical orthodoxy. And in the courts, the shaking theory was a pretty efficient package for prosecutors. It delivered them a perpetrator, usually the last person holding the baby. That person must have intended to hurt the child because it was said the violence required to cause the injuries could be equated to a car crash or a fall from a multi-story building. And you didn't need an eyewitness because the expert evidence of a doctor was enough to convince the jury. With all this in place, the number of convictions exploded. Miles Ferguson from Portsmouth is accused of shaking his baby to death. Daycare provider Audrey Edmonds was convicted in the death of seven-month-old Natalie Beard. Natalie showed signs of shaken baby syndrome. During the 1990s and early 2000s, thousands of mothers, fathers, nannies, and daycare providers were convicted of crimes related to shaken baby syndrome. I was charged with first-degree murder. I was facing 25 to life. Parents, they just lose their cool. And literally, it can take one shake. Parents lost access to their children. Many were charged with murder. In the United States, some were sentenced to death. On Thursday, Roberson is scheduled to be the first person executed in the U.S. for a murder conviction. A murder conviction linked to shaken baby syndrome. In one infamous case, a British nanny called Louise Woodward was tried for murdering the eight-month-old she was caring for. To Americans, it was like Mary Poppins was in the dock. His eyes were half closed. He wasn't focusing. And he was gasping for breath. The trial was televised, and Louise Woodward was found guilty. He's doing it. He's doing it. It's hard to know what Henry Kemp would have thought about how this syndrome evolved from an educational tool to a legal one. But I get the feeling he might have approved. Remember Ron described him as a crusader? He was in the business of righting wrongs and of dismissing shilly-shallying diagnoses that avoided hard work. He was a crusader. He avoided hard realities. I've got no doubt that crusader spirit still plays a role today in the tough way authorities pursue these cases. The intentions here are quite understandable. Admirable, even. Our system should be in the business of protecting those like babies who can't speak for themselves. But it should also be aware of overreach and of whether a medical diagnosis like this can establish wrongdoing beyond reasonable doubt. And about a decade after the prosecutions began in earnest, some people started to raise these difficult questions. And one of the first people to raise these doubts was a doctor all the way over in Sydney, Australia. I am Dr Mark Donoghue. I'm a medical practitioner. I work at Mossman Integrative Medicine. I'm not sure what else to tell you about. I'm getting old. By the early 2000s, the shaken baby thesis had well and truly reached Australia and prosecutions were coming up pretty regularly in courts all over the country. Mark's involvement came about by accident. He was asked by one of his patients, who was a lawyer, to look into a case they were running. There was a bleed into the eyes, a bleed into the subdural space, and they're being accused of violently shaking their baby. It was a pretty classic triad case. The child was born in a hospital and the child was alive, but was living in foster care. So Mark agreed to read through a huge stack of documents. And tucked away in there, he found something. All I did was look down and notice that there was a condition called steatosis where the liver is affected. It had been not noticed by anybody. Mark was surprised. Even as a GP, he knew that liver steatosis was associated with abnormal bleeding. And that by itself could explain why the baby had been affected. had suffered these hemorrhages. So Mark wrote a report. Then he gave evidence at the trial. He said being cross-examined was not much fun. There was a very high pressure. Are you absolutely certain? Would you be certain? Would you be prepared? And I said, look, it's a medical condition. I'm prepared to stand by what I see. Then something strange happened. And then the judge said, look. The judge turned to the prosecution's lawyers and said. Do you wish to divulge something now? And the answer was no, and I went through cross-examination again. Eventually, the judge stepped in a second time, saying. We should let him know. Letting me know was that one of their three experts had changed their mind and agreed with what I had found. What Mark's saying is quite shocking. One of the prosecution's own expert witnesses had changed their mind and now agreed with Mark. Perhaps liver steatosis was an explanation for the hemorrhages, but it didn't deter the prosecutor. They wanted to press on. What disturbed me about it was they were still going to give it a fair shot. Ultimately, the judge dismissed the case and the child, who'd been taken from the parents, was returned. But the whole thing left Mark feeling pretty unsettled, particularly at the prosecution's determination to secure a conviction, despite the doubts he'd raised. Had that one specialist not changed their mind and decided that that was a valid cause of bleeding and likely to lead to the findings that they saw, I suspect I would have been overridden and the person would have gone to jail. So it's always been on my mind. So Mark Donoghue, the doctor from Sydney, decided to do some science of his own. In 2003, he conducted what's called an evidence review. He searched for all the available medical literature on baby shaking between 1966 and 1998. He found more than 50 academic papers and ranked them based on the quality of the research. So did you find any evidence for the triad indicating shaken baby syndrome? I found what we would term very weak evidence, not level 1, 2, but level 3. Level 3 evidence is considered low quality. It essentially means the researcher has observed a phenomenon without really being able to test it. And the problem with level 3 studies is that they can't ever really test it. They can't ever really establish cause and effect. In this case, they don't answer the question, does shaking a baby cause the triad? Instead, what Mark found was that these researchers were drawing conclusions with high certainty based on bad evidence. And he identified, perhaps for the first time, something that's become one of the most persistent criticisms of shaken baby syndrome, circular reasoning. Now, I found this circular reasoning question a really hard one to answer. It's a hard concept to explain. So I've asked lots of experts for help. And I reckon this is the simplest way to express it. Imagine you're a researcher who wants to prove whether or not shaking causes the triad. So you go to a hospital and ask for all their files of children who've been shaken. You examine them all, and they all have the triad. Bingo. Proof. Except that the only reason these children are listed as having been shaken is because they're not shaking. The only reason they've been shaken in the first place is because they have the triad. So really, all the subjects of your study are infected with a kind of selection bias. And suddenly, your proof evaporates in a puff of logic. So what I found was that self-reinforcement, we found the following in these cases in hospital, and they were all abused. How do you know? Well, the triad was there. That's how we know. Therefore, the triad is true. And it was self-referential to the extent that virtually every paper assumed it to be true rather than proved it to be true. Mark pointed this out in his 2003 paper. And so the report was, there's not an evidence base for this. There's a strong opinion, and that opinion keeps on being reinforced by more papers saying the same opinion. The paper was published in the British Medical Journal. And perhaps naively, Mark thought it would be welcomed as part of a broader scientific debate. That's not what happened. The backlash was strong. 106 physicians wrote a letter to the editor dismissing his study. And the prosecutions continued. But when we caught up with Mark, he said his view hadn't changed. Well, to me, the idea was, oh, contribute something here. This is how weak or strong the evidence is. And instead, it kind of confirmed my bias that some people just know things to be true. And anyone who says they're not, it's more a religious battle than a science battle. And I thought that's what I ran into. On the face of it, that's a big statement, that a belief in shaken baby syndrome has become more of a religion than a science. Surely, 106 doctors writing to a medical journal have good science backing them up. So what is the state of the science behind this syndrome? And why is it so contentious? That's after the break. Now, to understand what happens to these babies, we need to understand some simple anatomy. This is Dr Jo Tully. She's one of Australia's key proponents of what she prefers to call inflicted head trauma. It's 2018, and she's addressing a room full of experts on brain injury. And this is a diagram of a baby's head with half of the skull bone removed, showing us the brain sitting within the skull. Jo is the deputy director of one of the key organisations in this story, the Victorian Forensic Paediatric Medical Service, which is based at Victoria's Royal Children's and Monash Children's Hospitals. It's this organisation that first identified Kabir and Deepika's daughter as likely being shaken. And leaving the surface of the brain, you can see in blue these small veins. And these are known as bridging veins. And these carry blood from the tissue of the brain up to a large vein that runs on the inside surface of the skull. And these bridging veins, are fixed to the inside of the skull. Jo's explaining how she says inflicted head trauma works. If you're squeamish, you might skip forward a minute or so. So let us think about what might happen to this baby when it's shaken in this manner. As the baby's shaken, the brain moves backwards and forwards and round and round inside the skull, tearing those bridging veins and causing bleeding around the brain. The small blood vessels in the back of the eyes are also torn apart. Resulting in bleeding at the back of the eyes. And you can see that in the picture on the bottom left. Those blobs are blood at the back of the eye. The brain being shunted backwards and forwards and round and round results in brain being damaged and bruised and a loss or reduction in consciousness which in a baby can be interpreted as them going to sleep. It also results in brain swelling. And this is the bit that can cause severe, severe and long-lasting damage. In a nutshell, Jo Tully is describing what practitioners call acceleration, deceleration and rotational forces. What she's saying accords with the mainstream medical view. I think that the science behind this is quite solid. This is Dr Christopher Greeley. He's the head of child abuse paediatrics at Texas Children's Hospital and a professor at Baylor College of Medicine in the United States. He's speaking at an online conference hosted by Stanford University in 2020. And so if we actually surveyed physicians, this was a paper by Narang and colleagues, and surveyed physicians, this was 682 physicians from around the country, and framed the idea of was this a real disease and identified the fact that most physicians, and I'll go here, most physicians agree that this idea of abusive head trauma is valid. You can see. Christopher says there's no debate that science backs up shaken baby syndrome. The only real controversy is whether a particular person shook a particular child. And I think that the debate is really, is this true for this specific person? That's necessarily, that is a legal debate. I believe this not to be a medical debate, but to be a legal debate with a white coat on. And I think this is true. The American Academy of Paediatrics says shaking is, quote, a significant cause of morbidity and mortality in the paediatric population, especially in young infants. And in the UK, the Royal College of Paediatrics and Child Health also strongly defends the diagnosis. In Australia, medical teaching materials say the triad is highly suggestive of abuse, and that parents shouldn't be believed when they say nothing happened. But the experiment that first cast doubt on shaken baby syndrome came way back in 1987. And it was conducted by a scientist, who actually thought her work would prove it. Her name is Anne Christopoulos. Dean Duhame, and that year she made a series of dolls to represent babies. They had wet cotton brains and flexible spines. Inside them, she put a set of accelerometers, and she got a bunch of Penn State football players to shake them as hard as they could. She then compared the forces generated with the results from those poor monkeys slammed into walls who helped inspire Norman Guthkelcher's theory. It turns out that the forces from shaking didn't come close. Duhame's paper said, Based on these observations, we believe that shaking alone does not produce shaken baby syndrome. Scientists have shaken mice, piglets and tiny lambs to try to replicate the effects claimed by shaken baby syndrome. But the anatomy's different, and the results didn't satisfy either side. This is Anne Christine Duhame talking about it. Criticisms of the science of abusive head trauma is that your models are wrong, animals are not children, we already know the answer from what we see clinically. I would say that those are arguments and it just spurs us on to try to get closer and closer to the truth. Every biomechanical study has shown that the force of the maximum shaking that can be exerted is less than a one-foot fall onto concrete. Remember, American Lawyer, has the power to do more than one thing. But many studies have also shown how unreliable confessions can be. When DNA evidence came along and allowed all sorts of previous convictions to be overturned, innocence projects around the United States found that 29% of those wrongful convictions, almost a third, had involved false confessions. But why on earth would anyone falsely confess to shaking their child? Now, think about a caregiver, a parent, a loving parent, who's just lost a child. Who's just lost a child. Here's Keith Findlay again. And they're in an interrogation scenario, whether it's by police or doctors. And the questioner, the interrogator, is saying to them, what did you do to this child? This was massive force. It couldn't have been accidental. This is the kind of thing we see with only caused by shaking, or shaking and slamming a child. The good and loving and caring caregiver is going to search their brain and say, geez, maybe I was a little rougher with him than I thought I was. Or, or, or, than I meant to be, or. Those are often taken as confessions in these cases. Other studies point out the kind of pressure that police and child protection can put on people to get a confession. They can threaten to take the child away, for example. They can also offer a shorter jail sentence in return for a guilty plea. And that reminded me of Kabir in the last episode. Remember he was told that if he confessed to shaking baby Dua, maybe, just maybe, she wouldn't be taken away. And put in foster care. And he told us he'd actually been tempted to lie. They're looking for an explanation. If you're not going to provide an explanation, they're not going to back out today. You have to provide an explanation. So at that point I was like, if you want me to lie, I'll lie. To keep our daughter with us, right? The science of this and the terminology used are constantly evolving. And these days it gets into some really complex areas. Now experts, talk about what specific kind of subdural and retinal haemorrhages they're looking for. Then there's a whole debate over whether shortfalls, say from a bed or a parent's arms, can produce enough force to create the triad. And how you can differentiate those from injuries from assault. All this has led the terminology to change. From shaken baby syndrome to abusive or inflicted head trauma. The words with or without impact are also sometimes added. It would take an entire other podcast series to explain all of this, so I won't. But I will tell you why I think it's so contentious. At stake in these arguments are the lives and well-being of the smallest human beings. Often it's whether they stay living with parents or are taken into foster care. And whether those parents go to jail for that alleged abuse. And this is important. Because doctors and scientists are not immune to the emotional stakes here. Perhaps it's for that reason that the argument over shaken baby syndrome is often not polite. In researching this podcast, we've heard people called unbelievers, denialists. I've heard the words dogma, religion, even cult. Those who dare to break with the orthodoxy are vilified and attempts are made to silence them. I'm telling you this because you can't understand the history here or the debate over the science without knowing how entrenched and emotional the two sides are. Through the months of reporting this story, we've watched documentaries, listened to conference sessions, attended online courses, interviewed multiple people. And what Keith Findlay is saying is no exaggeration. Things have got nasty. It's not much of an exaggeration to call it a war. Keith. Keith goes even further. And it's not just a war. It's like warring religions. I'll tell you, if I hear another validation based on confessions and convictions, I'm going to barf. This dispute involves every part of the science you've heard about. They have to go to confessions. And they tell us, here's what I did. Now, this is not a child. This is a piece of wood with an ant on it. The confession evidence, the biomechanics. Now, how in the world can you just forget about all of this? All of that, because somebody has an alternative hypothesis. It's a beyond reason to assume that every person who admits to shaking a baby is lying. I promised that I will not use the term, John, you ignorant slut. People have had their enemies sacked, deregistered, even prosecuted just for giving evidence. There were attempts to take away his medical license, and he was even prosecuted criminally. They're kind of a traveling group of freaks, honestly, who have this weird theory that they espouse. That is inconsistent, really, with common sense. You can ask your jury, do you think if you shook a little baby, you would hurt it, or that you wouldn't hurt it? The argument even extends to whether there's an argument at all. So if you have, make up a number, three, four hundred people who agree to the science, and one person who doesn't, does that really mean that there's a controversy? Does that really mean that there's new science? In this war, the supporters of shaken baby syndrome are outgunning the opponents. And pretty much every year, an American organization, the National Center on Shaken Baby Syndrome, holds a conference to gee up the troops. Doctors, lawyers, social workers, and child protection workers are invited. And the organizers aren't subtle about the message. At the 2019 conference, one talk was entitled, Liar, Liar, Pants on Fire, The Art of Obtaining a Confession. And another was, All Hat and No Cattle, The Controversy Surrounding Abusive Head Trauma. The National Center doesn't record these talks anymore, and they certainly won't give the tapes to journalists like us. But we have managed to obtain some. And at one particular conference, the dispute kind of reached a fever pitch. We were at the 11th International Conference on Shaken Baby Syndrome in 2010 in Atlanta, Georgia. The writer Sue Lutner was there. And Prosecutor Brian Holmgren had just given a talk on exams, examining defense testimony in shaking cases. And with each testimonial claim that they put up there, they had a picture of Pinocchio, and his nose got longer and longer. Keith Findley wasn't at the conference, but he heard about it. And they had a slide they put up, what they called it, the axis of evil. And the axis of evil listed people like Donahoe and others, the doctors who dared to speak out against them. And they labeled them the axis of evil. And at another session, they had a sing-along I just have the sing-along for you, okay? You can practice it right now. One of the pediatricians came out with a guitar, and they sang a parody version of If I Only Had a Brain. If I Only Had a Brain, from The Wizard of Oz, the words were, if I only had $10,000. No proof for them? Why should I care? I don't care. And in between, they had us, the audience, singing this da-da-da-da-da-da-da-da-da. basis for the claims and shaking cases, my opinions and demands. If I only get 10 grand. Imagine being a social worker in that room. Would you doubt, would you set yourself up for ridicule like that? That's what I, that's what shocked me about it all. I'm sitting there thinking, there are hundreds of people in this room who are going to take this knowledge back to the world and wed it to the power of the state to put people in jail who should not be in jail because nobody will be able to hear them say, you've got it wrong. I didn't shake my baby. There's no room after you've been in one of the, it was like summer camp and we are on the same side and we're going to sing along. When I started this project, I assumed doctors were scientists. I assumed that like scientists, they kept an eye on the literature and changed their views as evidence evolved. But in this project, quite a few people have told me that's just wrong. A lot of doctors don't have the time or the training to do science. Instead, they believe what they've been taught, what their colleagues just know to be true, and what they see with their eyes. Keith Findlay explains why. For many, many years, medicine was an authoritarian or eminence-based discipline. That is, young doctors or inexperienced doctors were taught this is the way it is, and they were taught not to question that. That when your authorities, your medical school, professors, your mentors in the professional world, whoever, when they tell you this is how it is, you accept it. Keith says that started changing in the 1990s when a new paradigm known as evidence-based medicine started to get traction. That's the approach Mark Donoghue took when he tried to get to the bottom of the science and faced a backlash from 106 other doctors. Mark himself doesn't think change will come soon. Deeply, we know that it takes, you know, 50 years to get a good idea in and 100 years to get a bad idea out of medicine. That's just the nature of our being a conservative medical community. But would that process be accelerated if one of this community's key authorities, if one of the founding fathers of shaken baby theory, changed his mind? Well, the year after that Atlanta conference sing-along, precisely that happened. Norman Guthkelch, the northern English doctor who first warned in 1971 that shaking could be dangerous, flipped. He was pretty much like, look, man, that was a hypothesis. We now have evidence that this can happen other ways, you know. They shouldn't be using my article like that. I mean, that was 50 years ago. That's Carrie Sperling, another lawyer. And in 2011, Carrie was the director of the Arizona School of Medicine. She was the director of the Arizona School of Medicine. She was the director of the Arizona Justice Project. And she was doing a lot of work questioning shaken baby cases. And quite by chance, she found out that Norman literally had moved in down the road. He was now about 90 years old, and he'd fled gloomy northern England for a sunny retirement in Arizona. So Carrie went to visit. He welcomed us in. I still remember him making us tea, very British, with the milk and everything. And we sat down. And we were talking, just asking him about his theory and how it came about. And then we said, well, we've got some examples. We've got people who have applied to us that claim they never shook the baby. The particular case Carrie was working on was of a guy called Drayton Witt. His son Stephen had medical problems from birth, pneumonia, fevers, vomiting, and lots of seizures. One day, when Stephen had a seizure, Drayton rushed him to the hospital. But he died. Drayton was charged with murder and given a long sentence. Carrie took up his case. And you look at these kids. Stephen was one of them. The baby in photographs before the autopsy is pristine. Not a single bruise on that child's body. And you just think, really? The physics of such is that you can grab hold of a baby and shake it. It's so hard that it's like a, you know, five-story fall or what they used to say, or a high-speed car crash, and not put a single mark anywhere on the baby. Norman was interested. He told Carrie he'd look at the papers. And it probably wasn't a week later when he called me and he said, you know, Professor Sperling, we have to get this guy out of prison. This baby did not die from shaking. This baby had an existing brain condition. Norman was astonished that his little theory from 1971 had spawned this legal juggernaut. He was just outraged. And then I started showing him all the research and people quoting him, and then coming to the conclusion of, if you see these three things, then it has to be shaking. And he's like, that's nonsense. That's not what I said. I never said, I said it was a theory. You know, I said, it's a, it's a hypothesis and they're using it to convict people. He was just livid about it. Norman agreed to appear as a defense expert for Drayton. And to cut a long story short, after a decade in prison, Drayton was freed. Carrie thought Norman's reversal would basically end the war. But. It didn't. And so I'm, I am more than perplexed. I'm, I'm just like, throw my hands up. I don't know what to do anymore. Instead of listening to Norman, the baby shaking proponents piled on him. They accused him of being senile. He was a lonely old man who got taken advantage of by blind defense attorneys. They were just beside themselves about his change of, of opinion. And it's just kind of sad what they were trying to do to his legacy, really. But if you're trying to keep an expert off the stand, that's what you do. You just attack their credibility. By this time, Norman Gutkelch was very, very old. But he clearly wasn't senile. A lot of babies have been ill. They're in a serious state of health. And he still had a. A strong grasp of the issues. And because nobody can figure any other explanation, this child is labeled as having been brutally assaulted by the caregiver. I don't think that's good enough to you. As for the case that flipped him, Drayton Witt's case. I think I used the expression in my report, I wouldn't hang a cat, on the evidence of shaking as presented. Back in Australia, this turmoil has all but passed us by. Most people aren't even aware of the controversy. But in 2021, a young man convicted of shaking his girlfriend's baby to death tried very hard to change that. He took his appeal to the highest court in the state of Victoria. And there, in front of three judges, he challenged not just the finding of guilt against him, he and his lawyers put baby shaking itself on trial. The young man's name was Jesse Venacia. Today, the former carpenter was jailed for eight and a half years after a jury found him guilty of child homicide. In a moment of frustration, he killed baby Caleb, robbing the child of a future and his mother of her only son. When Jesse lodged his appeal, he was serving his long jail sentence. A jury had found he'd shaken his girlfriend's 16-week-old baby. Now, the defence and prosecution both agreed that Caleb had been a very sick boy. He was born small and unwell, and a few days before he died, his head, which at birth was at the third percentile in size, had swollen at what one doctor described as a concerning rate to the 85th percentile. Caleb had spent three days in hospital getting treatment, and neither side at Jesse's trial suggested that any of this was the result of abuse. Then, three days after Caleb was sent home from hospital, Jesse was rushing him back. He insisted the little boy had just gone floppy and unresponsive in bed. Caleb had no bruises and no fractures, and his head was even bigger. When doctors examined Caleb, they found the triad. It was enough for the forensic paediatricians, including Dr Joe Tully, who you heard from earlier, to say it was ornithology. They were almost certain that he'd been violently shaken. At Jesse's trial, the jury agreed. sentenced him to prison. During sentencing, Justice Croucher was overcome with emotion. He broke down in tears, recounting the effect Caleb's death had had on his father, stopping several times to wipe away tears. At his appeal, Jesse's lawyers argued that the shaken baby diagnosis is no better than junk science. And to back up that claim, they relied on a landmark study by Sweden's Health Technology Agency. I'm not going to try to pronounce its name, but it's known as SBU, and it has a global reputation for its work assessing practices in health and social services. And in 2014, SBU decided to try to answer the shaken baby conundrum once and for all. It put together a team of more than 40 experts. And for two years, they worked to answer one seemingly simple question. The research question was, with what certainty can we say that these trial findings means that the child has been shaken violently? That's Anders Eriksson. He's a retired professor in forensic medicine and was one of the key authors of the SBU study. I am Anders Eriksson. I am 74 years old now. I retired. Anders and the team of researchers started by reading the abstracts of more than 3,300 papers in five different languages. Then they read 1,065 of those papers in full. You could say this was the entire corpus of shaken baby science as it stood back then. The SBU scientists then analyzed those papers according to evidence-based medicine studies. And then they read the abstracts of more than 3,300 papers in five different languages. And ended up with just 30 papers that were directly relevant to their question. But remember the circular reasoning we mentioned earlier in this episode? Well, 28 of the 30 papers just assumed shaking had caused the triad. Unfortunately, 28 of these 30 publications turned out to have a high risk of bias, which means very low scientific evidence. So in the end, only two papers were left. Both involved witnessed or confessed shaking. They showed some links with the triad signs, but it was complicated. So when the team reported their findings in 2016, they gave no comfort to the proponents of the theory. The main result of the SBU report was that the evidence to diagnose violent shaking from these three medical findings is very weak. So basically, we cannot say that these three medical findings means that the child has been shaken by luck. In essence, it's exactly what Mark Donoghue found 13 years earlier. So it's a circular argumentation that doesn't add any knowledge. It just confirms it's a self-fulfilling prophecy. To Jesse's lawyers, this evidence seemed compelling. So one morning in 2021, Anders found himself on Zoom in front of three eminent Australian judges. I remember that it was four o'clock in the morning here and I was stood in the court for four hours until eight o'clock. Then I had to leave. In evidence, Anders agreed with the defence that Caleb could well have died of an illness, not from assault by Jesse. So what happened? When an Australian court was confronted with Scandinavian science. It didn't go so well, at least not for the Swedes. The majority judges said they'd just got it wrong. I don't know if you read the final judgment, but they had a few things to say about your report. They said it was arguably not respectable. It didn't carry out. They hadn't done their own research, the judges said. The study only looked at other people's papers. Their findings were radical. They'd culled most of the papers and ended up with nothing. They'd asked the wrong question and didn't have any child abuse paediatricians or ophthalmologists on the panel. These are the criticisms made by the judges in that case. How do you respond to that? SBU is, in fact, one of the world's oldest and most respected health technology agencies in the world. So, I mean, who are these? Legal advisors, judges, who are they to tell us what science should do? I mean, it's off the wall. They are not qualified. Period. It's embarrassing. Ignorant and embarrassing. In making their decision, the judges had listened instead to proponents of shaken baby syndrome, including Dr. Jo Tully, who you heard from earlier. She'd been an important witness in the original trial. The majority judges described the prosecution's evidence as cogent and reliable, both scientifically and more generally. The third judge, though, took the opposite view. In a dissenting judgment, she wrote that, considering the state of the science, Jo Tully shouldn't have given evidence with the certainty she did in Jesse's case, and that his conviction represented, quote, a substantial miscarriage of justice. She said, Jesse should be immediately released from prison. But that judge was outvoted. Jesse is serving out the rest of his time in jail, and the majority judgment validated the Australian legal position that shaken baby syndrome is solid, credible science. In the United States, these issues play out in the media and the courts, as well as in conference rooms and academic journals. And things have changed. Our national investigates team has uncovered at least 21 people charged with crimes connected to the diagnosis of shaken baby syndrome who have had their convictions overturned since 2019. In New Jersey a couple of years ago, an appeal court found a shaken baby case had been based on junk science and prevented the theory being used in the trial. in the prosecution. Lawmakers say that the state's junk science law should force the conviction to be reconsidered. Why were most of their charges reversed? Evolving science, many experts say, proves the diagnosis is unreliable. In the UK, an appeal court laid down new guidelines saying triad cases should be approached with caution, basically because so much is still unknown. The very basis of the scientific theory that sent them to jail or jeopardized their freedom and family has been called into question. And in Sweden, the courts basically don't run shaken baby cases anymore. The Supreme Court said that to convict someone for shaking a baby, the medical findings are not enough. They are not reliable enough. But in Australia, the doctors making the diagnosis don't even acknowledge there's a significant dispute. So 20 years after Mark Donoghue wrote his paper and 15 years after Norman Guthkelch changed sides, and almost a decade after the Swedish study, in Australia, we're still jailing people with not enough questions asked. So who are the guardians of shaken baby syndrome in this country? And why have they had such a powerful voice? First time we met that doctor, she turned and looked at me and said, you did something and I'm going to prove it. Maybe we don't know what's going on. And maybe it's a mystery, not abuse. Maybe it's a mystery, and I think there are clinicians out there who are still unwilling to believe that a parent or carer will harm their child. It's target country. Target country? It's dangerous sort of medical territory, I think. They will be having dark thoughts, I can tell you, about this, but I'm retired. That's next time. If you found anything distressing in this episode, you can call Lifeline on 13 11 14. Diagnosing Murder is a production of The Age and Sydney Morning Herald. It's hosted and reported by me, Michael Bachelard. Ruby Schwartz, executive, produced the series. She's also the head of investigative podcasts. Julia Carr-Katzel was a producer on the series. Mixing by Martin Peralta. Editing from Patrick Elligott, the Ages editor, and Matthew Dunkley, the Deputy and Investigations editor. Additional editing by Catherine Winter. Special thanks to Lisa Muxworthy and Luke McElveen. Thank you. Subscribe to Diagnosing Murder to make sure you don't miss an episode. subscribe to The Age and the Sydney Morning Herald to read articles about this issue. Thanks for listening.