Thoughts on Dr Dewi Evans' response to the Channel 4 documentary

On Monday, a documentary Lucy Letby: Murder or Mistake aired on Channel 4. Much of it featured lead prosecution medical investigator and expert Dr Dewi Evans. On Tuesday, Dr Evans released a statement in response to the documentary. In this blog post I critique the statement.

Dr Dewi Evans' statement, 30 September 2025

Murder or Mistake, Channel 4, 29 September 2025

Danny Bogado, Rosy Milner and the Channel 4 team took on a challenging documentary (Murder or Mistake, Channel 4, 29 September 2025). Producing a 2 hours' documentary from evidence that took Cheshire Police several years to collect and led to the longest criminal trial in English legal history is in itself remarkable.

Comment: I shall watch those documentaries about the universe and dinosaurs with more respect from now on.

It was a privilege to be considered the chief prosecution witness in a trial that has led to the lifelong incarceration of England’s worst female killer. Her conviction would not have succeeded without Cheshire Police’s diligence and attention to detail, and the evidence of my fellow independent witnesses and the numerous nurses and doctors from the Countess of Chester Hospital. The evidence of all the witnesses was crucial.

Comment: The rebuke that frequently gets thrown at those who say Lucy Letby deserves a retrial is, “Think of the parents!”. It is strange that Dr Evans gives special mention to the police, to the Countess doctors and nurses, to his fellow independent medical experts, but not to the parents of the babies. Perhaps he is lumping them together in “all the witnesses” which of course includes Lucy Letby herself and a plumber.

Lucy Letby’s new barrister has broken new ground in launching an appeal via press and media, providing of course a very selective version of events. Mark McDonald helpfully summarised 6 issues on his agenda; the rota, the experts, air embolus, insulin, the neonatal unit, and the media. Despite his natural thespian skills and his considerable bravado Mr McDonald and his supporters failed to offer any evidence not disclosed at Letby’s trial, or would stand scrutiny at a clinical or scientific presentation or the rigours of cross examination.

Comment: Lucy Letby’s new barrister, Mark McDonald, has not broken new ground in launching an appeal via press and media. It is well understood that exonerations of those wrongfully convicted require public pressure on the Court of Appeal that comes from a concerted media campaign, whether driven by the lawyer in question or otherwise. There are plenty of examples of this. The below is from the book, Stolen Innocence:

Sue Stapely, also a solicitor, is now practicing as a public relations consultant. To persuade a Court of Appeal that a previous panel of such judges got it wrong, the argument must first be won in the court of public opinion, particularly in the legal press and broadsheets. Sue gave up a successful and financially generous partnership so that she could join another firm, Quiller Consultants, which would allow her one day a week – and however much more it would take – to mastermind the media campaign to free Sally [Clark]. And do it for nothing: pro bono. Her efforts led to the most sustained and effective media campaign in living memory.

Given that Dr Evans must see that, regardless of whether it could have been presented at trial, the new expert opinion from numerous eminent doctors, scientists, and statisticians that casts doubt on the verdicts should be taken seriously. He must also understand, given his medicolegal training, that new expert opinion counts, if deemed credible, as new evidence. In other words, he must see that the criteria for a retrial to be granted have been met. Given this, why would he want to criticise Mark McDonald for conducting a media campaign that he knows is necessary to first persuade the CCRC to refer his client’s case, then to persuade the court of appeal to grant a retrial? Perhaps it is because Dr Evans does so little with the media himself? 

The relevance, or not, of the rota, and Letby’s presence, has been flogged to death. Peter Elston, statistician and investment manager, just could not understand the difference between information that is statistically significant and that which is clinically significant. Statistics will point you in a certain direction. It does not prove that something took place. It’s why statistics did not play a part in the prosecution’s case. Presumably it’s why the Defence did not seek a statistical explanation either.

Comment: the relevance, or not, of the rota and Letby’s presence has not yet been flogged to death. That will happen in court at a retrial, assuming the prosecution wants or is allowed to show it to jurors.

As for Dr Evans' comment, "statistics did not play a part in the prosecution’s case", not only were statistics invoked by prosecuting barrister Nicholas Johnson KC in two of the first four paragraphs of his opening statement, but the CPS in its press release following the verdicts noted that one of the four types of evidence that convicted Lucy Letby was the roster chart with its uninterrupted line of X's. This roster chart was a statistical construct, albeit a greatly flawed one. Indeed, Prof Peter Green, of the School of Mathematics at the University of Bristol and former president of the Royal Statistical Society, has said, “The spreadsheet duty roster is almost a textbook example which I would give to my students of how not to collect and present data.”

Finally, when Dr Evans writes that I, “just could not understand the difference between information that is statistically significant and that which is clinically significant”, I’m pretty sure I do understand the difference and would invite Dr Evans to point me to evidence that I do not. More importantly, when he talks about information that is clinically significant, his only evidence at trial that Baby C had been attacked was that “he collapsed and died”. Can I ask him why he thought this information was clinically significant in pointing to intentional harm, and whether he will report to the police his suspicions in relation to the many millions who have died this century exhibiting the same symptomatology, wrongly diagnosed as heart attack or stroke.

Private Eye’s Dr Phil Hammond claims credit for exposing the concerns regarding what became the Bristol Heart Scandal. He conveniently overlooks the fact that the same magazine took over 10 years to admit it got it wrong in relation to discredited medic Andrew Wakefield and his flawed MMR / autism claims. His claim that my diagnosis of air embolus was because I “couldn’t think of anything else” reflects his clinical limitations, not unexpected given he is best known as a comedian who writes for The Eye. My diagnosis of air embolus was supported by my fellow independent witnesses, belatedly by the Chester paediatricians, and backed up by 18 publications.

Comment: I believe that Dr Hammond’s assertion that Dr Evans diagnosed air embolism because he “couldn’t think of anything else” is a reference to what is known as “diagnosis by exclusion”, a formal medical term. According to Google AI:

Expert witnesses for the prosecution in the Lucy Letby trial relied on a diagnosis of air embolism by exclusion, arguing that it was the cause of sudden collapses and deaths after all other medical possibilities had been ruled out. However, this method has been heavily criticised by other medical experts, leading to significant challenges in her appeals process.

As for his diagnosis of air embolism having been supported “belatedly by the Chester paediatricians”, Ravi Jayaram first came across the 1989 paper on air embolism well before Dr Evans came along, so it was the other way round. In fact, Dr Jayaram said under oath on different occassions that he first came across the paper three times, in July 2015, April 2016 and June 2016. But then this is the same Dr Jayaram who said that door swipe data had helped him clearly remember events pertaining to the case of Baby K. It turned out the door swipe data was the wrong way round. It’s like me saying, “I’ll remember that football game till the day I die because Arsenal scored five goals (psst, it was Liverpool who scored five goals not Arsenal). I’ll remember that football game till the day I die because Liverpool scored five goals”.

And as for Dr Phil being best known as a comedian who writes for The Eye, perhaps, given his performance in the documentary, Dr Evans might become best known, if he isn’t already, as a comedian who writes for lawyers ("Whatever opinion you're looking for, I have it!")

Shoo Lee may well be an eminent epidemiologist. His experience of air embolus is limited to reporting 3 cases from the 1980s. More tellingly he has no background of medicolegal matters because “I don’t do medico-legal cases. I don’t like them”. Bogado's programme did not have time to include my detailed critique of his 2 published papers.  If his most recent paper (published in December 2024) had been available before the trial its content would have been very useful - for the prosecution. His criticisms of my role, disclosed at the press conference, were all factually incorrect. Summaries from other members of the Panel were awash with errors, some of the weakest reports I have seen during my career.

Comment: you would have thought that if Dr Evans was going to write that Lee’s 2024 paper would have been useful for the prosecution, he might have wanted to explain why. Furthermore, if he is going to say Lee’s criticisms were “factually incorrect” and other panel members’ summaries “awash with errors”, he might also have wanted to say what these were. Otherwise, some might say these proclamations were the weakest they had ever seen.

Dr Evans has said elsewhere that the reason Lee’s 2024 paper would have been helpful for the prosecution was that it described characteristics that were identical to the Chester babies. In relation to Baby O, Dr Evans said,

"Given Dr Brearey’s description [“Small discoloured ? purpuric rash”] I believe it’s appropriate to note the comment from the paper by Lee & Tanswell in relation to colour changes. “Blanching and migrating areas of cutaneous pallor were noted in several cases and, in one of our cases we noted bright pink vessels against a generally cyanosed cutaneous background.” This description is not dissimilar to Dr Brearey’s observations regarding the discoloration he noted.”

On which planet are these two descriptions “not dissimilar”??? I suspect Dr Evans’ definition of the word “identical” might need questioning.

Given the time constraints it was inevitable that the documentary concentrated on just a couple of cases. The timing of Baby C’s collapse created confusion. In my very first report (7 November 2017) I raised concerns regarding the event late on Saturday 13 June 2015 that led to his collapse and death. Over the next 5 years the dates seem to have become confused, leading to the prosecution alleging that the assault took place the previous day. I recognised the confusion at the Trial, reaffirming my original concern that the fatal assault was late on 13 June. I believe that Cheshire Police, the CPS, and the Prosecution team should set the record straight.

Back in 2017, when I first raised concerns regarding Baby C’s demise, I was unaware of Letby’s presence on the unit, let alone that she was up close to Baby C at the time of his collapse. There is nothing in Baby C’s clinical records to confirm Lucy Letby’s presence.

Comment: in an addendum at the bottom of this blog are extracts from Evans’ three written statements about Baby C. There is also the matter of what Evans said to Radio 4 a year or so ago, as detailed by Tried By Stats here https://x.com/triedbystats/status/1973113966860836884. Both cast serious doubt on the accuracy of what Dr Evans wrote in his statement.

Dr Evans writes, "In my very first report (7 November 2017) I raised concerns regarding the event late on Saturday 13 June 2015 that led to his collapse and death” but I cannot see any reference in his first report to an event, only to not being able to exclude infection or include inappropriate management.

He also writes, "Over the next 5 years the dates seem to have become confused, leading to the prosecution alleging that the assault took place the previous day.” It was he, in his third statement, who alleged the assault took place the previous day. Saying it was “the prosecution” is disingenuous.

Then, he writes, "I recognised the confusion at the Trial, reaffirming my original concern that the fatal assault was late on 13 June.” His original concern, per above, was not that the/a fatal assault was late on 13 June, just that he could not exclude infection or include inappropriate management. At trial, under cross examination, his evidence of assault on 13 June was simply “the baby collapsed and died”, supporting the fact that he had not invoked an assault in his first written statement. Furthermore, having not mentioned it in any of his written statements or in the joint expert meeting, he said on the stand that Letby had also injected air into Baby C’s bloodstream. Then there are all his statements' inconsistencies, changes in opinion etc since the trial.

Dr Evans finishes with, "Back in 2017, when I first raised concerns regarding Baby C’s demise, I was unaware of Letby’s presence on the unit, let alone that she was up close to Baby C at the time of his collapse. There is nothing in Baby C’s clinical records to confirm Lucy Letby’s presence.” This latter sentence seems incongruous. Is he saying that there is nothing to say she was not in fact present on 12 June or at any point prior to that since Baby C’s birth on 9 June? Letby’s shift calendar which Evans sent me a year or so ago confirms that she wasn’t. I’m also not sure why he was still in possession of that document given his role had ended by then or whether he was allowed to send it to me.

Baby O’s demise was given considerable attention, and we were presented with some extraordinary explanations from British Columbia based neonatologist Richard Taylor and the Brighton duo of Neil Aiton and Svilena Dimitrova. According to them Baby O’s death was due to a cannula inserted into the baby’s abdomen 20 - 30 minutes before he died. This fails to explain why the baby was moribund, sadly at death’s door, before the cannula was inserted. For good measure Aiton claimed that the ventilator pressures used to resuscitate the baby were too high. This does not explain why the baby required resuscitation in the first place. He had never required resuscitation following his birth, and his collapse took place when in Lucy Letby’s care. The Taylor, Aiton, Dimitrova hypothesis has been widely condemned by others, including a pathologist who gave his opinion on a recent Panorama programme. Their opinion should be confined to the Donald Trump School of Evidential Science.

Comment: As Dr Dimitrova has stated on the record, Dr Taylor had not seen her and Aiton’s report when he spoke at the press conference. Dr Evans commenting on a “Taylor, Aiton, Dimitrova hypothesis”, when there is no such thing is a tad irrational. Perhaps Dr Evans' opinion should be confined to The Dewi Evans School of Rational Thought.

The controversy regarding insulin poisoning has been widely explored. Professor John Gregory recently endorsed (on the same Panorama programme) the opinion of his fellow paediatric endocrinologist Prof Peter Hindmarsh, stating that baby F and baby M were both poisoned with insulin. Insulin poisoning had been accepted by Letby’s Defence team at the trial, and indeed by Lucy Letby herself.

Comment: I suspect the parents of L and M may well be upset that Dr Evans got their children mixed up (some fact checking should perhaps have been done?) So Gregory endorsed Hindmarsh whose opinion that F and L had been poisoned with insulin was based on blood test results that the chemist Anna Milan said under oath could only have been due to exogenous insulin when in fact the guidance from her very own lab stated it could also be due to endogenous insulin that had bound to antibodies. The winner of the Most Tangled Web award goes to…

Criticism of the neonatal unit building was reasonable, and the unit has long moved to more suitable premises. Alleging that the department was “failing” or unsatisfactory is unfair and unreasonable. Survival rates, the best quality control of any neonatal unit, were as good in Chester as the ONS [Office of National Statistics] figures for England & Wales, even for the smallest babies. The Thirlwall Inquiry confirmed that staffing was similar to other units in the North West of England.

Comment: Dr Evans told John Sweeney that had there been the sewage problems at his hospital that there were at The Countess, the unit would have been closed down. When he says that sewage was not unsatisfactory, perhaps he saying that the sewage in Chester is more palatable than that in Wales?

Mark McDonald and his team failed to show any new evidence that would justify another appeal. Unhelpfully for him, this week’s Law Society Gazette article by Bianca Castro (26 September) states: “Conviction did not uncover anything new, a potential problem if the CCRC [Criminal Cases Review Commission] are to refer the case back to the Court of Appeal.” 

Comment: I may be wrong, but I believe Mark McDonald’s CCRC application comprises 1,000 plus pages of new expert reports, not a transcript of the documentary. And Dr Evans must also understand, given his medicolegal training, that new expert opinion counts, if deemed credible, which given the authors it clearly is, as new evidence.

The victims in the Letby Trial are the babies harmed and murdered by Lucy Letby, and their families. A campaign to release England’s worst female serial killer is beyond my understanding. It adds to the hurt and grief the families have suffered for the past decade.

Comment: Dr Evans is correct. Seeking to get to the truth is so much more shameful than forgetting to mention parents, getting their children mixed up, and gaslighting them.

Dr Dewi Evans

30 September 2025

Comment: can't dispute those

Addendum: below are some extracts from Dewi Evans’ three baby C written statements. They can be compared with what he wrote in his statement this week, as well as what he said about that case at trial and in various media interviews he has given since the trial.

Statement 1 (dated 7 Nov 2017)

The third X-ray is of the abdomen and is timed at 12.36 hr on 12 June 2015. I do not think it shows any significant pathology.

There was evidence of suspected sepsis with the increase of CRP from 1 to 22 [normal is less than 10].

One may never identify the cause of Baby C’s collapse. One cannot ignore the low platelet count or the presence of bile in his aspirates. Neither would not cause his collapse but both are markers that suggest he was an unwell baby. A raised CRP [of 22] similarly suggests that he was not completely well. There are therefore a number of features, allied to his very small weight, just 717 grams, that places Baby C at great risk of an unexpected collapse.

I have concerns regarding the unexpected collapse of Baby C at around 23.00 hr on 13 June 2015, given his reasonable stable condition immediately prior to this. I note also the presence of abnormal inflammatory markers [low platelets, raised CRP] and cannot exclude the role of infection in his collapse.

I would advise scrutinising the staffing present at this time. Given the presence of the inflammatory markers it’s not possible to state categorically that Baby C’s collapse represents inappropriate management at around 23.00 hr on 13 June 2015.

Statement 2 (dated 31 May 2018)

There was evidence of suspected sepsis with the increase of CRP from 1 to 22 [normal is less than 10].

The nursing entry found on page 336 notes that Baby C received his first feed at 23.00 hr [on 13 June]. He received 0.5 ml. “At around 23.15 Baby C had an apnoeic episode with prolonged Brady and desat. Crash call for neonatal team put out by myself. Resuscitation commenced”.

The second X-ray is timed at 22.38 hr on 10 June 2015.This confirms the presence of the two lines. On this X-ray there is evidence of consolidation of the left lung. This is consistent with pneumonia.

The third X-ray is of the abdomen and is timed at 12.36 hr on 12 June 2015.I do not think it shows any significant pathology.

One may never identify the cause of Baby C’s collapse. One cannot ignore the low platelet count or the presence of bile in his aspirates. They would not cause his collapse but both are markers that suggest he was an unwell baby. A raised CRP [of 22] similarly suggests that he was not completely well. There are therefore a number of features, allied to his very small weight, just 717 grams, that places Baby C at great risk of an unexpected collapse.

I have concerns regarding the unexpected collapse of Baby C at around 23.00 hr on 13 June 2015, given his reasonable stable condition immediately prior to this. I note also the presence of abnormal inflammatory markers [low platelets, raised CRP] and the presence of consolidation of the left lung [from 10 June chest x-ray] and cannot exclude the role of infection in his collapse.

I would advise scrutinising the staffing present at this time. Given the presence of the raised inflammatory markers it’s not possible to state categorically that Baby C’s collapse represents inappropriate management at around 23.00 hr on 13 June 2015.

Statement 3 (dated 26 Mar 2019)

Abdominal x-ray of 12 June [AXR] notes "loopy bowel. Distended." The diagnosis is "Suspected sepsis.”

The nursing entry found on page 336 notes that Baby C received his first feed at 23.00 hr [on 13 June]. He received 0.5ml. "At around 23.15 Baby C had an apnoeic episode with prolonged brady and desat. Crash call for neonatal team put out by myself. Resuscitation commenced”.

Baby C's glucose [sugar] was checked several times. I note a value "dropped to 1.7" recorded on 12 June.

The second x-ray, also dated 10 June, notes "extensive consolidation throughout the left lung. There is relative sparing of the right lung and the appearances are most likely to be due to infection.”

The 3rd x-ray is of the abdomen, and dated 12 June (363) and is reported as showing "marked gaseous distension of the stomach and proximal small bowel. This relates to the x-ray on page 391.

I had commented on the x-rays in paragraph 25 of my report of 31 May 2018. On reviewing my opinion I confirm that my interpretation of the 2 x-rays of 10 June is consistent with the x-ray reports. I agree that that the abdominal x-ray of 12 June shows gaseous distension of the stomach and proximal small bowel.

I note the gaseous distension of the stomach and intestine reported in the x-ray timed at 12.36 hr on 12 June. The degree of distension is considerable, even in babies who are receiving CPAP. It explains why Baby C was noted to be "very unsettled and fractious", and why he calmed down when CPAP was removed [around 09.00 hr on 13 June, page 333].

Whilst remaining suspicious about his unexpected collapse and his failure to respond to resuscitation I cannot ignore the presence of the low platelet count and raised CRP, both of which were present prior to his collapse, and are markers of infection. I had also noted the presence of left lung consolidation on the chest x-ray timed at 22.38 hr on 10 June. It's therefore probable that infection was a significant factor in Baby C's collapse during the late hours on 13 June 2015.

In the context of the clinical presentation of several other babies nursed at the Countess of Chester Hospital during 2015 and 2016 I am suspicious of the gaseous distension reported on the abdominal x-ray of 12 June, and wonder whether this represents inappropriate management, where his attendant inserted excess air into his stomach via his naso-gastric tube, doing so in the knowledge that it would cause the infant discomfort and distress.