The Guardian today on the safety of the Lucy Letby convictions

[Kittybythelighthouse]

This article was apparently months in the making but it was delayed by the reporting restrictions https://www.theguardian.com/uk-news/article/2024/jul/09/lucy-letby-evidence-experts-question

“A Guardian investigation has interviewed dozens of these experts and seen further evidence from emails and documents. Those raising concerns include several leading consultant neonatologists, some with current or recent leadership roles, and several senior neonatal nurses. Others are public health professionals, GPs, biochemists, a leading government microbiologist, and lawyers. Several of those still working in the NHS have asked to remain anonymous, fearing the impact if they are named.

These experts said they were acutely aware of the suffering of the families involved and did not want to reopen their trauma, but were so troubled they felt compelled to become involved”

The anguish suffered by the parents must be unbearable.

I’ve not read the entire thread, so apologies if I’m repeating a point already made.

I recall reading something about a person called in to repair a sink where bacterial contamination was found. One of the babies that sadly died had an infection caused by the contaminant. The sink was apparently removed. It seems the person fixing the sink was one of the few people called by Letby’s defence.

This.
I am 55 now. When I was 20 I had my first Nanny job. The baby was 6 months old.
I did the ‘cry it out’ method to get her to sleep.
she died of cot death.
luckily, no one else but me ever blamed me.
I went to her funeral, I am still acquainted with the family now - they greet me like an old friend on occasions our paths cross.

however.. it took me YEARS to stop blaming myself, and still, even now, there is a little bit of me that says what if..? What if I hadn’t let her cry it out..? What if she was trying to communicate something to me, that she didn’t feel well..?

I didn’t write notes, but had I done so, then would that have landed me in prison..?

This seems to be another misrepresentation. As far as I can tell they've written to the subsequent public inquiry - they don't say they wrote to the criminal trial at any point. Their letter is focused on how statistics can be used responsibly to identify suspicious behaviour (i.e. could this have been spotted and stopped sooner - which was also a big discussion after Harold Shipman's conviction), not questioning the veracity of her conviction. The questions they want answered through the inquiry are:

1. What local or national data-driven systems were in place to provide alerts to unusual patterns of events – in particular, neonatal deaths? Given that there are statistical systems that have been developed to identify unusual patterns of mortality (eg, MBRRACE, CQC Mortality Monitoring system) were any of these in use, covering both the Trust in question and Trusts more widely with regard to neonatal deaths.
2. If there was such a system in place, was an alert issued at any point during Letby's tenure?
3. What was the pattern of events, in particular neonatal deaths, before, during and after Letby's tenure? Using local data, can a change from the hospital’s past performance be detected? What light does national data shed on the long term performance of the hospital?
4. If a rigorous statistical monitoring system had been in place, would it have issued alerts? At what point?
5. What type of system could be implemented for monitoring neonatal adverse outcomes in the future?

This is all geared towards ensuring this never happens again rather than, as you keep implying, expert statisticians questioning whether it ever happened at all.

Yes. And what would ‘writing to the trial’ mean anyway? To the judge? And what the judge do? Start telling the jury on behalf of the defence that there are statistical issues that they must consider, i.e. judicial action for the defence?

And that’s without the risk of contempt or apprehended contempt, particularly if the report was published generally when the cases(s) was/were active.

Or write to the defence? Possible, I suppose, but then you’d have to ask why the defence didn’t deploy the expert opinion.

However, as you say, it looks like the RSS did none of these things and had no intention of doing them, in any event.

According to The Telegraph today:

The Royal Statistical Society (RSS) has also announced it will “convene a meeting” in the wake of the verdicts, stating that it was aware of “concerns” from RSS members and the wider community regarding the use of statistical evidence in the case.

That is actually a misrepresentation. The report ‘Healthcare Killer Serial Killer or Coincidence?”, written in 2022 in advance of the trial - it is linked below - is absolutely about how bad statistics, and poor understand of statistics, can (and indeed has, several times) lead to miscarriages of justice for health professionals. It was co-authored by Richard Gill, who stated that it was written in advance of the Letby trial for exactly the reasons I have stated. You can call him a crank all you want, but the Royal Society of Statisticians don’t agree with you. You can, of course, also call the whole organisation cranks. It just isn’t very credible.

rss.org.uk/RSS/media/File-library/News/2022/Report_Healthcare_serial_killer_or_coincidence_statistical_issues_in_investigation_of_suspected_medical_misconduct_Sept_2022_FINAL.pdf

To advise having an expert statistician present to explain and contextualise the statistics to the jury, that’s why.

Oh my goodness. I am so so sorry to hear about your loss 💐 💛

I hope that puts your mind at rest. The journalistic integrity of two of our primary newspapers of record is important and it’s important that the public feel they have a right to query them. However, I don’t anticipate a retraction. It’s clear the articles were well researched over many months, written by esteemed journalists, and not published carelessly.

That’s not how it works. 🙄

Thank God.

It was in a shocking state and very poorly run. Several babies - that had nothing to do with Letby - were killed, or left disabled, by clinical mistakes (e.g a baby was intubated by one of the doctors into his oesophagus rather than his trachea. The doctor - who incidentally spoke against Letby in the trial - blamed the machine when the desaturation alarm kicked in and didn’t bother checking the placement. The child died).

The RCPCH wrote a damning report in 2016 in response to the 2015 spike. The spike, by the way, wasn’t even highest spike in relative numbers in the UK at that time. If the spike in deaths points to a serial killer and not a cluster, there must be other hospitals harbouring serial killers, too.

Are you claiming that no institutions can write to individual lawyers and judges or just the RSS? Are you claiming that this is a legal stipulation?

It’s a rhetorical question. I know the answer.

Well indeed. Imagine if it did! It is what Richard Gill did though.

This thread is sad and mad in equal measure. The prosecutions opening statements are widely available online making it clear what evidence they were calling and why. They were clearly informed by the RSS report and the wider longstanding issues with citing statistical evidence.

There is no evidence that the errors the RSS highlight were made by the prosecution. The prosecution did NOT rely on cluster evidence to prove the children's deaths were unnatural. They presented each death independently at trial and tried each case separately. Absolutely no cluster data was used to suggest Letby was guilty - the evidence used to convict her was that each death was individually both a murder and that Letby committed it.

At appeal one of the grounds for appeal was that the judge had misdirected the jury on how they reached their verdict by telling them that whilst they must be sure in each case that the deliberate harm had been inflicted they did not have to agree on how the harm had been inflicted.

The defence argued this could lead to different members of the jury reaching their verdict by different means and thus not giving due consideration to the possibility the deaths were due to natural causes.

However, the appeal judges agreed this is not a requirement. There are many cases where the prosecution cannot prove the cause of death beyond reasonable doubt, including cases where a body is never recovered. They only need to prove that the defendent inflicted the harm that caused death.

Critically though in Letby's case the jury are directed they must find this in each case. It is likely at trial one where they acquit her on some charges and then cannot reach a verdict on others this is precisely because they were carefully evaluating the very complex medical evidence to see if they found it good enough. It is very clear they did not throw the book at her BECAUSE she had been present at all the alleged crimes. In many ways her path to appeal would be easier had she been found guilty on all charges and if the jury had been out for less time as it would suggest the judge had misdirected the jury.

The report gives general good practice advice on using statistical evidence in cases of alleged medical harm. It at no point mentions the Letby case and wasn't directly sent to the judge by the RSS (as you've repeatedly claimed). It may well have been inspired by it (though not all contributors appear to agree on this), but it - entirely appropriately - doesn't give any judgement or evaluation on the use of statistics in a case that hadn't yet been heard. At no point since the case have the RSS publicly said they think the use of statistics in the case was unsound (again, this would be wildly inappropriate, so it's not surprising that they haven't). Richard Gill, a single individual, did so. That isn't the same thing.

Again you repeatedly stated as true something that isn't, and are changing your story when it's shown to be untrue.

It doesn’t mention the Letby case. It doesn’t have to. We what it was written for and why because Richard Gill co-authored it and he has said so. Again, call him a crank if you want, but have a think about whether you want to claim that the Royal Society of Statistics are cranks too.

The Royal Society of Statistics as a body have not ever offered an opinion, positively or negatively, on the use of statistics in the Letby case, despite what you've claimed. They have a statement on it which is completely, studiedly, deliberately neutral: https://rss.org.uk/news-publication/news-publications/2024/general-news/rss-statement-on-the-statistical-aspects-of-the-lu/

You keep claiming that the RSS have things they haven't and implying, incorrectly, that they've endorsed what Richard Gill has said in the case, in the same way as you stated as confident fact that the note said 'they went' - and, in the same way, are now changing what you're saying and the significance of it now you've been questioned.

@MoreDangerousThanAWomanScorned it is the RSS’s report! Their name is on it!

So now The Royal Society of Statisticians are cranks, The Guardian are cranks, The New Yorker are cranks, The Telegraph are cranks, several leading consultant neonatologists, senior neonatal nurses, public health professionals, GPs, biochemists, legal experts, and a leading government microbiologist are all cranks.

How many more (and there will be more) are to be smeared as cranks and added to this list before some people will accept there are legitimate issues with this case? And that It is actually completely fine, in a modern democracy, for those issues to be properly explored and resolved?

https://rss.org.uk/RSS/media/File-library/News/2022/ReportHealthcareserialkillerorcoincidencestatisticalissuesininvestigationofsuspectedmedicalmisconductSept2022FINAL.pdf

It's their report that gives general good practice guidance on the use of statistics in medical harm cases. They have not ever offered any opinion, at all, on the Letby case. You keep deliberately conflating the two to be misleading.

I have never suggested that the RSS are, collectively, cranks. They're a professional body which advocates and campaigns for the better use and understanding of statistics as a general cause. Which is why they don't and haven't offered statements of opinion on ongoing criminal cases or on their opinion on verdicts given by a court.

Genuine question, what exactly are you saying here and why do you think this? I'm trying to understand. It sounds like you are saying that it would be a bad idea for a jury to receive statistical guidance from an expert but maybe I'm misinterpreting? From the perspective of someone who knows a bit about stats and cognition that sounds like a good idea. But I'm aware that I know little about the legal system so maybe there's something I'm not understanding here. Genuinely trying to get a clearer grasp of the arguments.

@Kittybythelighthouse can you please point to the actual evidence put before the jury in the trial which violates the guidance published before the trial by the RSS?

Please don't just cite the Guardian and New Yorker articles. Please cite the actual evidence that the jury themselves were shown that would have resulted in their being potentially mislead as to the statistical likelihood that either the deaths were deliberate or that Letby was the perpetrator. There is none.

The prosecution case was very carefully constructed to avoid requiring the jury to draw inferences that the deaths were the result of deliberate harm alone on the basis of cluster evidence, and likewise the case that Letby was the perpetrator was not solely based on her physical presence in each case but a large volume of wider circumstantial evidence both placing her at the scene, with witnesses observing her behaving suspiciously. At NO point is any statistical evidence adduced suggesting it is inherently unlikely she is not the killer given the frequency she is on shift or the frequency of deaths. The evidence of her presence shows that if all the deaths are deliberate killings and committed by one person she is the only person who had access to every single victim at the time they died. It is evidence of opportunity, not evidence that she is the likeliest culprit.

John Quigley and Richard Gill have badly misinterpreted how this evidence was used because they are statisticians and NOT evidence lawyers. Their meaningless assertions are well outside their wheelhouse - and whilst promoting best practice in the use of statistical evidence in trials is a worthy endeavour they should also concede that is exactly what happened in Letby's case. The continued publicity they are trying to draw to the case now is only hurting bereaved families.

She's saying judges don't just admit evidence because anyone writes to them and says listen to me - judges do not supply third party evidence to the court, they are meant to be impartial AND it is not their job to assess the quality of the evidence. The role of prosecution and defence is each to make the strongest case they can. If judges interfered by admitting new expert evidence that would cause chaos and practice directives exist to prevent it.

There is nothing to stop an expert writing to the defence saying I may be able to supply relevant evidence and Richard Gill did this and the offer was not taken up. It is very likely that taking up evidence from a statistician from the defence would have brought more heat than light to the proceedings.

So where would you position the Guardian, the Telegraph and the New York Times? Below reddit? And obviously, below you, person of great knowledge and insight.

You really think it would be a bad thing for the court to hear evidence from expert statisticians? Bloody hell.

Bigcoatlady has answered this.

Court procedure and evidential admissibility decisions do not work as you think. Because the courts are well regulated.

To those who think that this was a safe conviction I would like to ask

1. Are you satisfied that the jury of non expert lay people would be competent to understand and reason correctly about the evidence in this case
2. If yes, what is your justification for thinking this

There seems to be a lot of focus in this thread on establishing that the prosecution behaved appropriately. I recognise that this is important to establish and I'm somewhat persuaded of it now.

However the ultimate question, surely, is whether the conviction was actually safe, and a conviction is only as safe as it's weakest element. I am much less persuaded that a group of ordinary people could handle such a complex task, especially without explicit guidance, and I've seen relatively little engagement with this issue.

Again, this question is not intended to be combative. I genuinely want to understand people's reasoning.