Lucy Letby (2)

[Oftenaddled]

Not only the second thread of course, but maybe with them filling so fast it is time to start numbering them?

@EyeLevelStick if you say so. I just nicked it off reddit when someone else made the comparison and I thought it was a good point. I should've known it wouldn't land considering most on here don't understand analogies.

There you go again with your supercilious sneering.

Of course people understand analogies, but it's only the losers screaming from the rooftops isn’t an analogy.

Can you link to her appeal?

There are pros to the jury system, but it's neither realistic nor reasonable to expect a lay jury of people without relevant professional expertise to appraise huge volumes of evidence in fields where they don't have the experience or training.

Which professions should have made up the juries for the Letby trials? And how could they all have relevant knowledge across all of the wide-ranging areas covered? What expertise would neonatologists have about the specifics of nursing practices and protocols? Or pathologists about endocrinology? etc etc

Multiple professional bodies have been sounding the alarm about the expert witness system for years, in some cases decades. We don't have safeguards built in to prevent an expert giving genuinely believed but wrong specialist evidence. It should concern us all that Dewi Evans could give incorrect evidence to the court without it being corrected.

The defence were perfectly entitled to challenge the evidence provided by Dewi Evans via their own experts. And I know some posters might point out that the defence were refused permission to put forward their expert/s as the trial went along, but I’ve also sat through a fair few long trials where the defence experts have swept in at the end and completely blown the prosecution evidence out of the water. Why did the Letby defence decide it was better to do nothing than to have a go at that? Particularly if the quality of the prosecution evidence was as shockingly poor as people are saying it was (and we can be pretty certain Ben Myers would have known poor quality evidence when he saw it).

In an adversarial system, prosecution and defence have a duty to put the case as well as they can, which isn't the same thing as having a duty to make sure the court is as well informed as possible. There isn't a duty to present an expert witness to the court, hence Letby had at least one expert who wanted to challenge dodgy prosecution evidence but the court never got the benefit of hearing that.

So if you’ve committed a crime but want to plead not guilty, you should be compelled to put forward an expert who will inevitably undermine your case (because you actually did it)? That feels like quite a fundamental shift away from the principle of innocent until proven guilty.

Prosecution also have an incentive to do things like give shonky statistical evidence and rely on cod psychological tropes about behaviour, and the prosecution in this case were just doing their jobs when they did it. But it was poor quality evidence.

We can probably agree to disagree about the extent to which statistical evidence was used in the Letby case, but it would be ridiculous to not be able to consider someone’s behaviour as part of the evidence. Again, the defence are perfectly entitled to challenge this. In the Letby case the defence instructed at least one forensic psychologist - they could have called them to explain why there was nothing concerning or unusual about Letby’s behaviour (assuming that’s what their opinion was).

Then there's the issues with remedying MOJs. The CCRCs organisational problems are well known, but also there's the question of whether the test isn't wide enough. Justice Gap have some good stuff on this.

Not sure what you mean by whether the test is wide enough?

For some specific issues like prosecutions including poor statistical evidence, much as I respect the work of the RSS (their various publications have been linked to on previous threads and are very good) I'm not sure education is going to tackle the problem. Not when the incentive remains due to the adversarial system. In order to address that we might need more specific rules about what claims can be made, perhaps specific applications if a party wants to depart from RSS guidance.
(Sorry, that was long for a summary!)

The Letby defence instructed a statistician - why didn’t they call them to provide some challenge and clarity on the statistical position? I suspect it’s primarily because the actual data would look far worse for Letby than the rota chart ever did.

I don't think that's really what happened, though. You had some loud voices confident in Lucy Letby's guilt: Liz Hull, Caroline Cheetham, Judith Moritz, Shaun Lintern - and some voices that started out confident in the verdict - Phil Hammond, Josh Halliday. And they have either changed their views or gone fairly quiet, recently.

There's no sign that people convinced of Lucy Letby's guilt don't see the need to talk about it. The Mail built a whole new subscription podcast model on it. Moritz and Coffey (who claims to be ambivalent) have got through three Panoramas and two editions of their book. And that's before we think about various YouTubers, twitterati, redditors and tattlers who are firmly attached to the existing verdicts.

What has happened is that new developments, leaks, and new expert pronouncements have overwhelmingly undermined the safety of the conviction. There's a clear appetite for content that should support the convictions. But it seems the material is hard to come by

(Sorry for overthinking the football analogy, though!)

Why didn't her defence call any witnesses except a plumber. Because it would have not helped. In other words she's guilty.

Yes - probably because they couldn't then (but can now) explain the insulin cases. Fortunately new scientific research is an acceptable ground for review of a conviction

www.private-eye.co.uk/pictures/special_reports/lucy-letby-28.pdf

@Oftenaddled that someone had to come up with a whole new theory to explain it that no endocrinologist knew of prior to the trial tells you everything you need to know.

I don't see how a statistician could have addressed the prosecution claims without reference to the construction of the case. Myers requested this information and was denied it; and he was not permitted to ask witnesses questions bearing on it.

You see Myers discussing probabilities and data selection in his closing speech, though without technical terms, and I suppose it is possible this was with expert advice.

That said, I think an explicit statistical argument would have been a risk. People often tend to find statistical analyses counter-intuitive. From that point of view, I agree with you. If the raw numbers made Lucy Letby's position look bad, there was the risk of highlighting them while losing an inexpert audience with your analysis. That brings us back to the first question @IonianNerveGrip raises, of how a lay jury assesses claims.

That's science for you! The question is whether, faced with unusual test results in pre-term babies, people previously wrote them off as probable errors, assumed deviations in that age group, or followed up with appropriate forensic tests. Some mix of the three I suppose.

It needs to be remembered too that the tests conducted in the Chester cases did not measure Insulin directly, and that the new research in the field looks at what different tests measure, allowing for some variation. The workings of these tests aren't an endocrinologist's direct concern.

@Oftenaddled what's the overwhelmingly most likely reason the results showed what they did?

We really don't have the data to answer that, but I know that the defence has instructed a statistician, Dr Hilde Wilkinson-Herbots, who has contributed to two reports on the insulin cases submitted to the CCRC. So it seems likely they have considered this angle.

It's worth remembering that the jury was not told that poisoning was the most likely explanation. They were told that it was the certain explanation (despite Professor Hindmarsh having told a court previously that other explanations existed).

Which also brings us back to my question - what professions would be represented on a professional jury in the Letby case?

If there’s one thing that has been hammered home to me from these threads, it’s that very few people can ‘do’ statistics other than statisticians (and that includes most doctors as I understand it). So should the professional jury just be made up of statisticians in cases like this? Otherwise, how is the professional jury any more likely than a lay jury to be assessing statistical claims correctly?

And then what about the way the professional jury assesses claims relating to neonatal medicine or pathology or endocrinology or psychology or hospital plumbing?

Which professions should have made up the juries for the Letby trials? And how could they all have relevant knowledge across all of the wide-ranging areas covered? What expertise would neonatologists have about the specifics of nursing practices and protocols? Or pathologists about endocrinology? etc etc

In the Letby case. I doubt an expert jury would've been feasible because of logistics. An agreed expert panel model would've probably been easier to pull off. You left out the part of my post where I talked about what was more realistic.

The defence were perfectly entitled to challenge the evidence provided by Dewi Evans via their own experts. And I know some posters might point out that the defence were refused permission to put forward their expert/s as the trial went along, but I’ve also sat through a fair few long trials where the defence experts have swept in at the end and completely blown the prosecution evidence out of the water. Why did the Letby defence decide it was better to do nothing than to have a go at that? Particularly if the quality of the prosecution evidence was as shockingly poor as people are saying it was (and we can be pretty certain Ben Myers would have known poor quality evidence when he saw it).

But the poor quality and in one case incorrect evidence is a problem in itself. It's all very well saying get another expert to address it, but Dewi Evans having honestly misstated the Shoo Lee paper means it's possible the jury will believe the incorrect interpretation.

As has been pointed out before, being an expert in a field doesn't translate to being a persuasive performer in court. They're not necessarily the same skillset, and while we're anecdoting, someone being incredibly eminent in their field and also coming across as arrogant is very much a thing!

So if you’ve committed a crime but want to plead not guilty, you should be compelled to put forward an expert who will inevitably undermine your case (because you actually did it)? That feels like quite a fundamental shift away from the principle of innocent until proven guilty.

What a silly response. You've been told before on these threads to stop attributing things to people that they haven't said. As pointed out already, one way to address this is an agreed expert

We can probably agree to disagree about the extent to which statistical evidence was used in the Letby case, but it would be ridiculous to not be able to consider someone’s behaviour as part of the evidence. Again, the defence are perfectly entitled to challenge this. In the Letby case the defence instructed at least one forensic psychologist - they could have called them to explain why there was nothing concerning or unusual about Letby’s behaviour (assuming that’s what their opinion was).

The fact that poor statistical evidence was used at all is bad. As neither of us know how persuasive the jury found it, there's no basis to either agree or disagree about extent- we simply know that it was part of the prosecution.

And the fact is that the prosecution tried to draw inferences from her behaviour, which weren't backed up by expert evidence. Think back to the closing speech for example. It's all very well saying they could've called a forensic psychologist but that doesn't actually mean the jury are going to find that expert persuasive over some cod nonsense. And the prosecution weren't doing anything wrong within our system here either, which is the point.

Not sure what you mean by whether the test is wide enough?

Ie does it need to be less narrow. The Law Commission undertook and published a review last year.

The Letby defence instructed a statistician - why didn’t they call them to provide some challenge and clarity on the statistical position? I suspect it’s primarily because the actual data would look far worse for Letby than the rota chart ever did.

Doesn't address my point. The RSS have been raising the alarm for years about poor use of stats in legal matters, you must know this, and the prosecution did it again here. As with the other times you made this argument, merely saying an alternative expert can be called doesn't take away the problem of it happening in the first place. Once poor quality evidence is adduced, it becomes possible that a jury will find it persuasive.

ok well most of the country, if you read the book by chris morris he explains everything which i could not do.... i guess if you just stick to the bbc/sky news or listen to the horrendous daily mail podcast you might believe she actually done it but maybe have a think about how silent they are. It is not the exciting subject it used to be,

Not sure why you would go on about judges or being found guilty the whole court case was a farce, she had no chance. Some guy Malkinson i think, just got out of jail after 20 years every MOJ includes a court case.

The police wanted her prosecuted again, i wonder why that would be and why they were told to forget it as they have done enough damage to her life and caused panic with how biased the investigation was.

@IonianNerveGrip in response to your post -

In the Letby case. I doubt an expert jury would've been feasible because of logistics. An agreed expert panel model would've probably been easier to pull off. You left out the part of my post where I talked about what was more realistic.

Okay, so how do you deal with the inevitable deadlock and delays that would cause? Would the court have to force one side or another to capitulate after a certain amount of time and/or vetoes? I don’t often find myself standing up for Letby, but I really can’t see how a system like that would have served her fairly for eg the insulin babies.

But the poor quality and in one case incorrect evidence is a problem in itself. It's all very well saying get another expert to address it, but Dewi Evans having honestly misstated the Shoo Lee paper means it's possible the jury will believe the incorrect interpretation.

How material do you believe the misinterpretation of Shoo Lee’s paper was, and did that misinterpretation undermine the 13 (I think) other references cited by Dewi Evans in his report? Did the misinterpretation mean that a diagnosis of AE was absolutely excluded?

And in terms of poor quality evidence, who should decide what is good quality and what is poor quality within a judicial system? Expert evidence is basically just one person’s opinion - I can’t see how you could ever effectively (and fairly) control for quality, other than by letting both sides challenge the evidence provided and put forward their own instructed experts.

As has been pointed out before, being an expert in a field doesn't translate to being a persuasive performer in court. They're not necessarily the same skillset, and while we're anecdoting, someone being incredibly eminent in their field and also coming across as arrogant is very much a thing!

Agreed - as was ever thus, and ever will be. But the same would be true even in an agreed expert system. What if as a defendant the expert speaking to the strongest bit of your case was also the weakest performer in court or most likely to rile the jury up? At least in the current system lawyers are able to use their experience to instruct the expert that they believe can best support their client’s case.

What a silly response. You've been told before on these threads to stop attributing things to people that they haven't said. As pointed out already, one way to address this is an agreed expert

So you can’t see anything wrong with defendants having to agree to a joint expert, even if that expert is only going to be able to support the prosecution case and undermine the defence case? As per my response above, how would that have played out any better for Letby for the insulin babies? It would have been two whole life orders guaranteed before they’d even got started.

The fact that poor statistical evidence was used at all is bad. As neither of us know how persuasive the jury found it, there's no basis to either agree or disagree about extent- we simply know that it was part of the prosecution.

So the prosecution should not have been allowed to demonstrate that Letby had the opportunity to commit the crimes she was accused of? Hardly any time was actually spent on the rota chart during the 10 month trial, and as the jury didn’t convict her on all counts perhaps they didn’t find it as persuasive as some statisticians would like to think they did.

And the fact is that the prosecution tried to draw inferences from her behaviour, which weren't backed up by expert evidence. Think back to the closing speech for example. It's all very well saying they could've called a forensic psychologist but that doesn't actually mean the jury are going to find that expert persuasive over some cod nonsense. And the prosecution weren't doing anything wrong within our system here either, which is the point.

Do you think we should just remove any consideration of human behaviour from criminal trials then? Even though all crimes are fundamentally, and only, a result of human behaviour. What sort of expert would even be able to provide objective evidence on motive or intent?

Ie does it need to be less narrow. The Law Commission undertook and published a review last year.

I am still none the wiser what you actually think needs to change to make it ‘less narrow’.

The "real possibility" test is frequently considered too narrow, @rubbishatballet , and is a key area on which the law commission has consulted

https://lawcom.gov.uk/project/criminal-appeals/

The rota chart wasn't the only use of statistics at the trial, but it was a particularly fraught intervention because it purported to describe events where a suspicious incident had been identified, in a context where the jury was not permitted to learn how these incidents had been selected,

Clearly, it would have been perfectly possible to charge Lucy Letby with any possible crimes without a spreadsheet suggesting that only nurses' involvement was relevant and weighting the probability of each nurses's involvement equally by omitting contextual data.

So you can’t see anything wrong with defendants having to agree to a joint expert, even if that expert is only going to be able to support the prosecution case and undermine the defence case? As per my response above, how would that have played out any better for Letby for the insulin babies? It would have been two whole life orders guaranteed before they’d even got started.

The assumption I presume is that experts face questioning from fellow experts, not from barristers only, in this scenario. We know that Hindmarsh, for example, was willing to concede to another expert that other causes existed for these test results. We know that he conceded in court that he had guessed at adsorption rates. It's no be hoped that the system @IonianNerveGrip describes would have picked up on such problems, and that the judge in a subsequent case would not have been put in the position of misleading the jury on the existing science. If not - if something slips through and subsequent scientific developments undermine it? That would give grounds to review the case.

Before we start @rubbishatballet I want to stress that what systemic improvements could be made isn't the same point as whether there are problems to begin with. Just so we all remain clear about the known, factual, not a matter of opinion issues with this specific case. There's nothing wrong with concluding that this is the least worst option available, provided one doesn't in doing so minimise the issues.

Okay, so how do you deal with the inevitable deadlock and delays that would cause? Would the court have to force one side or another to capitulate after a certain amount of time and/or vetoes? I don’t often find myself standing up for Letby, but I really can’t see how a system like that would have served her fairly for eg the insulin babies.

You've not shown us how this would differ from the inevitable deadlock and delays in the system we already have. I know your legal experience goes back a bit, so you perhaps haven't seen what the court system has looked like over the last few years.

So you can’t see anything wrong with defendants having to agree to a joint expert, even if that expert is only going to be able to support the prosecution case and undermine the defence case? As per my response above, how would that have played out any better for Letby for the insulin babies? It would have been two whole life orders guaranteed before they’d even got started.

Of course there are potential problems in any system, although obviously if the Letby trial had been conducted under an agreed expert panel system the process would've looked so different that your two life orders claim is a wild guess. How on earth do you know who the agreed experts would have been in a system where roles like those Dewi Evans saw himself filling don't exist? You don't. @Oftenaddled has addressed the rest before me just now.

Do you think we should just remove any consideration of human behaviour from criminal trials then? Even though all crimes are fundamentally, and only, a result of human behaviour. What sort of expert would even be able to provide objective evidence on motive or intent?

It would be better if we were to avoid reliance on the sort of speculation the prosecution engaged in during the closing argument, yes (again not a criticism of any member of the team, they all did their duty in the system we have). If there isn't any expert evidence that would be applicable to a particular behaviour or motive, the jury can be told that.

So the prosecution should not have been allowed to demonstrate that Letby had the opportunity to commit the crimes she was accused of? Hardly any time was actually spent on the rota chart during the 10 month trial, and as the jury didn’t convict her on all counts perhaps they didn’t find it as persuasive as some statisticians would like to think they did.

Your 'perhaps' means nothing. None of us know how persuasive they found the poor statistical claims put forward by the prosecution on the cases for which they did convict, although it's instructive how often the people who believe she's guilty want to make assumptions about the jury's views here.

The prosecution were entitled to do as they did within the system we have, so its not a criticism of any individual, but we need structural change so this doesn't keep happening. I mentioned earlier the possibility of RSS produced guidance and an application being needed to depart from this, if we weren't using agreed expert models.

On the narrowness of test point, evidently you hadn't bothered searching for the document- @Oftenaddledhas linked.

How material do you believe the misinterpretation of Shoo Lee’s paper was, and did that misinterpretation undermine the 13 (I think) other references cited by Dewi Evans in his report? Did the misinterpretation mean that a diagnosis of AE was absolutely excluded?

Evans said in court that he relied on Lee and Tanswell for the specific rash which is diagnostic of air embolism. He had of course misunderstood their paper.

The other symptoms he mentioned in association with air embolism may certainly have been found in the 17 other papers he mentions. If these papers are about neonatal air embolism they will also feature in Lee & Tanswell / Zhou and Lee, which review a far greater number of papers. And indeed, the two Lee papers mention other symptoms that are consistent with (not diagnostic of) air embolism.

But because the other symptoms Evans mentions are all common symptoms, the findings are insignificant. Crying, non-specific rashes, postmortem air in vessels, quick resuscitation, slow resuscitation and failed resuscitation are all applicable to a much broader range of afflictions, obviously.

If it is just a matter of drawing a line to match two symptoms, like an infant school exercise, I wonder how many diagnoses could have been found to match?

It is of course impossible to offer positive proof someone hasn't died of air embolism after autopsy is completed. That is the ace up Evans's sleeve and is why his claim that there were no plausible alternative explanations, despite the pathologists' findings, needs thorough scrutiny.

How material do you believe the misinterpretation of Shoo Lee’s paper was, and did that misinterpretation undermine the 13 (I think) other references cited by Dewi Evans in his report? Did the misinterpretation mean that a diagnosis of AE was absolutely excluded?

It - specifically the “fact” that rashes (not just Lee’s sign) were diagnostic of AE - was critical to the conviction.

The appeal court judgment makes this very clear, because they repeat the misunderstanding to justify ignoring Lee’s evidence.

The police wanted her prosecuted because they thought she was guilty. The jury thought she was guilty. I don't think she will be freed any time soon. AI agrees.

AI can probably work from the time to resolution in British miscarriages of justice to support that pessimistic prediction, yes.

I wonder what the jury thinks now. The question they sent the judge about insulin timings suggests they'd have been very open to hearing the various challenges to emerge re Hindmarsh's testimony.