Lucy Letby case expert witness was under fitness to practise investigation

[ITMA2000]

Wow! Pot, kettle, black? Need a new investigation?
Lucy Letby case expert witness was under fitness to practise investigation during trial | Lucy Letby | The Guardian

And you’re missing the point that using a very small and unrepresentative data set is prone to throwing up anomalies.

Could be a statistically irrelevant cluster.
Could be one self- extubating baby.
Could be 4 babies badly intubated/ intubated with the wrong size tube by the same doctor.
Could be two tricky in-between sized babies for whom there isn’t a perfect fit ET tube.
Could be a defective batch of ET tubes.
Could be a killer nurse.

You can’t just leap to the latter, even supposing there was a killer nurse around. It would make it more likely, but still not a certainty.

And you certainly can’t use these incomplete data to prove that LL was a killer nurse in Chester a few years later, because that’s circular and illogical, and the stuff of witch trials.

I've only listened to the first few minutes at the moment but there's something about the Evans replies described as "juvenile point scoring and low key insults" - not sure what it could be.

They looked at 100 vented shifts not involving Lucy and tubes came out unexpectedly less than 1%. Then they looked at Lucy's and they came out 4 times over just 11 shifts.

If 4 tubes are coming out in some sort of anomaly occasionally then surely that would bump up the stats to far more than 1% overall? Otherwise we'd have to go many 100s of shifts where none came out to even out to 1% and then go "oh the four dislodgements anomaly just so happened to occur on her shifts after 400 shifts with NO dislodgements"...not really buying it sorry.

Why does she have much higher figures than every other nurse? Oh "we don't know, there's not enough data" how convenient. Again this is why the statisticians get ignored.

There isn’t enough data.

How convenient for you that all you need to hear is 4 out of 11 to reach a conclusion and you don’t need any other information to be firm in that conclusion, you can convince yourself of anything if you are happy to reach conclusions with very little data.

Source for the first part?

@kkloo it was in the Panorama doc

It's not "convenient" for me-it's because she only worked 11 shifts. Now the figures for COCH would be interesting. But then she'd moved on from merely dislodging tubes by then hadn't she.

What sort of info would I need? She's been compared against the other nurses, and people are trying to magic up any excuse for why her stats are wildly different, yet again!

I’d need to see a different source as even when panorama corrected that segment it was still misleading.

And yes it is “convenient” for you, imagine thinking it was “convenient” that people wanted to have all the facts before assessing a figure 😅

@Firefly1987 @kkloo She worked 11 ICU shifts in 2015 at LWH, which has a 16-bed ICU. Unless the ICU had only 1 vented baby per shift on average, 11 vented shifts can't be the right number.

The description of the review (upthread) talks about the review, and says nothing about reviewing 100 non-Letby shifts, or how these were chosen and how it was ensured that they're comparable to Letby's shifts on factors that affect extubation rates.

And, again, in a review of all event types, you have to compare the number of found events of all types, to the expected number of events of all types. You have to compare like with like.

You’re making the same mistake that Moritz and Coffey made in the documentary.

The <1% dislodgements in 100 vented shifts is from Thirlwall. A vented shift is a shift in which one baby was ventilated.

At LWH, with 16 NICU cots, in addition to the HDU cots, it’s reasonable to suppose that at least half of these, probably considerably more, contained vented babies during Letby’s 11 shifts. So let’s say 10 - 15 for easy maths. The rate is therefore 4 in 110-161 ventilated shifts while LL was present. This is quite close to that quoted background figure, and quite a long way from the 40% that you and Panorama insinuate.

https://www.thetimes.com/uk/crime/article/lucy-letby-medical-documents-baby-hospital-8d7m5qxc8

In case people can't read the article-

https://archive.ph/LUvF0

A premature baby’s chest drain was dislodged repeatedly while Lucy Letby was on duty during a training placement, according to medical documents seen by The Times.

Among the cases at Liverpool that Cheshire police has asked expert medical witnesses to examine is the case of one baby born in October 2012.
Medical notes reviewed by the experts record that the baby’s chest drain was dislodged once on October 26, twice on October 27 and once on October 29. The child’s breathing tube also fell out on October 29. Letby was on duty on all the days.
“It’s important to point out that chest drains can and do fall out, but not in my opinion with the frequencies in his case,” the expert reviewing the case wrote. “The number of chest drains this baby had over such a short period of time was extraordinary.”

In another case examined by the expert witnesses, a premature baby born at Liverpool collapsed in November 2012 after water from the ventilator circuit went down the baby’s endotracheal tube. The experts concluded that the reason for the presence of water in the tube was unexplained. Letby was again on duty.

More bad luck for the world's unluckiest nurse I guess.

@Firefly1987
who were the medical experts the police got to look at the files?

@kkloo does it matter?

Yes obviously it does.
If it were Dewi Evans or Sandi Bohin then we already know they ignore the obvious so I wouldn’t trust their opinion on it.

I think this is the problem where you're looking for any event which happened while LL was on duty and only when she was on duty. On duty means somewhere in the building on that day. You can consider the chest drain dislodgements as suspicious, but 14 years later it's unlikely whether you could pinpoint exactly where a trainee was at a particular time unless they were recording some notes which still exist. If you're going to make an accusation then you at least need to be able to say it's likely she was there in that particular room and was left unattended or unobserved long enough to dislodge the tubes.

Those incidents were about 3 or 4 weeks into her placement. It would be an extremely bold move for a supervised trainee in a hospital they didn't normally work in to do something like that deliberately. The risk of being caught would have been huge.

The thirlwell documentation about the 2012 placement doesn't say how many rooms there were at the time, only that it increased to 6 critical care rooms with 4 babies in each after the rebuild (12 ic and 12 hd) whereas in 2012 there were 16 ic beds. I'm assuming that the 16 ic beds couldn't be all in one room. Unless we're potentially blaming any event during that 11 week period on LL just because she was at work in the building I think there needs to be some actual positive evidence of her at least being in the right room at the right time.

@Firefly1987 you say she's the world's unluckiest nurse. Well is she? We could perhaps get some idea of that by looking at other comparable periods during that year and seeing if there were any unexplained events during those periods. I think that would be considered too costly an exercise though. They're no longer looking for suspicious events blindly, they're starting with a suspect and seeing if there's anything that could be a crime that could fit.

You can't say "what are the chances of it happening to her?" for any suspicious or unexplained event when the only days they searched were those when she was on duty and the only criteria for suspecting her was that she was at work that day.

Look again at the breadth and scope of the LWH review. 1632 care days, in a tertiary hospital, for
“any potential suspicious events”, non-blinded with a focus on a suspect. I’d be surprised if such a review found no events. How many events would you expect it to find for an innocent nurse?

“Following contact with Cheshire Police in Summer 2018, a review of case notes was undertaken for both periods when Letby worked at LWH. The process that was followed is outlined below (data relates to both time periods, 2012 and 2015). In order to review any potential suspicious events at LWH a review was undertaken. All incidents reported in the incident reporting system that included the name LL were examined. All deaths occurring on the days when LL was on duty were reviewed. The electronic patient record system (Badger system) was searched to identify any note entries made by LL. She had contact with 33 babies on a total of 324 patient days. Each of these days of care was reviewed by a neonatal consultant to identify any unusual or suspicious deterioration in the patient's condition. In addition, all babies who were on the neonatal unit on the days when LL was on duty were identified using the Badger system. Each of these days of care were reviewed by a neonatal consultant to identify unexpected cr unexplained deterioration in clinical condition. There were 257 babies on the unit during those days with 1,308 days of care requiring review”

I see you’ve swerved to a different type of tube, a chest drain, and then back to ET tubes.

On the latter point, where’s the original evidence (for example trial transcript or Thirlwall document) for this please? Condensate entering an ET tube is quite a common occurrence and there’s lots of literature on the phenomenon, if you’d care to search.

So an expert implying this doesn’t happen is either not an expert, is lying, or has been mis-reported.

Letby was found guilty because the jury believed she was not because there wasn’t someone arguing about statistics in her defence.

The jury’s part on this is over beyond the CCRC and or COA making assumptions on what impact certain things would have made or not.

For all we know a statistician might have made a difference, also extremely important to note that it’s not just statistics that are in doubt but all of the medical evidence, the testimony, the integrity of the investigation and so on

The jury believed what they were told, which was that she and only she was there for every single suspicious death or collapse. But the way the prosecution deemed whether an incident was suspicious or not depended on whether she was there. Do you see the problem?

The Guardian is not intending to be impartial here, and that's quite right.

This isn't a news story, where the idea is to report transparently. It's investigative journalism, where the idea is to seek out evidence that might inform the discovery of the truth.

Would you have suggested that The Washington Post ought to have been 'impartial' when investigating Watergate?

@PinkTonic

So you are saying that she was not there for the incidents that they said she was? If so she never said so in court, and nor did her team.

Likewise there was of course other people on shifts during all incidents

I remember the other nurses saying incidents were rare - which cases do you know about at the hospital that happened when Letby wasn’t on shift?

Also the medical professionals who reviewed the cases did not have explicit sight of who was on shifts. All they were doing was looking at how these babies might have died…..

As they have said they have been asked to review plenty files where they have said no that is not suspicious etc

Were they supposed to lie and say the deaths were not suspicious? As in their opinion the deaths/collapses were not normal

I am not sure how you think this was personal against Letby - the panel of experts who came out against the court experts did not even have access to the full medical details of each baby

There's obviously no audit of events that happened when Lucy Letby wasn't on shift, so please don't think this is comprehensive, but you can read about some of those events here and in the articles linked

https://www.reddit.com/r/LucyLetbyTrials/s/4f7fHi9bcL