Letby’s Staff Rota and the “Missing” Deaths

Letby’s staff rota and the “missing” deaths

Be careful what you wish for, Letbyists

CHRISTOPHER SNOWDON

I first wrote about the Lucy Letby case because people were saying the trial revolved around ‘statistics’ when it clearly did not. A now-famous spreadsheet that was shown to the jury towards the start of the trial showed Letby on duty when all the murders and attempted murders were committed. If they were indeed murders and attempted murders, it would seem that there is only one suspect. If there were no crimes, as the defence claimed (except, perhaps, in the case of the insulin poisonings), the spreadsheet is irrelevant.

As various mid/half-wits have pointed out from the outset, there were probably deaths and collapses at the hospital in 2015 and 2016 when Letby wasn’t on duty. If these incidents had been included in the spreadsheet, the 100% correlation with Letby would likely disappear. Maybe the police cherry-picked incidents at which Letby was present? Maybe Dewi Evans, who reviewed the cases in 2017, only identified incidents as suspicious if Letby was on duty?

You have to let your imagination run wild to find a reason why either the police or Evans would do this, knowing that it could lead to an innocent being thrown in prison for life, but the financial investment manager Peter Elston (who has been portrayed as a statistician by Letbyists) has been wanging on about it for two years, starting with a now-deleted blog post in June 2023. He appeared in the recent Channel 4 film about the case, saying that there were 17 [sic] neonatal deaths in the hospital in 2015-16 and asking rhetorically whether this meant that there were two serial killers?! It’s a fairly idiotic thing to say because nobody claims that there were no deaths from natural causes in the hospital, nor does anybody claim that Letby’s mere presence on the ward when a child died is proof of murder.

With the help of Richard Gill, who is a real statistician, Elston pumped the idea that the case is about statistics into the mainstream and that if a statistician had given evidence, the trial would have somehow turned out differently. It is far from obvious what a statistician could contribute since the question is not about numbers but about whether particular deaths and collapses were the result of intentional harm. The argument about statistics would make more sense if the prosecution had said something like “the chances of this happening to an innocent nurse are 1 in 60 million” (which is what happened in the Lucia de Berk case), but no such claim was ever made.

The idea that there were some equally suspicious/unexplained deaths when Letby wasn’t on the scene took a blow last year when details about an additional nine deaths were made public:

“Four of the deaths were babies born with a congenital problem or birth defect, another baby was sadly asphyxiated or deprived of oxygen at birth, the remaining four died of infection and their deaths were precipitated with a period of time consistent with infection, they did not suddenly and unexpectedly collapse and die.”

A related, and slightly less stupid, question is how the cases that were sent by the police to Dewi Evans were selected. Evans initially reviewed 17 deaths and 16 non-fatal collapses. Since the consultants had suspicions of Letby, they may have been more likely to consider non-fatal collapses suspicious because she was there, thereby leading to selection bias (alternatively, of course, it may have been her presence at so many unexplained collapses that made them suspicious of her in the first place).

Peter Elston put these questions to Evans in a series of e-mails which he has decided to publish online. It is a sign of how deluded Elston is that he thinks these e-mails make him look good and advance the cause of Letby’s innocence. They do not.

He begins as follows:

I am a fellow of The Royal Statistical Society, which I hope qualifies me as an “expert” with whom you would be prepared to communicate (see screenshot of my credentials below). I hope it does.

Anyone can become a fellow of the Royal Statistical Society for £165 a year.

Elston then asks several questions about the process by which Evans reviewed the medical notes and whether it was sufficient to avoid bias. Evans replies by saying what he has said many times before: that he asked for the medical notes for all the deaths and collapses in 2015 and 2016 and that he told the police that he didn’t want to know the name of the suspect (if they had one).

Avoiding bias was a crucial part of my remit and I proposed the following. At my first meeting with Cheshire Police (10 July 2017) I requested copies of the clinical records for all deaths and collapses from January 2015 to December 2016, and that the files should include ALL deaths and ALL collapses, not just ones that were ‘unexplained’ or ‘suspicious’. This 2 year window extended beyond the ‘peak deaths’ noted during this period. I told Cheshire Police that if they had any suspects that I did not want to know any names.

… I think it is indicative of the effort we all made to look at each case on its clinical merit, avoiding ‘fitting’ the clinical presentation to comply with Letby’s, or anyone else’s, presence on the ward. As I concentrated mainly on the clinical (medical) notes rather than the nursing entries, no nurse name ‘registered’ with me during my review of 33 notes conducted prior to Letby being named in July 2018.

… As for the number of cases I was dependent on the Police and Chester clinicians identifying the cases. Naturally the conspiracy theorists (and Richard Gill) won’t be satisfied with that, alleging “selectivity” no doubt.

It is difficult to know what would satisfy the ‘sceptics’ on this front other than the police giving Evans the medical notes of every infant treated in the hospital between August 2012 to July 2017 (the timeframe suggested by Elston). That would be a ridiculous workload and it would still have nothing to do with statistics. The important point is that Evans could have declared all of the incidents he reviewed to be non-suspicious if that was his opinion – and he did declare a large number of them to be non-suspicious. Of the 17 deaths of babies born in the Countess – of which 13 took place in the hospital – he flagged up “only” 7 suspicious deaths (the pathologist Jane Hawdon had identified five “unexplained” deaths in a less rigorous review six months earlier).

Letbyists have suggested that Evans only declared a death to be suspicious if Letby was on duty, but (a) he had no reason to do that, (b) he says he didn’t do that, and (c) he didn’t do that, as we shall see. But still the statisticians go on about the Texas sharpshooter fallacy. It is a non-starter, as Evans told Elston…

Where the statisticians get into a stew is this. They state, correctly, that an increase in annual neonatal mortality from 4, 3, 3 , to 10, 5 is within statistical probability. But it’s not the numbers!! It’s the cause of the deaths – unexplained, unexpected, suspicious, etc. This is why engaging statisticians would not have helped. Presumably Defence felt the same.

Elston continued to pester him about the incidents that were not included in the spreadsheet and so Evans decided to take another look.

Evans: As of this week I’ve decided to review all 82 cases other than the ones in the trial, therefore (82 – 17) 65. It’s taking some time but so far she seems to be present at all of the ones I’ve reviewed to date. Note that these cases are “selective”, so the conspiracy theorists will simply say that the cases were chosen because the clinicians “knew” Letby was on duty. And with idiots like Richard Gill around, and the Guardian newspaper giving an unfettered platform to a barrister acting on behalf of England’s worse serial female killer, the publicity will endure.

Elston: Were you asked to do this by Cheshire Police?

Evans: No

Elston: You were given 33 cases to review in July 2017, then another 28 later (total 61). Then you I think told Phil Hammond you reviewed a total of 82. Why do you need to review 65 (82-17) again?

Evans: What has this to do with statisticians?

Elston: You write, “It’s taking some time but so far she seems to be present at all of the ones I’ve reviewed to date.” Why are you reviewing them in terms of whether Letby was present rather than whether, per your other messages, incidents in your view were or were not suspicious?

Evans: What has this to do with the figures I’ve given you re Letby’s presence. You (and that abusive guy Gill) have been bitching for months that the spreadsheet did not contain information regarding the other deaths. I’ve given you the information regarding her presence and the other deaths, and you don’t seem to be able to respond.

When Evans looked at who was present for the other deaths, it did not make happy reading for the Letbyists…

Much has been made of the fact that the spreadsheet identifying Letby’s presence was limited to those deaths that formed the prosecution case, and that the chart should include all the deaths. The prosecution argued that the spreadsheet was presented to show that Letby was present at the time – at the scene of the crime if you like. Nevertheless I thought I should look at all the deaths and note the following. The deaths occurred between 31 March 2015 and 24 June 2016. (As noted earlier I’ve no record of any deaths during August 2016). I received copies of 17 deaths recorded between January 2015 and December 2016. Three of the infants were transferred and died elsewhere. I note the following:

7 deaths that formed the prosecution case: Letby present for all 7

3 deaths elsewhere: Letby present when 2 of the 3 deteriorated. One was Infant K – she was the one at Letby’s retrial. One was a very preterm infant. [The other was very asphyxiated at birth and died elsewhere several hours after transfer]

4 deaths associated with congenital abnormalities: Letby present in 3 cases. One had cardiac problems [Ebstein’a anomaly]. She was however stable but suddenly collapsed and died when in Letby’s sole care. One had multiple congenital abnormalities deemed to be incompatible with life. One had multiple congenital abnormalities but was deemed stable before suddenly deteriorating when in Letby’s sole care. Resuscitation was performed by a consultant paediatrician but was unsuccessful. [The other infant died within 2 hours of birth from conditions deemed to be incompatible with life.]

3 deaths associated with prematurity and infection. Letby was present in all 3 cases.

In summary: Letby was present when 15 of the 17 babies died. She was present at the point of deterioration for all 15 cases.

It is not quite correct to say Letby was present when 15 of the 17 babies died since three of them died in a different hospital.1 But according to Evans, she was present for the deaths of 13 of the 14 babies who died at the Countess and was on duty when two of the three babies who died elsewhere deteriorated. Of the deaths at the Countess, the only one for which Letby wasn’t present involved an infant who “died within 2 hours of birth from conditions deemed to be incompatible with life” and which was therefore very unlikely to have been a murder. Of the deaths that could have conceivably involved foul play, Letby was always present.

Evans also estimated that Letby was on duty for around 20% of the time. In short, she did 20% of the hours and was there for 92% of the deaths. What is a statistician to make of this?

It is at this point that Elston goes quiet, so Evans e-mails him again with another statistic.

Mortality rates at Chester for 2012, 2013 and 20014 were 4, 3, and 3. I calculate this as 0.5, 0.4 and 0.4% mortality per nursing shift. This compares with the figure of 0.3% when Letby was NOT on duty during the 13 months I’ve noted in my letter to you. The mortality rate of 9.2% when she was on duty is therefore a massive outlier.

What this means if anything is, I believe, a matter that is of statistical usefulness and outside my expertise. Its usefulness as something that is of clinical significance is a matter that needs exploring.

Elston still has no answer, so Evans e-mails again.

This the first time in the Letby case I’ve explored the possibility of using statistics. As I’ve noted the 9.2% figure for deaths per (Letby) shift seems a massive outlier compared to figures of <0.5% otherwise. At least the figures I’ve given you cover all mortalities over the specific period, and, indirectly, the years 2012 -14. (Indirectly as I don’t know whether Letby was on duty during one or more of those deaths.).

Whether it’s CLINICALLY significant or not demands a combined statistical/ clinical view.

After apparently receiving no reply again, Evans prompts Elston once more.

9.2% v 0.3%. The significance of these 2 variables (using actual figures not % of course) is what I need.

When Elston finally replies, he does something that is common amongst conspiracy theorists when they are losing an argument. He asks for an impossible amount of additional information.

Dear Dewi,

Thank you for your patience in awaiting my response. I can confirm that I have now looked through the data you sent me and your proposed statistical analysis. Whilst I can see from the way you have presented this data that it may seem like these numbers are significant, in order to carry out a robust and accurate statistical analysis, we would need to take into account a variety of other factors, such as:

1. How many nurses are there per shift?

2. Are all shifts the same duration and is there a difference between the number of nurses on day vs night shifts?

3. How many nurses are part time vs full time, and how many extra shifts did they do on average during that time period?

4. What are the differences in competencies between nurses?

5. We need to compare day with night shift data due to the known higher incidence of adverse events happening at night; as well as the proportion of shifts in terms of days vs nights that Lucy worked, as well as the other nurses.

6. We need to have a look at all the names of people on shifts when the deaths occurred. And not just the deaths/collapses that ended up going to the criminal court, but all deaths/collapses. For all nurses, as well as doctors.

If you can forward me the missing information above, I would be happy to input into the analysis of this data. If this information is not available to you, I’m afraid it will not be possible to analyse this data in a statistically valid way.

In subsequent e-mails, Elston changes the subject to air embolism and other issues that have nothing to do with statistics. Evans made one more attempt to get him to comment on his area of supposed expertise but to no avail…

I’m very disappointed in your failing to address the information re Letby’s presence during the deaths. The general whinge of the statisticians re the spreadsheet was that the prosecution had not disclosed information about all the deaths. You now have that information, and I’ve asked you a perfectly reasonable question.

What is the statistical significance if any of her presence when 15 of the 17 babies collapsed and died. (The other 2 babies sadly had unsurvivable conditions).

I also gave you information re her shifts (which I only received in September this year). I’m very aware of other variables and I don’t have the information. But that should not stop you given a qualified opinion.

… I fully expected you to respond to my 9.2% v 0.3% query, and would understand if your response was qualified. But ducking it all together? No use at all! And of course this aspect of the case is the only part for which you can claim any expertise.

What should we make of this information? Firstly, it should be obvious that Evans and the police did not deem a death to be suspicious simply because Letby was on duty. If they had, she would have faced many more murder charges. Secondly, the statistical association is incriminating and could have been used as further circumstantial evidence against the World’s Unluckiest Nurse. Thirdly, it was not used as evidence and statistics played no part in the trial.

It is an interesting fact that Letby was nearly always on duty when babies died in 2015-16, but whether or not she was on duty when babies died of natural causes is of no real relevance to the case. How could it be? If Letby had been at home with an alibi when all the babies who were not on the indictment died, it would still be irrelevant. The trial was never about statistics. It was about intentional harm.

••••

The Liberty Beacon Project is now expanding at a near exponential rate, and for this we are grateful and excited! But we must also be practical. For 7 years we have not asked for any donations, and have built this project with our own funds as we grew. We are now experiencing ever increasing growing pains due to the large number of websites and projects we represent. So we have just installed donation buttons on our websites and ask that you consider this when you visit them. Nothing is too small. We thank you for all your support and your considerations … (TLB)

••••

Comment Policy: As a privately owned web site, we reserve the right to remove comments that contain spam, advertising, vulgarity, threats of violence, racism, or personal/abusive attacks on other users. This also applies to trolling, the use of more than one alias, or just intentional mischief. Enforcement of this policy is at the discretion of this websites administrators. Repeat offenders may be blocked or permanently banned without prior warning.

••••

Disclaimer: TLB websites contain copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available to our readers under the provisions of “fair use” in an effort to advance a better understanding of political, health, economic and social issues. The material on this site is distributed without profit to those who have expressed a prior interest in receiving it for research and educational purposes. If you wish to use copyrighted material for purposes other than “fair use” you must request permission from the copyright owner.

••••

Disclaimer: The information and opinions shared are for informational purposes only including, but not limited to, text, graphics, images and other material are not intended as medical advice or instruction. Nothing mentioned is intended to be a substitute for professional medical advice, diagnosis or treatment.

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the official policy or position of The Liberty Beacon Project.