What the Letby trial tells us about justice

There is a recurring pattern of defence teams coming up with a counter-argument only after a first trial has resulted in conviction. Something needs to change.

Like many people in the UK and perhaps further afield, I have spent a significant amount of time wondering about the former nurse Lucy Letby. Some people are convinced that she is innocent of the multiple murders and attempted murders of babies in her care at the Countess of Chester Hospital. Others are satisfied of her guilt.

I simply don’t know.

I suspect that puts me in the majority. It’s a very uncomfortable majority because Letby is serving a whole life prison term (meaning that she will never be eligible for parole) for crimes which have sparked a huge debate over whether they ever took place.

Photo by Claudio Schwarz on Unsplash

It is just over 100 years since Lord Hewart, Lord Chief Justice, said “justice should not only be done, but should manifestly and undoubtedly be seen to be done.” It may well be that justice was indeed done in Letby’s case, but the circumstances are so complicated and the debate so intense that, right now, there can’t be too many people in the UK who are confident that justice has been seen to be done.

The problem lies with a criminal justice system that manifestly doesn’t deal well with scientific evidence. In the late 1990s and early noughties, several women were convicted of murdering their children (in each case, more than one child) based on the expert evidence of a senior paediatrician who told the jury that only in the most unlikely of coincidences would sudden infant death syndrome occur multiple times in the same family. It was several years before, in 2003, the justice system cottoned on to his errors and quashed a number of these convictions, including those of Sally Clarke, Donna Anthony and Angela Cannings.

Another medical miscarriage of justice was that of Ian and Angela Gay, a married couple convicted of manslaughter by force-feeding salt to a child they were planning to adopt. After 15 months in prison, the Court of Appeal quashed the conviction on the basis of new medical evidence that the child may have suffered from a health condition which caused his body to retain salt in the system.

A common feature in all these cases, and others, is prosecution based on a hypothesis of deliberate harm causing death and a defence that seems not to have been equipped until several years later to put forward a rival theory to explain that the death was due to natural causes. Perhaps the problem isn’t so much a matter of handling scientific evidence at trial; maybe it’s a case of defence teams not knowing how to get hold of counter-arguments first time around. But either way, it’s a very serious problem.

A trial is not a search for the truth. It is a contest between two opposing arguments. The time for seeking out the truth is before the trial starts. Once a trial is underway, the prosecution will present evidence that it believes is sufficient to provide a realistic prospect of conviction. In other words, the prosecution has decided that the evidence is sufficient for the jury to be sure of the defendant’s guilt (the phrase “satisfied so that you are sure” having replaced the more famous “beyond all reasonable doubt”).

It would be cynical to suggest that the prosecution doesn’t have to believe the case it puts forward; it only has to be satisfied that it can persuade a jury to believe it. But, where medical evidence is concerned, if the prosecutor has been presented with experts who all put forward the same explanation for what has happened, there is good reason for the prosecutor to accept what those experts are saying and present that evidence in court.

If the defence team hasn’t been able to find an alternative medical theory to put to the jury, it is highly unlikely that the trial will produce anything to fill the gap in their argument. The defence may be left fighting with little more than the defendant’s denial, coupled with a challenge that the prosecution’s case is pure speculation. But history shows (as in the cases of Clarke, Cannings and the others above) that juries can be persuaded of guilt after hearing medical experts testify confidently that death was the result of deliberate harm.

And let’s not forget: parents do sometimes commit murder, as do doctors and nurses. There will be times when the prosecution experts will be right.¹

The case of Lucy Letby is much more complicated than the historical cases I have mentioned. The prosecution against Letby alleged 22 counts of murder or attempted murder relating to 17 different babies using multiple different methods to inflict harm. She was found guilty in relation to 14 babies (seven of murder and seven of attempted murder) using five different methods of causing harm. During the trial, seven medical experts gave evidence – all called by the prosecution. It is important to state that there was also factual evidence pointing to her guilt, for example, an allegation (denied by Letby) that she falsified medical records relating to some of the babies and much else besides.

The defence had instructed experts and had at least one of them in court throughout the trial, but none of the defence experts were called to give evidence at the trial. That was the defence’s choice. No one was stopping Letby from calling expert evidence to refute the prosecution’s case. The obvious inference is that the experts were thought not to be of any help. But that is mere supposition. Letby’s lawyers have not said why they didn’t call their experts; they are prohibited from divulging the reason without Letby’s permission. So we can’t know, and won’t know, unless Letby gives permission for the reason to be made public – something she may yet have to do if she is to succeed in her application for a further appeal.

Since the trial, Letby has instructed a new legal team. They have brought together a panel of 14 international medical experts, who have prepared new reports. Those reports have not been made public, but selected parts of their findings have been disclosed at press conferences and by the individual panel members in interviews. We are told that the panel has found entirely innocent explanations for every one of the deaths and the deteriorations which were alleged to have been deliberately inflicted.

To the casual observer, the emergence of alternative hypotheses advocated by serious experts in their field must be enough to raise doubt. If not enough doubt to overturn her existing convictions, then at least enough to result in a retrial at which the new evidence can be explored. Indeed, one might think that a retrial would be ordered in very short order.

But it’s not that simple. For a start, the prosecution experts say that the defence experts have got a lot of the facts wrong. A re-trial is not going to be ordered without some preliminary testing of the new opinions to see if they pass a basic smell test. That is the job of the (much maligned) Criminal Cases Review Commission.

A more immediate obstacle is in the form of the legal rules which govern the appeal process. In much the same way that a trial is not designed to be a search for the truth, the appeal process can often be more concerned with whether the trial rules were followed rather than determining whether the right result was reached. A look at Letby’s first appeal shows how that can go.

Letby’s appeal against her conviction

Much has been made of the fact that Letby submitted an appeal which was dismissed by three very senior appeal judges. Their decision is set out in a 58-page judgment.

But an appeal is not a re-hearing of the case. The appeal court’s decision does not tell us that the judges reviewed the evidence and came to the same conclusion as the jury. As paragraph 13 of the judgment shows, Letby’s lawyers were not arguing that the evidence pointed to Letby’s innocence; they were primarily arguing that the judge made legal errors in his handling of the trial. (There was a further ground which I discuss below.) And, to be strictly accurate, this wasn’t even an appeal; Letby was seeking “leave” to appeal (ie permission to hold a full appeal hearing), which was denied.

If the defence was right that the judge had mishandled the trial, it would, of course, be justification to quash the conviction and start again. But the fact that the Court of Appeal rejected those arguments tells us only that they supported the judge’s handling of the case, not that they agreed with the jury’s conclusions. The guilty verdicts are not something on which the appeal judges opined.

It is important to remember that, whilst judges preside over criminal cases, it is entirely for the jury to decide matters of fact. That applies equally to expert evidence, no matter how technical or complex the expert evidence might be. So long as the evidence was legally admissible, it is for the jury to decide whether they accept or reject what the expert(s) have told them – something which the appeal judgment mentioned more than once:

the quality of a witness’s evidence … is quintessentially a matter for the jury to assess (paragraph 118)

we are wholly unpersuaded that it could not properly be left to the jury to decide whether they were sure … that the expert witnesses had rightly excluded other possible causes, had rightly relied upon a number of signs and symptoms … and had correctly [reached their conclusions] (paragraph 148)

And, in response to a suggestion from the defence that the expert evidence was too vague or weak to be admissible:

Such criticisms of the experts were … matters for the jury to consider and evaluate along with the other evidence in the trial (paragraph 38)

The test for admissibility of expert evidence is essentially that the evidence addresses relevant information which would otherwise be outside the court’s own knowledge and that the expert is competent and has expressed an opinion that is “reliable”.²

Letby’s appeal application also included a request to bring new expert medical evidence. But that, too, was rejected by the judges. The evidence the defence wanted to bring was from Dr Shoo Lee, who had written an academic paper which had been referred to by the prosecution. Dr Lee has since been appointed to lead the expert panel retained by Letby’s new defence team. But, at the time of the appeal, which was conducted by the original legal team, the defence wanted Dr Lee to give evidence to refute only one specific issue from the prosecution experts’ arguments: whether some of the babies had died as a result of a deliberately caused air embolus.

The appeal judges concluded that this new expert evidence was “misguided” (see paragraph 190) because it could only serve to undermine an argument “that was not the basis on which the prosecution witnesses reached their opinions” (paragraph 189). As I mentioned above, it is not enough to find an expert willing to challenge the prosecution’s own experts. An appeal court will want to be sure that it does so in a way that is relevant.

Somewhat worryingly, the appeal court also rejected the new expert evidence on the grounds that it was now too late to bring forward this evidence, saying:

there was ample time to obtain [this evidence] during the trial (even if not before), and no good reason has been shown why the applicant should now be allowed to adduce evidence which could have been obtained and adduced at the appropriate time. The interests of justice require a defendant’s whole case to be put forward at trial unless there is good reason why that could not be done (paragraph 186)

There is no doubt that a justice system needs rules. One cannot approach each and every criminal trial with a blank sheet of paper on which to develop rules designed to suit the unique features of that case. And rules governing new evidence are necessary. No justice system would survive if every defendant were free to treat the trial as a dry run in which they could put forward only part of their case, knowing that they would be able to have another go with different evidence if the first attempt didn’t work out so well.

Section 23(2) of the Criminal Appeal Act 1968 sets out the conditions under which the Court of Appeal may admit new evidence during an appeal. In support of their decision to reject Dr Lee’s evidence as too late, the appeal court quoted the words of Lord Bingham, when he was Lord Chief Justice, sitting in a previous case in which an attempt was made to bring belated expert evidence:

It would clearly subvert the trial process if a defendant, convicted at trial, were to be generally free to mount an expert case which, if sound, could and should have been advanced before the jury [my emphasis]
R v Jones (Steven) [1997] 1 Cr. App. R. 86, p93

The reasonableness of Lord Bigham’s statement depends critically on the inclusion of the word “generally”. It cannot be a blanket rule. And, happily, it isn’t. The Criminal Appeal Act specifically lists “a reasonable explanation for the failure to adduce the evidence in those proceedings” as a factor the court should have regard to. My (probably imperfect) research also suggests that other courts have shown a willingness to consider the strength of the new evidence as a relevant factor, albeit in very different circumstances to this case (R v. Solomon [2007] EWCA Crim 2633).

But I am still left feeling distinctly uncomfortable that Letby’s appeal court thought it wise to state that the untimeliness of the new evidence would still be a factor in excluding it, even where the result of including it would have been to undermine the murder conviction. No justice system can survive in a democratic society if it creates the impression in the public mind that, even if appeal judges conclude that new expert evidence would undermine a key part of a murder conviction, they will not allow an appeal unless the evidence is brought forward at the first possible chance.

It is also worthy of note that, although the new evidence related to only one of the alleged methods of killing, it could potentially have undermined more of the murder convictions. That is because it was a feature both of the prosecution’s case and of the appeal court’s decision-making that, once the jury had reached a decision in relation to murder (or attempted murder) by one method, that became a factor which could lead them to be sure that other methods of harm had occurred. Specifically, in deciding whether to allow the evidence of harm caused by air embolus, the judge cited:

In the context of ... the fact that … someone in the [neo-natal unit] deliberately sought to harm [two other babies] by adding insulin to their nutrition … I am satisfied that the evidence [relating to harm by air embolus] is admissible expert evidence for the jury to consider (paragraph 131 – my emphasis)

It puzzles me that the judge referred to insulin poisoning as a “fact” – something which is hotly contested – which he treated as a contributory factor in deciding to allow expert evidence on air embolus. Even more alarming (to me, at least) is that the Court of Appeal said:

[Letby] admitted that both babies had been poisoned by insulin, but denied that she was the poisoner (paragraph 30 – my emphasis)

The trial transcripts are not available so far as I can tell, but I have researched Letby’s evidence as reported contemporaneously by several news journalists in attendance at the time. When it was put to Letby by the prosecution that scientific evidence establishes that two of the babies were poisoned with insulin, she responded to the effect that “if that's the evidence", but she denied that she administered it. It doesn’t appear that, in giving her evidence, Letby ever volunteered that the babies had been poisoned.

Where does it all go from here?

Convicted criminals are not permitted to mount endless appeals. Having failed with her previous appeal,³ Letby’s hopes rest on being able to persuade the Criminal Cases Review Commission (CCRC) to accept that her new team of experts have provided good enough grounds to send the case back for a further appeal. She may also have to be doing this without the benefit of legal aid, which can only add to the hurdles that she faces.

Interestingly, the new defence team initially suggested that they would take the matter directly to the Court of Appeal, bypassing the CCRC. But, in February, the CCRC confirmed that an application had, in fact, been made to them by Letby’s lawyers.

The detailed opinions of Letby’s new experts have not been published and there is no reason why they should be. But a number of the experts have spoken at press conferences held by the defence team – something that we are not used to seeing in the UK – and some have given interviews to journalists.

It has become clear from interviews given to a BBC Panorama documentary that the difference of opinion between the experts for the prosecution and for the defence is not merely a matter of expertise, but also, in some cases at least, down to matters of fact.

In the case of Baby O,⁴ the prosecution argued that death had been caused by injury to the liver. The lead expert, Dr Dewi Evans, and the pathologist who examined the baby after death, described the cause as “blunt force trauma”.

In stark contrast, Dr Neena Modi, professor of neonatal medicine at Imperial College London, and now one of the defence panel, says the cause of the injury was “most likely shearing stresses during delivery”. But one of the original defence experts, Dr Mike Hall, who wasn’t called to give evidence, countered this by pointing out that the baby was born by Caesarean section, so no such stresses occurred.

Others of the new expert team have a different theory. Drs Neil Ayton and Richard Taylor, both neonatologists, suggested that the liver was damaged during the attempted resuscitation of Baby O. Dr Ayton describes how the liver might have been pushed in such a way that it was damaged by a needle that is known to have been inserted into the abdomen during the resuscitation process.

The needle theory was, in fact, considered during the trial. The pathologist who examined Baby O at the time found no evidence of a needle having damaged the liver. On top of that, the prosecution’s expert, Dr Dewi Evans, points out that, by the time the needle was inserted into the abdomen, the baby was already at death’s door. So even if there was needle damage in the final moments of the baby’s life, it still doesn’t explain (to him) why the baby needed to be resuscitated in the first place.

So, in the case of this one baby, competing hypotheses are:

1. a blunt force trauma theory, which the prosecution says fits the facts and is a complete explanation for what ensued;

2. a needle damage theory which can only explain the liver damage as an event that occurred after the baby’s resuscitation had begun, but which does not explain why the baby was in a state where resuscitation was called for in the first place; nor does it fit the facts found by the post mortem pathologist; and it was explored by the defence, unsuccessfully, at the first trial; and

3. a birth trauma theory, which Letby’s own defence expert says is not compatible with the baby’s method of delivery.

As this conflict of opinions demonstrates, it is one thing to have a theory about how a baby died; it is quite another to have a theory that fits all the known facts and will stand up to cross-examination in front of the jury.⁵

In civil litigation, it is normal practice for the opposing experts to meet in advance of the trial with the goal of narrowing down points of disagreement. By the time the trial begins, it may be much clearer what facts will need to be proved in order for one or other expert’s opinion to stand up.

The same process can be ordered by the court in a criminal trial, but it is not the standard practice. If such a meeting had taken place before Letby’s trial, we wouldn’t know what the outcome was because the defence decided not to call their experts so any areas of agreement would not have needed to be disclosed. In a criminal trial, the defendant is not required to prove their innocence: it is for the prosecution to prove guilt.

Conclusion

It is abundantly clear that the Letby case involves a complex set of interweaving arguments that need to be thought through very carefully indeed. It requires the sort of analysis that senior lawyers are more than capable of conducting. But can we rely on a jury to think through all the logical building blocks in the right order to form a conclusion and, at the same time, identify whether or not the expert medical evidence was up to scratch?

The cases of Sally Clarke, Angela Cannings and others mentioned earlier suggest that it may be asking too much. Some people are already convinced that Lucy Letby’s name should be added to that list. I remain unsure. But I am definitely uncomfortable with the way all these cases (Letby included) proceeded.

I think there is a strong case for saying that the criminal justice system needs to revisit the way that it handles complex expert evidence, probably at the initial trial and certainly at the appeal stage.

I have in mind the creation of a new procedure by which the court is permitted to usurp the role of the jury in relation to expert evidence. In much the same way that a defendant can argue for (some of) the expert evidence to be excluded (ie treated as inadmissible) on the grounds of reliability, I would like a court to be required to reject expert evidence if it decides the evidence is not of sufficient quality that it can be accepted by the jury.

I recognise that my suggestion would go against the principle that matters of fact are treated as solely for the jury to decide. It would not be a simple change to make. And it would doubtless be far more complicated to formulate in practice than my simple, one-sentence articulation of it suggests. But I think the criminal justice system needs to give it a go.

1. My father was a medical practitioner. He was called to give expert evidence by prosecutors in numerous criminal trials, across a range of different circumstances – but none, so far as I am aware, in which the key question was whether there had been a criminal act in the first place.

2. The factors determining reliability are fairly extensive, as set out in Criminal Practice Directions 7.1.2 and 7.1.3.

3. There have actually been two appeals. Letby was convicted in two separate trials and so was able to mount an appeal after each trial. Both were unsuccessful.

4. My description is based on interviews shown in the documentary from 29:30 to 41:00

5. The Panorama documentary also interviewed another pathologist. This second pathologist (who spoke on condition of anonymity and was voiced by an actor) had read the post-mortem report. The pathologist confirmed that there was no evidence found to support the needle damage theory, but also told Panormama that the blunt force trauma theory was “unlikely”. Interestingly, the broadcast interview does not include this pathologist offering a theory to explain what did cause Baby O’s death.