Why I am angry about the Letby murders

The cancer of blame is spreading

1.    Agony

“No parent should outlive their child.

I heard this maxim from an ICU consultant who had seen what the death of a child did to parents. I’ve seen it in friends. It hollows-out an adult life. There is no respite from the purgatory of grief.

How much worse to know that you outlived your child, not through bad luck or happenstance, but because of the shameless malfeasance of a nurse to whom you entrusted the care of your tiny baby.

For this we have no words. Nothing we can say will do justice to the pain of the families affected by Lucy Letby’s murders. We can have compassion, but little else to offer.

For the parents and families of Letby’s victims, as well as for her own family, who, lest we forget, have also lost a child, I am truly sorry.

2.    Anger

But this is not why I am angry.

I am angry because it is unlikely we will learn how to prevent this happening again.

Healthcare just doesn’t learn.

By contrast, there are some industries, notably the airline industry, that have learned how to reduce death and accidents. Matthew Syed has written eloquently on this in his best-seller Black Box Thinking.

Even when there are hundreds of deaths – including those caused by malfeasance – the airline industry learns collectively, makes systemic changes and monitors near-misses with ruthlessness. It wasn’t always like this; the industry has learned how to adapt over the last 50 years.

Here’s two statistics to back up this assertion:

1. the chance of being killed in an IATA registered airliner is about 1 in 4.2million flights. That’s equivalent to 5,000 YEARS of twice daily flights.

2. the chances of an inpatient suffering an adverse event in an acute care facility in an OECD country is about 1 in 10.

Personally, I know something about how easily this happens. My life, along with the life of a patient and her family, were all nearly devastated when I overdosed a child on insulin.

The terrible statistics about preventable harm haven’t budged in decades despite the heroic efforts of Martin Bromiley, James Titcombe and many others.

Why?

3.    Retribution

The airline industry worked out why. They found the cancer. And they named it.

Blame.

They worked out that a culture of justice – where no one is blamed for accidents – promotes reporting (and “speaking up”), which creates an open arena for discussion of risk and error. That openness, in turn, leads to a systemic approach to investigation and a reduction in harm through changes to systems.

With any sort of blame, those positive associations go into reverse. Blame makes people scared and secretive, stops them from speaking up. Difficulties are ‘hushed up’, change remains focused on individuals being ‘held to account’. Nothing changes and thousands more patients are harmed every year.

We have known this for decades.

And yet you only have to look at the reaction to Lucy Letby’s conviction to see that blame is the stick that many people want to weild, in order to find a solution to this problem.

4.    Blame

The people held out for blame for the extent — if not the fact — of the Letby murders, are the senior leaders who, nearly a decade ago, made some decisions which in retrospect may have been poorly judged.

To wit: the chief nurse has been suspended from her current role; the paediatric consultant who first raised concerns is suggesting that NHS managers should be regulated; the medical director and CEO have both put out personal defences of their actions; commentators have argued that there are just ‘too many managers’ and the NHS should instead be managed by doctors (a suggestion that was little in evidence when Harold Shipman was exposed, or during the Bristol heart surgery scandal); others have decided that these managers failed to implement effective whistleblowing policies, lorded over a culture of bullying and instead made those who reported concerns feel they were to blame for causing trouble.

In this last account, the managers are being blamed because of a tendency to blame. The irony would perhaps be amusing if the subject-matter is less terrible.

It is clear, from any reading of the senior NHS culture that pertains in and around the board-rooms of our trusts, that these senior leaders themselves lived in a context where blame was rife from above (NHS England, CQC) and around (the general public, MPs). You can be sure that these managers were blamed for everything: trust performance in general; financial issues;  ambulances queues; A&E waiting times; complaints in general; complaints in particular; the parlous state of the hospital buildings; the food … they were ‘held accountable’ for an endless series of problems, most of which they had little means to control.

In particular, they are now being blamed for putting ‘the reputation of the trust’ above patient safety concerns. Reframe this as their natural desire to avoid inevitable shame and humiliation when these extraordinary accusations were found to be imaginings … and the decision seems more rational, more personal and less corporate.

5.   Retrospection

The overarching consensus is that Lucy Letby’s extended period in paediatric ICU was the managers’ fault because they should have acted differently at a point when, as we now know, Letby was killing children. But we should remember that at the time they made these decisiosn they didn’t know that.

Sidney Decker, an Australian pilot, accident investigator and safety expert, has written an outstanding book on safety called the The Field Guide to Understanding ‘Human Error’.

One of the succinct and powerful diagrams in his book is this:

From inside this “tunnel of events”, no one can see what’s going to happen in the future. There are no counterfactuals in the present, only in retrospect:

Going back through a sequence, you wonder why people missed opportunities to direct events away from the eventual outcome. This, however, does not explain their performance.

Their performance – in other words, their decisions – is explained by the context in which they find themselves at the time of decision-making, which includes the information available to them in the moment, and their experience of similar situations.

They zigged because it was the right thing for them to do in that moment, even if, in retrospect we recognise it might have been better if they had zagged.

This is fundamental truth about safety: people act in ways that are appropriate given the context.

I mean, no one wants to kill a child, right? The chances of that are one in a million …

6.    Gestalt

So how do we react when we are asked to make a decision – whether to zig, or whether to zag?

There are many ways in which our decisions are influenced by non-rational factors. But when we make rational, weighted decisions, then we use some form of Bayesian probability. Succinctly, this means that we assign a probability at the outset, collect some data and then modify our probability based on what we find.

Let’s take a classic example from the Emergency Department: a patient presents with a cough. The doctor makes a mental short list of differential diagnoses each with an associated probability. These probabilities are provided by the doctor’s gestalt – the combined experience and expertise, honed during their career. Simplified, it might be: chest infection (95%), cancer (5%). So we do a test. The chest x-ray shows clear signs of pneumonia. Now the probability has shifted. Chest infection 99.9% ; cancer 0.1%.

Does the doctor now do a full-body staging CT scan to look for cancer? Of course not. 999 out of 1000 patients don’t need it and the CT will do more harm than good to that many patients. But inevitably, one out of every 1000 such patients will  have a chest infection which is caused by their cancer. The doctor will have missed it.

So now let’s look at the Bayesian calculation for our execs in the Letby case. A consultant appears in the office and says that there have been some unexpected deaths and there is suspicion about one of the nurses causing these. The exec uses their gestalt – their combined experience and expertise honed during their career – and comes up with a pre-test probability. Simplified, it looks like this: (1) the consultant making a fuss because they want more money for staff and equipment, or there is some other perfectly reasonable explanation for these horrible events (99.999999%) or (2) there a baby killer loose in our hospital (0.000001%).

At this point, the tests you do will be to look for an explanation inside that 99.999999%. And just like you won’t do a CT scan if the chance of cancer is only 1:1000, you won’t call in the police because it would do more harm than good by putting patients off coming for treatment in a hospital over-run by coppers.

In other words: murderous nurse? It simply can’t be true.

Put this another way: what would you think if someone came into your office today and accused one of your colleagues of lacing the coffee with arsenic?

You’d think the person reporting this was crackers, wouldn’t you?

7.    Learning

The problem we are faced with here is the natural desire for retributive justice. Someone must have made a mistake, so someone must pay – with their job, their reputation.

This is the fundamental problem. Real learning only takes place within a culture that promotes it. My work on this, with my colleague Judi Ingram, clearly shows that such a culture can be developed, with some effort, and that it benefits patients and staff.

There is a place for ‘whistleblowing’ and ‘speaking up’ – but both of these imply the need for mechanisms whereby the weak can speak truth to power. But the very need for such things is itself an indication of the cultural problem that needs to be addressed, namely that people don’t feel safe enough to report problems within a hierarchical institution.

Unless and until staff feel safe then encouraging a culture of reporting will likely be met with silence.

But if such a culture of reporting can be achieved – alongside the three other elements of a safety culture: a culture of openness, a culture of justice and a culture of improvement – then we have the chance to improve the protection of patients.

But if any of these four elements are missing, safety does not improve.

8.    The rub

So here’s the problem that makes me angry.

By calling for heads-to-roll and attempting to hang-out-to-dry those executives who were involved in the Letby case, we risk creating a chill in the hearts of the thousands of managers who toil in our healthcare institutions to make it a better place for patents and staff.

Creating a patient safety culture is the work of years; dedication by leaders at all levels to create a place where staff can report and talk about errors, concerns and near misses, and, where they can examine and implement systemic changes that will improve safety.

Leaders will not do that if there is a chance that the world will beat them for trying.

Better to keep your head down, avoid the difficult questions, and keep-on keeping-on. Or maybe just move to an industry where they hours are better and the personal risks are fewer.

It’s hard not to blame. Our natures seek out stories of personal accountability in order to find and punish those who we believe have caused harm. But what we actually need, like the airline industry, is better systems.

That’s worth repeating: we don’t need stories. We need systems

And for that we need leaders and managers who are unafraid develop and implement those systems.

Healthcare needs to learn to learn.

Without blame.