In this blog post, Liv System’s Nigel Scard talks with Courtney Grant, a Senior Human Factors engineer with Transport for London (TfL). Nigel and Courtney worked together for a number of years at TfL on a number of station and line upgrade projects. A few years ago, Courtney applied his Human Factors and research skills with great tenacity, to a serious healthcare related incident which impacted him personally.  This resulted in an important, lifesaving change to ambulance service procedures. In this interview Courtney describes this in detail and also describes his recent work in supporting the Chartered Institute of Ergonomics and Human Factors (CIEHF) in supporting the response to the Covid-19 pandemic.

Emergency response procedures

I understand you were involved in some work with London Ambulance Service (LAS) a few years ago related to their procedures associated with dealing with people showing symptoms of a stroke, so could you tell us about that?

Yes, a loved one collapsed whilst she was at work and the paramedics came out very quickly, but despite the fact she was unable to speak and unable to properly move one side of her body, they quickly concluded that she was suffering from an anxiety attack. Over an hour went by before they started to suspect it might be something more serious, such as a neurological event. They then administered the FAST test (Face, Arms, Speech and Time), but decided it was inconclusive on the grounds that she was unable to speak. They administered it twice more and came to the same conclusion each time but decided to take her to a Hyper-Acute Stroke Unit (HASU) to err on the side of caution. The HASU confirmed that she had suffered a massive stroke and she almost died. At that point I decided to look into the situation further and made a formal complaint to the LAS. They replied that they didn’t have any concerns about the clinical care provided. However, I saw certain things in the clinical care that were troublesome.

As a Human Factors specialist, I could see that there must have been an element of confusion here, but I wasn’t quite sure exactly what it was because healthcare wasn’t my domain at that point. But I was able to use my HF and research skills to get to the root cause of what happened. I began by speaking to some of the world experts on stroke, such as the clinical director for stroke, for NHS England at the time. The feedback I got from him was that any problem with speech should immediately be classified as FAST positive.

I also got a copy of the LAS patient report form. There were some basic Human Factors issues on there, which in my opinion could lead the paramedics to make this mistake. So, I went to one of the LAS board meetings, where I met Malcolm Alexander, who is the chair of LAS independent patient’s forum, and through him I was able to gain more access to the LAS and more information from them. The more information I got, the more apparent the potential for system-induced human error became.

This is starting to sound like a detective story!

Yes, that’s what it was like, I was having to piece together various bits of information to understand what had happened here. I also applied to the Parliamentary Health Service Ombudsman (PHSO), who upheld my case and confirmed that this amounted to service failure. I showed the PHSO exactly what the HF issue was on the patient report form, why it’s causing confusion and how the LAS could go about improving it. I did a mock-up of an amended patient report form, that would eliminate that error, by removing the ambiguity they were experiencing because of how it was designed. I also discovered that there was a clinical update which had been issued a month before the incident happened, where the LAS Deputy Medical Director expressed concerns about this same issue, stating that a number of paramedics were missing the signs of stroke because they were misinterpreting the speech element of the FAST test. So, this was not an isolated incident, this was a systemic problem.

I also raised the issue with Boris Johnson, who was the Mayor of London at the time. One of his Deputy Mayors then contacted the Chief Executive of the LAS directly, who then responded to the Deputy Mayor by listing out all the actions they would take to amend the situation. This included changing the patient report form in line with my HF recommendations, and making it very clear to their paramedics that the diagnosis of anxiety attack is a diagnosis of exclusion, in that they should not come to that diagnosis until more serious diagnoses have firstly been eliminated.

What happened is that they jumped to that conclusion too early, that’s where the human error first came in. Based on the availability heuristic, they jumped to that assessment whereas in fact they should have gone through the diagnosis of exclusion process.

I then had meetings with the LAS Medical Director and training team, to work with them to develop a training video based on this case study to help ensure this doesn’t happen again. The feedback that I got from the chair of the independent patient’s forum was that it was being well received by the frontline staff. I also went out on an ‘ambulance ride out’, where the senior paramedic who accompanied me mentioned how hard hitting that training was, and how useful it was to highlight the issue.

You mentioned the ‘availability heuristic’, so you think that was the underlying cause of that human error mode?

Yes, that’s based on the work of Daniel Kahneman, as described in ‘Thinking Fast and Slow’ as one of the common cognitive biases.

What would have been underpinning that though? Was it the fact that she is a relatively young person so that usually for someone of her age, when they see those symptoms, it’s usually caused by an anxiety attack?

Yes exactly. In fact, one of the paramedics said that whilst they eventually started considering a neurological event, they believed that it was highly unlikely given her demographic.

That might point to a flaw in the ‘recognition primed decision making’ approach to expert decision making, whereby experts make rapid decisions based on recognising patterns as a result of years of experience. That approach applied to this type of situation would actually have drawbacks. Maybe that’s what happened here, they saw a young person, exhibiting certain symptoms, and they had encountered this type of situation many times before, when it was a panic attack, so they thought they recognised the situation, but in this case it was something more severe.  

Yes, I believe this is what happened here. And this is why reflective practice is so important.

Supporting the CIEHF response to Covid-19

I understand that more recently you’ve been helping the CIEHF with developing healthcare procedures in response to the Covid-19 pandemic? Were any of the insights you gained from your involvement with the ambulance service helpful for this?

Well, it has been slightly different, but my experience with the LAS was insightful because it was my first deep dive into the world of healthcare. But the research question was slightly different. With the LAS, I was looking at a situation where highly experienced and skilled professionals got tripped up on the recognition primed decision making aspect. But what I found, at least with the work I was doing with the CIEHF during the first wave of Covid-19 was that, because of the unprecedented nature of the pandemic, we were in a situation where clinicians who specialised in one field had to move into a completely different area, namely caring for critically ill Covid-19 patients. So, we produced guidance mostly for people who were less familiarised with that particular domain.

There are a number of different Covid-19 pandemic related healthcare procedures that have been produced by the Chartered Institute of Ergonomics and Human Factors, which of those were you involved in?

As part of the CIEHF Covid-19 Expert Panel, I worked with Professor Paul Bowie’s team to develop good practice on how to produce work procedures. We put together general good principles to reflect the work as done rather than the work as imagined. We emphasised working with the end users to develop those procedures and tied it in with general HF best practice and principles as to how to develop procedures, down to using plain language, sentence length, line lengths, and so forth. I also worked with Professor Sue Hignett to develop the bedside procedures for Routine Tracheostomy Care and Ventilator Emergency Care for Covid-19 patients. Again, this was primarily aimed at clinicians who are used to working in a different domain, so the question was how can we lay out the information in a really easy to use format, to reduce the anxiety for the person using it, reduce the likelihood of them making errors and improve the efficiency with which they can respond to the patient.

There seems to have been quite a lot of people involved in producing these procedure guides. I understand they were divided up into gold and silver teams?

Yes, the gold team was a small team at the top and there were a number of different silver teams. The silver teams were comprised of people with a lot of healthcare expertise and people such as myself, predominantly coming from other industries, such as rail in my case, but with HF expertise.

What were your specific contributions?

Well, I drew on some of the experiences from projects we worked on together at London Underground, where we used to produce detailed equipment lists required for control rooms. When it comes to caring for Covid-19 patients, one of the challenges they had was if they had to move a patient from point A to B, how to ensure they have all the necessary equipment? What they were finding was that on part of the journey they might not have everything they need.  So, I applied a technique from our rail projects, of producing a list with all the equipment, including the number of items, illustrations and so forth, being really meticulous about it and removing any ambiguities when it comes to what’s needed to provide effective care.

Another aspect was my understanding of the literature, making sure all the things we were adding to the procedures were underpinned by sound HF rationale.

Looking at the procedures the graphic design looks very good. Did you have any input into the graphic design of the layout of the guides?

Well, the institute have their own graphic design team who designed the visual presentation of the guides, but I did review the visual presentation and had some input into that also. But the primary focus from an HF perspective was the content.

Is your involvement in this work ongoing or is it completed now?

The work on the bedside guides is completed but I’m now involved in developing the strategic guidance for the Covid-19 vaccination programme. I’ve been looking at the HF literature on the use of particular types of syringes such as pre-filled syringes, and if there is anything there which can help with the efficiency of the use of the vaccine. We’ve also been putting together a task analysis of the whole process in terms of bringing the patients in, doing the admin, vaccinating them, understanding where there is potential for errors, what the consequences of those errors could be, what are the existing error recovery strategies and what additional ones might be needed.

Have you noticed any differences in the approach to HF in the healthcare sector, compared to the rail industry?

My understanding is that HF integration in the healthcare sector is not as deeply embedded as it is in domains such as rail, or air traffic control, so there’s still a long way to go in terms of HF having more impact and working with healthcare professionals to adopt recommendations from HF practitioners. There does seem to be a demand for it though.

In the rail sector, there is a lot of focus on physical ergonomics, on the layout of equipment in control rooms and train cabs etc. Has there been much attention to that on the work you’ve been doing relating to Covid-19?

Physical ergonomics has been coming up a lot in relation to the vaccine roll out, we’ve been looking at accessibility issues for wheelchair users, do they have enough space for the turning circles, accessing the areas for the admin and then the areas for receiving the vaccine itself. Also screening off the areas where people receive the vaccine.