Hello and welcome to BJGP Interviews.

I'm Nada Khan and I'm one of the associate editors of the bjgp.

Welcome back to the first season of the BJGP podcast here in 2026.

And we're starting off this season of the podcast with a chat with Dr. Joy McFadyn.

Joy is a GP based in Swansea and clinical lecturer of Patient safety based at Cardiff University.

We're here to talk about the paper she's recently published here in the BJGP alongside her colleagues.

The paper is titled Critical Illness in Prisons A Multi Method Analysis of Reported Healthcare Safety Incidents in England.

So, hi, Joy, it's really lovely to meet you and to talk about this research, but yeah, just taking a step back, I think it's fair to say that the prison population is an underserved and probably fairly under researched population as well.

But you point out here in the paper that it's not only this, but that the prison population is actually at a much higher risk of early mortality as well.

So can you talk us through this at all?

Yeah, that's a really good point.

So we know that people who reside in prison, known as prisoners, will have very high rates of physical and mental health needs.

And as you say, there are concerns that they have rates of premature mortality, so they may die up to 20 years earlier than the rest of the population.

But they are a population which isn't necessarily the area of focus.

So even though we know the importance of supporting their healthcare as a public health concern, they are often underserved, they're quite vulnerable, and yet there hasn't been enough research to support them to have what we call equivalent health outcomes.

So there are lots of definitions of what is considered to be equivalence of care for people in prisons.

So the Royal College of General Practitioners Secure Environments Group, they have defined what equivalence of care is for people in prisons, thinking that they should have the same quality of care, the same level of staffing, the same resources as anyone who is residing in the community in order to get the same health outcome.

And currently that is not being realised.

And just as a background to all this work, how many of these early deaths do you think are preventable?

So we carried out a study which was called the Avoidable Harm in Prison Study.

So it was focusing very much on healthcare events where people were harmed or could have been harmed whilst they reside in prisons.

So our focus is very much on these patient safety incidents, reports and incidents themselves, and ultimately the findings of the other space of the study.

We haven't released yet they're still embargoed.

But we were seeing within our sample of patient safety incident reports, events where prisoners were undertaking significant harm.

So within our paper, we haven't seen any evidence of the deaths which could be considered to be avoidable.

But our focus was very much on events where without urgent treatment, there was a high risk of death.

And we considered many of those events to be avoidable.

And I guess all this is tied into what you're aiming to do here in this research, which was to look at and characterize patient safety incidents in the prison population and find opportunities to improve care.

So you used a really detailed approach here and looked at patient safety incidents reported in England and carefully examined and coded all of the incidents here.

But I really want us to talk through what you found, what were the main sorts of incident type.

And what I'm trying to get at is what really happened in these reports.

Yeah, thank you.

So we reviewed Originally up to 4,000 of those patient safety incident reports.

And then when we focused specifically on those events where someone was at very high risk of death if they hadn't received treatment, we were looking at conditions suggestive of heart attacks, strokes, status epilepticus, diabetic ketoacidosis, for example.

And what we saw is that most of the reports that were included for analysis, so about 100 of those reports, people in prison were not being able to access healthcare professionals when they needed to.

So in prisons, people will have an assessment when they arrive to the prison, which is an assessment of their healthcare needs.

They should also have access to nursing staff, GPS and allied healthcare professionals, as well as referrals to secondary care as needed.

And what we were seeing is that when there are events where someone was critically unwell, they couldn't access the staffing when they required.

So it's very much a nurse led service in the prisons.

And even when there were prisoners who had collapsed, nursing staff could not access the prisoners.

And that was for lots of different reasons.

Some of it was related to poor communication, that there's quite a reliance on the use of radios in our reports.

And so if people were trying to radio from one area of the prison to the healthcare teams, then there was too much radio traffic that their messages weren't getting through or they were using the wrong emergency codes.

So actually the nursing staff weren't aware of the urgency of when they needed to get there.

So there were lots of delays in actually having the healthcare teams arrive and assess the patients themselves.

But also when a decision was made that someone needed to be conveyed to an emergency department, for example, due to difficulties with staffing levels, there weren't sufficient prison officer numbers to escort them from the prison to hospital.

So there was significant delays.

So what we could see in some of the events is that someone had collapsed.

There was concern that this was suggestive of a stroke, they were dysphasic, they had facial palsy, they had tinnitus, headaches, et cetera.

And nursing staff had assessed, said, no, they're unwell.

Gps had said they need to be conveyed to the hospital and they weren't transferred until the following day.

So those types of delays were very evident as well.

So difficulty accessing the healthcare professionals in the first place and then a delay getting the correct treatment or management, even with conditions which are time critical.

That all sounds really shocking, actually.

But I wonder if we could just take a step back and, and could you describe to us what healthcare provision is like generally in prisons?

You mentioned about a nurse led care system, but how easy is it to access other healthcare professionals like GPs in prisons generally?

So I think there are two very different opinions in this.

So we have the access to the patient safety incident reports, which is telling us that it's very difficult for them to access healthcare professionals as needed within the prisons that we looked at for the avoidable harm in prison study, for example, we were only focusing on prisons where health care was delivered on site and the provision is very variable.

So different prisons may have NHS provision, but the majority is probably private provision as well.

So it's a commission service, there's a lot of competitive tendering and there are concerns by some that a focus may be more on cost saving than it is on quality provision.

So what we saw within our patient safety incident reports was evidence that it was very difficult to access the healthcare teams.

So even though healthcare provision should be delivered and there are nurses, you know, round the clock, they were having lots of difficulties accessing any types of healthcare provision out of hours.

Our instant reports was an overreliance often on some of the electronic E consulting systems.

So the use of System 1, for example, in prisons in England, and what we could see is that people were presenting with quite significant symptoms and instead of what we would have thought would happen is someone was picking up the phone and referring them in.

Lots of electronic tasks were being sent around teams without necessarily an overview as to who was completing those tasks or an overview of what that meant.

So our focus is very much on these critical conditions, but some of it was related to the management of long term conditions.

In the first place that if someone's diabetes was being managed appropriately, that they were having annual blood tests or having their blood pressure checked, they were making sure that they had sufficient insulin, for example, then there shouldn't have been an occasion where they were experiencing diabetic ketoacidosis and needed to be admitted.

Making sure that there's appropriate management of care, but also then that organisational factors.

Are there sufficient staffing numbers or are there not?

And part of the concern that we could see in our incident reports was the role of locums and agency staff who perhaps were not as familiar with prisons and prison health care systems.

And they would often forget their passes to even log into the system, so they couldn't see a patient's medical records.

They were not familiar with the need to actually call for help, how they called for help.

They didn't know that if an emergency code is coming through the radio, that meant they needed to grab the healthcare bag with all the emergency equipment and run towards a specific wing or whatever is needed.

But focusing very much on these emergency conditions, there was a concern that the locum staff were not familiar with the protocols, the policies of the prison.

They were not carrying out observations, they were not documenting efficiently what they had actually carried out with the person residing in the prisons and that was delaying care that was stopping them from being transferred to emergency departments when they needed to be.

Yeah.

And what's interesting here is that in this paper you looked at some major themes here around these different incidents.

Can you talk us through this and what were the main findings here?

So we were thinking about the different aspects and cogs within the healthcare system in the prison and how they all interact with each other.

And we use the CEAPS model, which is the systems engineering initiative for patient safety, and it has six main domains that we were trying to understand if thinking about our patient safety incident reports and the themes within it, as well as the contributory factors, so why these events were taking place.

We tried to then map them to the domains of the Systems engineering and initiatives patient safety model, which is ceps.

And what we could see were the different domains were prevalent throughout the reports.

So there is a concern about tools and technology.

For example, so I've mentioned about the emergency radios, but also the lack of certain tools.

So there wasn't a provision of life saving equipment in prisons.

So there were often reports from paramedics as well as people who reside in prison to advise that when there were events where a patient may have harmed themselves or there'd been an assault and an injury.

There wasn't life saving equipment within the prison, so no cannulas, no IV fluids.

Obviously there was going to be no consideration.

There would be blood products or anything of the like, but there was nothing that they thought would, would support major blood loss and hemorrhage.

There were also, in many of the prisons, no AEDs.

So if someone had collapsed, potentially having a heart attack, for example, and their heart had stopped, we know the evidence that they need to get the paddles on their chest, we need to restart their heart if it's in an appropriate rhythm.

But there was nothing of that, like in many of these prisons, to actually support that.

So if there is any type of delay in calling for an ambulance, an ambulance should be adhering to the same national guidance of the emerg response times.

That should still be actualized within a prison too.

But what was happening is that an ambulance was being called.

There was some confusion as to where in the prison the prisoner actually was, which wing of the prison, which area of the prison.

Once the ambulance was arriving at a gate, they couldn't actually come straight through because of security concerns that the ambulance might need to be stopped and searched to ensure that nothing was entering the prison that shouldn't be.

And that was, you know, causing significant delays.

And then when they were getting to patients who'd collapsed, for example, there were delays for them even conveying them out of the prison.

So there was a concern that the healthcare professionals were not saying to them, you just need to convey them now they need to go to an emergency department.

We do not have sufficient care for them here.

So that was the concerns about tools and technology, for example, and then thinking about the organisational aspects.

So that would be within a healthcare system, things like staff rotors.

You know, I've mentioned already that there were some concerns with sufficient staffing levels.

So there are concerns by people who work in prison.

There can be quite a high turnover, perhaps an over reliance on locum and agency staff.

People may become quite burnt out in the system and therefore they may leave the prison.

And for some GPs who work in prisons, it may not be there full time physician.

They may work elsewhere and then they may do a couple of shifts in the prison.

So there isn't necessarily that continuity of care and how that might impact on prisoner healthcare.

Then within the CEIBS model there's concerns about personal factors or person factors.

So these are the people working in the system as well as the patients themselves.

So one of our recommendations after reading all of the reports, is that perhaps they require more focused training for how to deal with emergency conditions and the response.

So what we saw is that people weren't prepared to have multiple emergencies happening at the same time, which unfortunately does happen in the prison.

So there were lots of reports in which there were concerns with substance use in parts of the wing, perhaps using the substance spice, for example.

And then there was a report that three, four, five prisoners were all unconscious at the same time.

They therefore all required set of observations, need to check their oxygen levels, probably be placed in the recovery position and observed carefully until they came round, or if they weren't coming round, they need to be conveyed to an emergency department.

And then thinking about the context of the prison, we think about the internal environment.

So knowing that within the prison, security constraints will often outweigh concerns with healthcare.

And that is an important balance that both the prison teams, the prison officers, the governors, need to balance alongside the healthcare professionals.

And so what we were seeing, for example, is that during any type of lockdown in the prison, so security concern, a wing needs to be locked down.

That means the prisoners need to return to their cells.

They cannot le.

If something happens where someone is considered to be critically unwell, they collapse, they are complaining of chest pain, they have symptoms suggestive of a stroke, for example, they haven't got access to their insulin, so their sugars are rising, they become unwell, etc.

What we could see is that the healthcare teams could not access the prisoners, they couldn't get to them.

So that's the constraints of the internal environment.

And then the external environment is like I was mentioning, about those commissioning gaps.

So concerns where care is not being funded appropriately, if that emphasis is on the cost of a service rather than the quality and the outcomes for patients, then perhaps they're not getting appropriate care when they should be.

And I think I'd suggest to anyone listening who's interested in this area, I'd suggest they go back to the paper and take a close look at box two, where you talk about the main recommendations for prisoner health as a result of this work.

But what do you think are the most important things that could change in practice here?

That would make a big difference.

I think it's really tricky and it will take a lot more work working with people in the prisons, so working with the prison officers, working with the governors, working with the healthcare professionals, working with policymakers and the funders, because when we think about that human factors approach, you know, thinking about the whole health care system, it all needs to be changed to make sure that it's safer for people in the prison.

And that's not something that you can necessarily be doing overnight.

I think staffing is probably one of the most important thing and appropriate funding.

So there needs to be sufficient attention that this is an area of public health that is worth investing in.

And what my concern and the concern that is shared by others is we shouldn't be accepting poor practice because someone resides in a prison.

We should all be working very hard to overcome that.

And it's not giving them, you know, special attention or any type of services that aren't afforded to anyone else.

This is knowing that by supporting this public health campaign and concern for people residing in prisons, we can help the wider community as well.

Thinking about the cost saving to the nhs, thinking about the millions of pounds that is spent on people being conveyed to emergency departments, having what we saw were avoidable hospital admissions, when actually if there was this focus on these long term health conditions and a recognition that if someone is critically unwell, they need emergency treatment and they need it now, I think that's what's needed.

So perhaps it's more of a culture shift as well as the funding provision, but it just needs to be changed.

We need to have enough attention that this is an area that we should be investing in and then we can see that it is safer for people who are in the prisons themselves.

Thanks, Joy.

I mean, it's really fascinating work and I just think it's great that you and your team team are shining a light, as we discussed on this really underserved community of people who quite frankly are receiving shocking sort of levels of health care.

And it's just really important to understand that a bit better to try to work out where things can be changed.

So, yeah, I think that's a great place to wrap things up.

But I just wanted to say thank you very much for your time here and it's been great hearing about this research.

Thank you very much and thank you.

All very much for your time here and for listening to this BJTP podcast.

Joy's original research article can be found on bjgp.org and the show notes and podcast audio can be found@bjgplife.com thanks again for your time and bye.