Professor Trish Greenhalgh: Before we talk about remote, I would like to talk about staffing

Matthew

Hello. In a thoughtful, balanced but important ways, damning report, the Commons, health and social care and science and technology committees have produced what will no doubt be one of the first of many investigations into the government's handling of the covid pandemic. One quote from that report stands out. It reads: ‘Decisions on lockdowns and social distancing during the early weeks of the pandemic - and the advice that led to them - rank as one of the most important public health failures the United Kingdom has ever experienced.’

My guest today is one of many scientific experts who found themselves frequently at the centre of the public gaze during the pandemic. Her insights and warnings have led to her being lauded but also attacked. She's also a leading expert in another area of controversy, the use of remote digital channels by clinicians and particularly GPs. And that use of virtual consultations is, of course, part of a wider debate about how the NHS can best take advantage of the massive and accelerating scale of innovation and digital data and medtech. My guest has also insights into how we help the NHS become one of the most innovative health systems in the world.

I'm delighted to be joined by Professor Trish Greenhalgh, professor of primary care at Oxford University. Hi Trish, how are you?

Trish

Yeah, I'm fine. Thanks for inviting me.

Matthew

Well, it's great to talk to you. I just want to start by asking you to give our listeners a kind of sense of the range of your work because you look at a whole number of areas, so tell us just about the kind of, where you're focussing your energy and where you have been over the last few years.

Trish

Yes. So, I was trained as a GP and before that I actually trained in hospital medicine, in diabetes. And so my very first research was looking at how GPs and hospital clinicians could work together in something called Shared Care for Diabetes. So, you know, that was 30 years ago, and since then I've done a lot of different studies which go right across the clinical spectrum. But I'm particularly interested in the organisation and delivery of services and also in how we can make services equitable. So I spent 25 years in the east end of London, for example, doing a lot of work on providing services that would be accessible by the full range of minority ethnic groups and social groups in that part of London.

Matthew

Let's start by talking about covid. What was it like first to find yourself so much in the public eye in terms of some of the things that you said? I mean, you were ahead of the game in many areas in predicting what might happen and what we should be doing about it. And as I say, that brought you a lot of attention, a lot of praise, but also being attacked. How was that?

Trish

Well, I think we all look back on the period of, you know, sort of March/April 2020 as the most extraordinary period sort of personally, professionally. It was absolutely unprecedented. I was a full-time academic then, as I am now, I wasn't doing any clinical work because my academic job had taken over. The first thing I did was volunteered to do clinical work and I was told that you'd be better off staying where you are and giving us the evidence that will help guide our clinical actions. So that was a great steer. I realised that I could do more and possibly save more lives if I applied my skillset to generating and distributing evidence on this new disease.

And quite by chance, I had been planning to welcome a group of Chinese professors to visit Oxford in the end of March 2020. And of course, they cancelled that trip because in about January 2020, they were in a city called Guangdong, which was the second city in China after Wuhan come to have a kind of covid outbreak. And so they were sending me all sorts of information and guidance that they had put together for the management of covid in China. And I think that's why I got together very early on with a couple of colleagues and wrote an article which was published in the British Medical Journal in March 2020, how to assess people with suspected acute covid in primary care in the remote environment that we were now doing, and we're doing it by the phone remotely. So, I wrote this piece, which was really a summary of the evidence we had based on Chinese data, and that was published very, very quickly within days of us finishing it. And within about six or seven days of the BMJ publishing that it had appeared in a NICE guideline, which was quite extraordinary because you'd think of NICE guidelines as things that are put together very meticulously by committees to reflect on the evidence and that usually takes forever. But this was a really fast produced, NICE guideline. So very, very quickly I suddenly realised: “Oh my gosh, I've become a bit of a source of authority on this. And yet I've never actually seen a patient with covid.” You know, it was all done on reviewing the literature that was being very rapidly generated and circulated as what we all now know are preprints in the literature.

So, then I suppose we got on to the mismatch between the evidence that came from China and the evidence that was coming out of the hospitals in the UK and what was going on in primary care because actually the vast majority of people with covid were either self-managing at home or they were being managed by their GP or by the Covid Clinical Advisory Service, the NHS 111 phone service. So that meant that all that data that came from people who'd made it into hospital wasn't terribly relevant. So, we then did some quick research. We did interviews with patients who'd been acutely ill. We did interviews and online focus groups and other kinds of online surveys and things with GPs, with nurse practitioners, with paramedics. And we started to put together a picture of what's happening in primary care to the people who don't make it to hospital and also to the people who do go on to make it to a hospital. And that was the RECAP study, Remote Covid Assessment in Primary Care, which had both a qualitative component and a very big quantitative component where we followed many, many thousands of patients from that first phone call through into hospital and those who sadly died, we were particularly interested in what had been the features of their illness right at the beginning that could have predicted, you know, the fact that they were going to do badly. And so, you know, we're still writing some of that up now, but it all happened very, very quickly, actually.

Matthew

Now we'll come on in a few minutes to the question of the broader question of your insights into how innovation takes place and the ways the best ways to bring technological and scientific progress into systems. But one of the things that was said quite a lot during covid was that an opportunity of it was to accelerate the pace because as you've just described the pace at which we had to learn, the way in which in, for example, ICUs, there were kind of WhatsApp groups with clinicians comparing interventions they were making and trying them out almost in real time and, of course, the development of the variety of vaccines that we've got. Do you think it's right to draw a conclusion from all of this that we could be speeding up the innovation cycle?

Trish

I think innovations tend to happen in times that are turbulent, you know, be it during war, for example, this is a very good example of situations where things get invented and tested very, very quickly. And I think, yes, in some areas of science the pandemic has, dare I say it, been quite good for science. And one example actually is platforms for randomised control trials. So, the idea is you're not just testing drug A against placebo or drug A against the best current drug that people usually give, you’re developing a platform where you can very quickly slot in any new drug that someone develops or wants to repurpose. The ethics are all in place. You just have to do a quick amendment. The infrastructure is there. The data collection is all there, and it means that you can then launch into a very rapid clinical trial. And of course, at times when there's a lot of covid about it, it doesn't take that long to amass the number of participants in the studies to be able to complete the study fairly quickly. And that's why we've not only got we got definitive evidence very early on that dexamethasone was useful in people who were very sick in hospital, and also the dexamethasone wasn't much use unless you were very sick in hospital.

We should also talk very quickly about vaccines. I'm not a vaccine expert, but I think we are all amazingly impressed at the speed with which vaccines were developed, tested, shown to be safe and then, sort of, rolled out. Now, having said that, it's the rollout that's been the most difficult, hasn't it? So, it brings me to my point that not all innovation is helped by pandemic situations. So, the kind of bench science, the platform studies, the studies that are readily put through this kind of protocol, can be greatly accelerated and have been greatly accelerated, and that's wonderful. There are other aspects of research that really didn't go so well. For example, the social science aspects of vaccine uptake, vaccine hesitancy, vaccine inequity, all that kind of thing. It's now been nearly a year since the first vaccines were approved and released. And still, there are some sectors of the population in the UK and certainly some whole countries where the uptake of those vaccines has been very low. So, you know, we could ask ourselves, why not? But certainly, just because there's a pandemic doesn't mean that every aspect of science is going to progress more smoothly and more rapidly. Quite the opposite.

Matthew

Just before we leave covid, I talked about that enquiry, the Health and Social Care and Science and Technology Committee enquiry. It is the first of many, no doubt. Do you have a sense of what we going to learn from these enquiries. I'll tell you my feeling is right now I feel we'll learn something from the historical record. So, you know, as it were, future generations will be able to look back and form a kind of reasonable consensus about what happened and why some countries did well and other countries did badly. I also think it'll help in lots of technical ways in terms of our understanding of some of the things we've just talked about, about how it is you respond to a crisis like this, how you develop new drugs and vaccines or whatever. What I feel slightly less positive about is a kind of political and societal implications of it, the sense of learning something about how covid held up a mirror to elements of our political system, elements of our society. I see, I don't want to be pessimistic, I see less evidence that that's going to take place. What's your sense?

Trish

I think it's we're all still reeling from that report, and I've read bits of it with a feeling of horror, actually. But I suppose to try and pull some learning from this. I think the point that I would make is not that scientists were wrong; I think there was a lot of good science being talked about in those key committees; there was also a lot of scientific hesitancy, the sort of natural, we don't want to overdo it, we don't want to overreact. You know, the public wouldn't cope with a lockdown, all those kind of things. But actually, there was quite a lot of good science being put on the table as well, you know, modelling and all the rest of it. The thing that bothers me is there wasn't deliberation over those different kinds of science that were being tabled. So, what you got was the scientists would very quickly come up with a recommendation, and the politicians and senior civil servants would say, oh right, that's it then, we better not do anything. You know, it looks like the scientists say we shouldn't do anything, whereas, you know, there's a there's a piece in the report with Matt Hancock, who was the secretary of state for health and social care at the time, saying, ‘I bitterly regret not pushing back because common sense showed us that this was an escalating pandemic.’ ‘Common sense’, he used that expression. That really we ought to have been acting and acting promptly.

Now, what you had there was science pulling rank over common sense, if you like. And some of us were saying: ‘You know what? I'm siding with common sense here. I'm not waiting for scientific evidence. It's all looking a bit bad. And all right, if we lockdown and if we all put masks on and it was unnecessary, then we can, y’know, someone will laugh at me, but fewer people will have died. But if we get this wrong, this could be catastrophic.’ And I was saying that back in March 2020, and so were various other people, including Dominic Cummings, interestingly, but there was no real pushback on the science and therefore the lack of deliberation, I think, can explain quite a lot of those knee-jerk reactions. I think the other thing that comes out, a little bit between the lines in that report, is that we had a particular prime minister who was not known for his attention to detail and had a bit of a track record of winging it. And I think he, as an individual, got up and said, ‘Oh, I'm still going into hospitals. I'm shaking hands with people who've got covid.’ And next thing you knew he was in hospital himself and that, you know, I just wonder if we had a different kind of leadership from the top, someone who was more into the detail, a bit more proactive, whether things would have been very different.

Matthew

Let's move on to another area of controversy, which you've talked about a little bit already, but that's the debate about digital or virtual care now. You know, I've watched with some dismay the same newspapers that are prone from time to time to get their columnists to bemoan the fact that the NHS is not as advanced and modernised as it should be, then turning on the NHS when it comes to encouraging people to use online consultations in the same way as they use online tools for all sorts of other things. So, I mean, what's your sense of what's going on in this debate? And how has it gone wrong? Because it certainly seems to have gone wrong.

Trish

It has gone wrong. I think that’s a good way of putting it. So first of all, let's get one thing straight. Healthcare is not banking. Right. Now, one's interaction with a clinician, and I speak as someone who's got a relative going through a pretty serious illness at the moment. One's interaction with the clinician is an ethical interaction. It's about making judgements about what best to do in a complex and potentially life threatening or health threatening situation. And those judgements relate to things physical, mental, emotional, social or whatever, which are extremely complex and completely unique. You know, it doesn't matter if you’re a 14-year-old with acne and other 14-year-olds have also got acne. The particular concerns that 14-year-old will have are going to be unique because of all sorts of other things going on in the individual's life, etc, etc. And that's a very simple example.

Now that means that, you know, as an estimate, something like 80 per cent of all clinical encounters cannot be protocolised and that is very different from banking. I was checking my balance just this morning. I hit the app. I knew exactly what I wanted to do. I wanted to go into my current account and see if I had enough money to make a purchase. And every time that goes through my head, I make the same couple of keystrokes as does everybody else. So, it is a much simpler, more straightforward task. It's not the same task. So, when I get on to one of those e-consultation forms and try to have a conversation with a doctor. I actually want a conversation; I don't want to just follow an algorithm. And we need to factor in that healthcare is a very complex human interaction. And it's also very risky. I mean, these are the things that are coming out at the moment. You know, remote was considered a really important thing at the beginning of the pandemic, as indeed it was. And now what's coming out in the newspapers is, you know, GP missed my cancer because I only got a phone call or I got a video call. So, we need to work out the balance of risk and benefit when we're thinking about remote consultations and when and for whom they're appropriate and through what medium?

Matthew

So how do we need to change this debate? You talked earlier about the sense that it would have been better at the beginning of covid if we'd had a more deliberative approach. We opened up to people the probabilities, the risks, allowed a freer debate about it. Do you feel the same way about this, that that what we need to recognise is that there are pros and cons around the greater use, of remote consultations, obviously enormous advantages, well, some advantages for clinicians in terms of their time, but much bigger advantages actually for individuals who choose to do it because you can do it from your own home and you can fit it in when you want to. But you know, all the risks also kind of in a sense inherent, which is that there will be things that will be missed. And in a sense, one of the challenges all this technology is that often it starts out with bold claims that we can avoid people needing to be seen face to face or avoid people going to ED or whatever. And then because of the risks, the algorithm changes, the calculus changes, and you end up finding that these forms of kind of triaging, or whatever, that many fewer people than you think are stopped from going to that ultimate destination. Which then leads people to say, well, was it worth doing it at all? So how do we work our way through this maze, Trish?

Trish

I think there's a couple of things to say before I answer that directly. One is that I think the development of remote services is another example of where the pandemic provided a massive impetus. And I think we probably made more progress in the last 18 months in terms of remote services than we'd made in the previous 20 years. So I think the pandemic really was a burning platform for innovation there. All sorts of really interesting things happened, but particularly some of the tech companies were very agile and they worked with frontline clinicians to develop very bespoke pieces of software that allowed them to do particularly video consultations in a way that really fitted in with the workflows and patterns of working. So that's one thing. The second thing to say about remote consultations is we've got to be very, very careful to distinguish between the medium of consultation and the workload and the staffing and all that kind of thing. Because what we've got at the moment is something like 10 per cent of the NHS workforce off sick.

So, we're understaffed massively. We've got a lot of GPs in particular taking early retirement. They're just exhausted. They've had enough. And when Matt Hancock was secretary of state, his big thing was that we will continue to do remote consultations in order to solve or help to solve that workforce problem. And so he assumed that technology was going to make the health service more efficient. It was going to free up clinicians to do other things. That was his that was his very line. Actually, of course, there was no evidence that remote consultations were going to make healthcare more efficient. And actually, there is some evidence that for some kinds of problem, they make it less efficient because of the double handling problem. Let me tell you, one more thing about remote consultations is that they are now being used, certainly in primary care, as a kind of gatekeeping system to the front door of general practice. And that is a great shame because only two years ago you could just walk into a GP surgery and book yourself an appointment, and you might not get one tomorrow, but it was perfectly possible to say, I'd like to see my doctor please in two weeks’ time or something. You can't do that anymore. You've got to go through what they call total triage, and the GPs are actually quite reluctant to remove that barrier, if you like, simply because they are so overworked, so exhausted, so understaffed. So, before we talk about remotes, I would like to talk about staffing and staff wellbeing and workforce and where all these extra doctors and nurses and pharmacists, etc, etc, are going to come from. Because if we solve that, then let's say we can have a much clearer conversation about the place of remote.

Matthew

Yeah, although I guess, Trish, those workforce issues, even if the government turns its full attention to them, will take a few years to have effect, and things are deteriorating now, not improving, and so we have to find some way of getting through the coming period. I mean, it feels to me like this is a kind of service design challenge that what we need to be thinking about is bringing clinicians and patient representatives and policy makers and others together and exploring how we can redesign services, but in ways which are deeply informed, almost ethnographically informed by people's experiences rather than by assumptions about what people want. And I was talking to someone the other day actually who said a very simple thing, but I thought it was absolutely right, which is in the end, when you are worried about something you want to meet somebody who can do two things. First of all, you can trust them, and secondly, they can actually do something for you. And it feels to me as one of the problems with the way in which we do triage is that we expect people to be satisfied with someone they may not trust, but also worse, somebody who can't actually do anything for them, really, except refer them down the line.

Trish

I think that's true. But I think we also need to shift our expectations, and it reminds me of a dear departed relative of mine who, it must be, 30 years ago, he was diagnosed with diabetes at the age of about 70 and couldn't believe that he was going to be seeing the nurse. Now, this was a guy who had the mildest of diet-controlled diabetes. He was overweight and had all that kind of thing. He was a very simple, straightforward case for a for a diabetes trained practice nurse. And he went along and he was bitterly upset that the doctor wasn't going to be managing his diabetes. And I put a lot of pressure on this relative of mine and talked up the whole idea of nurse-led diabetes care. And within three or four months, he'd completely come around to the fact that this nurse actually knew an awful lot about diabetes, that she was able to support him with his diet and his exercise, etc, etc, and came to trust her to use your terminology. So, I think this all very well of patients say, Well, I've got to see someone that I trust and you've given me a paramedic or you've given me a GP when I really want to see a neurologist. Actually, it may well be that this is partly to do with public expectations, which could shift quite considerably to accommodate a much wider range of health professionals dealing with people's problems.

Matthew

No, I completely agree with that Trish. There's no question that quite a lot of people who knock on the door of the GPs could be seeing someone other than a GP and getting a good, faster and possibly more kind of human experience as a consequence of that than getting a few moments with a hard-pressed GP. I think my point is we need to think about how we build that trust. And I think part of the problem we got at the moment is there's a kind of national debate, which is people are palming you off. And if you feel you're being palmed off, this is not the best circumstance for you to start to trust people and to understand what they have to offer. So that's one of the ways in which this debate has gone wrong and one of the ways in which I think we need to try and think of ways in which we can pull the public back into this a bit more because the public is in a state of suspicion towards this now.

Trish

Yeah, can I just come back to you with that? I think that's true. I'm just going to share a little bit of the research data that we've got on this. One of the things that our patient advisory groups and the patients that we've interviewed say is a thing that upsets them the most is not being able to select what they believe is the most appropriate type of consultation. And so they get told that what they're going to have is a phone back from the healthcare assistant. And they absolutely know that because it requires a change of prescription medication that they have to see either a GP or nurse prescriber. And they know that. But the protocol is telling them that they've got to have this intermediate step and they, the patient, say to us that is what is inefficient, is that they, the patient who understand what's wrong with them can't actually have agency, if you like, in making the selection. And then the answer comes back from some of the people who design the system, saying, ‘Well, if everybody had agency, then they'd all flood in and want face-to-face appointments’. So, we can't do that because we're not staffed to do it. And so, yeah, there isn't. This is a wicked problem. This is a wicked problem? There's no easy solution.

Matthew

It is. I remembered a similar problem in public policy, and the conclusion we came to was lots of signposts, but no, no entry signs. And I think that's kind of what you have to do here. But it's, the doing of it is a different matter, but it's something we're going to have to work on. And that takes me to the I regret to say final because I could talk to you for hours Trish. But the final thing I want to turn to, which is the broader question of the way in which we think about scientific technological innovation. And I want to, I guess, suggest a thesis to you here, Trish, which is that, my feeling, and this is not just about the health service, I see this in all sorts of settings, but that we tend to view innovations too much as kind of single changes in complex systems without thinking about the overall system within which that innovation is going to take place. And we need to look at things with a more systemic lens and that the reason that often things go wrong is simply because all sorts of unexpected effects happen in complex systems, and people were expecting a very simple thing to happen, and it doesn't turn out like that. Now that's the first part of the thesis.

The second part of the thesis is that one of the reasons we should try to adopt a kind of principle of subsidiarity or devolution when we come to decision making because it is almost impossible to fully understand the complexity of these systems when you're sitting a long way away from them, which is why it is that people in the centre often have a very simplistic view of how innovation takes place. And so, I'm not for a moment suggesting that we don't recognise that there's the best way of doing things. There is often the best way of doing things, and we don't need to ensure that where there is technology that really works, that we encourage people to take it up. I get all of that and you know, we have an apparatus to make sure we do that. But also, we need to understand that if we are talking about innovations that require this systemic lens, we need to let people reasonably close to the frontline be involved in that process just because of the sheer complexity of it.

Trish

I think what you say about complex systems and the idea that letting local people and departments sort things out locally as far as they can applies to any aspect of a complex system. But there's something that I'd like to bring in, which is technology, because technology complicates things in an interesting way. For example, it is almost always the case when you introduce a technology into a workplace, in particular, that the person who's putting in the work into that technology is not the person who gains the benefit from that technology. So, for example, you know, when I was a GP, they brought in all these electronic records and suddenly I had to put in coded data to make sure that someone else in a different building, in a different department, you know, all the rest of it, could press a button and get aggregated data on what was going on with my patients. So, it made work for me, but it saved work for someone in the public health department. Now, one of the things we need to do when we introduce technologies is try and map how the work is changed across the system because that technology might well save money overall. But if it isn't saving money for the person who's doing the work, they're not going to be very motivated to do it. So that just kind of underlines what you were saying about complex systems and subsidiarity. The other thing is another aspect of technology and particularly computerised data, which is that the farther that data has to travel, the more it loses its meaning.

Matthew

Interesting.

Trish

Now I'll give you an example of that is, you know, if someone came to me as a GP having had some penicillin and came out in a red rash, I might well write free text in the notes, patient came in with a red rash day after penicillin, etc. But I might also put some coded data which just said penicillin allergy. That person then goes into hospital two years later, and the doctor then looks on the summary care record and sees penicillin allergy. But they don't know where that data came from. They don't know what happened when the patient had penicillin because that bit was in the free text and didn't get transmitted. And all that context, you know as the GP I knew that was the first time the patient had had the penicillin, so it probably was a true allergy, etc, etc, etc. But because that data has now travelled and been coded and decoded, it doesn't mean the same thing, and the hospital doctor may or may not trust it. So, I think you're right about local. Local is good. On the other hand, standardisation is also good, and so you get that tension between standardisation and contingency. But I would think about that. Anyone listening who thinks what we really need is what Richard Granger used to call ruthless standardisation right across the NHS, with everyone using the same system in the same way and coding everything in the same way. No, we want something a bit rougher, a bit messier, a bit more locally sensitive, a bit more patient centred.

Matthew

Thanks, Trish. It's been absolutely fascinating, and there's a lot of areas I'd like to have gone into in more depth, but Trish, thank you so much for joining us.

Trish

Thank you for inviting me.

Outro

This episode of Health on the Line is part of our new Integration in Action initiative, which brings together podcast events and case studies from our members on themes affecting the health and care system. Visit NHSconfed.org for more information about us and to register for events and webinars that delve deeper into the issues explored in this podcast. And save the date for NHS ConfedEXPO, the premier event in the health and care calendar taking place on the 15th and 16th of June 2022 in Liverpool.