Professor Gillian Leng: ‘There’s nothing that says PAs are 100% safe in primary care’

In a relaunch of Pulse’s ‘Big Interview’ series, editor Sofia Lind speaks to Professor Gillian Leng, leader of the Government’s review into physician associates (PAs). Commissioned by the health secretary in November, Professor Leng’s review has seen her engage with over 1,000 people to look into the safety and effectiveness of PAs in the workforce. Speaking at Pulse LIVE Birmingham last week – ahead of the report’s release – Professor Leng shared exclusive insights into her findings.

SL: The report was expected in early June, but when can we now expect to see it?

GL: It’s right to mention timing, because I had very much hoped we would have published for this event, because I could then be having a much franker discussion with everyone about recommendations. That’s not to say that I can’t say anything, but I am not going to give away what the recommendations are going to conclude. The report is due to publish at the end of the month or the very beginning of July – that’s where we’re scheduling it at the moment. I’m really keen to get it out as soon as possible. It’s been a massive amount of work over just about three/four/five months, but I know that there are people who are really waiting for the findings. Particularly GPs who employ PAs, and PAs themselves because they are very much finding their lives are on hold. So I am personally committed to publishing it as soon as possible.

SL: The debate on PAs has heated up over the last few years, but they have been around in general practice for a while now. Could you tell us a little bit about the history?

GL: Yeah, some of you will know the history, but physician associates came from a model in the United States. We piloted their role, and the first ones were introduced around 2002 but we still didn’t have very many in post at about 2010. However, I think it was about 2015 when Jeremy Hunt decided it would be a useful addition to the workforce in primary care, and at that point, about 1,000 were trained to work in primary care. So that’s kind of been the rollout.

What I’ve seen during the period of the review is, of course, tension; tension about what they should be doing, and what they shouldn’t be doing. But, I’ve visited some practices that employ them, so I’ve seen how they work, as well as hearing from people who think that they don’t have an appropriate role seeing undifferentiated patients in primary care. And it’s a very tense topic. There are very polarised views around the roles of PAs so I have tried to get as much evidence as possible and listen to as many people as possible. I’m absolutely clear that the recommendations will not please everyone, but I’ve taken what – I hope – is a pragmatic, sensible approach to move things forward. That’s what I hope you’ll see when you read the report.

SL: So what specific evidence have you found regarding PAs working in general practice? 

GL: Well, you probably won’t be surprised to know that there are no randomised controlled trials that have been carried out in general practice. There are no controlled studies of any sort, so the data is relatively limited. There are studies, but they’re pretty low quality – they’re pretty small. We’ve looked at everything we could find. We’ve looked at data, nationally and internationally, and because the studies are small and low quality, it’s not easy to draw any definitive conclusions from data.

But there are some studies that show, for instance, that PAs are better in some areas than GPs – in others they’re less effective. Things like referral rates seem to be similar to GPs. PAs tend to have slightly longer appointments. They tend to give slightly more advice – that might be correlated to being given more appointments. So there are bits of data that one can draw upon, but it’s not a comprehensive, perfect evidence base. 

That didn’t surprise me. I wasn’t expecting there to be a perfect evidence base. Having worked at NICE for 20 years, we produced lots of guidelines where the evidence base in many clinical areas isn’t perfect, and what you then have to do is look at perspectives that relate to the evidence base to decide what needs to happen next. So on this occasion, I’ve drawn upon patient perspectives, on clinical views, on expert opinions, both nationally and internationally, and looking at the needs of the workforce. 

So all of those bits of information are being triangulated with the centre, with a core around safety and effectiveness. But there is not black and white evidence around safety and effectiveness in primary care.

SL: Can you say anything about what the evidence does say?

GL: I’ve given you a flavor of what it says. You know, there’s nothing that says PAs are 100% safe in primary care, but there’s equally nothing that says they are 100% unsafe in primary care. There’s a lack of definitive data, but I am confident that we’ve looked as hard as we can. I don’t think anywhere, anyone else around the world has looked so comprehensively for information about one particular staff group. And I have to say, I don’t think any other staff group has had this level of scrutiny. It would be interesting to see what happens to other professionals if we try to find data on their safety and effectiveness. It’s quite unusual to have this level of scrutiny. 

So this is why I was saying when I have then had to take judgments based on the best available information that I have got, and it’s partly about evidence and data, but it’s only also those other factors: patients, experts and about the workforce requirements. And I told you that I’ve been to visit some practices where they employ PAs. I’ve also had a whole host of submissions from people across the healthcare system. Where practices are using PAs, they’ve been clear it’s because they couldn’t fill the roles. They couldn’t fill the roles with GPs. They really couldn’t. 

So that’s why I said I had to look at workforce considerations, and I have also picked up a number of issues that relate to training of doctors. There are clearly significant differences in the ways doctors are now trained than when I was trained. It’s things like modernising medical careers, the rotations that people now have – it’s quite a fragmented system. It’s quite unsatisfactory, and the rotations are not helpful for quite a lot of resident doctors. 

That hasn’t been the scope of my review to pronounce on training of doctors. But I have been liaising specifically with Chris Whitty, who is leading that review on post-graduate doctors training, because there are linkages. Some of the causality around ‘why can’t you get medical staff working in particular places?’ links to how we’ve started training doctors. There are a number of factors at play. So I think it’s important that moving forwards, we look at the picture in the round. I’ll stop there, because I don’t want to say too much about training of doctors.

SL: I imagine you heard some different things speaking to GPs and speaking to PAs. What do GPs think that PAs should do? And what do PAs think that they should do in general practice?

GL: Again, mixed views. Mixed views. I mentioned, you know, the RCGP scope of practice, which is quite limiting. I’ve been to visit practices where PAs are doing a lot of the roles that GPs carry out. So that’s sort of the context. But we ran a survey. I don’t know if any of you here have responded to it, but we produced a bespoke survey to inform the review. That was directed at people who were either PAs or people who worked with PAs. I suspect some others also replied, but that was the core group that I was trying to get hold of. 

One of the questions had a list of different types of activities that might be carried out in primary care. People were asked to say: Do you think these are appropriate for a PA or not? And PAs responded, and GPs responded. And without giving you that long list of different activities – because I can’t remember it – the striking thing is the difference in perspective. So whereas PAs are confident that they can do the vast majority of those activities that you might see in primary care, the response from doctors was much less supportive of those roles being carried out. 

So we saw quite a divide in views on what could be carried out – should be carried out. That resonates with research that others have done in the past, so we triangulated those findings with others. You might say: ‘Well that’s not very helpful then, is it?’ There’s a difference in opinion, but it’s interesting to unpick why that is, and the perspectives that PAs have been given in their training. They’ve probably been encouraged to take on roles, and they would like to take on roles. They’re a committed part of the workforce and they are keen to contribute. And then we have the medical perspective, which is actually the training is more limited. This group of staff is not going to pick up things that we might be aware of. They don’t have the same knowledge base, so that is probably why we got that divide. And probably the answer is somewhere in the middle.

SL: One concern that we’ve heard from GPs is about their supervision of PAs and potential liability for any mistakes that do happen. What evidence did you hear on that topic?

GL: Well, you’re right to say there’s concern about supervision, because I think there has been confusion about what that meant. The GMC, who is now, of course, the regulator for PAs, has put out some advice more recently, saying what the difference is between being a named supervisor and a day-to-day supervisor. So there’s more advice on what that supervision should be. 

Then there’s what I’ve observed in the practices I went to visit, and there were different mechanisms for doing that. Sometimes there were GPs who were seeing patients and supervising at the same time, so the supervision was responding to queries that might be coming through in real time. There were other practices where there was a GP who was not seeing patients but was available to respond to PA queries. So I’ve seen different models, and I’m not giving you any judgments on those. I’m just telling you what I’ve seen. And of course, the GMC has now provided more advice, and I think that’s helpful – and in terms of the PAs now being part of a regulated profession, that also helps in terms of accountability. 

SL: There wasn’t any pause of PA recruitment during this review. But have you seen any evidence of a shift anyway?

GL: Yes, I think things have changed during the course of the review, and I think that there has been a definite pause in recruitment. The comment I made at the beginning about there being a need for this, it’s because living with uncertainty is hard, so I want that uncertainty to end by producing the review. And I’m also aware that the scope of practice that was produced by the RCGP has also had an impact, and that some PAs are no longer employed in primary care as a result of that – it’s not a result of the review, it is a result of that.

SL: Yeah, that makes sense. But since PAs were introduced in the UK, the employment situation for GPs has changed quite significantly and is now a lot more challenging. Have you found any evidence that PAs have had a direct impact on that situation?

GL: There’s a short answer to that, and I think the short answer is no. I don’t think I have. I don’t think so.

SL: So PAs are not to blame for GP unemployment?

GL: No, I don’t think they are to blame. But I think the link is the one that I mentioned earlier, which is when practices have struggled to employ GPs, they have been able to employ PAs for whatever reason. I don’t know why, but they have definitely filled those gaps. I visited practices that said they would not be open right now; they would not be able to provide a service for patients without PAs. That’s what they told me.

SL: That’s really interesting. So we have already touched upon several times that this debate has been quite toxic at times. Do you think there is anything that the UK should have done differently in terms of the rollout of this profession in general practice? 

GL: Well, if there’s time for comments or questions from the floor, I’d be interested to know what you think about the rollout. I don’t know whether you’re a group here that’s very keen or the opposite of PAs. But if you want my perspective in general on the rollout of PAs, it seemed to be going quite well at the beginning. There were lower numbers, and there didn’t seem to be much objection to the role, either in primary or secondary care. 

The challenge happened with the increasing numbers of PAs, with the workforce plan and with Covid and probably subsequently with junior doctor strikes, when PAs became used in roles they weren’t originally being used for when they were substituting for doctors. That’s where the big tension has come from. 

But if I’m being systematic: any change done well requires a clear vision for what the future state should be like. I don’t think we have that described. I don’t think that was ever set out. We have leadership that will communicate that and will talk to people about what the issues might be and how it’s going to work. And you have good, practical local implementation, looking at what workforce challenges are, looking at what the roles might be. Because most people have told me, even some people whose opinion is generally quite anti-PA, will say, there is a need for people to help doctors, but we weren’t convinced that the rollout has been done in a useful way. So there is that workforce planning element at a local level that properly thinks through ‘what might a service model look like with a PA in it?’ And I think fundamentally, it’s those three things – vision, leadership, and then local implementation – where we could have done with the wisdom of wonderful hindsight, and have done it a bit better.

Want to hear more from our interview with Professor Leng? Check out the newest episode of Pulse in Focus: The Podcast for GPs, for clips from the day, discussion of her findings, and analysis of what the final report might recommend regarding PAs in the workforce. Listen here.