The art of lay reviewing in health research: some pointers

A little while ago someone suggested that I write something about my
approach to lay reviewing in health research.

Not so much a ‘how to’ guide as there are lots of these and very good
they are too. More, an insight into the questions I ask myself when
forming an opinion on an application for funding. Whether it be funding for a specific research project or a new research organisation.

For lay reviewing is an art, a matter of applying one’s judgement to
help a funder decide whether what is being proposed is not just going to work from a scientific point of view but from a patient and public one too.

This week I spent two days sitting on a panel for the Welsh Government
which was selecting their new Biomedical Research Centres and Units.
The lay reviewers were excellent. As was the way in which their review comments were listened to and respected by the panel Chair and other members.

So here’s my ten pointers or should I say pointers for ten!

Q1. Who will benefit from this research?
Has the team clearly set out how their proposal will addresses patient need and expressed what this need is? How do they know and how have they come to this view (see Q2 below). Normally I like to see some data here about numbers of people affected, limitations of current treatments and perhaps a sense of what this means to patients, their families and the health and social care system. Without this I think you are within your rights to ask
whether what is proposed is a patient priority? There may be other
valid reasons why the work should be undertaken but is that made
clear and the reason for not including patients and the public a sound one.

Q2. Have there been any pre-nuptials?
These days I would expect some evidence being put forward in the application that the team has already involved patients and the public to a) identify what they are doing is a priority, b) help design the project and it’s methodology and c) develop a public involvement plan that fits with the project. Time and again the evidence shows that public involvement falters during the course of a project if not enough time is given over to it at an early stage and, ideally, as the project is being put together. And there’s
lots of ways in which researchers can do this. Some funders like the
Wellcome Trust offer researchers small pots of money to develop plans around what they term ‘engagement’ and NIHR’s Research Design Service (RDS) does similar.

Q3. How much of the public involvement plan is in the ‘future perfect’ tense?
Let’s face it. This is a proposal so there’s going to be a lot
of ‘will-ing’ expressed. To an extent you can’t help but take some of
this on trust. But you should be look for some assurances that this trust is not misplaced. Clear statements about ‘how’ they will approach the task is one. Also, whether clear goals for public involvement are set out and a timeline for achieving these. At the very least, these are then things that can be monitored and tracked as the work progresses. You would also want to then see some involvement of patients and the public in this reporting and monitoring.

Q4. Will the public involvement operate under a glass ceiling?
Here we are really talking about issues around governance. Ask yourself howthis team or organisation will be making decisions? Then ask yourself
whether you can hear the patient voice in this? And will patients and
the public be able to raise concerns with the Board or leadership should they need to? Often the difficulty in public involvement is that when things go wrong it’s difficult to elevate the issue to a place where the issue
can be considered and decisions made. At the very least I would expect
patients and the public to be on the Board or equivalent forum. You might also want to ask yourself whether it is clear who is going to lead on public involvement within the team: a junior researchers or the Principal Investigator. That can be pretty telling on how important public involvement will be considered.

Q5. Is public involvement off-label?
I nearly called this ‘the dreaded advisory board.’ For it is the easiest mechanism for research teams or institutions to use as a place to put all those troublesome patients and the public and perhaps others as well. People that they are told they should work more closely with but don’t really want too. Don’t get me wrong, some of these boards work really well? The secret of their success – and the questions you should ask – are: where do they fit in with the governance? Who will chair them and sit on them? How are they supported? And what they will do? Even at the start of a project I would expect applicants to know the sorts of questions they need this group of ‘experts’ to address.

Q6. Is it a one-person operation?
A great friend and colleague of mine once said to me that when asked by people how they can find a way of not working with the usual suspects will say: ‘have you seen the unusual suspects?’ It is inevitable for the same names to crop up in people’s public involvement plans. Particularly if the applicants fall into the same condition area or part of the world. All I would say is that these days, and with the networks now available to people, there seems no excuse for researchers or research teams not to make an effort to source a wider view. Also, it concerns me when quite so much emphasis is put on one person not least because of the stress that can mean (see also Q8).

Q7. Is it charity or people they are after?
So we know how this one goes. Under the public involvement plan there is an eloquent description of the wonderful relationship people have formed with ‘x’ charity – ‘the leading voice of ‘x’ in the country.’ Of course, charities are very important in making research happen. They can bring patient voice but also important funds. But they can also be highly corporate in their view. The problems arise when this singular relationship with a charity is unaccommodating of other views. So, how will they reach the patient voice that might not be represented by the named charity?

Q8: Are they making a rod for their own back?
Let’s be realistic, no team or organisation or company for that matter has money to do everything? So, we should beware of the team that says it’s going to do 40 focus groups to write a patient leaflet as much as we should the team that says they will put it on the website for people to find and edit over a few days? Assuming we are content with the answers to Q1-7, what I really want to see now is a clear expression of the issues or questions that patients and the public will be asked to address and the methodology for doing this. But it seems important in this day and age that the plan feels appropriate, measured and a good use of money. To be honest I think you can tell quite quickly from an application – especially in its tone and style – when researchers and public colleagues have reached a state of mutuality in how they intend to work together.

Q9: ‘Is it a house built on sand?’
Nice research, shame about the budget. Actually, that should be: nice public involvement plan, shame about the budget. What you really want to see here is a clear presentation of the budget that will be provided to support public involvement. In particular; expenses and reimbursement of public involvement colleagues; the running of meetings and events and; learning and development. If it is not there, we have every right to ask why and where it is.

Q10. ‘How will patients and the public learn about this work when the results are out?’
These days everyone is asking about the impact of research. Quite rightly so. And most funders in their forms will ask teams to set out how they intend to disseminate the results of their research when it is completed. Cue two lines about publication in academic journals and presenting at international conferences. Or a line or two about building a website! But from a public involvement point of view neither really cut it. Unless things have changed vastly since I last looked, the public are not known for their readership of academic journals or attendance at scientific conferences. Websites also have their limitations. But there are a vast array of old and new tools in communications that are not expensive and remain better ways of reaching people – from articles in patient group newsletters to using social media. ‘Know thy audience’ is the motto here. Because if you know your audience you will likely know the places they are more likely to look for information.

So, there you go. It’s not intended to be exhaustive and i am sure many other questions could just as easily be included. In fact you have suggestions, please do add them as a comment to this piece.

One last point, And this is about behaviours.

It is simply that we all have our own way of doing things. In my book, politeness, decency and respect goes a long way. And I tend to think that the vast majority of people whose work lands in front of me are trying to do a difficult job in often trying circumstances. My role is therefore not to trip them up but to help where I can to make what they are doing even better.


5 thoughts on “The art of lay reviewing in health research: some pointers

  1. Simon – at Q6 I think you’ve omitted the larger point. Suppose researchers ask themselves: if there are two sources of trial evidence, shall I only use one? The answer is obviously no. It’s the same with lay input – you should NEVER rely on a single representative, no matter how suitable they are. To do so is to introduce the risk of bias into the whole project, something researchers should have learned to avoid, especially when it’s the easiest thing in the world to ensure calibration by recruiting a second lay voice. I certainly feel uneasy whenever I’m the only “PPI” person present – it smacks of tokenism, but, worse, of lack of understanding of good methodology
    All the best, John Walsh


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s