How to write the PPIE section of an NIHR application

The question every reviewer is actually asking

Strip away the assessment criteria and the PPIE section is marked against one question: has involvement changed this application, or is it decoration?

That is why "we will involve patients throughout the study" scores so poorly. It is unfalsifiable. It commits to nothing, evidences nothing, and could have been written before the research question existed, which reviewers can usually tell. The sections that score well do the opposite: they name a specific thing that is different in the protocol because someone outside the research team said so.

One concrete example beats three paragraphs of principle. "We had proposed a 12-week follow-up; our contributor group told us that people finishing treatment are least able to travel in the first month, so follow-up moved to weeks 6 and 16" tells a reviewer more about how this team works than any statement of commitment could.

The six things the section has to cover

1. What involvement has already happened. Before this application was written. Who, how many, recruited how, and when. If the answer is "none yet", say so plainly and explain what you are doing about it, because a reviewer will notice the absence far faster than they will punish the honesty.

2. What it changed. The heart of the section. Name specific changes to the research question, outcome measures, design, materials, or recruitment approach. If involvement confirmed a decision rather than changing it, say that too, and say how you know it was a real test rather than a nod.

3. What will happen during the study. At which points, doing what, with what influence. "Reviewing participant information sheets before ethics submission" is a plan. "Ongoing involvement" is not.

4. Who these people are, and how you found them. Reviewers are alert to involvement drawn entirely from people already close to research. If your contributors are all from an established patient panel at your own institution, that is worth acknowledging alongside what you are doing to reach beyond it.

5. Support, training and payment. Payment at or above NIHR's published rates, how it will actually be paid, what training or induction contributors get, and what accessibility support is available. An unpaid PPIE plan reads as a plan that will not happen.

6. How you will know whether it worked. The part most applications skip. How will you record involvement activity and its impact across the study, and report it at the end? Naming a framework such as GRIPP2 signals you intend to evidence impact rather than assert it.

Phrases that quietly cost you marks

"Patients will be involved throughout the study." Says nothing. Replace with the specific points and the specific decisions they will have influence over.

"A patient representative will sit on the steering group." One person, no support, no mandate. Reviewers read this as tokenism because it usually is. Two contributors, briefed and paid, with a named person responsible for supporting them, is a different proposition.

"We consulted with patients and they were supportive." If involvement changed nothing, either it was not a real question or you have not reported it fully. Both readings hurt.

"Participants will be involved in the research." Participants are the people the research is done on. Contributors are the people it is done with. Mixing the two suggests the distinction has not landed, and it is the distinction the whole section rests on. Our guide to PPIE versus engagement versus experience covers where the lines sit.

"PPIE costs are included in staff time." This tells a reviewer contributors are not being paid. Give PPIE its own budget line.

Writing it when you have not done much involvement yet

This is the position a lot of teams are in with three weeks to a deadline, and the wrong move is to write the section as though the involvement happened. Reviewers have read thousands of these. Vagueness where specifics should be is the tell, and an application that appears to overstate involvement is worse off than one that is candid about a gap.

What works instead: do something small and real now, and report it accurately. A single well-run session with six people, on a genuine open question, produces enough specific material to write a credible section. Approach a relevant charity or support group, pay people properly, ask about the thing you are actually uncertain about rather than presenting a finished design for approval, and record what changed.

Then be explicit about the sequence: this is what we have done so far, this is what it changed, this is the fuller programme funded by this application, and this is why the two are proportionate to each other. A reviewer will accept a modest starting point with a credible plan far more readily than an implausible claim of extensive prior involvement.

How much is enough

Proportionate to the study, and reviewers do calibrate. A small feasibility study is not expected to show a standing panel. A large trial with a five-year follow-up is expected to show more than one consultation session.

The failure mode is not usually too little activity, it is activity concentrated entirely at the start. Involvement that stops once the application is submitted reads as involvement done for the application. Spreading contact across the study, at points where something is genuinely still open, is both better research and better scoring.

Budget is a signal here too. A PPIE plan with no money attached will not survive contact with a busy study. We set out the components and realistic figures in budgeting for PPIE in NIHR grant applications, and there is a fuller discussion of what drives the total in what does PPIE cost.

The last read-through

Read the section back and ask: could a reviewer point at one sentence and say "that is a thing that changed"? If not, the section is not finished, however well written it is.

Then check the section against the rest of the application. PPIE plans that contradict the budget, or describe activity the timeline has no room for, are common and easy to spot. The involvement described in the plan, the money in the budget, and the time in the Gantt chart should be visibly the same programme.

Finally, if you can, have a public contributor read it. Not to approve it, but to tell you whether it describes something they would recognise as involvement. That is the readership, after all.