My colleague, Steve, can’t be the only viewer in the last month to have turned vegan after watching Simon Amstell’s film Carnage. This combines humour with polemic to make us meat-eating, dairy-swilling people think twice about what this is doing to us and the world we live in.

But it is unusual for a film—or a book or a research paper—to change behaviour in this way. Especially something as fundamental as what we eat. We know just how hard it is to make and sustain lifestyle changes. The same is true for ingrained professional practices and culture—”the way we do things here.” But those of us in the business of producing and promoting evidence have to overlook this knowledge to carry on doing what we do. We need to believe that people act rationally on the basis of what is known to be effective. And that new evidence on what works or what harms will result in people changing practice and behaviour.

Set aside for a moment what we know of drivers other than research which influence clinical practice.   This includes formal and informal sources of knowledge, as described so well in groundbreaking work on clinical mindlines. What interests me though is how the makers of the film Carnage could know what impact their film had on my colleague Steve and others who watched it. Numbers of downloads, sure. Critical and public reviews and ratings, yes. But the golden metric here would be eating habits changed after viewing—and attributed to that viewing alone. This would be almost impossible to measure.

Lucky perhaps that the entertainment business does not set such store on assessing impact. But it makes me think what a high bar we set ourselves in health and health services research. We aim to fund and promote research which not only adds to knowledge but leads to change in practice. And to demonstrate that research has made that change.

At the NIHR Dissemination Centre, we have a pool of over a thousand clinical, managerial, and public raters who help us select the most relevant research to feature in our NIHR Signals. We produce three or four of these from the many hundred reviews we draw down every week. These are short critical summaries of recent research, with NHS context and commentary. Each has a title with findings upfront. We focus on research which could change practice or make clinicians think twice about current practice. So we are fairly confident that we are selecting research with a message—and that could lead to action.

But it is very hard to track whether research has made a difference. There are some sources we can look to—for instance, clinical audit can tell us how many bedbound stroke patients are getting intermittent pneumatic compression stockings. Recent NIHR research showed that this can reduce leg clots by a third—featured in a recent Signal. So an increase in the proportion of these patients using such sleeves over time would demonstrate effect. Similarly, research on approaches to confirm placement of nasogastric tubes could be considered in relation to changes in reported “never events” with tubes misplaced into the lungs. But we cannot tell what part research has played in these changing behaviours. We do not know exactly how research reaches frontline staff. This is particularly true for the kinds of research which sit outside clinical guidelines and formal mechanisms to get evidence into practice.

There is a growing science of measuring impact, now embedded in academic funding formulae. And an emerging evidence base on what may increase research uptake, including the kinds of formats likely to appeal to decision-makers. We can monitor metrics on downloads of research reports and social media provides other ways of tracking interaction and engagement. But that is only part of the picture of impact. We also need the individual stories for people and organisations on why particular research hit them at the right time in the right way to support particular decisions. And we need to learn from filmmakers, advertisers, and other persuaders about how to engage hearts and minds.

Carnage is a well-crafted narrative, which uses humour and drama to make a compelling evidence-based argument. Like Jonathan Swift’s satire over two hundred years ago proposing that Irish people sell their children as food to combat English complacency on Irish famine, it shows what we have normalised. It entertains—but it also has the power to shock. These are some of the approaches that might be needed to drive home important messages from health research. Simon Amstell, could your next mockumentary tackle antimicrobial resistance?

Tara Lamont is deputy director of the NIHR Dissemination Centre.  The views expressed here are her own.

Competing interests: None declared.