Insights into human behaviour can help to improve health care services

The Behavioural Insights Team lead on several pieces of research as part of the Health Foundation’s Behavioural Insights Research Programme – testing behavioural interventions that can improve efficiency and reduce waste in our health care services. We spoke to their Senior Advisor Hannah Burd about how research into human behaviour can generate insights to improve service delivery and help people live healthier lives.

How can behavioural insights research help improve health care services?

Behavioural science is about understanding how people behave in the real world and using those observations to design policy in a way that goes with the grain of human nature, rather than against it. It acknowledges that we are heavily influenced by cues from our social environment when making decisions, without even being aware of it.

That’s relevant to health research because so many of the challenges around health are behavioural in some way. It’s only when we take into account what really influences people to behave the way they do, that we can design and deliver better public services.

Can you give us some examples of how insights have been applied?

One project we worked on tried to reduce missed appointments for hospital specialists by sending patients reminder text messages. It’s easy to think people are just wasteful or lazy when they don’t turn up to appointments, but we looked into the research about what’s really going on. Mostly people just forget and life gets in the way – on the day itself, other things simply take priority.

We experimented with different messages, using randomised control trials. One just pointed out the social norm of behaviour, that nine out of ten people attend their appointments. Another said that the consultant was looking forward to seeing them. But the best performing option was the one highlighting that each appointment costs the NHS around £160. Just sending that message reduced missed appointments by 25%. Across the UK that could potentially save millions of pounds in wasted staff time and help many patients be seen more quickly. It really shows that the detail of communications to patients can help to change behaviour.

Another example of how behavioural insights could be used more is in reducing obesity. There’s lots of research which shows that how you display items affects buying decisions, supermarkets placing sweets at check-outs for example. One Australian research project rearranged the drinks in a hospital canteen. The ones with high sugar content were still available, but no longer displayed at eye level. Sales of sugary drinks dropped, but the overall number of drinks sold stayed the same.

What are you working on at the moment?

I’m leading a project funded by the Health Foundation’s Behavioural Insights Programme on increasing efficiency in NHS procurement. We think a lot of the current variation in how people buy things in the NHS is being driven by people’s environment. When busy people in hospitals need to order things they don’t have time to find the cheapest option, and current procurement portals don’t make it easy to search. If everyone bought products at the cheapest prices it’s estimated that the NHS could save around £5 billion.

So we have partnered up with a new ‘Amazon-style’ procurement platform being launched in Guy’s and St Thomas’ NHS Foundation Trust, which is much easier to modify. We’re working with the hospital and the developers to try out different ways of framing best value products so that it’s more likely that people buy those. We’re also trying to take waste out of the procurement pathway. For example, if we message people about delivery charges in a certain way, can we encourage them not to pay for next day delivery unnecessarily?

What are the particular challenges for behavioural insights research? 

Getting access to data can be a challenge. We try and use information that’s already been collected wherever we can, but getting permission is sometimes disproportionately difficult. Protecting patient level information is vitally important, but for many of our projects the information we wish to use has no identifiable patient data, yet the information governance process can slow things down. We also, quite rightly, have to go through ethics review before we can run randomised control trials involving patients. But those processes are often set up for clinical trials, testing new drugs for instance, so require high levels of documentation. Often our trials affect administrative details, yet we still need to go through the same level of scrutiny.

While we welcome ethical review before trials go live, we think there could be a lot of value in a more proportionate route for gaining ethical approval for our less intensive trials. This could unlock one of the major barriers to running well-evidenced service improvement projects using behavioural insights.

Where do you think learning from behavioural insights could have the biggest impact in the future?

Focusing on the behaviours that prevent people getting ill in the first place will have a big impact – such as finding evidence to support interventions to reduce obesity.

And with such a big rise in long term conditions, the focus has to be there too. Some great ideas came out of the Realising the Value programme that need to be tried and tested more widely. Things like offering peer support, which is shown to help keep people healthy and taking their medication. How can we encourage people to access those services in the first place?  Can we make it more likely that practitioners offer them?  There’s a big need to change how we do things inside the health and care sectors.