It’s March 2016 and the Huffington Post has just described behavioural science as ‘cool’. In fact it’s so cool that, last September, the Obama administration set up its own Behavioral Science Team to help it design ‘human-centred’ federal programmes based on behavioural insights of the American people. The subsequent rise of Donald Trump could, at face value, suggest that American behaviours are not an exact science. More likely, it proves once again that it’s hardwired emotions, not knowledge, that drive our behaviours. It’s why we eat cake when we know it’s bad for us and watch television when we’d be better off at the gym. As humans, we so often refuse to act in our best interests because of non-conscious processes and influences that shape our behaviour. And these dictate our attitudes to health and medicine as much as they do reality TV and political referenda. Behavioural science tries to break down those influences to unlock new ways to drive better behaviours and improved outcomes. Will it work for the US Democrats in November? Who knows. But can we make it work in healthcare? Yes. We. Can.
Real-world behaviours
Behavioural science has been used in healthcare for many years, but it’s the current zeitgeist for many in pharma. The battle to understand and address real-world behaviours is, like a real estate billionaire on Super Tuesday, taking centre stage. And, like a US electorate seemingly fed up with the old guard, it’s the inadequacies of institutionalised methodology that’s driving the agenda. “There’s a large gap between clinical trial evidence and what happens with medicines in everyday clinical practice,” says Fiona Hammond, managing director, Hammell. “In the real-world there’s variability in how doctors prescribe, who they prescribe to and the advice they give to patients. From a patient perspective, there’s variance in how they use a drug; are they using it optimally or doing something different? The gap between RCT evidence and real-world practice all boils down to behaviours. The challenge is to understand those behaviours and determine the best ways to change them. Through big data, we can now access huge amounts of data on clinicians and patients to build a picture of individual behaviours. This undoubtedly has value, but it doesn’t tell you why people behave the way they do. It’s all very well knowing the ‘what’, but if you don’t understand the ‘why’, it’s very difficult to change behaviours. That’s the gap that behavioural science strives to fill.”
It’s no surprise that pharma companies are increasingly trying to understand the behaviour of all stakeholders in the healthcare environment, knowing it could be a catalyst to improve the communication dynamic between clinicians and patients and drive better outcomes. But behavioural science is broad. “It encompasses health psychology and behavioural economics and is influenced by aspects such as beliefs, cultural setting and social groups. Each aspect has deep scientific and academic roots,” says Fiona. “The challenge for pharma is to take an holistic approach. Often, companies focus on just one behavioural science – but by doing this, they risk missing out on other significant factors that explain why people behave in a particular way. It’s like trying to establish what causes cancer but only focusing on smoking as the single risk factor. To maximise behavioural science, companies need to understand the whole picture. If pharma’s methodology is as robust as it can be, it really will be able to change behaviour and drive patient-centricity.”
On its own, data is just that: data. The value comes when you translate data into evidence
The science of compliance
The past few years have seen a growing awareness of how behavioural science can drive outcomes by improving medicines adherence. However, there is variability in approach. “The most progressive companies are embracing behavioural science and trying to address adherence holistically,” says Clare Moloney, director of clinical strategy, Atlantis Healthcare. “However, some attempt to solve it with ‘reminder services’ or solutions that don’t look at people as individuals but segment them into groups or types. Unfortunately, broad segmentation doesn’t really get to the core of what’s driving an individual to behave in a particular way. It’s only by identifying individual drivers that you can begin to think about behavioural change.
“There’s a strong health psychology evidence-base that can be drawn upon to help tackle the challenge. For example, there are many known drivers of non-adherence such as individual perceptions about illness or beliefs about treatment. Further research into the disease area and treatment can help outline subsets of key drivers for those particular populations. But it’s important to recognise that adherence is just one part of a bigger self-management picture for an individual. Keeping a patient on treatment may be a KPI for pharma, but if that’s all you focus on the results will be limited. Companies need to understand that every person that’s taking their drug is an individual with their own life – and their treatment, for the most part, is just a small part of that life. We therefore need to understand that individual to be able to help them manage their condition and affect real behaviour change. It’s not just about giving people information – it’s about supporting them with tools and techniques to help them set goals, manage mood and overcome unhelpful thoughts. Health psychology allows us to use proven psychological behaviour change techniques and develop evidence-based interventions to make patients better self-managers.”
Define intervention
The dynamic between clinicians and patients is crucial in driving behavioural change. As such, pharma focuses heavily on developing interventions that support HCP decision-making and enhance patient interaction. Facilitating behavioural change is a key aim but the main goal is to drive better real-world outcomes. Achieving the former doesn’t necessarily lead to the latter. “Good interventions can contribute to behavioural change amongst physicians and that, in turn, can drive positive outcomes,” says Dennis O’Brien, CEO, Lucid. “But understanding behaviours is just the start. When you develop an intervention, you need to measure whether it really does change behaviour and whether that change improves outcomes. Objectives and metrics are key. Brand managers are very specific around programme objectives – the audience they want to reach and the messages they need to convey. But what they’re sometimes missing is the purpose; the ‘why’. At board level, companies are positioning themselves as ‘patient-centric’ and committed to ‘improving lives’. What we need is for brand leads to define how their programmes deliver against those goals. For example, if we’re aiming for a particular outcome, what specific behaviours should we target? When you’re specific about the behaviour, you’ve got a chance of measuring it. How do you do it? First, articulate – clinically – the outcomes you want. Secondly, be clear on the specific behaviours that contribute to that. Then take time to understand why those behaviours aren’t happening – and create interventions to address the challenges.”
The best interventions, says Dennis, will be personalised to address individual challenges or barriers. “There’s literature out there that describes interventions that change behaviour – but you have to tailor them for every programme and individual. That’s not as difficult as it sounds. If you’ve a good programme, an audience that’s willing to participate and well-designed digital interventions, you can create self-selecting programmes. It’s not quite one-on-one personalisation but you can develop a suite of interventions to address a particular behaviour – and let individuals choose the ones most appropriate to them.”
The gap between RCT evidence and real-world practice all boils down to behaviours
Real-world data
It’s clear that understanding – and responding to – behaviours on both sides of the HCP/patient divide is a crucial piece in the jigsaw for pharma. “Used properly, behavioural science really can lead to changes in practice,” says Dr Mariam Bibi, principal, real world evidence, Double Helix Consulting. “It offers evidence-based techniques that can help patients adhere to their treatments and encourage HCPs to implement guidelines or engage with educational programmes. These actions can have a significant impact on patient outcomes – yet they all depend on nurturing the right behaviours. So how do we get there? The trick is to make the link between real-world data and real-world evidence.
“On its own, data is just that: data. The value comes when you translate data into evidence. In the context of behavioural science, there are swathes of information out there that can be linked to help understand individual behaviours. What’s the current standard of care? What are the trends? Why are particular patient groups, HCPs or individuals behaving in a certain way? These can be bolstered by collecting more bespoke real-world data. Collectively, this information can be used as a baseline to inform interventions or make recommendations on how to implement a change in practice. You can then set up studies to monitor specific endpoints and evaluate whether the intervention resulted in the desired behavioural change. In the process, you’re again using real-world data to measure whether the intervention worked and if the ROI is sufficient to roll it out further. The caveat is that it’s very difficult to change behaviours. As humans we’re hardwired to behave in a certain way and our emotions are hugely influential. Implementation science is therefore also critical. But the lifeblood needs to be data; measure the baseline, develop the intervention and then use real-world data to evaluate the effects.”
The Trump card
And so there we have it: behavioural science is ‘cool’ but difficult. Yet done well it’s also proving to be rather effective. That’s why it pays for pharma companies to look outside and identify specialist partners that can understand both the science and the art of developing innovative interventions that can change entrenched behaviours and deliver better outcomes.
So is it time to look beyond the status quo and try a new approach? Perhaps the US electorate will give us a clue. And maybe the policies developed by the Democrats’ behavioural science team will be resonant in the end. Because the Americans surely can’t elect Donald Trump as their next President? No. They. Can’t… can they?
Sometimes human behaviour can surprise us. It makes sense to keep on top of it.