As part of Bill Clinton’s successful bid to become US president in 1992, campaign strategist James Carville coined the phrase “The economy, stupid” as one of three messages on which the Clinton team should focus. The other two messages were “Change vs. more of the same” and “Don’t forget health care” but they were largely forgotten, and the ‘economy’ phrase became the slogan for the whole campaign.
In the summer of 2024, elections are here again, this time in the UK. I’m writing this blog as a manifesto for behaviour change for road safety, for whichever new government the UK has early in July. It is to be read as an addendum to the more wide-ranging manifesto from the Parliamentary Advisory Council for Transport Safety (PACTS), to which TRL lends its full support.
The four measures suggested in the PACTS manifesto are essentially about system thinking. A National Strategy with targets, a Road Safety Investigation Branch, Graduated Licensing for new drivers and adoption of the EU Vehicle Safety Regulations. Behavioural science and behaviour change cut across these areas. Having a behavioural scientist as the Department for Transport’s Chief Scientific Advisor presents an opportunity for the leadership organisation in road safety in the UK to embrace the evidence in this domain more than ever before.
My assertion is that behaviour change is best implemented within a system approach if the goal is to actually change behaviour. This assertion is not new. Others have said it before I have, and I have said it a lot before now. Sometimes though, repetition is helpful. We must move away from a model of behaviour change that focuses far too much on ‘the individual’ and far too little on the wider set of system components that can have a much bigger impact.
Change vs. more of the same
A widely used framework for behaviour change is the COM-B model. COM-B highlights the mutual importance of capability and motivation to perform a behaviour, and the opportunity to do so, in determining the behaviour's expression. The reasoning for this is sound. For example, a driver who understands how to use their car's Intelligent Speed Assist (ISA) function may still not want to. Similarly, a driver motivated to use it may be unable to figure out how to. Finally, no matter how capable and motivated someone is, if there is no ISA fitted to a car, then it cannot be employed. Behaviour change approaches based on COM-B and similar models are increasingly being applied in road safety, and guidance is available for practitioners. One question though is 'what do behaviour change road safety interventions focus on?'
My reading of the literature is that we often ignore opportunity approaches, for the easier but less effective capability and motivation approaches. Evidence from across road safety supports this. Frank McKenna for example, wrote a narrative review in 2010 summarising education in road safety and other fields. He noted that although educational approaches are popular with policy makers, and usually have public support, they have "…one tiny inconvenient problem…" that "Across a wide array of public health, the hoped for benefits of educational measures are not realised." A very recent 2024 systematic review of existing meta-analyses shows further that across many domains interventions focused on giving people easy access to the desired behaviour (opportunity) are much more effective than those focused on adjusting behaviour at the individual level through provision of things like knowledge and skills (capability) and attitudes and beliefs (motivation). Of the individual level factors, habit appears to be the most important determinant of behaviour.
It is time for us to stop expecting behaviour change interventions focused on individuals to solve everything in road safety. It is the system that needs change. Want people to stop speeding? Then have the technology to help them with this (ISA) present in all cars; also, have the system default to ON, rather than OFF, as is the case for many vehicles. In short, make it easy for people to do the thing you want them to do, and ideally make it difficult for them to do the thing you don't want them to do.
It’s the system, stupid (and we've done this before)
There are examples in the UK of us taking a system-wide approach. Consider for example the hazard perception test, delivered in the UK as part of the theory test for licence acquisition. We knew in the 1990s that novice drivers lacked this skill. It would have been easy to have simply asked people to try and get better at hazard perception themselves, before passing their test. Instead, the decision was taken to do two things. First, training materials that worked to improve the skill were created and made available. Second, and critically, the licensing system was changed in 2002 to require that people have the skill before they are allowed to enter the driving population. I think that the hazard perception test is a masterclass in how behaviour change science and policy can come together in a system-wide approach. We wanted people to engage with training in hazard perception, so we made it accessible, and we also made it a requirement of getting a licence.
Where are we now? As a counter-example to the hazard perception test, consider the current 'ready to pass' campaign, designed to encourage people to delay booking their driving test (in an effort to make learner drivers more prepared for their test, and hopefully safer new drivers). These materials are all useful for some people, undoubtedly. However, the campaign suffers from the limitations noted in the evidence discussed above. It provides people with motivation (reduced stress, better chance of passing) but fails to address the opportunity component. The motivation provided does not seem to be overcoming the competing pressures persuading people to book their tests early; notable examples include the long waiting times for tests at the time of this blog being written, and the wider cultural pressures to get driving as soon as possible (mobility, rite of passage). Think how much more effectively we could lengthen the time people spend learning before taking their test (the desired behaviour) if things were simply designed with this in mind. Want people to take longer to learn? Then implement a minimum learning period, within a wider graduated licensing system. In such an approach, behaviour change researchers could focus on helping people to adapt to new frameworks, rather than being expected to produce alternatives to proven systems.
Behaviour change within, rather than instead of, the system
If we wish to make meaningful improvements to road safety in the UK, and keep our often-mentioned-by-the-Minister status of "having some of the safest roads in the world", then we need to move away from seeing behaviour change as an alternative to system-wide approaches, and towards seeing it as part of them. All the components of the COM-B model are useful. As a psychologist I tend to be most interested in research to uncover the secrets of capability and motivation. My interests as a researcher, however, should not take precedence over what works best to improve outcomes. Nor should politics.
If we want to change how people behave, then paradoxically we need to focus less on the people and more on the systems in which their behaviour is expressed. We know the system-based measures that will get the dial moving on road safety. Now we need policy makers brave enough to enact these measures in full.
