Four building blocks of a successful behaviour change campaign
This article, written by A/Prof Peter Bragge was first published in Apolitical.
The ultimate aim of policy is to ensure that within the resource, logistical and political constraints in a given area, services and interventions have the maximum impact on the greatest number of citizens.
Often, this is dependent on human behaviour; some examples in the health sphere include appropriate use of antibiotics, not presenting to Emergency Departments with minor ailments and ensuring that children are vaccinated according to the government schedule.
Given the pivotal role of human behaviour in optimising policy outcomes, governments are increasingly collaborating with behaviour change experts to achieve their policy aims.
In mid-2016 my colleagues and I at BehaviourWorks collaborated with the Victorian Department of Health and Human Services (DHHS) and our communications partner The Shannon Company on a campaign to reduce unnecessary calls to the Victorian Emergency Ambulance Service (accessed by dialling “000” in Australia).
Around 40,000 calls annually — almost one in 10 — didn’t require an emergency ambulance response, but managing and responding to these calls took precious time from genuine medical emergencies.
To meet this challenge we used an established approach to behaviour change, grounded in decades of research and applied across hundreds of projects. In the end we were able to deliver real change that, according to our stakeholders, has saved lives.
Below I draw out four insights from this work with relevance to policymakers planning large behaviour change initiatives.
Large-scale behaviour change needs multi-sector buy-in
To figure out how we could cut down the number of unnecessary calls, we first held a day-long, Chatham-House Rules dialogue with 18 people representing the Victorian Government, Ambulance Victoria, hospitals, alternative “non-emergency” health service providers, telecommunications authorities and media strategists.
Prior to the dialogue they were sent a rapid review of evidence and practice to inform deliberations on key decisions. Questions included:
- For what period of time can the Government invest in behaviour change?
- Should a campaign target the whole community, or specific “frequent flyer” problem groups? What are the risks/benefits of each of these options?
- What alternatives to 000 are available for non-urgent health care needs?
- Do the community know where to find them?
- How well resourced are they?
There was not universal agreement on all of these questions; however, all participants understood and supported the key outcome; commitment to a statewide mass-media campaign to reduce unnecessary calls to 000.
This underlines the value of convening an in-depth discussion of the key issues with representatives of all vested groups.
If you are not convinced, consider the counterfactual: how many initiatives do you know of that have been scuttled by influential stakeholder groups who weren’t consulted at the beginning?
Naming undesired behaviours makes them more likely
As my colleague and I wrote in The Conversation, decades of behavioural science research has established something that we all know from personal experience: being told to “stop it” often doesn’t work. In fact, it can have the opposite effect, because naming a bad behaviour sends a signal that many people are doing it.
Research has shown that this phenomenon, known as “social norming”, has a powerful influence on behaviour due to our innate (and sometimes unconscious) need to be doing what most others are doing.
In our evidence and practice review, we found that this theory held for inappropriate use of ambulance services; when examples of inappropriate use were highlighted in a “don’t do this” campaign, there was a small increase in calls to 000. For this reason, a positive (do this) campaign message, “Save Lives. Save Ambulances for Emergencies” was used instead of a “stop it” campaign.
It seems counterintuitive to be silent on the behaviour that needs to stop, but for most negatively framed behavioural messages there is a positively-framed alternative, and the research shows that positive messaging can be effective, and eliminates the possibility of norming the wrong behaviour.
Attitudes and beliefs shift before behaviour changes
The Theory of Planned Behaviour, developed over 30 years ago by Icek Ajzen, is one of the most influential in the history of social psychology. It states that attitudes are a key predictor of behavioural intentions and resulting behaviours.
In other words, if you want to change people’s behaviour, start by changing how they think about a given topic. If attitudes need to change first, behaviour change campaigns need long-term investment.
Having unsuccessfully tried short-term strategies targeting specific groups in the community, DHHS recognised this need.
Our campaign ran from March 2017 to April 2018.
The first wave of the campaign aimed at shifting community attitudes to what ambulances should be used for. A follow-up campaign, Meet the Team (below), highlighted the role of pharmacies, the nurse-on-call service and local general practitioners as alternatives to using ambulance services for minor ailments.
The takeaway policymakers should take from this is to shift focus away from the behaviour itself to the attitudes and beliefs that may be driving it.
Consider questions like:
- Why would people think this is ok?
- Could people be misinterpreting something in the way that a service or program is being communicated?
- What other messages or attitudinal signals might people be responding to?
The takeaway policymakers should take from this is to shift their focus away from the behaviour itself to the attitudes and beliefs that may be driving it.
Monitoring and evaluation is critical
Academic knowledge is a more robust foundation for action than “first principles” or “gut feelings” (which continue to underpin a depressingly large amount of behaviour change effort and investment). But this scientific knowledge must be underpinned by robust monitoring and evaluation — especially for large, community-wide investments.
The effectiveness of the 000 campaign was measured using repeated large-scale panel surveys covering campaign awareness, attitudes towards ambulance use and knowledge of alternative health services.
Because dialling 000 is a rare behaviour at an individual level, population-level data was accessed on a range of variables including calls to 000, calls to alternative healthcare services such as “nurse on call” and attendance at emergency departments and medical clinics.
By collecting these data we were able to demonstrate that the campaign was effective in positively influencing both attitudes and behaviour.
Specifically, there were approximately 50 fewer calls requesting an emergency ambulance in Victoria after the campaign. This prompted Ambulance Victoria CEO, Tony Walker, to state publicly that: ”In my mind, it has helped save lives”.
Behaviour change en masse is challenging. Governments work tirelessly and often under considerable time pressure. many will say there is no time to waste on deliberative discussions and monitoring and evaluation plans.
But if time is of the essence, there is also no time to get it wrong. In fact, the value of applying the above principles is only multiplied.