Behavioural Insights trials - one hit wonders or sustainable and scalable solutions?

21 April 2015Simon Raadsma


Tagscase study, health, nsw, policy, trial

Lessons in practical applications of BI from the health domain.


My first trial in the Behavioural Insights Unit (BIU) focused on helping public hospitals develop creative ways to support patients' use of private health insurance (PHI). BIU was approached by the NSW Ministry of Health (MoH) to assist them with gaining a better insight into why some hospitals have significantly lower than average PHI use.

Westmead Emergency Department was our first trial site. I undertook detailed fieldwork and workshops with patients and frontline staff to identify barriers and solutions. Through some of the strategies listed below, we saw a 7 percentage point increase from when we implemented the interventions compared to the previous year's data, far surpassing our initial expectations.

One of the most interesting findings for me was that the results continued to increase after I reduced my intensive involvement in the project and handed full operational responsibility back to the hospital. This indicated that the model was sustainable. But was it scalable?

To answer this question, I was seconded to MoH to first test whether the model was replicable. I chose two other hospitals, Auburn and Fairfield, both of which had low rates of PHI use. Both sites showed similar outcomes to that of Westmead (3.6 percentage point increase in Auburn and 2.7 percentage point increase in Fairfield). For example, the graph in Table 1 shows that after April I reduced my intensive involvement at Auburn emergency department, however, the results continued to increase, therefore proving that the model is both sustainable and replicable.


Table 1: Proportion of patients admitted from the emergency department using their PHI – Auburn Hospital

Image not found 


To date, these three trial sites have generated a net benefit of over $7 million to Western Sydney and South Western Sydney Local Health Districts.

However, even though the model proved to be sustainable and replicable that does not necessarily mean it is scalable. For a small Unit like the BIU it would be unrealistic to try and implement this model in the 220+ hospitals in NSW. Given this challenge we had to come up with a way to share the lessons learned with the Local Health Districts and motivate them to implement the changes. The answer to this challenge was to develop a set of guidelines which would capture the process we went through, the interventions we developed and the lessons we learned.

The Guidelines are currently in development and will draw heavily upon the UK Behavioural Insights Team’s trial methodology and EAST principles. The Guide will also include the tools I developed to Understand, Build and Test the interventions. The high-level solutions implemented include:

  • Easy – simplified admission processes for staff and patients
  • Attractive – guaranteed no out-of-pocket costs for patients when they elected to use their PHI
  • Social – motivating the revenue and clerical staff to work as a single team
  • Timely – timely prompts for patients and performance feedback for staff

One year on, I have achieved what I set out to do and in the process gained some meaningful insights into taking trials to scale. Three insights I would like to share with you are:

  • Results build the case for change – In order to motivate decision makers to implement this model in their hospitals we had to achieve compelling results. I chose hospitals with low PHI usage rates in order to demonstrate large effects. I also looked for executive staff like Shaun Drummond, Chief Operating Officer at Westmead, who would champion our work and be supportive of new ways of doing business. 
  • People are the biggest variable – in order for the model to achieve results it requires a champion on the ground, building relationships and driving change. Without this there is a limited chance that results would be achieved and especially sustained. Jenny Hart, Clerical Manager in Westmead ED, is a great example of this. She took on additional responsibilities and ensured that the solutions we co-designed were implemented. Jenny was nominated for the Premier’s Award for Individual Excellence and Achievement for her involvement in this project.
  • Train the trainer models work – I used high-performing individuals to help train other hospitals. Staff value and trust advice from their peers more so than outsiders. I had some one from Auburn seconded from Westmead and had the Project Manager from Fairfield spend time training at Westmead and Auburn. The first-hand experience they gained from this model really helped to expedite the change process at their hospitals.  

I have just returned to the BIU after my 12-month secondment to MoH and looking forward to working on obesity and hospital efficiency trails.