William Roissetter

 Early research suggests that self-management in COPD goes far beyond just medication, and that patients have diverse needs and goals.

In partnership with the Montreal Behavioural Medicine Centre, we designed an intervention to support patients with COPD. The product helps patients to understand and implement behaviours so that they can self-manage their condition.

In our early research with patients, we quickly came to understand that effective self-management for COPD goes far beyond just medication, and that patients often have needs and goals that necessitate wide ranging self-management strategies and behavioural adaptations. In other words, it’s not easy making changes. So we focussed on things like

• Ways to be more physical active within limits

• Dealing with the mental stress of a chronic condition like COPD

• Adhering to medication regimens

• Self monitoring your condition and wellbeing

This project required a nuanced understanding of motivation, and the difference between external and autonomous forms. Where some rely on persuasion, we used established evidence-based techniques like motivational communication to help people understand their options and come to a place where they can take informed action about health-oriented behaviours. By combining that knowledge with expertise in behavioural science and human-centred design we’re able to design products with the power to truly support change in people’s lives.

The way we approached this challenge was to select a target behaviour. Then work through a behaviour change framework we developed to turn theory into things that patients could interact with. This would take the form of a chatbot, that we could quickly prototype and iterate so that we could put our ideas into the hands of patients and get their responses.

Due to behaviour change design taking more time to test than most other usability testing, we adapted our user research to be longer to take into account any changes to behaviour using our service. We ran diary studies over a period of weeks to try and measure how well our interventions had succeeded.

After a year and a half of working on the service it was deemed ready to move into a more formal testing phase for software as a medical device.