She began by asking why people behave in ways that harm their health. Her answer is that short term rewards are very powerful drivers of behavior. Changing these behaviors is one of the goals of both web and mobile app delivered projects. Examples include apps that try to get people to exercise, quit smoking, lose weight or change diets to improve blood pressure or heart failure.
Michie stated that to change behavior requires four steps:
- understand the behavior you are trying to change. She sees too many apps and projects where the authors have too superficial knowledge of the behavior they are trying to change.
- consider the full range of options for intervention
- use a systematic method for selecting behavior change
- evaluate and re-evluate the interventions, nothing ever stays the same.
In order to understanding behavior, one needs to see that behavior emerges in the context of the interaction of capability, opportunity, and motivation. She refers to this triad as COM-B. In her review of behavioral change applications, she that too many focus on motivation without addressing opportunity or capability. We over emphasize the reflective components of behavior rather than the impulsive ones.
#med2 @SusanMichie – we've tended to over-emphasise the reflective components of behaviour rather than impulsive ones ncbi.nlm.nih.gov/pubmed/15454347—
AnneMarie Cunningham (@amcunningham) September 23, 2013
She says that we are so enamored by the internet and smartphones that we look at some of these apps as a panacea. We are at risk of focusing too much on the delivery of the treatment rather than its content. She compares this to emphasizing on the size and color of a pill rather than its biological effects. In order to study the content of behavioral interventions she studied every possible intervention and categorized them into 93 techniques in 16 groupings. By creating this schema, we can describe interventions with a common language in a replicable way.
She then highlighted Txt2Stop a mobile phone app that helps people quit smoking. Importantly, Txt2Stop was proven effective in a RCT.
Michie says we need to analyze successful interventions to find out which components make it successful so we can learn and extend our successes. With Txt2Stop she was impressed with its ability to engage users. She says these interventions need to be open source so future researchers and developers can build on successes to make incremental improvements and dissect successes to determine what makes them effective.
This Keynote, was in many ways an ideal opening talk because it empowered everyone to look at behavioral change projects in a more scientific and rigorous way. Great choice.
Some of her peer reviewed work on this schema can be found here.
Post written by Dr. Joel Topf, eAJKD Advisory Board member.
Check out all of eAJKD’s coverage of Medicine 2.0’13.