BACKGROUND
Despite showing strong evidence of positively assisting individuals, a common problem in the field of digital health is high attrition and low engagement. Non-healthcare, for-profit digital ventures like Facebook, LinkedIn, and Twitter leverage behavioral experiments to increase the time that users spend on their platforms. To our knowledge, digital health organizations have not conducted these types of experiments, and there are no published data indicating whether principles of behavioral economics can be leveraged to increase member engagement in digital health platforms.
OBJECTIVE
This three-arm, randomized controlled trial tests whether three types of behavioral prompts - nudges, present bias tips, and future gain tips - can increase engagement among users of two well established digital health courses designed to treat anxiety and depression.
METHODS
New members registering for anxiety and depression courses on the Evolution Health platform will be randomized into one of three arms. The first control arm will feature user home pages as-is, without behavioral prompts. The second arm will feature user home pages with a tip-of-the-day section containing directive content (reminders). The third arm will feature user home pages with at tip-of-the-day section containing social proof, present bias, and fresh start effect content. The third arm will also feature a “To-Do” item check list. Each arm will measure engagement in various platform components, specifically: goal setting, journaling, community engagement, gamified cognitive behavioral therapy course sessions, as well as other interactive tools.
RESULTS
The three arms will be compared to determine whether prompts and nudges could increase engagement and decrease attrition. If nudges and tips are deemed successful at increasing engagement, demographic characteristics of members will be tested to see if they are predictive. The experiment was designed in August 2021. Data collection is currently underway.
CONCLUSIONS
This three-arm randomized controlled trial will be the first to compare three distinct types of behavioral prompts and nudges in two digital health courses designed to treat mental health issues. We expect results to generate insights into which types of behavioral prompts work best within the populations. The insights will then be used to apply prompts and nudges to the platform’s additional addiction-focused courses. Based on results of the experiment, insights will be applied to using artificial intelligence to train the platform to recognize different usability patterns and make engagement recommendations to specific users.