Healthcare Innovation

Persuasion Engine

Presenting healthcare interventions with experiences customized to users' specific needs, skills, beliefs, interests, and quality of life.

Poor medication adherence poses serious and unnecessary health risks, and costs the US healthcare system over $100B and 125,000 deaths annually. Non-adherence has been estimated to cost the US healthcare system between $100B and $289B annually in direct costs. Studies have consistently shown that 20 to 30 percent of medication prescriptions are never filled, and on average, 50 percent of medications for chronic disease are not taken as prescribed. Rates of medication adherence drop after first six months. Strong evidence suggests that benefits attributable to improved self- management of chronic diseases could result in a cost-to-savings ratio of approximately 1:10.

There is an immediate need to provide a consumer with individualized care plans & personalized interventions for medication adherence. Patient engagement undertakings must be aimed to address personal outreach and care management via a variety of methods, modalities, technologies, and education initiatives.

Rather than a "one experience fits all" approach, healthcare applications should present interventions with experiences customized according to users' specific needs, skills, beliefs, interests, and quality of life, and should be able to persuade them towards a behavior change.

Our Approach

The Persuasion Engine uses various machine learning approaches, which enables it to fine tune itself with users' preferences and to make personalized recommendations over its usage. It leverages classic behavioral theories (such as Trans-Theoretical Model, Cialdini's Influence Principles) to identify patients with propensity towards change. It then provides hyper-personalized and persuasive recommendations tailored to be strongly appealing towards a behavior change. While patients often have healthcare apps on their device, their actual usage is far less frequent. To address such scenarios, the Persuasion Engine uses principles of crowdsourcing for clustering-based ranking of similar patients to predict a user's present stage of behavior. Persuasion Engine is a classic example of user modelling and personalization. While the current use case is in Healthcare, it is easily replicable to other domains such as Public Sector, Retail, etc.

Business Engagements

Pilot interest from Conduent's businesses for the use case of chronic illness management, and potential transfer to Pharma Life Sciences.

Technical Accomplishments

"A System and method for Personalized and persuasive recommendations for Chronic Disease," Pending Evaluation.