Development of a Chatbot for Delivering Long-Term Motivational Interviewing for Improving Exercise Adherence in Hemodialysis Patients

There are approximately 600,000 individuals in the U.S. with kidney failure undergoing maintenance hemodialysis (HD) therapy. HD patients have several co-morbidities, including muscle wasting, cardiovascular disease, and bone disorders, with very high medical costs, averaging ~ $93,000/patient/year. Lifestyle interventions such as exercise and nutritional modification are often prescribed to improve their health, but compliance with these interventions is extremely poor. The overarching goal of the study is to develop an accessible, generalizable, and adaptive digital health solution for promoting long-term exercise adherence among the patients with kidney failure undergoing maintenance hemodialysis (HD) therapy. Motivational Interviewing (MI) is a counseling approach demonstrated to promote positive behavior change but is often impractical to conduct because of the administration costs (e.g., qualified professionals or frequency of sessions required). A possible alternative would be to develop an automated conversational agent to deliver the MI. The proposed study will bridge theories in behavioral sciences (e.g., MI and stages of behavioral change) and Natural Language Processing (NLP) to develop a long-term MI conversational agent, LotMintBot, for exercise adherence. This agent will use MI to increase HD patients’ intentions and commitments to exercise programs in HD patients. This work will complement the Jump ARCHES grant our team received in 2021, which aimed at developing conversational agent to deliver brief MI for improving the intentions and acceptance of COVID-19 vaccination. Read more.