OHSU

Automated Health Coaching


Investigator: Holly Jimison
Affiliation: OHSU Department of Medical Informatics and Clinical Epidemiology
Funding Period: 2005 - 2009
Funding Source: Intel BAIC
View related topics

Abstract

Most health care actually takes place in the home, especially for people with chronic diseases and conditions of aging.  Often the most effective components of the management of these conditions have to do with improving health behaviors, such as exercise, diet and adherence to agreed-upon treatments.  With escalating health care costs, it is important to focus on technologies that facilitate home-based care with less expensive personnel, trained to provide coaching for improving health behaviors.  We currently have several ORCATECH projects that are ready to be adapted from monitoring and diagnosis efforts to actual home-based interventions that would have near-term impact on the quality of life of elders. For example, in addition to cognitive monitoring and the early detection of decline, we now have the ability to provide a cognitive health coaching intervention, with a communications system suggesting tailored cognitive interventions, such as computer games and activities that specifically address diagnosed deficits from our monitoring system.  This type of intervention requires evaluation for efficacy and effectiveness.  In this project, we are proposing to develop and test a modular health coaching software system that would enable current ORCATECH efforts (such as interactive exercise, medication adherence, socialization, and interactive cognitive activities) to be delivered as interventions within a holistic health coaching tool.

The long-term goal of this project is to understand the optimal way to provide automated coaching and tailored feedback to participants who were engaged in a coaching plan.

Results of efforts in the first year of the project

The initial year of the project was spent developing use cases for a coaching system and defining the requirements and goals of the coaching system. Initial attempts to implement a coaching interface through the Shared Care Plan had to be abandoned when the company decided not to complete their work on the project due to other commitments.

Results of efforts in the second year of the project

In the second year the project focused on investigating, through focus groups and a study with health coaches, the best ways to communicate with elderly clients to encourage target. It also included updating the ORCATECH Management Console to include a coaching interface, and creating a Spry-based client for playing games and receiving coaching feedback. The system was field tested with 13 of the Living Laboratory subjects to determine if this is a scalable, efficient and effective approach to providing coaching to the home. We learned that to keep people engaged over a longer period, we need to have protocols that progress in time, as well as protocols that include additional activities, such as sleep management, physical exercise, and nutrition. In addition, we learned from the professional health coaches in the simulation study that our coaching messages had to always be positive (no matter what the performance) in order to keep people active on the computer.

Third year goals

In year 3 we plan to modify and supplement our database of coaching messages by 1) ensuring that all messages have a positive spin, 2) adding messages for new protocols, and adding greater variability to all message types. We will also modify the coach interface, based on our early experiences, to make it easier to characterize each participant’s status at a glance. After these system updates, we will perform a final quality and usability test, followed by a new deployment to our existing 13 participants and the remaining 17 to be recruited for the full study. We will follow these 30 participants for an additional 6 months, measuring usage, engagement, cognitive performance (computer game metrics and neuropsychological test scores), monitoring data changes, and user satisfaction. Coach measures of efficiency and perceived quality will also be monitored during this period.