Collaborative Patient Portals: Computer-based Agents and Patients' Understanding of Numeric Health Information
Project Final Report (PDF, 329.47 KB) Disclaimer
Disclaimer
Disclaimer details
Project Details -
Completed
-
Grant NumberR21 HS022948
-
AHRQ Funded Amount$299,602
-
Principal Investigator(s)
-
Organization
-
LocationChampaignIllinois
-
Project Dates05/01/2014 - 04/30/2017
-
Technology
-
Care Setting
-
Population
-
Type of Care
-
Health Care Theme
Portals are intended to support patient-centered care, but they often pose challenges as they present health information and clinical results with little context. Older patients may struggle to understand results if they have limited numeracy and memory problems; they interpret data in the context of their own goals and knowledge. Face-to-face visits allow clinicians to help patients understand their numeric information using both verbal and nonverbal strategies that provide context. However, with the trend toward decreasing patient contact time, clinicians are challenged to provide this support.
This project investigated whether patient portal messages, delivering cholesterol and diabetes screening results, enhanced with context would improve patient comprehension and risk perception. The researchers designed messages to add context verbally, graphically, or visually. They also evaluated whether the new formats improved attitudes toward and intentions to perform self-care behaviors.
The specific aims of the project were as follows:
- Identify portal messages that improve comprehension of clinical test results.
- Identify messages that make test results more actionable.
Researchers enhanced patient portal context: 1) verbally, by adding low, borderline, or high-risk labels to provide context and promote emotional processing; 2) graphically, by embedding results in color-coded number lines and emphasizing the order of low- to high-risk results; and 3) with videos using a real or an animated virtual computer agent (CA) provider to discuss results and their implications with verbal and nonverbal cues. A total of 144 older adults were randomly assigned a fictitious patient profile scenario and a message describing results for that patient in either the standard or enhanced portal formats. After viewing each scenario, participants responded to a series of questions that measured memory, perceived risk, emotional reactions to the risk information, and intentions to perform self-care activities to reduce risk of disease.
Findings indicated that older adults’ memory for test results was improved by enhanced compared to standard message formats, with the video format being most effective. However, the enhanced formats did not directly improve emotional response, risk perception, or behavioral intentions. Results also indicated that the CA was as effective as the video for supporting memory and emotional response. However, participants preferred the videos because the provider was perceived as more relatable, personable, conversational, natural sounding, or as explaining more, despite both formats delivering identical content. In the future, researchers plan to investigate other types of communication in portals, such as patient stories about self-care, which have been found to influence patient attitude and performance of health behaviors.
Disclaimer
Disclaimer details