The purpose of this study is to compare the use of a virtual agent vs. a human agent when onboarding oncology patients over the telephone to Remote Patient Monitoring (RPM) devices. RPM devices are instruments that a patient can use to measure their own weight and vital signs. Both the virtual and human agents will be available by telephone to instruct the patient on how to use the RPM devices to measure weight, blood pressure, heart rate, temperature, and oxygen saturation. Patients will be randomized to either the virtual or human agent, have assessments of their medical and oncological history, overall well-being, body measurements, and vital signs, and will complete questionnaires about their experience.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
The virtual agent is an interactive audio agent that is similar to voice agents that interact with callers in many industries today. They create a very human-like interaction in contrast to more traditional virtual call agents that can only respond to menu options (e.g. "choose "1" for appointments, choose "2" for questions about your bill", etc.). The virtual agent is engineered to accomplish the very specific task of onboarding the patient on use of the RPM devices. It is trained to understand normal human English speech, detects emotional tone and frustration of callers, and is trained to deescalate when appropriate.
The human agent will onboard the patient on use of the RPM devices.
Duke University Medical Center
Durham, North Carolina, United States
Usability of virtual agent vs. human agent for onboarding patients to RPM devices as assessed by the System Usability Scale (SUS)
The SUS is a questionnaire that assesses perceived usability. It includes 10 five-point items with alternating positive and negative tone. The final score is a number 0-100 with a higher score indicating better usability.
Time frame: Day 1, Day 2, Day 3
Description of the patient experience when using a virtual agent in the context of teaching a patient as assessed by the patient experience survey.
The patient experience survey includes 3 open-ended questions about experience with the training process, input on using monitoring devices at home, and any other important information for the researchers to know. Descriptive analysis will be conducted within each arm.
Time frame: Day 3
Identification of opportunities to improve the use of a virtual agent in the context of teaching a patient as assessed by the patient experience survey.
The patient experience survey includes 3 open-ended questions about experience with the training process, input on using monitoring devices at home, and any other important information for the researchers to know. Descriptive analysis will be conducted within each arm.
Time frame: Day 3
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.