NCT02668705

Agent-Enhanced Document Explanation

Research Ethics and Safety Promoted by Embodied Conversational Technology

• Status: Terminated
• Phase: N/A
• Study type: Interventional
• Enrollment: 101 (actual)
• Sponsor: Boston Medical Center · Academic / Other
• Sex: All
• Age: 21 Years
• Healthy volunteers: Accepted

Summary

Both the main and sub-studies will investigate how consent forms and other health care documents are explained to patients. The findings will be used to create a computer program that can successfully explain consent and other health-related forms to individuals who have difficulty reading and understanding consent forms on their own as a supplement to the normal consent process done in research settings.

Detailed description

The main study in this protocol is a 3 arm (HUMAN vs. ECA vs. ECA + HUMAN), randomized, between-subjects experimental design. There is one sub-study, the Images study, also included in this protocol. Subjects who are eligible for the main study can choose if they would like to participate in this sub-study. This is a within-subjects design, where subjects will be randomized to different conditions within the system. One additional sub-study, called Stance, also a within-subjects design, where subjects will be randomized to different conditions within the system, was also included in the original protocol,but has since been closed to enrollment. Data collection for this sub-study was the same as for the current sub study (Images), and the main study (Consent Advocate). Eligible subjects will be given the option to participate in the sub-study in addition to the main study. Subjects may opt out of this sub-study, but not the main study (as in, a subject cannot do the sub-study and NOT do the main study, unless there is some legitimate reason to stop after the sub-study). Due to the nature of the content, the order of the studies must be: Images (sub) Consent Advocate (main) In these studies, baseline data is collected by the RA verbally administering the instruments to the subjects. Then, the subject will sit in front of a touch-screen monitor to interact with the character on the screen. A member of the research team will be available at all times to assist with any technical issues, or if the subject needs to take a break. Once the on-screen interaction(s) are complete, the RA will administer any post-system data collection instruments. Due to the amount of time required for data collection and completion of study protocol (each study takes approximately 60-120 minutes), the subject can choose to complete the 2 studies in 1 or 2 study visits, depending on study subject and staff scheduling needs. Main Study: Consent Advocate In this study, the participant will be randomized to one of three arms: 1) a sham research informed consent form will be explained by a research assistant; 2) a sham research informed consent form will be explained by an ECA, or 3) a sham research informed consent form will be explained by an ECA and then questions will be answered by a research assistant. Baseline Data Collection: 1. Sociodemographics 2. Need for Cognition 3. Patient Activation 4. Trust in Medical System 5. REALM Post Data Collection: 1. BICEP (primary outcome measure) 2. Trust in Agent 3. Instructor Evaluation Permission to audiotape the interaction will be obtained prior to conducting this portion of the sub-study. The audio recording will be used to analyze what questions the participant asks. If assigned to the ECA and RA condition, the audio recording will be used by the research assistant prior to conducting the live consent process. A knowledge test will be conducted after the consent process to ascertain what was understood. This study should take approximately 60 minutes to complete. Sub-study: 1\. Images. The ECA will explain a consent form but will use images to represent potential risk as opposed to descriptive language only. Understanding will be measured by a knowledge questionnaire. (n=50) Baseline Data Collection: Sociodemographics Need for Cognition Post Data Collection: Instructor Evaluation Informed Consent Scales Knowledge (of sham consent form) This study should take approximately 60-90 minutes to complete. If a subject participates in the sub-study that requires the same data collection forms (i.e. sociodemographics, need for cognition, etc), these forms will only be completed once for each subject, not for each study they participate in.

Conditions

• Mental Competency

Interventions

Type	Name	Description
OTHER	Embodied Conversational Agent	A computer generated character that can administer informed consent to a potential research subject.

Timeline

• Start date: 2015-09-01
• Primary completion: 2016-09-01
• Completion: 2016-09-01
• First posted: 2016-01-29
• Last updated: 2016-11-25

Locations

1 site across 1 country: United States