Virtual Patient Advocate to Reduce Ambulatory Adverse Drug Events (Massachusetts)

Project Final Report (PDF, 574.55 KB) Disclaimer

Project Details - Ended

Project Categories

The transition process from the hospital to the ambulatory setting is non-standardized and frequently poor in quality. One-in-five hospital discharges is complicated by an adverse event within 30 days, often leading to an emergency department visit or re-hospitalization. Nationally, about 25 percent of hospitalized patients are readmitted within 90 days. Many readmissions stem from errors that can be directly attributed to the discontinuity and fragmentation of care at discharge.

This project developed a post-hospital system to address these issues. In prior AHRQ-funded studies (UC1 HS 014289 and U18 HS 015905), the project team developed a paper-based tool, the After Hospital Care Plan (AHCP), to deliver the Re-Engineered Hospital Discharge (RED), a set of recommended activities to be performed at the time of discharge. Subsequently, in order to make the AHCP electronically available, a virtual patient advocate (VPA) named “Louise” was developed. This computerized animated character emulates the face-to-face conversational behavior of an empathic provider.

This project adapted and tested Louise to interact with patients after hospital discharge in order to address the problems of discontinuity and fragmentation of care that occur in the transition from the hospital to the ambulatory setting. The “post-hospital Louise” system educates patients on self-care and medication use, assesses patient understanding and adherence to medications and appointments, and monitors for adverse events in the days after discharge.

The specific aims of this project were to:

  • Program the VPA, a computer-based, interactive, animated character, to offer patients with limited health literacy or health education advice on self-care and medication use during the transition from hospital to ambulatory care. 
  • Design and implement an ambulatory care plan using the VPA to educate the patient and respond to questions. 
  • Evaluate the intervention in the ambulatory setting. 
  • Build a robust dissemination program that will introduce this system into a health care system that is a member of a national test bed. 

The project team developed and programmed the post-hospital Louise system for use by patients in the post-discharge period. The system was pre-tested with 13 patients who had recently been hospitalized, and then revised based on this testing. Analysis of pre-testing system interactions and interviews showed that patients thought the system was helpful and easy to use.

The system was evaluated with a randomized controlled trial (RCT). The control group received typical care for discharge, and a post-discharge visit within 2 weeks. The intervention group received the RED, which included post-hospital Louise interactions between discharge and the first post-discharge visit. Of the 24 participants in the intervention group, four used the system post-discharge. The study showed that the system educates on self-care and medication use; assesses, monitors, and promotes patient understanding and adherence to medications and appointments; and monitors and reports adverse events. The project team concluded that the post-hospital Louise system has great potential to be part of the health care system. Upcoming efforts will focus on a large-scale RCT to examine statistical significance in reduction in adverse drug events between those who use the post-hospital Louise system and those who do not. The larger sample size will support more in-depth data analysis, including stratification of the sample by health literacy status.

Virtual Patient Advocate to Reduce Ambulatory Adverse Drug Events - 2011

Summary Highlights

  • Principal Investigator: 
  • Organization: 
  • Funding Mechanism: 
    RFA: HS07-007: Ambulatory Safety and Quality Program: Enabling Patient-Centered Care (PCC) Through Health Information Technology
  • Grant Number: 
    R18 HS 017196
  • Project Period: 
    September 2007 - August 2011
  • AHRQ Funding Amount: 
    $1,180,772
  • PDF Version: 
    (PDF, 191.96 KB)

Summary: The transition period between hospitalization and the first post-hospitalization ambulatory visit has a high risk for medical errors. The objective of this project was to expand the use of an animated conversational agent to assist patients during this transition. In prior AHRQ-funded studies (UC1 HS 014289 and U18 HS 015905), the research team developed a paper-based tool, the After Hospital Care Plan (AHCP), to deliver the Re-Engineered Hospital Discharge, a set of recommended activities to be performed at the time of discharge. Subsequently, in order to make the AHCP electronically available, Dr. Jack and his team developed the Virtual Patient Advocate (VPA), a computerized animated character named "Louise" that emulates the face-to-face conversational behavior of an empathic provider.

Louise offers health education, advice on monitoring and self-care, and assessment of medication dosing and adherence. To meet the needs of an ambulatory environment, the team modified the content, logic, layout, workstation, AHCP, and training manual. The team also developed links between the VPA system, Boston Medical Center's electronic medical record (EMR), and the ambulatory providers' information technology systems. They conducted a series of qualitative evaluations with potential users and clinicians. Once the beta version of the VPA was prepared, the team pre-tested the system with potential users and clinicians, made modifications pursuant to findings, and conducted a randomized controlled trial with subjects who were at high risk of adverse drug events.

The participants in the randomized trial were instructed to: 1) check in with the Louise via computer following discharge from the hospital and on a regular basis before the first post-hospital visit; 2) bring to the first post-discharge visit the result of the online interactions, which is a list of items to discuss with the clinician; and 3) meet with the Louise after the ambulatory visit for instructions on any medication regimen changes made during the office visit. The team evaluated the intervention by comparing process outcomes (i.e., enrollment, adherence, attrition, fidelity, therapeutic alliance, and patient activation) and clinical outcomes (i.e., patient and provider satisfaction, patient knowledge of self-care and medications, adverse events, and pharmacist interventions) of those using the VPA to outcomes for a usual-care group. Concurrently, the team pursued dissemination of the VPA by introducing the system to other interested health care organizations.

Specific Aims:

  • Program the VPA, a computer-based, interactive, animated character, to offer patients with limited health literacy or health education advice on self-care and medication use during the transition from hospital to ambulatory care. (Achieved)
  • Design and implement an ambulatory care plan using the VPA to educate the patient and respond to questions. (Achieved)
  • Evaluate the intervention in the ambulatory setting. (Achieved)
  • Build a robust dissemination program that will introduce this system into a health care system that is a member of a national test bed. (Achieved)

2011 Activities: The focus of 2011 was data entry, cleaning, and analysis. A 1-year no-cost extension was used to complete technology development and patient recruitment. The project timeline was therefore adjusted to allow time for data analysis in 2011. As last self-reported in the AHRQ Research Reporting System, project progress was on track and project budget spending was on target. This project was completed in August 2011.

Preliminary Impact and Findings: A total of 47 patients enrolled in the study. Sixty-two percent of enrolled patients were female, 64 percent were between the ages of 40-69, 66 percent were black, and 81 percent were single, divorced, separated, or widowed. Of enrolled patients 70 percent screened with high health literacy. Twenty-three enrollees were randomized to the control group and 24 to the intervention group. The study cohort was followed from discharge until their first appointment with their primary care provider, a time of approximately 2 weeks. Of the 47 randomized patients, four logged into the system. The four patients logged in an average of eight times each, to generate a total of 31 alerts. Fifty-five percent of the alerts related to a possible side effect. Other alerts included inability to pick up medications, intentional non-adherence, and appointment rescheduling.

The project team hypothesizes that patients may have used the system only if they were concerned about their health. None of the four users were re-hospitalized or had an emergency room visit. Dr, Jack reported that the study was limited by participants' lack of access to the Internet in their homes, which was crucial for the success of this project. Dr. Jack would like to conduct a fully-powered randomized controlled study following this pilot study.

Target Population: Adults, Low-SES/Low Income*, Medically Underserved, Racial or Ethnic Minorities*

Strategic Goal: Develop and disseminate health IT evidence and evidence-based tools to support patient-centered care, the coordination of care across transitions in care settings, and the use of electronic exchange of health information to improve quality of care.

Business Goal: Implementation and Use

* This target population is one of AHRQ's priority populations.

Virtual Patient Advocate to Reduce Ambulatory Adverse Drug Events - 2010

Summary Highlights

  • Principal Investigator: 
  • Organization: 
  • Funding Mechanism: 
    RFA: HS07-007: Ambulatory Safety and Quality Program: Enabling Patient-Centered Care (PCC) Through Health Information Technology
  • Grant Number: 
    R18 HS 017196
  • Project Period: 
    September 2007 – August 2011, Including No-Cost Extension
  • AHRQ Funding Amount: 
    $1,180,772
  • PDF Version: 
    (PDF, 325.13 KB)


Target Population: Adults

Summary: The transition period between hospitalization and the first post-hospitalization ambulatory visit has a high risk for medical errors. The objective of this project is to expand the use of an animated conversational agent to assist patients during this transition. In prior studies, the project team developed a paper-based tool, the After Hospital Care Plan (AHCP), to deliver the Re-Engineered Hospital Discharge, a set of recommended activities to be performed at the time of discharge. Subsequently, in order to make the AHCP electronically available, Dr. Jack and his team developed the Virtual Patient Advocate (VPA), a computerized, animated character that emulates the face-to-face conversational behavior of an empathic provider.

Following hospital discharge, patients have access to VPA via a Web portal, and are encouraged to use it before their first visit with their primary care physician (PCP). The VPA offers health education, advice on monitoring and self-care, and assessment of medication dosing and adherence. To meet the needs of an ambulatory environment, the team is modifying the content, logic, layout, workstation, AHCP, and training manual. The team is also developing links between the VPA system, Boston Medical Center’s Certified Commission for Health Information Technology-certified GE Centricity electronic medical record (EMR), and the ambulatory providers’ information technology (IT) systems, as well as conducting a series of qualitative evaluations with potential users and clinicians. Once the beta version of the VPA is sufficiently prepared, the team will pretest the system with potential users and clinicians, make modifications pursuant to findings, and conduct a randomized controlled trial (RCT) with subjects who are at high risk of adverse drug events. The system has now been completed, debugged, and tested with hospitalized patients. The RCT, a test-of-concept trial, began in August 2010. The outcome data will be collected after the first postdischarge appointment with the responsible clinician.

The participants in the randomized trial will be instructed to: 1) check in with the VPA via computer following discharge from the hospital and on a regular basis before the first post-hospital visit; 2) bring to the first postdischarge visit the result of the online interactions, which is a list of items to discuss with the clinician; and 3) meet with the VPA after the ambulatory visit for instructions on any medication regimen changes made during the office visit. The team will evaluate the intervention by comparing process outcomes (i.e., enrollment, adherence, attrition, fidelity, therapeutic alliance, and patient activation) and clinical outcomes (i.e., patient and provider satisfaction, patient knowledge of self-care and medications, adverse events, and pharmacist interventions) of those using the VPA to outcomes for a usual care group. Concurrently, the team is pursuing dissemination of the VPA by introducing the system to other interested health care organizations.

Specific Aims:
  • Program the VPA, a computer-based, interactive, animated character, to offer patients with limited health literacy or health education advice on self-care and medication use during the transition from hospital to ambulatory care. (Achieved)
  • Design and implement an ambulatory care plan using the VPA to educate the patient and respond to questions. (Achieved)
  • Evaluate the intervention in the ambulatory setting. (Ongoing)
  • Build a robust dissemination program that will introduce this system into a health care system that is a member of a national test bed. (Achieved)

2010 Activities: The intervention was launched after significant development, testing, and refinement of the VPA system. The study team continued testing the workstation to identify areas that needed improvement or additions. A variety of clinicians and patients reviewed the system and made recommendations about how the system should be modified. Additionally, testing for the VPA portal included assessment of how information flows from the patient to the clinical team and back to the patient. A codification system was designed to categorize alerts. Additional diagnosis scripts, diagnosis pages, medications, medication scripts, primary care providers, and pharmacies have been added to the existing selections. Integration of the workstation with the Boston Medical Center EMR was completed.

Following extensive testing, the research team began to recruit patients for the RCT. Due to delays with development, the length of time to run the RCT was truncated to 5.5 weeks. A total of 47 patients were randomized; 23 to the control group and 24 to the intervention group.

Grantee’s Most Recent Self-Reported Quarterly Status (as of December 2010): Project milestones and aims are mostly on track, while project spending is on target.

Preliminary Impact and Findings: The RCT was completed and data analysis is in progress. The study cohort was followed from discharge until the first appointment with their primary care provider, a time of approximately two weeks. Of these 47 randomized patients, four logged into the system. The four patients logged in an average of eight times each, to generate a total of 31 alerts. Fifty-five percent of alerts were related to a possible side effect. The grant team hypothesizes that patients may have only used the system if they were concerned about their health. Of the 47 users, 17 were re-hospitalized or had an emergency department visit. None of the four users were re-hospitalized or had an emergency room visit.

Strategic Goal: Develop and disseminate health IT evidence and evidence-based tools to support patient-centered care, the coordination of care across transitions in care settings, and the use of electronic exchange of health information to improve quality of care.

Business Goal: Implementation and Use

Virtual Patient Advocate to Reduce Ambulatory Adverse Drug Events - Final Report

Citation:
Jack B. Virtual Patient Advocate to Reduce Ambulatory Adverse Drug Events - Final Report. (Prepared by Boston Medical Center under Grant No. R18 HS017196). Rockville, MD: Agency for Healthcare Research and Quality, 2012. (PDF, 574.55 KB)

The findings and conclusions in this document are those of the author(s), who are responsible for its content, and do not necessarily represent the views of AHRQ. No statement in this report should be construed as an official position of AHRQ or of the U.S. Department of Health and Human Services.
Principal Investigator: 
Document Type: 
This project does not have any related resource.
This project does not have any related survey.
This project does not have any related project spotlight.
This project does not have any related survey.
This project does not have any related story.
This project does not have any related emerging lesson.