Background and Significance: Adverse drug events (ADE) are the most common of all healthcare associated adverse events. Transitions between inpatient and ambulatory care can lead to ADE and avoidable healthcare utilization affecting up to 20% of patients. Insufficient monitoring has been identified as a particularly common cause of preventable and ameliorable ADE. The Joint Commission for Accreditation of Healthcare Organizations (JCAHO) has called for a process to "accurately and completely reconcile medications across the continuum of care." Inpatient computerized medication reconciliation tools have shown great potential, but little is known about successful medication reconciliation strategies immediately post-discharge. One promising solution to outpatient medication reconciliation is the use of secure electronic communication (SEC) between patients, accessing their personal health record via a web portal, and providers within an electronic health record. Surveys have consistently found that patients want to communicate with their primary care providers via SEC, and SEC is rapidly becoming a standard feature in electronic health records. A SEC-based approach to outpatient medication reconciliation following hospital discharge has the potential to improve patient safety, but important questions must be addressed, including: What information needs to be included in the SEC? What elements of the SEC are essential to its design? How can SEC be best incorporated into the workflow of the primary care office practice? Will patients and physicians find this approach acceptable? To address these questions, we propose a one-year pilot study at VA Boston with the following specific aims:

Aim 1: To evaluate the primary care environment's receptivity to secure electronic communication (SEC) for medication reconciliation and design a prototype SEC. We will use qualitative methods to characterize the primary care clinic environment - the physical space and facilities, the workflow, the resources, as well as the knowledge, attitudes and skills of staff and, most importantly, the patients. Specific approaches will include in-depth interviews, direct ethnographic observations, and visualization methods Aim 2. To develop a prototype SEC for medication reconciliation through a participatory design process, involving primary care clinicians, staff and patients working collaboratively with the research team, and to test the usability and acceptability of the prototype SEC among a sample of patients and primary care clinicians.

Research and Development Plan: With extensive experience in designing and evaluating health care informatics innovations, we have assembled a multi-disciplinary team with expertise from primary care internal medicine, clinical informatics, health services research, qualitative research, clinical pharmacy, nursing and industrial engineering. This team will carry out a formative evaluation and develop a prototype of a secure electronic message to facilitate medication reconciliation following hospital discharge. The proposed CIMIT project will leverage the resources and expertise of the e-Health QUERI (a national VA program for innovation and evaluation of the VA's e-Health programs) at VA Boston and other participating VA facilities nationwide. The figure below shows a very rough conceptualization of the SEC that will guide the qualitative research and participatory design of the prototype. This project will lay the foundation for a future rigorous evaluation of this approach to medication reconciliation and improving patient safety.

Adverse drug events (ADE) are the most common of all healthcare-associated adverse events1 and commonly occur after hospitalization when multiple changes to medication regimens take place in the context of poor patient education, inadequate information transfer and delayed continuity of care. Medication reconciliation - i.e., any intentional process to resolve discrepancies between what medications the patient is taking and what medications the health care team thinks the patient is taking - has been designated a "National Patient Safety Goal" since 2005.2-6 Recent studies examining inpatient computerized medication reconciliation tools and pharmacist-facilitated medication reconciliation programs at hospital discharge have shown promising results. However, little is known about successful medication reconciliation strategies following discharge, representing an untapped point of leverage to improve medication safety.

With recently awarded seed funding, and in collaboration with the Department of Veterans' Affairs (VA) nationwide Medication Reconciliation Initiative, we are developing an informatics-based tool for medication reconciliation following hospital discharge by secure, e-mail-like communication: the Secure Messaging for Medication Reconciliation Tool (SMMRT). In this proposed study, we plan to pilot the SMMRT prototype through My HealtheVet (MHV), the VA's secure personal health record and web portal, among 50 Veterans at VA Boston, with the following two Specific Aims:

Specific Aim 1: Refine the existing SMMRT prototype and integrate it into MHV's secure messaging web-interface, with authentication of secure two-way communication between patient and primary healthcare team. We will use a participatory and patient-centered process involving the primary care clinicians, pharmacists, nurses, other staff members and Veterans working collaboratively with the research team to establish the usability and acceptability of the prototype SMMRT.

Specific Aim 2: Pilot the SMMRT prototype, refined through Specific Aim 1, among 50 Veterans with primary care relationships and hospitalized at our facility. Veterans will be recruited prior to hospital discharge, enrolled in MHV if necessary, and trained in the use of secure messaging and specifically the SMMRT tool.

This study will benefit from a recently funded preliminary study (Veterans' Engineering Resource Center seed funding, $25,000), in enabling us to analyze the primary care environment's receptivity to SMMRT-based post-discharge medication reconciliation, focusing on staff activation and education, workflow optimization to incorporate staff and clinicians managing SMMRT's incoming and outgoing messages, and coordination with MHV staff to ensure continued successful software integration.

This proposed CIMIT-funded study will lead to a refined and strengthened intervention to enhance the preliminary use of SMMRT, as well as qualitative results that will constitute the foundation of an investigator-initiated research proposal alongside a full-scale implementation of SMMRT into MHV for Veterans nationwide.

Inclusion Criteria for Staff and Clinicians:

• Staff at Jamaica Plain and West Roxbury VA.
• Trained in use of secure messaging.

Inclusion Criteria for Veteran pre intervention focus groups:

• Over 18.
• Receive medical care at the Jamaica Plain or West Roxbury VA.

Inclusion Criteria for Veteran intervention sample:

• Over 18.
• Inpatient at West Roxbury VA.
• Home computer access.

Exclusion Criteria for Staff and Clinicians:

• N/A

Exclusion Criteria for Veteran pre intervention focus groups

• Non-ambulatory
• Not receiving medical care at Jamaica Plain or West Roxbury VA.

Exclusion Criteria for Veteran intervention sample.

• Non-inpatient.
• No home computer access.