Medication discrepancies, defined as unintentional differences found between patients' medical records and patients' reports of the medication they are taking, occur frequently after hospital discharge, predisposing to adverse drug events (ADEs), emergency department visits and readmissions. Resolving medication discrepancies - medication reconciliation - is mandated at every care transition, but little is known about intervention strategies to improve medication reconciliation in the post-discharge period, when patients may lack prompt access to primary care and are at high risk for ADEs. To address this gap, the investigators developed and pilot tested the Secure Messaging for Medication Reconciliation Tool (SMMRT), with a pharmacist communicating with Veterans to review medications and reconcile discrepancies after hospital discharge via Secure Messaging (SM), within My HealtheVet (MHV), VA's patient portal. The objectives of The SMMRT Trial are therefore

To optimize the end-users' experience with SMMRT through usability testing and refinement of the tool; To conduct a randomized controlled trial (RCT) of usual care vs. usual care plus MHV Training vs. usual care plus MHV Training plus SMMRT to reduce hospital utilization; To evaluate how Veterans and staff perceived the impact of SMMRT on routine clinical practices and, specifically, on Veterans' interactions with their primary care providers.

Study Procedure of Work Done at Richard L. Roudebush VAMC in Indianapolis

Aim 1 of SMMRT will be conducted at the Human-Computer Interaction & Simulation Laboratory within the HSR&D Center for Health Information and Communication at the Richard L. Roudebush VAMC in Indianapolis, IN. The work in Indianapolis will include a formal usability test of SMMRT to access and improve usability for pharmacists, nurses, and patients. Ten VA pharmacists, 10 VA nurses, and 10 VA patients will participate in the usability tests in Indianapolis. They will assess how well the SMMRT interface supports their respective medication reconciliation tasks. Pharmacists who work for the Roudebush VAMC and have some responsibility for medication reconciliation in their normal work will be eligible to participate in the study. Roudebush VAMC nurses who work as part of a Patient Aligned Care Team (PACT) will be eligible. Eligible patients must be taking at least 5 medications and must have been discharged from the Roudebush VAMC within the last 30 days. Veterans will be excluded if they are cognitively impaired, based on the Callahan screener. Patients will be recruited regardless of prior MHV use. No patients will have previous experience with SMMRT, as it is a new feature in MHV. Usability testing sessions will be about 30 minutes and will include audio and video recordings of the computer screen movements and the participant's face. A standardized script is read to the participants to instruct them to do certain tasks that will evaluate specific usability issues in the SMMRT. The participants enter reconciled medications into the SMMRT and the completed work is sent to the research team for analysis.

Although the investigators intend to complete all data collection procedures for Aim 1 at the Indianapolis VA, the investigators did note in the submitted grant proposal (Section 4.g.1) that the investigators would recruit (VA) Boston pharmacists for Aim 1 data collection in the unlikely event that the investigators cannot recruit sufficient numbers from Indianapolis. If that unlikely contingency arises (i.e., the investigators determine that there is a need to recruit pharmacists from VA Boston), the investigators will submit an amendment to the VA Boston IRB with appropriate consent and authorization forms at that time.

Study Procedure of Work at VABHS. The Study has received approval for a Project Modification.

The essential scientific changes of the proposed modification are (1) switching from a three-arm to a two-arm RCT; (2) changing the main outcome measure from hospital utilization to medication discrepancies; and (3) reducing the sample size from 1400 to 240 subject. The second specific aim of the Project will become the following:

Aim 2. To conduct a two-arm RCT to evaluate the effect of SMMRT. This trial will compare Usual Care (UC) with UC plus the bundled MHV/SMMRT Intervention among Veterans discharged from the VA Boston Community Living Center (CLC) and from the VA Boston acute inpatient setting of West Roxbury. The primary outcome measure will be medication discrepancies detected 30 days after discharge. Secondary outcome measures will be 30-day hospital utilization (combined readmissions plus emergency department use) and Veterans' self-efficacy in medication use. The primary hypothesis is that SMMRT will result in a reduction in 30-day medication discrepancies as compared with UC.

The third specific aim of the Project remains unchanged:

Aim 3. To evaluate the SMMRT intervention for potential future implementation. In anticipation of future implementation, we will use qualitative methods to identify features of the intervention that are most and least effective, as well as elements that are most and least likely to be accepted into clinical practice.

Inclusion Criteria:

• Veterans age 18 years or older

• Having a VA primary care provider (PCP) at any VA facility in VISN-1

• Planned discharge home (as opposed to another facility)

• Computer and internet access

• Anticipated to be discharged with at least 5 medications.

  • Having a VA PCP will be defined as having seen the provider within the past two years
  • Planned discharge home will be ascertained from the Veteran's nurse; approximately 75% of VA Boston discharges are to home
  • The nurse will also provide number of anticipated discharge medications

Exclusion Criteria:

• Cognitive impairment (as determined by the Callahan screener)

Not sharing individual patient data.

Comparable.

30-day follow-up

30-day follow-up telephone calls