Secure email, also known as secure messaging (SM), between patients and healthcare team members is an important tool to communicate and deliver care. SM is also of widespread interest given its inclusion in the meaningful use of health information technology. Clinician barriers to access and use a SM system efficiently can lead to less enthusiastic SM adoption and lack of fit between the technology and clinical work. VA implemented SM expecting SM to become a key component of Patient Aligned Care Teams (PACT) and new models of care. The SM application is tethered to, but not integrated into, VA's electronic health record. There are continued questions about how SM fits into clinical workflow. Understanding how healthcare team members respond to SM and resolve patient messages can benefit VA by identifying opportunities to enhance the quality and effectiveness of online communication.
The purpose of the study is to describe messaging communication and clinical workflows associated with use of SM, and to identify issues affecting efficiency and effectiveness in primary care. Study aims include: (1) examining the content and complexity of secure messages, exploring how team members route and respond to patient online communication; (2) mapping clinical workflows for common message types, reflecting variations in message triage and team experiences, and (3) exploring PACT staff attitudes about and experiences with SM.
Using a mixed methods approach, the study employed three complementary data collection methods at two settings, Boston and Portland VA Medical Centers. For workflow analysis, four PACT teams at each site were interviewed; teams were selected based on variation in message volume and the proportion of messages completed by the provider versus other team members. We conducted 29 interviews with members of 8 teams, querying about clinical role, how SM was used in daily practice, and attitudes towards and experiences with SM. Data were analyzed using a semi-structured template, focused on elements of workflow, defined as a sequence of activities by multiple individuals related to SM use. Process maps were created to visually represent SM work processes. For secure message analysis, 500 message threads from each site were coded and analyzed in depth. We examined 25 threads from each of 20 teams selected based on message volume and timeliness of completion. We examined message content, and staff responses on each thread. For healthcare team views about SM, all PACT staff at both sites (N=534) received 2 email invitations by the site PI to complete an online survey.
Workflow interviews found that team members believed SM was valuable for patients, providing an alternate channel for communication, but were mixed in assessments of how well SM worked. Some staff preferred not to use SM while others were enthusiastic users. Some staff were not comfortable with computers, while others found SM burdensome. Motivated users often took on a greater role within workflows, e.g., opening all messages. Because SM is not integrated into the electronic record (CPRS), clinicians juggle windows to locate pertinent information such as test results to copy and paste into the SM screen. Challenges were identified at the organizational level, including SM accounts used by family members; message handling when staff were off duty, manual copying of messages into the electronic record, and lack of workload credit for using SM. Across the teams interviewed, two broad workflow patterns emerged, with some teams viewing SM primarily as another means that patients could use to contact the clinic, with responses often handled through the telephone. Other teams used SM to more systematically respond to patients.
A total of 234 PACT team members responded to the online survey request (44% and 43.4% at Boston and Portland, respectively), with 8 declining participation for a total of 225 surveys for analysis. Twenty-three percent of responses were from MD, DO or NP, 27% RN and 17% LPNs. Sixty percent used SM at least daily, and 48.2% spent 1 hour or less per week handling messages. A total of 27% perceived messages to be urgent at least monthly, 51.6% felt messages were too complex to handle at least monthly, and 55.9% handled a message from a caregiver monthly. One-third felt the team had some or much difficulty managing current volume of messages. Overall, 48% felt SM made it easier to provide care, and 81% thought SM offered value for patients. About half felt that SM had lead to fewer telephone calls to handle, and 23.6% believed SM had reduced in-person visits.
Message content analysis was completed on 925 message threads that had been initiated by a patient. The distribution of primary message topics were as follows: 47.5% medication refill or renewal; 17.7% appointment issue; 12.9% medication issue; 11.6% health issue; 10.4% test issue; 7.2% referral; 6.3% administrative issue; 4.7% for your information; 4% technical or SM issue; 3.9% appreciative; 2.5 self-monitoring data; and 1.4% life issue.
The clinical workflow for SM is complex and includes variations in implementation across different clinic settings. Direct involvement of patients as users also contributes to its complexity. Based on interview and survey data, clinicians and team members generally viewed SM as providing value to patients, but challenges remain. We identified eight workflow patterns that emerged in SM implementation, using a multi-level technology-fit framework. To improve fit, design and implementation strategies should encompass both technical improvements and organizational practices, and be sufficiently comprehensive to address workflow issues at different levels. While this study was based on data from a single organization and SM system, which influenced issues identified to some extent, the multi-level framework and analysis of fit are applicable for any system and organization. With meaningful use likely driving the number of patients using SM and the volume of messages, organizations need insights about how to implement SM systems to ensure that access to provider is improved and online communication leads to more effective, timely, patient centered and safe care.
This project sought to improve the meaningful use of SM. With an in-depth view of clinical workflow using SM, message thread content as well as input from a large number of PACT team members, the study provides valuable lessons learned that can be applied to clinician user acceptance of a novel technology and improving user task performance. Understanding SM tasks and workflow will support effective implementation and adoption of SM to support team-based care, laying the groundwork for future studies, including simulating the impact of SM on workload and designing new uses for SM and other patient-clinician communication digital tools.
External Links for this Project
- Shimada SL, Petrakis BA, Rothendler JA, Zirkle M, Zhao S, Feng H, Fix GM, Ozkaynak M, Martin T, Johnson SA, Tulu B, Gordon HS, Simon SR, Woods SS. An analysis of patient-provider secure messaging at two Veterans Health Administration medical centers: message content and resolution through secure messaging. Journal of the American Medical Informatics Association : JAMIA. 2017 Sep 1; 24(5):942-949. [view]
- Ozkaynak M, Johnson S, Shimada S, Petrakis BA, Tulu B, Archambeault C, Fix G, Schwartz E, Woods S. Examining the Multi-level Fit between Work and Technology in a Secure Messaging Implementation. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium. 2014 Nov 14; 2014:954-62. [view]
- Shimada SL, Petrakis BA, Zirkle BA, Rothendler JA, Ozkaynak M, Johnson S, Tulu B, Fix GM, Simon S, Gordon HS, Zhao S, Woods S. Content of Patient-Provider Secure Messages at Two Veterans Affairs Medical Centers. Paper presented at: Communication in Healthcare International Annual Conference; 2015 Oct 5; New Orleans, LA. [view]
- Shimada SL, Petrakis B, Zirkle M, Rothendler J, Zhao S, Ozkaynak M, Johnson S, Tulu B, Feng H, Fix G, Simon SR, Gordon HS, Woods S. Patient Aligned Care Team (PACT) Experiences with Secure Messaging at Two Early Adopter Facilities. Paper presented at: VA HSR&D / QUERI National Meeting; 2015 Jul 10; Philadelphia, PA. [view]
Treatment - Observational