More than a century after the Flexner Report urged standardized medical education in the United States, schools again seek to align curricula with the evolving needs of the contemporary medical student and health care system.1 Current areas of reform in medical education include increased technology use, the inclusion of competency-based training, and the need for a greater understanding of complex health systems.2 The Carnegie Foundation for the Advancement of Teaching and the Association of American Medical Colleges have called for curricular changes domestically, while multicountry consortia like the Commission on Education of Health Professionals for the 21st Century have proposed international reforms.2–5 Most proposals include elements of longitudinal patient experiences (LPE), or curricula designed to allow students to work with one or more patients at multiple points in time, often across distinct clinical settings and/or phases of illness. Medical educators are not alone in calling for a greater emphasis on LPE; students themselves, including those in traditional clerkship models, wish to benefit from care continuity and systems-based learning in their interactions with patients and clinical teams.
Though competency-based, outcome-driven study of longitudinal clinical experiences versus immersion clinical experiences is limited, there is considerable rationale for integrating LPE into third-year curricula. Accreditation Council for Graduate Medical Education competencies express the need for proficiency in working with patients across health services and in communities; requisite proficiencies include well-developed interpersonal communication skills, the ability to acquire competency in systems-based practice in a complex health care network, and the capacity to coordinate care across practice settings.6 LPE affords students greater exposure to the complexities of patient care before entering residency training programs. Further, notifying students about patients in real time simulates the experience of responding to patient needs, mirroring both residency and practice while promoting prioritization and time management skills.
Despite strong advocacy for LPE inclusion in traditional and novel curricular models, following patients over time and across health systems remains a challenge. Patients are cared for during arranged and emergent visits with varied professionals in inpatient and outpatient settings, which are difficult to track even for involved practitioners, let alone students in a role that spans practice and observation. To simplify and enhance the experience of following patients, we describe a pilot program that uses a novel visit notification tool (VNT) that integrates with the electronic medical record (EMR) to facilitate LPE within a traditional block clerkship curriculum at the Massachusetts General Hospital (MGH). With support from the MGH chief information officer (CIO) and two faculty advisors, two third-year medical students developed and implemented this pilot based on their interest in patient continuity beyond that which occurred during block rotations.
Harvard Medical School in Boston, Massachusetts, has roughly 165 third-year students. Approximately 45 are sent to the MGH for their Principal Clinical Experience (PCE), which comprises all required clinical third-year block rotations: internal medicine, neurology, obstetrics–gynecology, pediatrics, primary care, psychiatry, radiology, and surgery.
In July–August 2012, two of us (S.N.E., K.E.K.), third-year medical students at the MGH who wished to follow patients longitudinally, sought a method for immediate notification of patient visits to facilitate this experience. We located an existing mechanism for patient tracking by inquiring about similar efforts within the MGH.7 The VNT was originally developed to support participation in the Medicare Care Management for High Cost Beneficiaries (CMHCB) demonstration project. The tool allowed CMHCB case managers to receive real-time notifications regarding all patient visits within the Partners HealthCare system. One of us (K.J.), the MGH’s current CIO, originally constructed the tool in a Microsoft Access database and designed it to extract three pieces of “loose” information from the EMR: (1) real-time census data, (2) future schedule data, and (3) a registry that associates health workers to a known patient. The VNT is now used as a Health Insurance Portability and Accountability Act–compliant, Web-based module with EMR integration that can be accessed using a clinical identification and password.
The CIO modified the tool to allow third-year students to create accounts. During the initial eight-month pilot period (August 2012–May 2013), a secure, password-protected electronic patient cohort was created for one third-year student (S.N.E.) in August 2012 and for a second student (K.E.K.) in October 2012. Students would add a patient to their cohort if (1) they had participated in the patient’s care and (2) they were interested in following the patient over time. Students obtained verbal consent for longitudinal follow-up across health services from all patients prior to adding them to their cohort. (In accordance with Partners HealthCare policy, students are afforded access to protected patient health information if immediately involved in patient care and for educational purposes.)
The pilot students received e-mails from the VNT at the end of each week summarizing all upcoming scheduled appointments for their cohort patients at all Partners HealthCare–affiliated sites (see Figure 1 for sample e-mails). Dedicated e-mails were sent about individual patients if any of the following occurred: arrival to, admission to, or discharge from the emergency department, as well as admission to or discharge from any inpatient hospital service. Separate e-mails were sent at the beginning of the week to notify students of any last-minute additions and/or cancellations. Additionally, the tool was able to display patients currently admitted to Partners HealthCare–affiliated hospitals, as well as recent admissions or discharges. Students were able to access the Web-based tool at any time to view or modify their existing cohort.
The pilot students received approval to visit cohort patients from their current rotation’s supervising preceptors. This was achieved through an e-mail or in-person conversation explaining the pilot and offering a referral to faculty advisors (M.d.M., D.E.W.) for questions. The students’ current required rotations and related patient care duties took precedent, and students often visited cohort patients outside of rotation hours. Students generally spent one to two hours per appointment, emergency department visit, or procedure, and less than one hour per week, averaged across all PCE weeks, visiting their cohort patients.
The pilot required minimal resources. As the VNT had already been developed and implemented within the EMR, no additional software or programming was required. However, an information technology expert with investment in the project was critical. Further, faculty advisors from distinct departments who hold leadership roles within clinical clerkships (e.g., clerkship director) and student codevelopers were necessary. In our case, the faculty advisors provided support and refinement of the students’ project plans and structure for PCE integration. No additional financial resources were required. Time requirements for students varied from two to three hours per week for the three weeks prior to pilot implementation for concept development and approval and less than one hour per week during the pilot for programmatic refinements and troubleshooting, largely related to expansion. Institutional review board guidance was sought at the beginning of the pilot, but the effort was deemed as educational improvement using tools that were available to clinicians for patient care and learning, and thus was not in the purview of human subjects research.
LPE integration into the third-year clerkship curricula
Each pilot student added approximately 20 patients to their cohort throughout the eight-month pilot. The added patients were from all PCE clerkship rotations. Students were notified about all patient activity, though students only followed 3 to 5 patients consistently. Students reported that the choice to follow patients closely depended on the frequency of patient visits and the student’s availability during visits.
When students received weekly e-mail notifications, they planned to see patients of greatest interest if they did not have required clerkship-related duties at that time. Prior to attending an outpatient appointment, students would most often send an e-mail to the patient’s provider to explain the pilot and request that they be allowed to observe and participate.
Initial concern was raised among faculty that students would find it difficult to balance required clerkship-related duties with the desire to participate in cohort patients’ care. However, the pilot students did not report running into this difficulty; students most frequently visited patients in the emergency department during off hours and generally arranged outpatient appointments with advance notice. There is no indication from this pilot that students were dishonest with respect to LPE pursuit; for example, they did not use time away from a required clerkship that had been approved for the purposes of this pilot for personal aims. Further, several unsolicited comments were received from residents and attending faculty members who felt that the project was innovative and added to the students’ knowledge.
If patients were admitted to hospitals or seen in clinics outside of the Partners HealthCare system, we could not track this activity. In these situations, students would attempt to contact the treating physicians from other systems, which was often a time-consuming endeavor that necessitated additional approvals.
Pilot student assessment and integration of learning
The pilot students felt that the VNT made it significantly easier to follow patients over time and across clinical settings. Their appreciation of chronic illness care developed as they met patients over multiple clinical encounters. In their case reports, written as part of their voluntary self-evaluations, the pilot students indicated gaining a greater understanding of the integrated nature of patient care, seeing patients travel through multiple specialty clinics and practice settings (Box 1). However, as a limited pilot, participation was not part of any clerkship evaluation. Allotting flexible time earmarked for visiting cohort patients or other individualized educational endeavors within required clerkships may have allowed students to attend a greater number of longitudinal encounters without extending their clinical hours.
Box 1 Case Report of a Longitudinal Patient Experience Facilitated by the Use of a Visit Notification Tool, Massachusetts General Hospital, August 2012 to May 2013a Cited Here...
Mr. S was admitted to the surgical intensive care unit after surviving a motorcycle crash. Following immediate postoperative stabilization, he was transferred to the neurosurgical service where he remained for over three weeks. After two operative procedures and appropriate recovery in the hospital, Mr. S was transferred to an acute inpatient rehabilitation facility where he continued to improve. He began recovery from full left-sided paralysis and regained the ability to speak and swallow. Four months after his initial hospitalization, Mr. S was readmitted for surgical repair of a brachial plexus injury he had sustained at the time of his accident. The patient was seen in-clinic by neurosurgery and later by neurology and psychiatry.
The third-year medical student involved in Mr. S’s care first met him in the intensive care unit and again on the neurosurgical service. Because of his prolonged hospitalization, the student got to know him and his family well. She continued to visit him in the hospital when her time on the surgical consult service ended and, with his permission, added him to her patient cohort when he left the hospital. The student was notified via e-mail of the time and location of Mr. S’s outpatient appointments and was able to arrange to see him in the neurology and neurosurgery clinic. At his outpatient appointments, Mr. S would talk about the gains he had made in his rehabilitation program, often showing videos of his progress. The notification tool made it possible for the student to attend the operative repair of the patient’s brachial plexus injury, and when he was once admitted to the emergency department because of concerns of infection, the student was immediately notified. The experience of working with Mr. S over time and across services allowed the student to participate in the patient’s course of recovery and to recognize the financial and caregiver burdens of those recovering from long-term illnesses.
aText adapted from a pilot student’s case report, written as part of the student’s voluntary self-evaluation. Patient name and identifying details have been changed.
Institutional expansion of the LPE VNT
When we (S.N.E., K.E.K.) informally polled all MGH third-year students in February 2013, more than 90% (n = 26) of respondents noted they were attempting to follow patients longitudinally without any formal mechanism, and 100% (n = 28) noted they would be interested in using a tool that supports this experience. Third-year students at the MGH had “picked up” patients across all required clerkships, with slight weighting toward clerkships such as internal medicine and primary care (Figure 2). Considering the universal demand for a tool such as the LPE VNT, it was made available to all MGH third-year students during the final three months of their clinical year. From March–May 2013, 19 additional accounts were created for other third-year students; thus, roughly half of MGH third-year students were using the LPE VNT. As expected, gaining access to the tool late in the year limited its utility in terms of tracking patients longitudinally. For this reason, at the beginning of the next clinical year the tool was offered to all MGH third-year students.
Extended expansion of the LPE
Although the students who piloted the VNT had a positive experience and felt that the tool added to their clinical education, there is limited evidence on the objective value of LPE, particularly of hybrid curricular models that embed LPE in traditional block clerkships. Developing a body of evidence that supports the value of LPE could be instrumental in extending the potential benefits of longitudinal, integrated patient experiences to a majority of medical students.
Challenges in establishing LPE curricula
The generalizability of the VNT in terms of both software and EMR compatibility and curricular integration has not yet been validated. With respect to software and EMR compatibility, most EMRs should be amenable to the approach used here. Census and scheduling data are readily available at any hospital, and matching census data with the patient or student data in the registry is a straightforward task. In small numbers, the matching can be done by hand, and anyone with moderate knowledge of Microsoft Access or Excel would be capable of automating the imported data. Contractors or savvy information systems personnel could construct more elaborate systems (e.g., systems that provide more choice in types of notifications, systems that integrate with secure scheduling software, etc.). Furthermore, several clinician client programs are available for purchase, including the Morrisey Concurrent Care Manager application, which the Partners HealthCare Integrated Care Management Program uses to follow large groups of patients in disease-based cohorts. Similar programs have been used to track clinical trial participants and to serve as safety or adverse events reporting systems.
Curricular integration is more difficult. Most medical schools employ a clerkship-based model centered on immersion experiences in “core” specialties. The VNT crosses the artificial patient care boundaries in these models and contributes to both care continuity and collaboration in patient and student experiences. However, VNT implementation requires the support and flexibility of faculty and staff. To derive the full benefit from the tool, collaboration is required among faculty who may be unaccustomed or even unwilling to allow students to step away from nonrequired clinical duties for the purpose of attending LPE appointments. Sensitization to the model and buy-in regarding the importance of LPE are critical at all levels.
Finally, ethical issues including consent and the expectations of non-MD providers have been raised. Though no formal informed consent was signed for this pilot, each student explained the structure of the pilot and received verbal consent from a patient before adding the patient to her cohort. The faculty at our institution, both the project and clerkship faculty, felt that it was important for students to be granted access to patient information for patients that they were following longitudinally as an extension of their existing access to medical records for clinical care and educational purposes. Each student also clarified with each patient that she was a medical student, and not an independent care provider, and reemphasized her role upon repeat encounters. All encounters where students observed patient appointments or procedures were conducted and supervised by a physician faculty member.
The VNT outlined above empowers student health care providers with automatic notification of patients’ hospital admissions, emergency department visits, and upcoming appointments. The additional clinical experiences facilitated by the VNT allow students to experience the process of diagnosing and treating new or chronic conditions, particularly those in the students’ areas of interests. Students who use the VNT are encouraged to contact care providers, which supports the development of the professional and interpersonal communication skills necessary to succeed in contemporary clinical medicine.
Perhaps most uniquely, the ability to see patients in a variety of inpatient and outpatient settings across specialty services locates the patient experience within the health care system and allows for systems-based education to be applied to complex health systems. Extending the contact that students have with patients and their families allows students to begin to develop a critical sense of ownership over their patients’ outcomes and well-being, a foundation of professional competency.
Notification tools like the one outlined above may enhance a hospital’s existing EMR and facilitate longitudinal educational goals across all clinical clerkship models. In addition, such tools allow and encourage medical students to take on a more central, meaningful, level-appropriate role in the care of their patients, and offer more opportunities for learning contemporary medicine to support and improve health systems.
Acknowledgments: Sincere thanks to Dr. Stephen Calderwood, Dr. Alberto Puig, Katie O’Brien, and all of the faculty and staff of the Massachusetts General Hospital and Harvard Medical School Principal Clinical Experience for their support of this work. Thanks also to Benjamin Robbins for providing the informal connection to the Massachusetts General Hospital technology office that made this project possible.