Incorporating the human touch: piloting a curriculum for patient-centered electronic health record use

ABSTRACT Background: Integrating electronic health records (EHRs) into clinical care can prevent physicians from focusing on patients. Despite rapid EHR adoption, few curricula teach communication skills and best practices for patient-centered EHR use. Objective: We piloted a ‘Patient-centered EHR use’ curriculum, consisting of a lecture and group-observed structured clinical examination (GOSCE) for second-year students (MS2s). Design: During the lecture, students watched a trigger tape video, engaged in a reflective observation exercise, and learned best practices. During the GOSCE, one of four MS2s interacted with a standardized patient (SP) while using the EHR. Third-year students (MS3s) received no formal training and served as a historical control group by completing the same OSCE individually. All students completed post-GOSCE/OSCE surveys. The SP evaluated GOSCE/OSCE performance. Results: In 2013, 89 MS2s participated in the workshop and GOSCEs during their required Clinical Skills course and 96 MS3s participated in individual OSCEs during their end of year multi-station formative GOSCE exercise. Eighty MS2s (90%) and 88 MS3s (92%) post-GOSCE/OSCE surveys were analyzed. Compared to MS3s, significantly more MS2s rated their knowledge (19% vs 55%) and training (14% vs 39%) as good (≥4/5 point scale, P < .001 for both). Most learners (85% MS2s and 70% MS3s) thought training should be required for all students. SP ratings on GOSCE/OSCE performance was higher for the 20 MS2s compared to the 88 MS3 controls (73.5 [SD = 4.5] vs 58.1 [SD = 13.1] on 80 point scale, P < .001). Conclusions: A short workshop and GOSCE were effective in teaching patient-centered EHR use. This curriculum is now a permanent part of our Clinical Skills course. Clerkship students who did not receive our curriculum may have been exposed to negative role-modeling on the wards. To address this, training residents and faculty on patient-centered EHR use skills should be considered. Abbreviations: EHR: Electronic health record; EHR: Electronic health record; SP: Standardized patient

With increasing EHR adoption, physicians must balance clinical efficiency with maintaining meaningful interactions with patients [7][8][9][10][11]. In 2012, the Alliance for Clinical Education recommended medical schools develop clear sets of EHR-competencies and formally train students on these skills to prepare them for clinical practice [12]. The American Medical Association also recommends teaching physicians EHR-related communication skills, and recent editorials have called for curricula to teach these skills [13][14][15][16].
Despite these mandates, few trainees are formally taught how to use the EHR in a patient-centered way, and assessment methods are lacking. As a consequence, trainees rely on experiential learning and the hidden curriculum during clinical rotations to develop EHR-related communication skills, which may expose them to negative role modeling.
To address this gap, we piloted a patient-centered EHR use workshop and group-observed structured clinical examination (GOSCE) targeting second-year medical students (MS2s). Objectives were to: (1) highlight benefits of EHR use in clinical care; (2) identify barriers to patientcentered EHR communication; (3) introduce best practices; and (4) allow for skills practice and feedback.
Study aims were to: (1) compare GOSCE performance of MS2s, who received the curriculum, to individual OSCE performance of third years (MS3s), who received no formal training; and (2) compare MS2 and MS3 post-GOSCE/OSCE survey scores assessing knowledge, attitude, and skills.
Ethical approval was granted by the University of Chicago IRB.

Methods
Investigators reviewed relevant literature to develop and implement the 'Patient-centered EHR use' lecture and GOSCE for MS2s at the Pritkzer School of Medicine in February 2013. Following the GOSCEs, written surveys and consent forms were distributed by non-faculty staff to all students. Participation had no impact on performance. Because all lectures are recorded and available online, lecture attendance is not required or tracked. However, we used an 'intention to educate' analysis assuming all MS2s received the lecture, either in person or remotely. This approach underestimates the true effect of actual participation in the curriculum. The University of Chicago Institutional Review Board approved this protocol.

Curriculum design
The one-hour MS2 lecture was integrated into the Clinical Skills course. This was modeled on Kolb's experiential learning theory to engage students in experiencing, reflecting, thinking, and acting to reinforce knowledge and skill acquisition [17] (Figure 1). Students participated in a 'concrete experience' by watching a trigger-video illustrating poor EHR communication skills [18]. They engaged in 'reflective observation' during a structured video-debrief discussion. The lecture provided 'abstract conceptualization' of patient-centered communication benefits and best practices. Lastly, students participated in the GOSCE for 'active experimentation' and feedback.

Lecture
Students were taught the 'HUMAN LEVEL' mnemonic (Table 1) summarizing evidence-based best practices for patient-centered EHR use, such as 'Using the "triangle of trust"' to position the screen where patient and provider can both see it. The lecture concluded with a second video depicting best practices [19]. Details of our full curriculum are published and accessible for reference [18,20].

GOSCE/OSCE development
To provide students with a high-fidelity outpatient scenario, investigators worked with the University of Chicago's EHR trainers to design a mock-chart in the training environment with five years of clinic notes and studies. During the GOSCE/OSCE, students logged into the mock chart, addressed the standardized patient's (SP) chief concern, reviewed the EHR for relevant clinical information (i.e., labs, notes), and provided appropriate counseling. The SP received two hours of communication and evaluation training on how to use the SP evaluation tool, which is described below, to provide feedback on EHR

Concrete Experience
Students watch a 'trigger tape' video on EHR-related communication barriers

Reflective Observation
Students reflect on the video and participate in a structured debrief

Active Experimentation
Students participate in a Group OSCE on patient-centered EHR use for skills practice

Abstract Conceptualization
Students are introduced to key concepts and best practices during the patient-centered EHR use lecture Figure 1. Applying Kolb's experiential learning cycle to teach patient-centered electronic health record (EHR) use. We based our curriculum on Kolb's experiential learning cycle. 17 Second-year students (MS2s) participated in a 'concrete experience' by watching a 'trigger tape' video and engaged in 'reflective observation' during a structured video-debrief discussion. They attended a lecture introducing key concepts for 'abstract conceptualization' and participated in a Group Observed Structured Clinical Examination (GOSCE) for 'active experimentation.' utilization. All MS2s and MS3s were evaluated by the same SP, who was blinded to the student's training status.

MS2 group OSCE
Due to resource constraints in the preclinical curriculum, 89 MS2s participated in 15-minute 4-5 person group-OSCEs within one week of the lecture. One student interacted with the SP, remaining students observed, and a faculty member facilitated SP, peer and faculty feedback at the end of the case. The SP completed a written evaluation of the student's performance using the SP evaluation tool.

MS3 individual OSCE
In May 2013, 96 MS3s, who received no formal training on patient-centered EHR use, participated in the same OSCE as part of an end-of-year formative clinical OSCE experience. All MS3s participate in this multi-station OSCE exercise and receive feedback on their clinical skills with no formal grade assignment. All MS3s interacted with the SP individually and were given feedback using the SP evaluation tool.

Evaluation of GOSCE/OSCE performance: SP evaluation tool
To assess student GOSCE/OSCE performance, we developed an SP evaluation tool based on the HUMAN LEVEL mnemonic highlighting best practices. Likert-style questions (15 questions, 1-5 scale; one question scored as no = 0 or yes = 5) assessed EHR-related communication skills with a score range of 15-80 points; higher scores indicated better performance.

Surveys
All students completed paper-based post-GOSCE/ OSCE surveys consisting of 20 Likert-style questions pertaining to knowledge, skill, confidence, and training on patient-centered EHR use. Likert responses at the high end (≥4 on a 5 point scale) were grouped to dichotomize data and report outcomes.

Statistical analysis
Scores on the SP evaluation tool were summarized as means with standard deviations (SD), and the average scores compared between groups via two-sample t tests. Comparison of the score distribution between groups was assessed using Kolmogorov-Smirnov tests. Post-GOSCE/OSCE survey responses were tabulated and presented as frequency counts and percentages and compared between MS2 and MS3 groups using chi-square tests. Comparison of average scores between MS2s who directly interacted with the SP and MS2s who had no direct interaction was performed via two-sample t tests. Statistical analyses were conducted with Stata Version 13 (StataCorp, College Station, Texas).

Results
All 89    . In addition, beyond comparison of means, general differences in the distribution of scores between MS2s and MS3s were assessed using Kolmogorov-Smirnov tests and significant differences were consistently found. Results are summarized in Table 2.
Survey results of MS2s who directly interacted with the SP during the GOSCE (n = 20) showed no significant difference in self-reported skill (2.

Discussion
To our knowledge, this is the first study to show that a formal lecture and GOSCE are effective in teaching medical students how to use the EHR to promote patient-centered communication. We found that MS2s who received a brief one-hour lecture performed better on a GOSCE than more clinically experienced MS3s who received no formal training. We chose to deliver the curriculum to MS2s to allow them to incorporate GOSCE feedback to improve communication with real patients during their clerkships. Due to the success of the pilot, our curriculum has been formally incorporated into the MS2 Clinical Skills Course. Importantly, our findings suggest that the current method of relying on experiential learning during clerkships is not sufficient to master these patientcentered EHR use skills.
It is worth considering why MS2 confidence and self-reported skill levels were not higher after the curriculum, despite higher ratings by SPs. First, MS2s are still novices at interacting with patients, and this exercise may be inherently complex as it requires effective use of the EHR while integrating patient counseling. Students may require repeated interactions and re-evaluation to achieve a higher level of confidence. It is also possible that MS2s forgot the skills after the workshop or did not come to the lecture. Historically, approximately half the class watches recordings of lectures in lieu of attending class, but this is not tracked due to privacy reasons. Overall, our lecture attendance rate is consistent with those reported in the literature [21]. While we conducted an 'intention to educate' analysis assuming all students received the curriculum, it is possible that some students did not watch the recorded lecture or that watching a video was less effective than in-person participation. Further studies should investigate if these learning strategies impact GOSCE performance.
Our curriculum has potential uses outside of medical student training. Currently, most physicians are required to attend mandatory EHR training sessions that focus on EHR functionality but do not address patient-doctor communication. Given the focus on patient experience in healthcare, incorporating best practices of patient-centered EHR use into mandatory EHR training can improve the quality of patient-doctor communication. Presently, we have expanded our patient-centered her-use curriculum to internal medicine and pediatrics residents and faculty; validation of our evaluation tools is underway. Lastly, we are exploring the use of direct observation in real clinic interactions to assess MS3, resident, and faculty's patient-centered EHR use.
Our study has limitations that open avenues for further research. MS2s participated in GOSCEs, which is common practice in undergraduate medical education due to limited time and financial resources. Future research should look at MS2 individual OSCE performance, as our study suggests active OSCE participation may promote higher rates of knowledge acquisition, perhaps due to the individual feedback obtained. Another limitation is the self-reported nature of responses. While student survey responses were in line with SP evaluations, validating SP evaluations and parallel evaluations of student GOSCE performance by faculty observers is needed. Future studies can also evaluate the impact of direct participation in the live-lecture versus watching the recording online. In spite of these limitations, our findings provide a real-world estimate of the effectiveness of this curriculum, which is valuable to other medical schools that do not require lecture attendance. Moreover, since we used an 'intention to educate' analysis, assuming all students received the lecture; our findings may under-estimate the true effects of receiving a formal curriculum.
In summary, we successfully implemented a new curriculum and GOSCE that led to increased knowledge, skills, and performance in patient-centered EHR communication. This innovation can be expanded to teach residents and faculty how to remain focused on the patient while managing the demands of the EHR. from the University of Chicago's EPIC EHR Training Center.
This paper was previously presented at the following meetings: Society of General Internal Medicine Midwest meeting 2013 and 2014, Society of General Internal

Disclosure statement
No potential conflict of interest was reported by the authors.