A Virtual Curriculum to Improve Patient Education Skills of Internal Medicine Residents

Abstract

Patient education is a crucial component of a physician’s responsibility, and effective patient education can improve patient health outcomes and satisfaction. However, there is currently no formalized training for residents to develop and practice these skills at our large, academic internal medicine residency program. We created a virtual, case-based, interactive session for all residents to practice patient education skills and receive real-time feedback. Residents were given three scenarios: heart failure (HF), breast cancer (BC), and chronic kidney disease (CKD), and role-played as the physician, patient, and observer. The intervention was studied with single-group, pre-post intervention surveys. The session was virtual due to restrictions related to the COVID-19 pandemic. Mean Likert scale scores were compared by paired Wilcoxon rank-sign tests. The sessions were attended by 177 residents; 95 completed both pre- and post-session surveys (53.6%). Participants felt significantly more comfortable teaching patients about HF, BC, and CKD pathophysiology and treatments after the session (HF: pre-median = 4, post-median = 4, p = 0.0032; BC: pre-median = 2, post-median = 4, p < 0.0001; CKD: pre-median = 3, post-median = 4, p = 0.0016). There was a trend towards a significant increase in belief that teaching patients about common diseases should be integrated into the residency curriculum, but this did not reach statistical significance (pre-median = 4, post-median: 5, p = 0.0548). A targeted session for residents to practice patient education improved resident comfort with explaining three common diseases. These data suggest that the session was effective in a virtual format, demonstrating feasibility to be adapted in the increasingly online realm of patient encounters.

1. Introduction

Patient education is a core component of a physician’s responsibility. Residency offers a unique opportunity to develop education skills; residents serve as teachers for other residents, medical students, and patients. Effective physician–patient communication can improve health outcomes, health literacy, and patient satisfaction [1,2,3,4]. Surveys have demonstrated that residents believe patient education training is an important part of residency curriculum [5,6]. Participating in interventions to develop patient education skills further improves attitudes towards the importance of such training, and participation can identify deficiencies and encourage improvements in skills [4,6,7]. Efforts have been made to develop patient education programs featuring residents as teachers. Such interventions can improve metrics on self-evaluation and have demonstrated high resident satisfaction [8].

Utilizing the six-step approach to curriculum development by Kern et al., we first identified a need for a formal patient education curriculum for internal medicine residents at our large, academic, US urban residency program [9]. We conducted an informal needs assessment of the learners through organic discussions with residents and faculty members, followed by a formal needs assessment with a pre-intervention survey. Our educational goals and objectives were that residents would have improved confidence in teaching heart failure, breast cancer, and chronic kidney disease pathophysiology and treatments to patients. Next, we discussed instructional strategies and chose targeted, specific content for residents to incorporate in patient interactions. Our educational methods featured interactive scenarios with each resident taking turns role-playing as the physician, patient, and observer. This method was selected to provide residents opportunities to practice delivering counseling, to experience the patient’s perspective of receiving education, and to witness the interaction while keeping in mind specific goals for feedback. We implemented the session as a required component of the residency curriculum to maximize participation. Finally, we obtained evaluations and feedback through a post-intervention survey. Conducting this session in a fully virtual format, with a focus on high-impact diseases, and the use of structured role-based feedback are all innovative features highlighted in our intervention.

Although multiple patient education curricula for residents have been studied, ours provides a unique, innovative approach with our virtual format [4,6,8]. During the COVID-19 pandemic and its aftermath, there has been a greatly increased demand for and utilization of telehealth services. Therefore, it is increasingly important for internal medicine residents to feel comfortable conducting virtual patient education, thus providing the rationale for conducting this study [10,11]. A survey by Wong et al. demonstrated a critical need for formal training in telehealth for internal medicine residents. Of 156 residents surveyed, 90% reported no prior experience with video visits, and 88% were not prepared to deliver care using either video or phone visits [11]. New tools are also needed to evaluate the successful integration of telehealth competency into medical education [12]. Previous studies have demonstrated that combining case-based discussion with direct observation can increase self-perceived competence in telemedicine skills, and we followed a similar structure [13]. Many valuable nonverbal methods of communication, including body language, are challenging to provide virtually. Hosting our patient education session virtually provided an opportunity for innovation in medical education, and for residents to practice adapting these crucial skills to the increasingly online world of patient encounters. Our specific aims were to first demonstrate the efficacy of improving resident comfort levels with teaching patients about three common internal medicine diagnoses, and secondly to evaluate if resident beliefs regarding the importance of participating in structured education surrounding patient education changed after attending the session.

2. Materials and Methods

2.1. Context

We developed a virtual, case-based, interactive session for residents to practice patient education skills and receive real-time feedback. At our large, academic, US urban internal medicine residency program, residents receive one half-day per month of dedicated outpatient curriculum, hosted in groups of 35 residents each. These sessions are mandatory for all residents, and are traditionally held in person, although they were held virtually during the COVID-19 pandemic (2021–2022). We held our 45 min session during this dedicated block to ensure maximal attendance.

2.2. Intervention

All sessions were held virtually. The session began with a brief overview of the goals and objectives, informative resources, and tips to improve patient education skills. For most of the session time, residents participated in role-playing skills modules. The scenarios were developed based on clinical guidelines available from professional organizations (American Heart Association, American Cancer Society, National Kidney Foundation), and expert consensus through discussion with practicing experts at our academic institution. Content validity of the patient education scenarios was assessed for relevance and accuracy through review by senior physicians at our academic institution. Residents were divided into groups of three and rotated through specific roles of patient, physician, and observer for each of three sessions on breast cancer, heart failure, and chronic kidney disease. These three conditions were chosen because they are very prevalent and commonly encountered by internal medicine physicians, and can be grave and life-limiting, but the complexity may not be intuitive to patients. A presenting situation was provided to each resident. In addition, the patient received a list of questions to ask their physician, the physician received some clinical background to fill in any knowledge gaps about the disease, and the observer received specific points to address when providing feedback. Each module consisted of 2 min to preview the scenario, 5 min for the interaction, and 3 min to debrief, to simulate the time constraints of a patient visit. The session concluded with a brief wrap-up session in which residents were invited to share their experiences.

2.3. Evaluation

All residents received electronic surveys to complete before and after the session, conducted on the proprietary software of our university. A unique participant identifier was used to match pre- and post-session surveys. The pre-session survey asked about prior patient education experiences during medical school and residency, comfort with teaching patients about breast cancer, heart failure, and chronic kidney disease pathophysiology and treatment, and whether targeted education regarding how to teach patients about common diseases should be an integrated part of the residency curriculum (Supplementary Material S1). The post-session survey again asked whether the session improved comfort with teaching patients about breast cancer, heart failure, and chronic kidney disease, whether targeted education should be integrated into the residency curriculum, if the session improved general comfort and the resident’s understanding of the pathophysiology and treatment, and free-text take-home points and feedback (Supplementary Material S2). All multiple-choice questions on both surveys were graded on a five-point Likert scale, with 1 being extremely uncomfortable or strongly disagree and 5 being extremely comfortable or strongly agree.

2.4. Data Analysis

Medians and interquartile ranges are reported, as the data were not normally distributed, as calculated by the Shapiro–Wilk test. Wilcoxon rank-sign tests were conducted for each paired pre- and post-test response, as the data were not normally distributed. We compared each resident’s pre- and post-session Likert score responses using the unique identifier created, with an alpha of 0.05. Statistical analyses were performed using GraphPad Prism 9.3.0 (GraphPad Software, San Diego, CA, USA, www.graphpad.com, accessed on 15 September 2025).

3. Results

A total of 182 residents in the residency program received information about the exercise, with 177 participating in the exercise. Of these, 95 residents completed both surveys (53.6%), 37 did not complete any surveys (20.9%), and 45 completed only one survey (25.4%) (Figure 1). Only data from residents who completed both surveys was included.

On average, in the pre-survey, residents reported “neither agree nor disagree” with the statement that they had targeted education regarding how to teach patients about common diseases in medical school (median 3/5, interquartile range (IQR) 2–4), and reported “somewhat disagree” with the statement that they had similar education in residency (2.46/5, 0.90, 2/5, 2–3) (

Table 1. Median, interquartile range (IQR), p-values, and Cliff’s delta for responses to pre- and post-session surveys. All questions were scored on a 5-point Likert scale (1 = extremely uncomfortable or strongly disagree, 5 = extremely comfortable or strongly agree). p-values are calculated by Wilcoxon rank-sum tests.

Question	Pre-Session Median	Pre-Session IQR (Q1–Q3)	Post-Session Median	Post-Session IQR (Q1–Q3)	p-Value	Cliff’s Delta
I have had targeted education regarding how to teach patients about common diseases in medical school.	3	2–4	Not asked	Not asked
I have had targeted education regarding how to teach patients about common diseases in residency.	2	2–3	Not asked	Not asked
The patient education session improved my comfort level in explaining pathophysiology and treatment of common diseases to patients.	Not asked	Not asked	4	3–4
Learning and practicing how to teach patients about common diseases improved my own understanding of the pathophysiology and treatment of these common diseases.	Not asked	Not asked	4	4–4
I feel comfortable teaching patients about breast cancer pathophysiology and treatments.	2	2–3	4	3–4	<0.0001	0.93
I feel comfortable teaching patients about heart failure pathophysiology and treatments.	4	4–4	4	4–4	0.0032	0.50
I feel comfortable teaching patients about chronic kidney disease pathophysiology and treatments.	3	3–4	4	3.5–4	0.0016	0.71
I believe targeted education regarding how to teach patients about common diseases should be an integrated part of the residency curriculum.	4	4–5	5	4–5	0.0548	0.52

). In the post-survey, residents reported “somewhat agree” with the statements that the patient education session improved comfort with explaining pathophysiology and treatment of common diseases to patients (4/5, 3–4) and learning and practicing how to teach patients about common diseases improved their own understanding of the pathophysiology and treatment of these common diseases (4/5, 4–4).

On the pre-survey, on average, residents responded with “somewhat uncomfortable” regarding teaching patients about breast cancer pathophysiology and treatment. In the post-survey, residents responded “somewhat comfortable” to the same statement, a statistically significant improvement (pre-session median 2/5; IQR 2–3; post-session median 4/5; IQR 3–4; p < 0.0001) (Table 1). Similarly, there was also a statistically significant improvement between the same two statements with regard to heart failure pathophysiology and treatment, with residents responding “somewhat comfortable” on the pre-session survey and “somewhat comfortable” on the post-session survey (4/5, 4–4, 4/5, 4–4, p = 0.0032) (Table 1). There was also a statistically significant improvement in comfort with teaching patients about chronic kidney disease pathophysiology and treatments, with an average of residents responding “neither comfortable nor uncomfortable” on the pre-session survey and post-session survey, compared to responding with “somewhat comfortable” on the post-session survey (3/5, 3–4, 4/5, 3.5–4, p = 0.0016) (Table 1). The Cliff’s delta effect size was largely positive for each comparison (Table 1).

Residents were also asked if they believe targeted education regarding how to teach patients about common diseases should be an integrated part of the residency curriculum. Residents responded “somewhat agree” initially, and “strongly agree” after the session, but this did not reach a statistically significant increase in extent of agreement between the pre- and post-session survey responses (4/5, 4–5, 5/5, 4–5, p = 0.0548) (Table 1). Common feedback received was that more time should be dedicated to these sessions, either through a longer session or multiple sessions during the residency curriculum.

4. Discussion

Our virtual patient education skills curriculum improved residents’ comfort with teaching patients about the pathophysiology and treatment of common diseases, and our data suggested a large positive effect size, corresponding to a meaningful improvement. While other studies have demonstrated the efficacy of patient education curricula in improving skills, our session was completely virtual, yet still effective at improving residents’ confidence with patient education of the three core diseases practiced in this module [6,7,8]. Additionally, participating in this module drew attention to the importance of teaching patient education skills during residency, as has been shown in other similar curricula [6].

Some virtual models have been investigated, one of which is the virtual patient model, a computer-based scenario of a clinical case. The virtual patient model is used rather than a traditional lecture. It has demonstrated higher learning experience levels and can be used to improve competency for surgery and psychiatry residents, suggesting that virtual encounters can be helpful for residents to augment in-person learning [14,15,16,17]. Our session differs from this model, studied by Close et al. and McKendy et al., in that residents participated in the scenarios as physicians, patients, and observers, rather than using an online patient, and demonstrated improvements in comfort when teaching patients [14,15]. Courteille et al. compared the efficacy of a virtual patient to that of a recorded lecture and assessed knowledge retention both immediately and after 2 months [16]. While our tool had residents role-play to better engage in the learning experience and practice giving effective feedback, future iterations of this intervention should evaluate long-term knowledge retention, as Courtielle et al. did [16]. Jones et al. and Kirkland et al. created similar telemedicine curricula involving a full patient encounter and studied it with pre- and post-session surveys, finding similar results to ours; however, we focused primarily on delivering patient education on challenging diagnoses [13,18]. As telemedicine and online patient encounters become more common, residency programs may consider targeted sessions on patient education in the virtual realm.

Several studies conducted during the COVID-19 pandemic investigated providing virtual training on patient education to residents in several subspecialties. The rapid transition to virtual patient care was followed by many attempts to educate trainees to conduct effective patient education virtually, a critically important need identified even by residents themselves [11]. Our study adds to this growing literature base, as virtual patient encounters continue to be common in the post-pandemic era. Wong et al. created a longitudinal objective structured clinical examination (OSCE) training for 56 internal medicine residents to practice telehealth in the primary care setting [19]. Additionally, Lawrence et al. utilized a design thinking approach of empathizing, defining, ideating, prototyping, and testing to design novel training tools for internal medicine residents during this time [20]. While our study involved several more residents than both of these interventions, the interventions of Wong et al. and Lawrence et al. had several unique features, including a longitudinal curriculum, evaluation of objective measures unique to telemedicine, such as adjusting the camera and e-messaging, incorporation of attending physicians, and a structured process to elicit feedback regarding the delivery of these sessions. These features provide opportunities to improve future iterations of our intervention. We add to this growing literature base of the unique experiences of several academic internal medicine residency programs adapting to the increasingly virtual nature of patient encounters and education.

There are several limitations to our study. First, not all residents responded to both surveys and were thus excluded, which is an inherent bias of the survey modality. Additionally, voluntary participation leads to bias in participant selection. In the future, providing a specific time for residents to complete the pre- and post-session surveys during the session can improve the response rate. Second, survey questions were asking for subjective improvement in skills rather than knowledge acquisition and were not pilot tested or statistically evaluated for reliability or validity. Additionally, we did not include demographic data in the survey; therefore, we were unable to perform subgroup analyses. This data was not collected primarily to preserve anonymity in the survey collection, but in future iterations, we would like to obtain information, particularly regarding postgraduate year and previous medical training and clinical experience that may influence comfort with patient education. Lastly, our evaluations relied on self-reported comfort levels without objective measures of skill acquisition. We hope to address this major limitation with the future implementation of the modules, as we hope to also measure objective skills improvement. Nonresponse bias is also an inherent limitation of the survey modality, with the response rate of 53.6% of residents completing both surveys, and those who felt more engaged may have been more likely to respond, inflating positive outcomes. Additionally, we did not assess whether this session resulted in improvement in real-life communication with patients. Longitudinal surveys or standardized patient encounters may help to evaluate if skills have been applied after participating in the modules. Lastly, there was no control group; future iterations of this curriculum can compare it to non-case-based sessions, or to in-person sessions, to evaluate the non-inferiority of virtual interventions.

Next steps include evaluating preference for online curricula, establishing patient education skills as a core part of our residency curriculum, and creating evaluation tools to demonstrate objective knowledge acquisition in addition to subjective improvement. First, while residents reported an improvement in skills through this workshop, demonstrating the feasibility of this virtual session, having half of the residents participate in the modules virtually and half in the same modules conducted in-person could establish non-inferiority of the virtual experiences and evaluate preferences of residents for virtual vs. in-person learning. We also hope to establish a patient education skills curriculum as a standardized educational session for all residents, since the data from this study support its efficacy. Common feedback was that there was insufficient time dedicated to this exercise, and some residents suggested hosting multiple sessions integrated throughout the curriculum. The current outpatient curriculum is divided by subspecialty, so having residents practice the heart failure case during the cardiology block, breast cancer during the oncology block, and chronic kidney disease during the nephrology block, along with other similar modules, can help with longitudinal learning and reinforce the importance of regular practice. Additionally, we plan to refine our evaluation tools to investigate objective knowledge acquisition. Some examples include asking residents to identify medical jargon to avoid or selecting sentences that are at an appropriate reading level for patients. Finally, evaluating data from multiple iterations can help elucidate trends by postgraduate year since residents earlier in training may benefit more from this session compared to residents nearing graduation. While we intend for the intervention to be applied in other specialties, such as family medicine or pediatrics, and other institutions, we understand that the limitations we have discussed may limit the scalability at this time; therefore, a major goal of our future iterations of this curriculum and its evaluation is ensuring that it is easily adoptable by other institutions and adaptable to several specialties. Our long-term goal with this effort is to provide a structured patient education curriculum to all residents during their postgraduate training, since patient education is a foundational component of medical practice, and for this intervention to be adopted by other residency programs. As medical education moves towards teaching communication and other soft skills, in addition to the traditional curriculum, structured efforts for residents to practice delivering patient education and receive real-time feedback can provide safe spaces for physicians to practice these skills with their peers.

5. Conclusions

We created, implemented, and evaluated a virtual, interactive, patient education skills module for residents at a large, academic, internal medicine residency program. After the session, we evaluated the residents with surveys, and the residents reported an improvement in their perceived comfort teaching patients the pathophysiology and treatments of three common diseases. We demonstrated the feasibility and efficacy of a virtual skills module that can easily be integrated into the residency core curriculum.