A Visual Arts Activity to Support Residents as They Care for “Difficult Patients”

Longitudinal relationships with patients offer opportunities to provide high-quality care and can be a source of job satisfaction for health care professionals.¹  There are certain patients, however—often described as “difficult” or “heartsink” patients—who can be more challenging for their physicians and other care team members. These relationships can have the opposite effect on job satisfaction by challenging us, making us doubt ourselves, and leading us to thoughts, feelings, or actions that are discordant with our personal and professional values.^2-5  It is important, therefore, that medical training include strategies for working with “difficult” patients.

Teaching empathy skills is one effective strategy. Asking students to imagine the context in which their patients live has been shown to improve care.⁶  Empathy has also been a powerful tool when trying to overcome the confluence of race and the “difficult” label when caring for people with sickle cell disease. Imagining a patient’s reaction to living with extreme pain can help physicians understand why they might behave in ways that can be challenging.⁷  A focus on increasing empathy for difficult patients and/or caregivers helped pediatric residents interpret problematic behavior as reasonable reactions to intense emotions.⁸  Of note, empathy training in residency programs occurs during a time when empathy (and resident well-being) tends to decline.⁹  Given the probable bidirectional relationship between physicians perceiving their patients to be difficult and their own well-being, finding ways to build self-efficacy in caring for challenging patients may also reduce physician distress.¹⁰ 

There are several communication tools that can help residents build capacity for caring for challenging patients amid the pressures of training. Other-oriented perspective-taking is an intentional process during which one person suspends their own perspective in order to imagine a situation from another’s point of view.¹¹  This is especially important when the clinician perceives the other person as difficult. Their own emotional reaction can create a barrier, hindering connection and care. A related concept, “exquisite empathy,” is a skilled form of clinical empathy that requires the health care professional to lean in with tenderness and clear boundaries to the perspective of the patient.⁴  It allows health care professionals to be present and attuned while not confusing the patient’s experience with their own.⁴ 

Other-oriented perspective-taking requires this ability to shift perspectives. Therefore, in designing this session, we relied on the Empathy to Compassion model as a conceptual framework.¹²  The model explains how we move from reflexive mirroring to other-oriented perspective-taking before we can respond with sympathy or compassion. The described activity uses art to help learners understand the difference between sympathy, in which the focus of concern is one’s own feelings about the other’s situation, and compassion, in which the other’s perspective and needs are the focus.¹²  It takes advantage of parallels between the art viewing process, and the concept of other-oriented perspective-taking, in which one suspends their own perspective as they seek to understand another’s point of view, whether that be the artist or a figure in the work of art. This article discusses program development, feasibility, results from initial participants, and initial evidence that this approach is adaptable to other settings and groups of medical learners.

All second-year residents were required to participate in this learning session from 2017 to 2022 (N=142). Those who were on vacation, post-call, or participating in rotations without available coverage were excused. The sessions were conducted in multiple settings with different audiences, including (primarily) in a Midwest university’s art museum as part of a longitudinal curriculum for all internal medicine residents.¹³  Additionally, the session was adapted and conducted via Zoom with 12 international multidisciplinary participants as part of an annual conference.¹⁴  Instructors were faculty with expertise and training in using arts-based methods, and group facilitation.

The 1.5-hour museum-based resident teaching session was held during the workday. The session began with Visual Thinking Strategies followed by an improvisational theater game to encourage open and curious listening.^15,16  This was followed by a sequence of 4 activities during which participants: (1) recalled and reflected on a patient they considered difficult; (2) chose a piece of art they thought might be meaningful to the imagined person; (3) took turns sharing why the particular piece of art was chosen for the challenging individual; and (4) engaged in discussion to identify themes and perspective shifts resulting from the activities. In 2020, the session was carried out virtually with a group of participants at an international conference. In this setting, instructors replaced the Visual Thinking Strategies and improvisational theater game with a short introduction to the virtual meeting experience, an ice-breaker activity, and brief didactic on the concept of “difficult” or “heartsink” patients. Then, participants engaged in the same sequence of 4 activities. They selected the art objects from a collection of images from our university’s art museum. Detailed instructions for each step in the in-person and virtual sessions are outlined in Table 1.

After activity 1 and again after activity 4, participants rated their level of discomfort as they anticipated their next visit or conversation with their challenging patient/person. Discomfort was reported using a scale of 1 to 10 (1=no discomfort and 10=worst imaginable discomfort). During the first session, participants responded to a prompt regarding their primary emotion when anticipating an interaction with their difficult person. At the end of the session, participants wrote a comment on if/how their perspective had shifted. The pre-post survey is available as online supplementary data.

Pre- and post-session ratings of anticipatory discomfort were analyzed in Microsoft Excel using 2-tailed paired t tests. Three authors (S.L.F, M.A.Q., A.B.Z.) subjected the free-text responses of the participants’ primary emotion when anticipating an interaction with their difficult person to content analysis. All responses are included in Table 2, organized across the Empathy to Compassion model.¹² 

The university’s Health Sciences Institutional Review Board reviewed this project and determined that it qualified for exemption under category 45 CFR 46.101(b)(1): research conducted in established or commonly accepted educational settings and involving normal educational practices.

Sixty-five second-year internal medicine residents out of 142 possible attended in-person sessions at the university’s art museum. Twelve participants at the 2020 International Conference on Communication in Healthcare attended the virtual session.¹⁴  In all, 75 of 77 total participants completed pre-post discomfort ratings (response rate 97.4%). As shown in Figure 1, participants on average reported a reduction in anticipated discomfort after completion of the program. Three individuals (4%) reported increased anticipatory discomfort. Learner engagement during sessions was high; during group discussions, residents connected around the care of shared patients and supported their colleagues about common challenges. The comments written during the first session regarding anticipatory emotions included an array of negative emotions (Figure 2). At the first session’s completion, written comments from residents aligned with several themes within the Empathy to Compassion model (Table 2).¹² 

The cost of this intervention was minimal; in-person museum-based sessions were held at our university’s art museum, which does not charge admission, and it was delivered by 1 to 2 faculty members for 10 to 15 participants per session. Materials included pencils and paper to record pre- and post-session ratings and written reflections. Leaders of this exercise require skills in small group facilitation but not expertise in art history or analysis.

Based on participant responses, this activity reduced the distress felt when anticipating an encounter with a challenging person. While a variety of visual arts–focused activities, particularly Visual Thinking Strategies, have shown promise in medical education, this novel activity leverages art to purposefully invoke other-oriented perspective-taking and reduce distress.¹⁵  We were surprised that the virtual adaptation was also effective, because the virtual activity excluded some of the other museum-based activities described in Table 1.

This report augments the evidence that humanities-informed educational techniques, whether visual arts or reading fiction or poetry, provide us opportunities to engage in other-oriented perspective-taking. Narrative techniques can help to mitigate bias, for instance.^17,18  Importantly, these effects can be durable. While we did not formally evaluate durability, a year after engaging in this session, a third-year resident posted to a social media platform an image of the art object they chose for their patient (Figure 3)¹⁹  along with the following quote:

This activity can be adapted to other settings and instructors. One co-author, after participating in a workshop on this activity at a national meeting, was able to adapt it successfully for a group of medical students at her own institution, suggesting that it is generalizable to other instructors and learner groups. If one doesn’t have access to an affordable local art museum, art pieces can be viewed on display at a hospital or purchased as a set of art postcards laid out in a classroom, for instance. Even using a variety of art pieces viewed in PowerPoint was an effective format for teaching this session.

In this activity, participants are first asked to become mindfully aware of their own emotional response to a challenging individual, and then asked to focus on searching for something they imagine would have meaning for that individual. This approach builds self-awareness and then induces a state of curiosity about the other person. The act of choosing a gift for another invites feelings of tenderness and love. These elements—other-focus, tenderness, and love—create an experience of compassion. Compassion cultivation techniques, such as loving-kindness meditation, have mental health benefits, reducing distress and increasing altruistic behaviors.^20-22  This may explain the reduction of anticipatory discomfort we found in participants at the end of sessions.

We acknowledge limitations of our evaluation. While our assessment focused on the immediate experience and post-activity shifts in “discomfort,” we did not assess changes in clinical performance after this session, such as how the activity impacted learners’ interactions with patients in the clinical setting. Additionally, the activity was initially embedded in an ongoing course that not all residency programs conduct. While we have described how this activity works well as a standalone session, the lack of a curricular space for implementation could be a barrier to implementation in other programs.

Future research should include directly measuring the impact of this intervention on clinician distress, behaviors, and competence in working with “difficult patients.” Helping clinicians develop skills to effectively navigate more challenging relationships with patients can help them maintain well-being and reduce burnout. Since evidence suggests that physicians with lower mood states rate a higher proportion of their patients as difficult,¹⁰  there may be a “vicious cycle” in which exposure to the distress associated with challenging relationships can lead to lower mood. This, in turn, primes the physician to see other encounters as difficult. Finding ways to break that cycle by leveraging empathy and compassion may provide relief.¹³ 

This article describes a simple, feasible activity to build empathy skills and reduce distress when anticipating challenging clinical interactions.

The authors would like to acknowledge Linda Baier Manwell, MS, for her writing and editorial contributions to this paper, and Christina Hughey, MD, for sharing her reflection on the activity.

The online supplementary data contains the surveys used in the study.

Funding: The authors report no external funding source for this study.

Conflict of interest: The authors declare they have no competing interests.

An early version of this activity was presented as a workshop at the Alliance for Academic Internal Medicine meeting, March 19-22, 2017, Baltimore, Maryland, USA, and a later version was presented as a virtual workshop at the International Conference on Communication in Healthcare, September 9-11, 2020.