Background

Healthcare providers must develop personal connections to deliver optimal patient care, and some keys to forming these connections are compassion and empathy [1]. Compassion is recognizing emotional distress by others (or oneself) and the resulting desire to reduce suffering [2]. This is contrasted with empathy, or personally experiencing others’ feelings and emotions [3]. For providers, compassion and empathy contribute broadly to positive patient health outcomes and patient satisfaction [4,5,6], but repeatedly experiencing patient suffering is precarious for provider wellness. For medical students, empathy erosion notably peaks during clerkship in medical school with the heightened stress of transitioning to patient care [7,8,9]. Compassion fatigue, a type of burnout experienced by caregivers supporting suffering individuals [10], is also prevalent in medicine [11, 12]. Reduced provider compassion and empathy contribute to suboptimal patient health outcomes [13, 14] and to provider burnout (described below), which motivates academic healthcare centers to address provider wellness.

Wellness in medical education

Chronic stress is pervasive in medical occupations and associated with decreased compassion and empathy. Startling levels of depression, suicidal ideation, burnout, and alcohol abuse occur during student and resident training, and debilitating stress continues throughout physicians’ careers according to many US-centered studies [15,16,17,18]. Interventions that offer providers practical tools to manage stress during training and throughout their later careers could impede burnout [19]. Self-care activities such as resilience building and stress-management training have recently gained prominence [20], and reviews of medical-school interventions suggest that organization-sponsored programs are infrequent but very potentially effective in reducing burnout [21,22,23,24]. Self-care can also augment the development of compassion for self and others [25,26,27,28,29], thereby providing benefits to both healthcare providers and patients.

Mindfulness training is one approach to facilitate wellness. Mindfulness is a teachable aspect of awareness in which practitioners pay attention on purpose, in the moment, and without judgment [30]. This ability to refocus attention and address daily distractions or anxiety can positively influence cognitive appraisal and emotional reactions, alter and lower an individual’s perception of stress, and improve well-being [31,32,33,34]. Mindfulness has shown promise to address stress and support empathy with medical students across varied intervention structures and evaluations [35,36,37,38], and training can also improve physician-patient communication [39].

Despite the promise of interventions like mindfulness training [24], the systematic inclusion of wellness training in undergraduate medical education challenges many programs because of the reallocation of time from traditional content that would be required to adopt this training [40]. This prompts a need to identify and evaluate wellness programs that are both effective and efficient, and an intervention specifically aimed at improving both patient care and provider self-care is appealing because of competing curricular demands. A promising approach is compassion training, which can cultivate and improve compassion for both others and oneself through structured training [41,42,43].

Intervention conceptual framework and purpose

Here we report the results of a three-year pilot study of a brief elective course for medical students designed to foster compassion through developing mindfulness skills. Our educational intervention was modeled after the Stanford Center for Compassion and Altruism Research and Education’s (CCARE) Compassion Cultivation Training (CCT). This secular, eight-week educational course aims to strengthen the qualities of compassion, kindness, and well-being. CCT has wide-ranging success in alleviating negative emotional states, increasing compassion, and improving interpersonal skills for healthy adults [44,45,46]. A University of Louisville School of Medicine (ULSOM) faculty member (author JBK) with ten years of meditation practice completed CCT instructor training with CCARE in 2013. In 2015, ULSOM began offering a two-credit-hour elective following the CCT syllabus available through CCARE [47] with content adapted to medical students’ experiences, including compassion exercises specifically focusing on patient interactions where appropriate (e.g., exercises to help students understand that compassion is a process that can unfold in response to suffering, practicing compassionate thoughts and actions in response to patient interactions, focusing on making positive differences in the lives of patients, etc.). Weekly, two-hour sessions included pedagogical instruction, guided group meditation, mindfulness training, group discussion, listening and communication exercises, practical exercises related to weekly compassion themes, and 15–30 min of daily home meditation.

ULSOM was one of the first programs to implement and study CCT as an intervention for medical students specifically. Our research was built on a conceptual framework using social learning theory, in which behavior is modified through observation, modeling, and reflection on responses [48]. Social learning also places importance on the setting where learning takes place and the community of learners that influence the individual learner’s experience [49]. The CCT faculty instructor modeled compassion cultivation and mindfulness skills, including loving-kindness meditation. This meditation method can increase well-being, connectedness, and compassion through consciously wishing wellness to oneself and others, including for difficult relationships [50, 51]. By observing and mirroring the trained instructor, students learned in the group setting to modify their stress and compassionate behavior toward self and others.

The CCT elective objectives were to help students feel more connected to others and to promote wellness strategies early in their medical careers. We predicted that this brief educational intervention would improve students’ mindfulness traits and foster a sense of well-being and heightened compassion. Furthermore, we predicted the reward of positive interactions with others and/or reduced stress, coupled with the group reflection on these experiences, would provide internal and external reinforcement of self-care behaviors.

Methods

Study design

The full eight-week CCT course was given in 2015 and 2016, and an abbreviated five-week course was held in 2017. Preclinical second-year (M2) students were eligible to participate in the 2015 course while both M2 and fourth-year (M4) students were eligible to participate in 2016 and 2017. Study participants were recruited via email invitation to CCT-enrolled students.

A mixed-methods design evaluated the impact of the CCT elective on students’ mindfulness skills, stress management, and compassion. All students completed a pre/post-test of the Kentucky Inventory of Mindfulness Skills (KIMS) [52], which is a validated, 39-item self-assessment instrument with a Likert-type scale ranging from one (never or very rarely true) to five (almost always or always true). This inventory assessed four different mindfulness facets, including: observing, describing, acting with awareness, and accepting without judgment. The KIMS has been validated for internal consistency, retest reliability, and correlation with similar self-assessment tools [53].

Completing the pre/post-test KIMS was part of the elective curriculum as a self-assessment. However, submitting identified data for research was optional for students, so only matched pre/post data were included in the analyses. Students in the 2015 sessions completed paper-based KIMS for both pre/post-tests in class. Students in the 2016 and 2017 session completed a paper-based KIMS pre-test in class and the post-test electronically, after the final session was complete.

In addition to the KIMS, students were invited to complete a follow-up, open-ended questionnaire through an electronic survey platform in June 2016 (2015/2016 cohorts) and February 2017 (2017 cohort). Students were asked to provide subjective feedback to better understand their course experience, their sustained use of skills learned in the course, and their perceptions of utility, benefits, and transferrable skills from the course.

Data analysis

Individual pre- and post-test KIMS scores were calculated for each of the four mindfulness facets. Responses from 2015/2016 were combined for analysis as the cohorts experienced the same scheduling format. Scores from 2017 were analyzed separately since the course was held over fewer sessions. Two-tailed, paired t-tests were performed in SPSS (IBM SPSS Statistics for Macintosh, Version 24.0. Armonk, NY) to determine whether mean post-test scores were significantly greater than pre-test scores for each mindfulness facet. Completed pre/post-tests without identification for pairing were not included in the analysis.

De-identified qualitative response data from all three years were evaluated independently by two authors (SS, LAW) using a directed approach to content analysis to identify themes among the data [54]. After an initial review of responses, a priori codes were established by both coders for each question. Responses were consistent across cohorts, so qualitative data were analyzed in aggregate. Emergent codes were incorporated to re-analyze the dataset independently by each coder. No substantial differences emerged between coders. Categorized data were then assigned to and collapsed into themes across questions by the coders jointly, which resulted in a set of final themes for each topic.

Results

Participants

Forty-five students from M2 and M4 participated in the CCT elective over three years (Table 1). Twenty-five students completed the full eight-week course and twenty completed the abbreviated five-week course. Respondent sample sizes are noted and indicate that not all students answered all questions because of variation in surveying and the optional nature of the data collection. Demographic breakdowns are not reported to preserve student confidentiality.

Table 1 CCT enrollment and elective structure showing research participation in the mindfulness inventory and subjective/qualitative feedback
Full size table

In the full-length course, 44% of students (N = 11/25) completed identified pre- and post-test KIMS that could be matched for paired analysis. In the abbreviated cohort, 80% of students (N = 16/20) completed identified pre- and post-test KIMS. Thirty-eight students total provided qualitative feedback on course evaluations and/or the qualitative post-course survey.

Mindfulness inventory

For the full-length 2015/2016 cohorts, post-test mean scores for all KIMS mindfulness facets were higher than the pre-test values, with significant improvements in observing and accepting without judgment (Table 2). For the abbreviated 2017 cohort, no significant differences were seen between the pre- and post-test scores. Relatively large standard deviations across all KIMS subscales in both groups suggest that students started with variable baseline mindfulness levels, and among-student differences were generally maintained after the training.

Table 2 Mindfulness inventory comparing pre/post score means with a paired t-test analysis
Full size table

Student perceptions

Sources of stress

Students’ major sources of stress were reported to be academic/clinical responsibilities, interpersonal relationships, debt, and personal self-esteem/mental health struggles (N = 16). Students described “feeling inadequate” and that work and family “can be stressful on the best of days.” Students reported difficulty addressing daily stress prior to CCT: they previously addressed stress through physical activity and by taking breaks from academic work, but unhealthy coping strategies included substance abuse and avoidance. One student reflected that, “I tended to ignore [stress] until I became overwhelmed or until the stressor was removed and then I felt useless.” Another student reflected that, “Sometimes I’d have meltdowns.” Only one respondent reported working with a support network during times of stress before CCT, and only 25% of respondents (N = 4/16) had had any previous mindfulness experience. Of the students who had previous training, all but one had been provided this opportunity through a structured university program, suggesting that students were not receiving similar training elsewhere.

Applying CCT skills

Students overwhelming described positive experiences from the CCT elective (Table 3). Most reported using CCT skills daily or often after the course (N = 20/27, 74%), and no student reported “never” practicing CCT skills after the elective. Students continued meditating (N = 26/34, 76%), expanding compassion (N = 10/34, 29%), using mindfulness (N = 6/34, 18%), and breathing for stress management (N = 5/34, 15%). Students found that pairing the stress-mitigation, mindfulness, and compassion-building skills with deliberate practice was useful and effective. Many valued the ability to recognize and think through emotions they felt. They benefited from the group dynamic of CCT, the realization that other students have similar issues, and the safe space to discuss common struggles with stress management and patient care.

Table 3 Qualitative analysis of continued student skill use (N = 34), application (N = 29), and transfer of skills (N = 26)
Full size table

Likewise, students reported broad use of stress-management and compassion-building skills in academic, clinical, and personal settings (Table 3). Students used skills for personal stress management and reported that the course had affected their experiences by allowing them to evaluate sources of stress and take deliberate steps to focus on the present moment (be that studying or interacting with a patient). Several students reported positive applications of skills during patient interactions, such as: using mindfulness skills to refocus and be present with a patient, expanding compassion by consciously wanting patients to be relieved of an ailment, and increasing patience, listening, and empathy during difficult interactions.

Student motivation

Students initially enrolled in CCT to learn stress management skills and to prevent compassion fatigue so that burnout would not compromise their patient care (Table 4). They perceived these benefits and indicated that CCT also contributed to their professional identity formation through self-introspection about becoming a physician and caring for patients. One student described how CCT forced them “to reevaluate why [we] are in medical school,” while another gained “perspective on self-care and the importance of taking care of yourself while also taking care of patients.” One student simply described the course as “life changing.”

Table 4 Qualitative analysis of student enrollment motivation (N = 16) and integrating the course into required curriculum (N = 27)
Full size table

Students were unsure whether CCT training should be part of their required curriculum. Those who supported requiring this content suggested that the experience would be beneficial and low risk, and it would help students focus on the human aspect of medicine. Students who did not support requiring CCT suggested that students need to be open-minded to this type of training. They were concerned that this could create resentment, which along with larger class sizes, could weaken the group dynamics and thus effectiveness of CCT training.

Despite these reservations, 17 of 27 respondents (63%) felt that this intervention should be a required course for all medical students. When asked when the most beneficial time to hold the course would be, responses varied: fourteen (53%) suggested in preclinical years, ten (37%) suggested during clerkships, and three (10%) said anytime. Multiple respondents recommended CCT during preclinical years with follow-up sessions during clerkship rotations “once you’ve realized that patient care can be stressful and you still have [opportunities] to practice the techniques.”

Instructor importance

Finally, a central theme that emerged among respondents was that the instructor was critical for student buy-in and training success (Table 5). Students reported that the instructor’s enthusiasm and conscious effort to model CCT skills helped them develop this skill set. The scientific foundation of CCT methods presented by the instructor also initially validated the experience. Few students desired changes to the course. Those who had suggestions requested logistical changes, such afternoon sessions to accommodate clerkship rotations or requesting more comfortable spaces for meditation exercises.

Table 5 Qualitative analysis of student feedback regarding the faculty instructor (N = 20)
Full size table

Discussion

Our study demonstrates that a brief educational intervention can foster compassion and mindfulness skills in medical students participating in the elective. Students reported using CCT skills for personal well-being and stress management and for patient interactions. Students had overwhelmingly positive perceptions of the course and felt CCT was worth their time investment. This intervention ultimately addresses student wellness with threefold benefits: brevity, practical stress-management skills, and potential improvements to clinical care.

Modeling in medical education

Mindfulness is widely accepted as a teachable strategy to self-regulate attention [55,56,57,58,59,60,61,62]. In this study, the faculty instructor modeled meditation and mindfulness, and students applied these in their own lives as supported by social learning theory [48]. In clinical learning, role modeling is an essential conduit to learn humanistic aspects of care such as compassionate bedside manner and favorable physician-patient interactions [63]. If we hope to advance systematic efforts to decrease burnout and increase resilience, teaching faculty must purposefully model interpersonal and cognitive skills such as compassion and mindfulness. Social learning theory further proposes that imitation more likely occurs when the model is perceived as like oneself. Some student comments that the faculty instructor was “relatable” accordingly suggest that students benefited from a professional role model who was similar. This supports other studies of successful mindfulness instructors in which “embodiment, empowerment, non-reactivity and peer support” defined the instructor-student role ([64], p., 172). Without dedicated, professional role models, students may not view these qualities as legitimate, necessary clinical skills.

Wellness training

Our study aligns with previous research in mindfulness education showing that brief interventions have reported pre/post increases on mindfulness, with different interventions showing variable effects [38]. A mindfulness intervention looking specifically at the intervention’s brevity (4 weeks) of an adapted Mind Body Medicine (MBM) course found that students’ mindfulness increased and stress decreased [36], and thus short interventions can be effective. Another recent study that also looked at CCT with various healthcare providers found improvements in self-compassion and mindfulness scores [65], which aligns with the KIMS results and student feedback in our study. Their study did not find a reduction in interpersonal conflict or burnout with the instruments used, which contrasts the self-reported feedback from medical students in our study. Qualitative data may capture a more comprehensive understanding of how training effects participants. Other wellness programs for medical students have also used multiple-skill approaches for medical students, such as a combined reflective writing and MBM workshop [66]. This experience focused on building practical skills to address stress while using these strategies to better understand the source of stress and its effects on individual wellness and patient care, which is paralleled with CCT’s multiple-strategy approach to mindfulness, well-being, and compassion.

The CCT program is a valuable self-care training option because of its focus on expanding compassion. Previous studies using loving-kindness meditation with providers also found improvements in well-being and feelings of connection [43, 67]. In this study, students reported various ways that they expanded compassion with themselves, patients, and others. The specific improvements of the “observing” and “accepting without judgment” mindfulness facets could be linked to students’ reported applications of compassion. Students practiced patience during difficult interactions with family, friends, and relationships, and they deliberately focused on others and not just themselves and their troubles. The ability to withhold judgments about patients and to be aware of others’ thoughts and feelings may help improve patient interactions, and it could also help reduce some of the unidentified stress that students feel.

Patients want their providers to be compassionate [68], and patients who perceive their provider as person-focused also perceive their care to be higher quality and are more satisfied [69]. A lack self-compassion in healthcare providers may be linked to higher burnout and/or less compassion for others [27,28,29]. The erosion of empathy that students often feel when they get to the clinic can be addressed with compassion training in a way that is different than stress management alone by teaching students the importance of cultivating compassion skills with each patient interaction [70]. Furthermore, recent research examining multiple properties of empathy found that aspects of cognitive and affective empathy (aspects which, similar to compassion, comprise understanding others’ emotions and pain) actually increased as students progressed through medical school [71]. These outcomes highlight the complexity of reported medical student empathy decline [8], reveal an intervention opportunity for academic healthcare centers to develop provider compassion, and also reinforce the importance of assessing students with multiple instruments, such as in this study. Qualitative analysis is especially useful to gain insight into the complexity of compassion and medical students’ relationship with this construct.

Practical implications

Even as academic medical programs attempt to reduce systemic sources of stress for providers, it is unrealistic for all stressors to be alleviated from variable sources. Thus, the practical solution for providers is learning skills to cope with inevitable stress. The positive perceptions of CCT and its utility suggest that students who had opted into the elective found this type of training worthwhile, which supports other qualitative studies of stress management with medical students [72]. CCT students reported various benefits to deliberate practice, and these rewards prompted students to continue using the skills. The follow-up with the 2015 cohort occurred over one year later, and this longer-term (although not long-term) insight suggested that these perceptions were maintained. Students also discussed using skills they learned with patients, which emphasized clinical applications of CCT and potentially enhancing patient interactions.

Differences between KIMS results for the 2015/2016 and 2017 cohorts suggest that course structure could be crucial to developing mindfulness skills. The 2017 cohort experienced an accelerated course to accommodate scheduling constraints, which may have limited students’ ability to improve mindfulness skills over the elective duration, especially for students experiencing additional pressures and responsibilities as they progress through their training. However, positive qualitative feedback and perceptions of the course were consistent across cohorts despite differences in course duration and also sampling time (one year out versus immediately following the course). Furthermore, although KIMS scores did not change significantly across all mindfulness facets, individuals report deliberately practicing other skills in qualitative feedback (e.g., demonstrating acting with awareness by focusing on one patient at a time). For students who were uninterested in mindfulness, developing self-compassion and other self-care techniques may have also provided alternative benefits to CCT students. Thus, CCT may have meaningful impact that scales do not capture, reinforcing the importance of qualitative assessment.

Role modeling was an important component of CCT success. A practical consequence is that medical schools must ensure instructors are well trained, genuine, and relatable to medical students. Formal training to teach mindfulness and compassion provides instructors with a scientific foundation and ensures that instructors can translate their experiences to students. Representation from top faculty leadership in a program like CCT also develops these skills among campus leaders and demonstrates institutional support for well-being and compassion.

Students struggled with some of the same concerns as medical educators, such as whether wellness training should be integrated into required curriculum for all students [38]. Broad implementation of CCT for all medical students could be limited by several factors. Faculty interest and funds to train with CCARE for CCT may be limited, and instructors with the appropriate skillset and background may not be available to meet the number of students. Increasing the size of the class could reduce the intimacy and safe-space dynamics created by limiting enrollment size. Forcing students who are uninterested in mindfulness training could also lead to resentment, and thus negatively change the experience of all students in the course. Although some students expressed that they had not realized how helpful the course would be until they were in it, there was a serious concern that an influx of resistant students could adversely affect mindfulness training and compassion cultivation. However, undergraduate medical education likely provides a more practical, comprehensive setting for teaching these skills to all students versus more diverse and potentially stressful residencies, which perhaps requires medical educators to validate compassion training’s relevance and worth to students.

Limitations

This study was limited by its small sample size and short-term analysis, although these were inherent in the design to create a brief, intimate CCT experience. Not all students participated in the study, similar to other studies of compassion development in health professions with low response rates [73, 74]. This study was also limited by variation in data-collection approaches across cohorts that did not include comparison groups. Students’ perceptions of application and compassion were derived from self-report data without scale measures, and pre-clerkship students had limited patient interactions. Students self-selected into the elective and therefore may have been more likely to find benefit in this training. Alternatively, as students suggested, some students who would not opt in to this training could be immensely helped by it.

Future directions

We intend to expand CCT to a larger group of students, ideally through multiple course sections to maintain small enrollment size. Long-term follow-ups that include more formal burnout and compassion measures to improve robustness will suggest if skills from this brief intervention are retained, particularly in comparison to control groups who have not had similar wellness interventions. Comparison groups that receive similar amounts of interaction with an attending in a different setting could also help determine whether CCT itself or personalized attention from an attending is particularly effective. Because students do not receive training to expand compassion elsewhere, institutions must provide similar training opportunities to students and foster a climate of well-being. In their foundational report on burnout interventions, West et al. [24] charged the medical education community with understanding how individual interventions can be combined with organizational solutions, which have been shown to be more beneficial [75]. We are addressing this call by offering CCT and similar organization-directed programs to students, residents, and faculty physicians through Being Well, a comprehensive, institutional-led initiative at ULSOM to increase health, resilience, and compassion across the continuum.

Conclusions

CCT was well received by medical students who opted into this elective and subsequently described positive impacts on academic and clinical stress. During the elective, some students increased their mindfulness skills and many reported improved interpersonal and patient interactions. Incorporating a brief educational intervention like CCT to simultaneously improve self-care and expand compassion could help academic medical centers address competing challenges of supporting student wellness and patient care.