Having the “Headspace” for Compassion Toward Self and Others: A Qualitative Study of Medical Students’ Views and Experiences

ABSTRACT Phenomenon: Debate about compassion exhibited by healthcare professionals has escalated, following a perceived decline over recent years. At the same time, a growing interest in self-compassion has emerged, which is seen as facilitating compassion toward others. Little research has explored, in depth, what compassion to self and others means to medical students. Therefore, a study was designed to address this gap in knowledge. Approach: A qualitative study was conducted, involving students from all 4 years of a graduate-entry medical school in the United Kingdom. Focus groups were used to obtain the views of students on compassion for self and others (patients). Care was taken to achieve variation within the sample in terms of age, gender, and year of study. Focus groups were completed between September and October 2016. An inductive thematic analysis was performed. Findings: A total of 31 students participated in 4 focus groups, each lasting between 60 and 90 minutes. Having the cognitive freedom—“headspace”—to be aware of and respond to one's own and others’ difficulties and distress was identified as an overarching theme within the data. This was underpinned by the themes developed during analysis: (a) bringing humanity into the workplace; (b) compassion as a variable, innate resource; (c) zoning into an individual's current needs; and (d) collective compassion. Students talked about the importance of being adaptable and responsive to situational factors in relation to self-compassion and compassionate care. They also highlighted the contribution of role models in promoting compassion to self and others. Insights: It is important for medical educators to explore ways of enhancing students’ compassion to self and others during their training and beyond. Integrating approaches to “well-being” into the curriculum can create opportunities for self-compassion development, but rigid protocols could derail these efforts.


Introduction
In recent years, there has been a drive to enhance compassion within modern health systems (e.g. Hearts in Healthcare -New Zealand; the Schwartz Center for Compassionate Healthcare -America; Compassion in Practice -UK), following accounts of poor examples of care, such as those reported at Mid Staffordshire NHS Trust -(www.theguardian.com/society/2013/feb/06/mid-staffs-hospital-scandal-guide). This is an area that warrants consideration because links between compassionate care (CC) and positive patient outcomes have been raised. 1 In addition, being compassionate towards oneself has received increasing attention as a means of facilitating compassion towards others. [2][3] These two aspects of compassion (towards patients and the self) are explored in this paper.
Compassion is an essential aspect of healthcare expected by patients, practitioners and professional bodies they represent. 4 It involves action taken in response to an awareness of and drive to alleviate suffering or distress. [5][6] Empathy is often used as a synonym; yet empathy involves vicariously experiencing another's feelings, be they positive or negative, whilst compassion entails being concerned and moved to improve another's situation. 7 Compassion has been depicted as a beneficial characteristic that supported the survival of offspring and allowed for collaborative relationships within groups. 8 As we evolved to nurture and protect our kin, 9 the ability to show care towards the self was also said to have developed, through feeling attached to others. 10 Like compassion outwards, self-compassion places an emphasis on common humanity, alongside self-kindness and mindfulness. 11 Self-compassion is proposed as a means of facilitating CC, as it expedites interpersonal interactions, forgiveness and perspective-taking. 3 Self-compassion is credited with reducing feelings of threat, instead producing a sense of soothing, well-being and resilience. 8,12 It should be noted that its role in addressing staff stress or improving patient care is not conclusive. 10 Nevertheless, neuroscience has started to take an interest in this area. Preliminary experiments suggest it is possible to transform neural networks so people can respond compassionately rather than being overwhelmed and withdrawing from another's sorrow or pain; hence, compassion may be malleable and augmented (through activities such as contemplative techniques). 7 However, evidence about teaching compassion to healthcare professionals (HCPs) (whether this can be done and, if so, how) is limited and often focuses on nursing. [13][14] It is not clear how those wishing to enter a career in medicine perceive compassion towards self and others. Addressing this gap in knowledge will highlight ways to support medical students with being compassionate to self and others. This is an area educators have been charged with neglecting. 15 We need to understand this topic in greater depth because compassion forms part of professional codes of practice (e.g. in the UK it is included as a value in the National Health Service Constitution, 16 and it is part of the American Medical Association's Principles of Medical Ethics 17 ). It has also been noted that compassionate acts can bring benefits to their providers, in terms of mental and physical well-being. 18 Hence, it could help motivate HCPs in their role, when faced with external challenges (e.g. job insecurity, targets, time constraints).
Following a review of the literature, Sinclair and colleagues 19 suggested that an empirical understanding of compassion was underdeveloped. This study will contribute to addressing this gap in knowledge by offering an in-depth insight into how students' understanding of compassion to self and others might be shaped by their medical training.
Specific questions we aimed to explore were: • How do students define and view the meaning of CC?
• What are students' definitions and perspectives on self-compassion and its importance in patient care?
• What role do educators play in supporting the provision of compassionate patient care by medical students, as perceived by these trainees?

Methods
Design: The ontological lens for this work was that of subtle realism; 20 accepting an external reality exists that can only be accessed through the perceptions and interpretations of individuals. In line with this stance, the methodological approach informing the work was, as described by Sandelowski, 21 'Qualitative Description. ' We aimed to stay close to participants' perceptions and concerns within the analysis. Nevertheless, findings represent our "configuration of segments of coded data assembled as a novel whole" 22(p. 1406) Focus groups were seen as appropriate for data collection because they allow people to explore complex topics together; to express their individual views, but to also compare and reflect on one another's comments. 23 Approval was given by the University of Warwick's Biomedical and Scientific Research Ethics Committee.
Setting and sample: Students from a single UK medical school, which runs a four year graduate-entry programme, took part. A purposive sampling approach was adopted, seeking the views of those able to shed light on the topic. 24 We aimed for variation within the purposive sampling in terms of year of study, gender and age. A sample size of approximately 30 was chosen for feasibility reasons and to gather statements from enough participants to allow for an in-depth understanding of the topic; it has been noted that rich data can be provided following 3 focus groups with between 10-60 participants in total. 24 Data collection: All students at the recruitment site were sent an invitation via email and social media. Individuals who wished to be involved received an information sheet. It had the title 'Medical students and compassion in healthcare: A qualitative study.' They had time to read this document to make a decision about taking part. Participants completed a demographics form, to capture details on year of study, age, gender and religion. All gave written consent to their involvement and to the researchers using de-identified quotations when disseminating findings.
Four focus groups were held in total, one with each year of medical students. It was felt that this would provide an in-depth understanding from a range of perspectives. We were able to gather a good balance of data "to tell a rich story", whilst not being engulfed by material to an extent that deep engagement with what people said was impossible. 25 One researcher acted as facilitator, using a topic guide and following up interesting points raised by participants. The topic guide was adjusted after each group to enable unexpected areas to be explored further. Table 1 outlines questions asked and added for each group. Analysis: An inductive thematic analysis was conducted. Transcripts were divided into smaller content units (codes). These were then collated into similar threads or patterns of ideas (themes), to provide a nuanced account of the data. 26 We used the following stages to do this: 1) Becoming familiar with the data; 2) Developing initial codes; 3) Collating codes into themes; 4) Reviewing the themes; 5) Labelling themes; 6) Writing up results. 27 Members of the research team met on four occasions to carry out initial coding of each transcript. In a further meeting, initial codes from all four focus groups (n=162) were written on separate pieces of paper and similar codes were clustered into themes and sub-themes. Once themes had been established, they were compared to the original data to ensure they reflected the content of focus group discussions.
Rigour: Guba 28 proposed four criteria that should be considered by qualitative researchers in pursuit of a trustworthy study; credibility, transferability, dependability and confirmability. In our work, credibility was established by having more than one person involved in the analysis, from a range of backgrounds (see below).
In addition, open-ended questions were asked that allowed participants to raise issues of pertinence to them. Transferability was addressed by describing in detail how data were collected and analysed, and by relating findings to the existing literature. Dependability was ascertained by using a well-defined approach to analysis, and treating all data in a systematic manner. Confirmability was determined by creating an audit trail of how we moved from raw data to codes and final themes; members of the research team kept a journal during data collection, in which they reflected on emerging ideas and contextual information relevant to the on-going analysis (e.g. how they felt members of the group related to each other and interacted).
Reflexivity: Members of the research team consisted of an individual who was knowledgeable of literature related to CC and had experience in qualitative methods, and medical students who were interested in the topic of compassion within the context of their future career. This mixture of backgrounds allowed for varying perspectives to be raised during data collection (e.g. on questions to ask participants) and analysis (e.g. challenging/having to explain one another's interpretations of what participants said). We came to the research believing that participants would recognise the role of compassion in healthcare. Conversely, we were not sure how they might perceive self-compassion and its contribution to being a doctor. Focus groups were conducted by students to avoid power relations that may have occurred had they been led by a member of teaching staff.

Results
In total, 31 students took part, across the four focus groups. Their average age was 26 years (ranging from 21 to 45). Further details about their background can be found in Table 2.

TABLE 2 here
During analysis, the authors remained grounded in the data, by referring back to transcripts when developing themes. We also discussed the context of data collection; in a group setting with peers, which may have shaped the responses given. With this in mind, 4 themes were advanced by the research team. They depict compassion towards self and others as: Theme 1: Bringing humanity into the workplace -a personal characteristic that someone carries into their professional role; Theme 2: Compassion as a variable, innate resource -a reserve that can be drawn upon and fostered (or exhausted) to ameliorate workplace interactions; Theme 3: Zoning into an individual's current needs -centred on improving the situation of an individual (including the self); Theme 4: Collective compassion -having wide reach and influence. Each of these will be described, with quotations provided to support our interpretation of the data.
Those taking part have been given an identifier based on the focus group (FG) they attended, and their seating position within this (P1, P2 etc).

Theme 1: Bringing humanity into the workplace
Participants linked CC to being holistic. Compassion was described as part of someone's character looked for during application to medical school. They often used the term empathy interchangeably with compassion. However, differences were recounted. Compassion was depicted as having an active element, rather than just feeling with a patient. It was proposed that an individual's motivation behind an action made it compassionate (or not), based on whether someone was looking to support another: They mentioned how doctors could become desensitised to pain to avoid feeling emotionally overwhelmed. As a consequence, they did not always feel compassionate when individuals presented with 'minor' ailments (e.g. a broken finger, a cold). Participants also noted that compassion was a struggle when fatigued: FG4P5: "Being compassionate is one thing, being able to communicate that is completely different…you may lack that ability purely because of stress, having done a string of night shifts." In terms of self-compassion, students said it involved acknowledging that they were only human and unable to solve everyone's difficulties, which could be difficult to accept: FG3P5: "I think as medical people, we become less and less compassionate towards ourselves…you get into the system of treating people and trying as much as you can, I feel that you then kind of go -'right I'll do the best I can!' and you can't always…and that's why it's hard to know when to stop. We then care less and less about ourselves." For students who had progressed further in the course, their concept of self-compassion appeared more clinical in nature, centring on being a good doctor (e.g. enabling them to safeguard where appropriate and to prescribe safely). This profession-oriented ideal contrasted with students earlier on in their training, whose notion of self-compassion related to individual rituals of self-care to overcome things like exam stress. However, all groups recognised selfcompassion as an important contributor to CC, talking about how it enabled staff to look after others effectively: FG4P2: "We said about the doctor on call being busy and that's somebody that's stressed and probably hasn't got a lot of self-compassion and then goes to see patients in like a bit of a flap because they're…not being kind to themselves."

Theme 2: Compassion as a variable, innate resource
Compassion was depicted as difficult to teach because it was regarded as innate. At the same time, it was said to involve particular skills (e.g. being non-judgmental and responsive to patient cues) that participants felt could be developed: FG1P3: "I think in a way…some people are born more naturally compassionate than others…but I worked in substance misuse for 2 years and I can tell you I learnt a lot about compassion in that…to be compassionate to people who maybe you wouldn't be so easily compassionate to on a general basis. I learnt a lot through that cause you've got to try and put aside any kind of personal judgment." Drawing on life experiences was said to help in nurturing CC. Consequently, participants said undertaking a graduate-entry course was beneficial. Some did believe that being older could make self-compassion less accessible because they were more likely to ruminate and be self- The variable nature of compassion called for flexibility and consideration of what was required at a particular moment in time, which students could find draining. Likewise, it involved making an emotional investment: FG3P3: "…compassion kind of costs you something as well. I remember when I was sat in the clinic with a doctor that was just breaking bad news in this cancer clinic. She was breaking bad news to people all afternoon. This lady came in and she had to tell her that she had cancer, and after she walked out you could see that the doctor was really wiped, just because it was emotionally, she was emotionally involved with the patient." Consequently, cultivating self-compassion was regarded as important to prevent burnout. This was something they thought medical schools should promote. At several times during data collection, self-compassion (e.g. accepting one's flaws, being kind towards oneself and noncritical) was said to be considered only following a crisis. Therefore, they thought it needed to be legitimised within their training:

Theme 3: Zoning into an individual's current needs
A recurring refrain across focus groups was the significance of connecting with patients on an individual level to provide CC, a process whereby staff "zone into that person's situation" (FG3P6). To practice in this way, students talked about having to quell any biases so they did not prejudge patients. This could be hard because they came to clinical situations with their own views, shaped by their background and medical knowledge acquired during training: FG1P7: "…I was interacting with people that I'd never imagined 10 years ago, 15 years ago, I would be interacting with and it's that process of actually sitting down and having a conversation with someone that you may have a prejudice against naturally, but actually that's what forces you to re-evaluate your value system…" Participants stated that certain disciplines (e.g. psychiatry) had the luxury of getting to know patients on a more personal level. They also felt that the work patterns of some specialities allowed for self-compassion because they were more structured (e.g. 9am to 5pm), giving people time to reflect on their own needs. It was suggested that a complex mixture of factors had to be considered when thinking about how to show compassion towards a specific patient, including their physical state, their cultural background, their family's needs and current time pressures. This resulted in one person making the following comment: FG3P5: "I don't think being compassionate is hard. I think actually it's working out how you can be compassionate to that particular person is the difficult bit." Alongside treating patients as individuals, students had to recognise they were not identical to their peers and had their own strengths and weaknesses. This was necessary so they did not feel immobilised by an inner critic. Without this self-acceptance, it could be hard to request help for fear of appearing weak: FG2P5: "…in medicine…it's quite easy to be hard on yourself because everyone around you is perfect…if you've got something wrong or you don't understand something…it's hard to say to somebody that you're feeling that way."

Theme 4: Collective compassion
Many people talked about external factors affecting CC, including one individual who introduced the term "collective compassion": FG3P3: "…you can have collective compassion -so like, managing resources might not look like the most compassionate thing to do for the one individual patient, but actually if you're looking at the bigger picture, it might allow more people to have access to treatment..." Collective compassion incorporates the notion of CC being more than a single doctor's duty; participants stressed it should be a concern of all health service employees, including nonmedical staff (e.g. porters, car parking staff): Furthermore, participants argued that healthcare systems must treat staff in a humane way, not like machines devoid of emotions. Yet they sensed that wellness support, provided during training, did not transition into work settings: FG1P1: "I think more needs to be done in the healthcare system to look after the doctors, to look after the nurses… FG1P7: …perhaps we should have mindfulness directors…who is not your consultant but you can go and see them and they are responsible for the well-being of everyone…" They also described a lack of collegiality in their medical school due to the promotion of academic rivalry; for example, posting exam results for everyone in a year group to see, which was not regarded as compassionate: FG2P9: "…they send emails to everyone…this person got distinction, this person did the best…they rank us all essentially and then tell everyone how all their friends have ranked…that makes it even easier to sort of compare yourself to others and think 'well if they can do it, then why can't I?" Overarching theme: Needing headspace for compassion to be activated The 4 themes described above highlight that for these medical students, although they mentioned an emotional element to CC, this was affected by and secondary to the mental effort they associated with providing compassion to self and others (see Figure 1). and being open to a patient's personal concerns and experiences. At the same time, they had to accept their limitations and not be overly self-critical. They noted that compassion to self and others was not always prioritised, especially when exams focused heavily on factual information rather than the way in which care was delivered. Participants believed that compassion, as part of training, came second to learning about the body and technical skills.
They thought it was not necessarily fully integrated into the curriculum, as lecturers tended to divorce the human side of care from the physiological: FG2P6: "…we're just kind of encouraged to get good at sort of our clinical skills first and then sort of patient compassion comes second…" Students described how headspace peaked as they progressed in medical school and became familiar with knowledge and theory, which allowed them to then focus on compassion.
However, they were concerned that this space to contemplate compassion would diminish postqualification, as they took on more responsibilities: FG3P5: "It does seem that the higher people get up, the less compassionate they become. I don't know whether that's because they've seen so many more people or they have so much more to deal with..." Consequently, this was an area that students felt they could be prepared for during training: FG1P3: "…medical school could teach us how not to lose it [compassion] as we go on and as we get to the stresses of economic constraints and time constraints...I think that might be an avenue to explore…" Participants were encouraged to practice mindfulness during their training to expedite headspace for compassion. Not all students were comfortable with this activity. Therefore, it was recommended that educators promote a range of options so trainees could discover what worked for them in freeing up cognitive space for compassion: FG2P10: "Like zerobalancing.
FG2P8: Or yoga or something else -something that teaches that self-love idea, and being balanced, definitely… FG2P9: …have a variety of choices within that, so it could be joining a sports team, it could be doing mindfulness, it could be doing yoga or a craft."

Discussion
This study provides a novel insight into compassion from the perspective of medical students.
They depicted it as an active process requiring intellectual effort that could be curtailed by competing demands on their thinking capacity. In all focus group discussions, the notion of 'needing headspace for compassion to be activated' was emphasised. The cognitive aspect of compassion has been referred to by others, [29][30] but not in relation to medical students specifically, who have unique demands on their time and thinking. Our data showed that students thought this was not just about the ability to decipher how someone was feeling and devising a plan of action to help; it also appeared to call for the mental scope to recognise what was happening inside and outside of one's sphere of focus and to respond accordingly. In our data, having adequate headspace underpinned both compassion to self and others. It could explain why students experience times when it is harder to be internally kind and nonjudgmental, and to display these attributes to patients. Below, we highlight the following three key areas in participants' narratives that relate to the existing literature: forms of care, compassion progression and role-modelling.
The prominence that students gave to cerebral activities as part of compassion is reflected in the following four forms of care delineated by Post et al: 31 routine, cognitive empathy, affective empathy and compassionate. Some of the students taking part in our focus groups appeared to describe cognitive empathy -a somewhat detached approach to a patient's situation to avoid feeling overwhelmed by it. Others were reflecting affective empathy, which "describes the professional's ability to both understand what the patient is going through (cognitive empathy) and experience of a non-verbal resonance with the patient emotionally." 31(p. 875) There were also points during data collection where participants expressed CC according to Post et al.'s 31 taxonomy, said to represent an intensification of affective empathy in the presence of suffering. It is suggested that students can move along these levels through observing role models and gaining experience. 31 However, an emphasis on knowledge and technical skills as part of medical education, at the expense of relational aspects, can be a barrier to CC. 19,32 Participants said their ability to provide CC increased as they progressed through medical school. They suggested this was because they had more headspace for compassion as they became familiar with biomedical aspects of their work. This contradicts previous research suggesting that medical students become more competent in technical skills and factual information over time, but decrease in compassion and empathy. 15,33 Our participants did fear that compassion would decline as they took on more responsibility within their career, as their headspace became congested with work pressures. This concern stemmed from seeing senior staff who failed to notice and address a patient's suffering when on clinical placements.
Good role models are depicted as those exhibiting effective communication and respect towards colleagues, who comfort patients, express passion for their work 34 and remain compassionate in the face of difficult circumstances. 15 Engaging in self-compassion could be added to this list. However, self-compassion was not often observed among HCPs by students involved in our focus groups. It should be noted that the idea of self-compassion has been challenged by Sinclair and colleagues 10 for lacking: a) a strong empirical basis, and b) distinction from self-care and self-awareness. Nevertheless, it was a concept that our participants felt held resonance for them. They could see its potential merit in enabling them to be a holistic rather than technical practitioner, who had the psychological strength and space to cope in times of difficulty.

Limitations
It was anticipated that having medical students facilitate focus groups would help to encourage disclosure. However, participants may have presented in a manner that was seen as socially desirable to their peers. Consequently, further research could explore this challenging concept of 'headspace' using one-to-one interviews. People volunteered to take part, so may have been those most interested in the topic. Nevertheless, they provided a range of views that enabled us to develop an overarching and associated themes. The outcome of a qualitative descriptive study "should be the identification of a story, which the researcher tells about the data in relation to the research question…", which offers new insights. 26 (p. 403) We feel that this was achieved with the data collected for our study. That said, the study took place at one medical school, so findings might not be the same elsewhere.

Areas for future research and next steps
We propose that the themes outlined in this paper represent a useful starting point for discussion with students about the role of compassion in how they approach their own and others' health.
It can be used to explore situations when they find it difficult or easier to express or experience compassion. In the past, the wellbeing of doctors was said to receive inadequate attention as part of training, with an emphasis placed on relinquishing one's own needs in a competitive learning environment. 35 Competition among medical trainees, mentioned during our focus groups, has been raised as a barrier to CC by others. 4 Recently, activities have been introduced into medical school curricula to foster resiliency. This is often based on the practice of mindfulness, 35 which has been linked to better patient care, 36 Benefits of mindfulness may depend on participants' motivation. 44 For example, in a study by Aherne and colleagues, 45 students were clear that mindfulness was not for everyone and that making it compulsory could be counterproductive. Their research found that students felt other options for coping and stress reduction should be explored, a notion echoed by our focus group participants. This needs to be considered by those planning medical curricula to ensure that students are exposed to a range of techniques to help them have adequate headspace to be compassionate to self and others.

Conclusion
Medical students' views of compassion in healthcare highlighted the centrality of headspace. seek to understand patients' suffering rather than distancing themselves from it. Selfcompassion may be one means of helping HCPs to remain compassionate in the face of external stressors present within health services. However, this resource or ability to be selfcompassionate needs to be cultivated during medical training by emphasising its importance in enabling students to attend to their own and the suffering of others.
We would like to thank all the students who gave up their time to participate in this study.  Medical students' views of compassion in healthcare highlighted the centrality of headspace; being able to notice and respond to their own and the distress of others called for the cognitive freedom to think, unrestricted by overriding concerns they associated with their training (e.g. the accumulation of scientific knowledge). It meant being flexible and aware of immediate circumstances and challenging preconceptions.

Needing headspace
Theme 1: Bringing humanity into the workplace CC requires the cognitive capacity to notice and decide to act to alleviate distress/suffering; self-compassion entails accepting one's limitations and that mistakes can happen Theme 2: Compassion as a variable, innate resource CC is faciltated by having time to reflect, without too many competing demands; self-compassion helps with managing emotional aspects of the job

Theme 3: Zoning into an individual's current needs
CC calls for understanding an individual's needs and responding to their cues; self-compassion means being aware of one's strengths and weaknesses Theme 4: Collective compassion CC means recognising that one is not an island because care is delivered in a team; self-compassion requires a supportive environment to flourish