Beyond Compassion Fatigue: Motive-Based Approaches to Sustaining Compassion in Palliative Care

Abstract

Compassionate care is vital to quality palliative care and integral to the provision of spiritual care at the end of life. But is sustaining compassion costly for healthcare providers (HCP), impacting their emotional and spiritual wellbeing, professional purpose, and moral self-image? Concerns about the costs of compassion for carers have gathered pace in a growing healthcare literature on compassion fatigue. Critics, however, argue that compassion fatigue lacks adequate conceptualisation, querying whether it fits with HCP’s own perceptions and suggesting it lacks utility for identifying interventions. This article contributes to this debate about moving beyond compassion fatigue, by bringing new psychological research on compassion as a motivated choice to bear on these questions and demonstrating its potential for illuminating interventions to support compassion in palliative care contexts. It proposes a focus on motive-based interventions which serve to tip the cost–benefit analysis in relation to compassion and thus support HCP motivation to feel and act compassionately. A key implication of this approach is that sustaining compassion is not up to individuals alone, as can often seem to be the case with ‘self-care’ paradigms. Rather, there are multiple ways institutions and society can play a role in motive-based interventions to sustain HCP compassion and wellbeing. The final section explores one example in the form of institutional support for spiritual care education.

1. Beyond Compassion Fatigue: A Review of the Debate

Is compassion costly for carers? Is it especially costly for those who work in the challenging context of end-of-life care? Does providing compassionate care, day in and day out, often in response to intense suffering and over prolonged periods of time, negatively impact healthcare provider (HCP) wellbeing? Accounts of ‘compassion fatigue’ (Figley 1995; Stamm 2009) answer in the affirmative. While compassion holds many benefits for patient care, they argue, the story is reversed for the carers themselves for whom compassion brings significant costs. These costs accrue over time and impact several dimensions of carer wellbeing, including physical, mental, emotional, and spiritual. Moreover, proponents of compassion fatigue argue that these costs of compassion not only impact HCP wellbeing, but also undermine the very capacity of carers to sustain compassionate patient care over time; they deplete or numb compassion, as it were, putting patients at risk of poorer health outcomes (Sorenson et al. 2016). Given these serious consequences, they worry compassion fatigue could be a leading factor contributing to the exhaustion and burnout of HCPs.

If true, these claims are concerning. Responding to their seriousness, studies on compassion fatigue in healthcare have grown exponentially over the past decade. Together with studies on other stressors, these form part of a growing body of literature on HCP needs and wellbeing. Analysis of publications on both burnout and compassion fatigue among HCPs up to 2019 found that, while publications started in 1978, steep growth in the number of publications was observed in the last decade (Sweileh 2020). Studies on a wide variety of workplace stressors in healthcare have “emphasized that the physical, emotional, social and spiritual health of healthcare providers is impaired by cumulative stress related to their work, which can impact the delivery of healthcare services” (Sinclair et al. 2017, p. 9). Ninety studies have been conducted specifically on compassion fatigue, as well as the related idea of compassion satisfaction (Sinclair et al. 2017). Another integrative review on compassion fatigue in the healthcare literature found 307 articles (Sorenson et al. 2016). The rapid increase of research in this area has seen compassion fatigue become a significant area of contemporary healthcare research.

A similar literature is starting to explore compassion fatigue and related concepts in the palliative care HCP population. A 2021 scoping review found 20 studies, covering themes such as the conceptualisation of compassion fatigue and compassion satisfaction, measurement, consequences in relation to providing care for patients with life-threatening conditions, predictors of compassion fatigue among palliative care HCPs, and interventions to reduce it and support wellbeing (Baqeas et al. 2021). The authors also identified gaps in this literature and concluded by calling for further research on compassion fatigue in palliative care regarding the effectiveness of specific interventions and the impacts of compassion fatigue on palliative care HCPs’ work.

In line with this call for further research, some have argued that palliative care is a particularly pressing area for research on compassion fatigue given the uniquely challenging nature of the demands placed on HCPs in this context. Opening framings of articles contain rationales such as: “Palliative Care nurses are predisposed to distress, because they are surrounded by seriously ill or dying patients on a regular basis” (Cross 2019, p. 21) or “Prolonged contact with these patients predisposes PCHP to emotional and psychological distress such as compassion fatigue” (Baqeas et al. 2021, p. 1). Constant exposure to patient suffering, grief and bereavement are often highlighted, noting that palliative care workers “empathise with the losses their patients are experiencing in the dying process and often feel a sense of personal fatigue that they cannot help their patients; working on the edge between ‘life and death’ cultivates an acute awareness of the fragility of life” (Slocum-Gori et al. 2013, pp. 172–73).

If these calls for further studies of compassion fatigue are heeded, palliative care research looks set to head down a similar path to the broader healthcare literature on this topic. But should it? Before following this path, it is worth pausing to consider where it might lead, especially given the central importance of compassion to palliative care and the provision of spiritual care at the end of life. There are several reasons to resist making compassion fatigue central to future research about palliative care HCP wellbeing.

Critics of compassion fatigue argue that it lacks adequate conceptualisation and explanatory power, querying whether it fits with HCP’s own perceptions as well as suggesting it lacks utility for identifying interventions. The Professional Quality of Life Scale (ProQOL) is the most common measure used for assessing compassion fatigue, compassion satisfaction and burnout, with hundreds of studies relying on its brief definitions (Stamm 2009). Critics argue that the conceptualisation of terms in many of these studies lacks depth and has changed over time (Baqeas et al. 2021). It is also unclear how the definition of compassion fatigue employed in the ProQOL relates to the concept of compassion itself, as Sinclair et al. contends, it “does not assess any of the elements of compassion” (Sinclair et al. 2017, p. 9). Other critics question whether there is sufficient evidence for the construct validity of the scale (Geoffrion et al. 2019).

Systematic and integrative reviews of the compassion fatigue literature often conclude by advocating further conceptual analysis, for example, Sorenson and colleagues write that “as the definition and use of the term compassion fatigue has evolved, a need for a well-developed concept analysis has also become evident” (Sorenson et al. 2016, p. 456), which they seek to address (Sorenson et al. 2017). However, others such as Sinclair and colleagues are more pessimistic: “[t]he conceptualization of compassion fatigue was expropriated from crisis counseling and psychotherapy and focuses on limited facets of compassion” such that “after two decades of scholarship and more than 350 peer-reviewed publications on this topic, there is still no broadly accepted definition of compassion fatigue [and] its relationship to compassion is uncertain” (Sinclair et al. 2017, pp. 9, 10). Critics thus conclude that “compassion fatigue rests on a most fragile foundation” (Ledoux 2015, p. 2046) and has essentially “become a contemporary and iconic euphemism that should be critically reexamined” (Sinclair et al. 2017, p. 9). The terminology of compassion fatigue, they contend, should be relinquished in favour of a more precise discourse on HCP occupational stress. Conceptual models are needed that can better distinguish between the varied work-related stressors faced by healthcare providers and causing widespread emotional fatigue and burnout—many of which have less to do with compassion than with workplace issues (Sinclair et al. 2017).

Another key reason to question compassion fatigue is that empirical research indicates the compassion fatigue paradigm may not align with HCPs own perceptions and practice of compassion. One study found that HCPs avoided mentioning compassion fatigue, with some HCPs averse to the idea due to the causal connection it seemed to imply between compassion and burnout (Singh et al. 2018). HCPs thought it important that seeming barriers to compassion were not treated as excuses to avoid compassion, but as challenges that could be overcome, including by cultivating compassion through intentional actions, even in the absence of feelings of compassion (Singh et al. 2018). As the researchers conclude: “[o]ne of the unanticipated findings from this study was participants’ aversion towards associating the language of barriers to compassion as they felt it applied an absolute term to the nonconditional adaptive nature of compassion” (Singh et al. 2018, pp. 2093–94). We will return to this idea of the plasticity of compassion in later sections.

Related to this focus on HCP perceptions, Sinclair and colleagues have undertaken an ambitious program of research to deepen the conceptualisation of compassion in healthcare by drawing on patient and HCP perceptions of compassion to produce an empirically grounded model (Sinclair et al. 2016, 2021). Similarly, Baguley asks why so many studies of compassion in healthcare focus on barriers and fail to examine the strategies HCPs themselves use to successfully maintain compassion over time, often in the face of serious challenges in their environment (Baguley et al. 2020). The utility of reputed compassion fatigue interventions—many of which focus on self-care and personal resilience—has also received increased criticism, with some researchers seeking to go “beyond compassion fatigue” to offer alternative models able to illuminate more targeted, evidence-based interventions (Fernando and Consedine 2014, p. 289).

The above criticisms suggest that despite its rapid growth over the past two decades, several issues continue to plague the compassion fatigue literature and its reliance on the ProQOL. Considered together with the longstanding commitment of palliative care to the provision of compassionate care, these criticisms give reason for researchers to pause before multiplying more of the same studies of compassion fatigue now within the palliative care HCP population.

Yet, at the same time, the popularity of the idea of compassion fatigue suggests it has tapped into salient worries and concerns about HCP wellbeing, particularly around the emotional and psychological aspects of caring. It rightly highlights that compassion is rarely cost-free and raises important questions about the wellbeing of carers and what can be done to support them. How can we think better about these potential costs of compassion for carers—sharpening definitions, explanatory models, measurement tools and uncovering potential interventions?

Future research on compassion and HCP wellbeing should set aside the framework of compassion fatigue and take up new conceptual resources. In this article, I argue that palliative care researchers concerned about the costs of compassion would benefit from engaging with new research in psychology which has found evidence to support a ‘motivational model’ of compassion. I examine this model and show how it can help sharpen our understanding of what is going on when we feel and act compassionately—and when we fail to do so—and thus illuminate potential interventions for supporting HCP compassion and wellbeing. I argue that the shift to motive-based interventions proposed by this model is a welcome move beyond the frequent emphasis on individual self-care to instead illuminate a broader range of factors, including many to do with institutional support, which feed into the ‘motivational mix’ out of which any choice to feel and sustain compassion must come.

2. The Costs of Compassion

Before turning to new work in psychology on the motivational model of compassion, I first outline a working definition of compassion and its benefits and costs, especially in healthcare and palliative care. This provides helpful background to appreciate how something as positive as compassion might be seen as a threat to carer wellbeing.

One frequently cited definition holds that compassion is “the feeling that arises in witnessing another’s suffering and that motivates a subsequent desire to help” (Goetz et al. 2010, p. 351). This pairing of emotion and motivation is common to definitions of compassion in recent work in philosophy and psychology. For example, Cameron writes: “Compassion is an emotion of care: it is elicited by the perception of harm to others, and it motivates actions to reduce suffering” (Cameron 2017, p. 262). Some definitions emphasize a connection to action as well. Sinclair and colleagues found that HCPs themselves defined compassion as “a virtuous response that seeks to address the suffering and needs of a person through relational understanding and action” (Sinclair et al. 2016, p. 193).

Compassion then, when successful, comprises a sequence of emotion, motivation, and action. First, the emotional dimension—sometimes termed compassion, sympathy or empathic concern—comprises, following the work of C. Daniel Batson, “an other-oriented emotion elicited by and congruent with the perceived welfare of someone in need” (Batson 2011, p. 80). Compassion is thus a painful emotion (negative valence) congruent with the perceived negative welfare of the person in need and their painful emotion. It is also ‘other-oriented’ in the sense that the emotion is focussed on the other’s welfare: feeling for them. If our feeling of concern at another’s pain starts to switch to a more self-focussed painful emotion, we would no longer consider this compassion (it would be something like ‘vicarious distress’). To count as compassion, the emotion must be other-oriented.

Second, the motivational dimension. Compassion doesn’t stop at feeling for the other, it produces motivation to do something about it. Batson and colleagues conducted over 50 empirical studies that support the hypothesis that there is a causal relationship between feeling compassion for someone and having an altruistic motivation to help them. Compassionate or altruistic motivation here is defined as “a motivational state with the ultimate goal of increasing another’s welfare”, i.e., of removing the compassion-eliciting need (Batson 2011, p. 80). It is a motive to help the other person for their own sake (not for some other reason, for example, politeness or egoistic motivations).

Third, this altruistic motivation produces behaviour aimed at achieving the goal of helping the person in need. We thus see the full trajectory of compassionate emotion and motivation completed in compassionate actions. There are, of course, many ways compassion can fail along this trajectory, but this account gives us a working definition of what successful instances of compassion comprise.

The benefits of compassion are well-established, especially in healthcare, with compassion widely identified as an important element of quality, holistic healthcare practice (McSherry et al. 2020). Reported patient outcomes associated with compassionate care include alleviated suffering, enhanced well-being and enhanced care (Sinclair et al. 2016). There is a strong expectation that frontline HCPs across a range of interdisciplinary teams will provide compassionate care. This is even more so the case in palliative care, where a recognition of the importance of compassion has been central from its beginnings (Larkin 2016). Palliative care has also led the way in recognising the spiritual needs of patients and linking compassion and spirituality in understandings of patient-centred care (Puchalski and Ferrell 2010). I examine spiritual care further in Section 5.

However, compassion is also typically thought to come with some costs for the carer. There are four broad categories drawn from the psychological literature on compassion which are useful for analysing these costs: material costs, cognitive costs, emotional costs, and a further area of costs that can arise when trying to avoid the other ones, which we can term moral or spiritual costs.

First, compassion can have material costs since it motivates us to devote resources to help those we perceive to be in need. Such attempts can impact our finances and time. For example, a palliative care nurse might skip breaks or work overtime without pay to provide care for a patient whose suffering elicits their compassion. Second, compassion can involve not only physical effort but also an intensification of mental labour in pursuit of its goal of increasing the other’s welfare. Helping is rarely simple: it typically involves a measure of uncertainty and worries about the inefficacy of one’s help—all of which can be cognitively taxing on the carer. Third, compassion has an emotional cost associated with experiencing a painful or negative emotion. It is not hard to see why worries might then arise with respect to contexts such as palliative care where HCPs are routinely exposed to suffering and caring for multiple patients. Intuitive ways we tend to think about emotion might lead us to think their emotional ‘energy’ or ‘resources’ could become depleted or used up to the point of no return, affecting their psychological wellbeing and making them more susceptible to burnout. It is these kinds of worries about the emotional costs of compassion that drive concerns about compassion fatigue.

At the same time, compassion and altruistic helping is often also thought to come with emotional benefits for carers, sometimes termed compassion satisfaction or joy. Indeed, some defenders of compassion argue that compassion is ultimately a warm, emotionally rewarding emotion. Some further claim that, as such, the above problem of emotional costs only attaches to empathy (which they tend to define narrowly in terms of experience sharing or ‘feeling what the other feels’) and not compassion (defined more broadly to include motivation to help and action). For example, Bloom claims: “Compassionate helping is good for you and others. But empathic distress is destructive of the individual in the long run” (Bloom 2014). In the psychological literature, Klimecki and Singer contend that compassion fatigue should be relabelled empathic distress fatigue (Klimecki and Singer 2012). Too much empathy is the problem. However, while it may be tempting to draw a sharp distinction between empathy and compassion, and to cast compassion as the cost-free solution to the ills of empathy, recent empirical results temper too strong a distinction. Studies show that people similarly predict both empathy and compassion will involve emotional and cognitive costs, with one set of studies even finding that people “reported compassion to be more cognitively taxing than empathy and objective detachment” (Scheffer et al. 2022, p. 172). Such results suggest we still need to grapple with a better understanding of the costs of compassion.

The final category of costs is somewhat different in that these costs arise from attempts to avoid the previous sets of costs (material, cognitive or emotional), creating further potential moral and spiritual costs for carers. As one study put it: “Regulating compassion thus has a cost of its own: it forces trade-offs within a person’s moral self-concept” (Cameron and Payne 2012, p. 225). For example, if we consider the context of palliative care, compassion is seen as vital to patient-centred care and HCPs are strongly motivated to feel compassion and see themselves and their colleagues as compassionate people. As such, cases in which HCPs find themselves decreasing their compassion to avoid its perceived costs can precipitate crises around their professional identity. They may wonder if they are failing their patients or whether they can still do their job. Such questions in the caring professions touch upon a person’s moral self-image. If the ability to feel and act compassionately towards patients facing life-threatening illnesses is integral to one’s deeply held values and conception of a good person, then a sense of fatigue or failure at this could cause doubts about whether one is a good person after all. Guilt could also be associated with a sense of moral failing. Brémault-Phillips and colleagues reported that HCPs can experience moral or spiritual distress when “they perceived discord between their morals and values, and the demands placed on them by work, colleagues, families and patients” (Brémault-Phillips et al. 2015, p. 488). Calls to investigate the relevance of moral distress or moral injury among HCPs have also grown during the COVID-19 pandemic (Borges et al. 2020).

Having outlined some of the potential costs of HCP compassion, it is not hard to see why worries might arise about the way such costs could accrue over time. In the next section, I explore the role these costs and benefits play in giving rise to compassion, or lack thereof. I argue that new research in psychology on the motivational model of compassion helps us better understand that these costs are not static and fixed, nor are they value neutral. Rather, the balancing of anticipated costs and benefits in a given situation very much depends on our top-level goals and values, and how we prioritise them.

3. New Research on Compassion: The Motivational Model

A significant development in compassion science (Simon-Thomas and Goetz 2017) has been the proposal of a new ‘motivational model’ of compassion, which draws upon evidence from recently developed ‘free choice’ methods for studying compassion. Free-choice methods go beyond traditional self-report measures of occurrent state or trait compassion to instead study how participants respond when presented with choices to opt in or out of feeling compassion in various scenarios (Scheffer et al. 2022). Developing a motivational model of compassion to reflect these new empirical findings has been the prime aim of recent work by C. Daryl Cameron and colleagues (Cameron 2017; Cameron and Rapier 2017; Scheffer et al. 2022). They distinguish the motivational model from another prominent model of compassion that has tended to view it as “a limited-capacity resource that cannot be extended indefinitely” (Cameron 2017, p. 263). It is this model which is often implicit in discussions of compassion fatigue, where compassion is depicted on analogy with a reservoir or ‘gas tank’—we have a certain amount, but when repeated over time it can ‘run out’ with damaging consequences.

But what if compassion isn’t like this? What if, as Cameron and colleagues propose, it is better viewed not from “the perspective of a gas tank” but “from the perspective of a driver, who can make active decisions about whether to accelerate, slow down, or if needed, change course”? On this analogy, compassion “isn’t the fuel that allows driving to happen: it’s the very act of driving” (Cameron et al. 2017). The motivational model of compassion argues that compassion should be thought of as an active choice: it involves a dynamic decision process which occurs in each compassion-relevant situation as well as iterating over time to form habits of compassion or callousness (Cameron et al. 2017).

What exactly is meant by ‘choice’ here? Research on instrumental emotion-regulation (Gross 2014; Tamir 2016) has shown that contrary to popular stereotypes of emotions as automatic mechanisms beyond our volitional control—something that just happens to us—people can and do control their emotions using a variety of strategies depending on their motivations and goals. They do this both directly in the moment, for example, through cognitive reappraisal (e.g., “oh, that injury is not as bad as I thought, up you get“ or “really, that’s your own fault”), and indirectly, for example, by selecting the situations they allow themselves to get into in the first place (e.g., crossing the street or changing the channel to avoid an appeal from a charity).

These insights from the emotion regulation literature, employing free choice methods (Cameron 2018), have already been utilized in the study of empathy avoidance and the way people control whether and how much they empathize with others (Cameron et al. 2019; Cameron and Payne 2011; Shaw et al. 1994). One intriguing example for the healthcare context are studies which have shown that physicians down-regulate their pain empathy response to patients within 300 milliseconds (Decety et al. 2010). Thus, Zaki argues that while much research on empathy has emphasized its automaticity, it is crucial to recognize the “key role of motivation in driving people to avoid or approach engagement with others’ emotions” (Zaki 2014, p. 1608).

Like Zaki’s account of empathy as a motivated phenomenon, the motivational model of compassion claims that people can and do exercise this kind of control over their compassion (choosing to approach/increase or avoid/decrease compassion in various situations), and that they do this strategically based on their motivations, i.e., whether they are willing to feel compassion. Similar methods have been used by Cameron and colleagues in studies of compassion, presenting participants with choices to feel compassion or not. Across several studies, for example, Scheffer, Cameron and Inzlicht “found that people opted to avoid compassion when given the opportunity…and opted to feel compassion less often to the degree they viewed compassion as cognitively costly” (Scheffer et al. 2022, p. 172). Studies have found this compassion avoidance effect with respect to anticipated financial costs, cognitive costs and emotional costs (Cameron et al. 2016).

Together, these studies provide evidence that people choose to avoid compassion when they anticipate it will be costly (Scheffer et al. 2022). The results suggest we need to take a closer look at the many cases where compassion breaks down and ask, as with empathy, whether people fail to feel compassion “not because they are unable to do so, but rather because they are unwilling” (Weisz and Zaki 2017, p. 210), and thus making motivated choices to avoid compassion-eliciting situations (as we’ll see, in some contexts, this avoidance may be for very good, self-protective reasons).

Summing up the benefits of these new studies of compassion, Scheffer and colleagues argue that “[s]elf-report measures of trait or state compassion can be useful in diagnosing either retrospective or current experiences of compassion but speak less to how people might systematically shape those feelings using emotion regulation strategies” (Scheffer et al. 2022, p. 190). These studies bring into focus this strategic choosing and shaping of our compassion. They show that the seemingly automatic nature of many of our emotional responses should not fool us into thinking there are no motivated choices at work in them; both can be the case when it comes to emotion.

The motivational model shifts attention from compassion per se (as in self-report measures) to the underlying motives of people who feel and act compassionately. It asks about their goals and intentions in various compassion-eliciting situations (Batson 2022). Focussing on these motives, it finds that in any given compassion-relevant situation, we are inevitably faced with a ‘motivational mix’. Motives rarely come alone. Indeed, since compassion-eliciting situations typically give rise to altruistic motivation to help, they also prompt a cost–benefit analysis of possible courses of action to do so. This inevitably brings in further motivations that can conflict with and even override the altruistic motive. As Cameron and Rapier explain: “situations that tend to elicit compassion often activate competing goals, such as avoiding financial cost, avoiding exhaustion and fatigue, maintaining well-being, maintaining moral self-image, and abiding by social norms” (Cameron and Rapier 2017, p. 378). We are usually dealing with multiple, competing motives, pulling us in the direction of different goals and potential courses of action to pursue them.

For example, if Greg is confronted with an image of a suffering child in a charity advertisement in his social media newsfeed, he may feel compassionate emotion and an altruistic motive to do what he can to help alleviate the child’s suffering. Genuinely valuing the child’s welfare as an end-in-itself, he may feel motivated to donate financially or to sign a petition. But, even with this altruistic motivation, there will be other motives in the mix. For example, Greg will likely also be egoistically motivated to minimise effort, whether physical, cognitive or emotional, and so to continue scrolling (Scheffer et al. 2022). Or, another motivation could enter in, if a higher order goal for Greg—one important to his identity and moral self-image—is to be a compassionate person who never makes excuses to avoid the pain of others, especially when no one else is watching. In that case, he will have another principle-based motive to help—one that may even trump his other motives to avoid the emotional and financial costs of feeling compassion in this situation.

What matters most, according to the motivational model, is which motive is dominant. What determines whether Greg chooses to feel compassion or not, as well as the intensity and duration of the emotion, is how he values and prioritises competing goals in the motivational mix. This is what the motivational model captures by describing compassion as an active choice or dynamic decision process. It is to say that “compassion requires control, defined as the active management of competing goals” (Cameron and Rapier 2017, p. 379). It refers to the way people assign value to competing courses of action, balancing the costs and benefits associated with each option based on their priorities.

It should be noted that proponents of this model insist that these compassion choices do not need to be conscious or deliberate (Cameron and Rapier 2017). People compute the value of various action paths for pursuing their goals extremely quickly, balancing costs and benefits without conscious attention, and they use these predictions to regulate their compassion based on their priorities.

The motivational model thus adds a significant new dimension to the analysis of the benefits and costs of compassion outlined in Section 2. It reveals that our cost–benefit analyses of compassion are far from a neutral calculus. They are determined by our top-level goals and values, how we prioritise them when they come into conflict with each other, and thus how much effort and cost we’re willing to endure in pursing compassion before deprioritising it and switching to other goals. Balancing the costs of compassion will “depend on the competing goals and priorities of the individual, in the interest of maintaining effective pursuit of long-term goals, values, and self-identities” (Cameron and Rapier 2017, p. 395). It also means, crucially, that changes in our motivations and how we prioritize compassionate motives can produce very significant changes in our compassion.

4. Implications: Motive-Based Interventions and the Role of Institutional Support

How does the motivational model of compassion explain cases where compassion appears to flag and fail in contexts of repeated caregiving, i.e., in the kinds of failures of compassion motivating the growing literature on compassion fatigue? Can this model better explain why compassion can be so hard to sustain and what can be done about it?

According to the fixed-capacity explanation, compassion fatigue occurs because, as a person shows compassion again and again, they experience more of the costs of compassion: material, cognitive, emotional, and so on. As these costs accrue, the explanation runs, carers deplete their capacity for compassion (or the processes that underpin it such as attention and imagery) and eventually come up against an inherent limit on how much they can give (Cameron 2017). Beyond this limit, compassion begins to ‘run out’ and eventually fails. What can be done? The fixed-capacity model tends to be somewhat fatalistic—once compassion is running low, there seems to be little that can be done except to pace oneself by taking breaks and engaging in forms of self-care that might help ‘refill the tank’.

The motivational model offers a different explanation of apparent cases of compassion fatigue—one which points in the direction of interventions to sustain compassion that go beyond individual self-care. The motivational model suggests that in contexts where people are engaging in repeated care, they begin to predict or anticipate greater costs associated with feeling compassion. For example, they may anticipate increased physical or emotional exhaustion, or they may predict greater challenges to upholding their sense of professional duty or their moral self-image if they fear they won’t be able to deliver compassionate care to all their patients. As a result of these anticipated costs, people make a pre-emptive choice to avoid or limit their exposure to compassion-eliciting situations.

The motivational model thus shifts the explanatory focus from experienced costs to anticipated costs; “the claim is not about experienced emotions in helping situations. Rather, it is about anticipated emotions” (Cameron 2017, p. 266). The motivational model explains that compassion sometimes fails not because people hit their capacity limit, but because anticipated costs motivate people to want to decrease or avoid compassion. As noted, people have a variety of direct and indirect strategies to bring about this kind of emotion-regulation, including not only pre-emptive situation-selection, but also cognitive reappraisal or even dehumanising the person in need (Cameron et al. 2016). What often looks like a limit on our capacity for compassion may in fact be a strategic choice to regulate our compassion in order to avoid its perceived costs.

It should be noted that in offering this alternative account, the motivational model is not claiming that choices to avoid compassion necessarily explain what is going on in all cases where compassion appears to flag and fail. It could be that both capacity limits and motivated choices play a role. Further study is needed (Cameron 2017). But initial evidence leads Cameron to argue that in at least some cases, failures of compassion “may be explained by motivational factors, rather than basic capacity limitations on how much compassion people are able to feel” (Cameron 2017, p. 266).

At this point, it is crucial to note an important distinction in considering the application of this explanatory model in healthcare and professional caregiving contexts. In many everyday situations, people’s choices to avoid compassion are made due to egoistic motives overriding compassion’s altruistic motive to help, as in the example of Greg above and his desire to avoid effort costs. However, motivated choices to avoid compassion can also be adaptive, made with the ultimate altruistic aim of helping others still firmly in view. HCPs are typically highly motivated to feel compassion and have strong altruistic motivation to care for their patients, seeing this as core to their professional identity, vocation and moral self-image (Parkes et al. 2010). However, where HCPs anticipate significant costs, they too are likely to make pre-emptive choices to defensively regulate their compassion to be able to sustain it. For example, an intensive care nurse may perceive the costs of compassionate care while working in an overburdened hospital during a pandemic will be steep, impacting her physical, psychological, emotional, and spiritual wellbeing. Her choice to sometimes avoid compassion-eliciting contexts for short periods or reduce compassionate engagement with some patients may be a wise, professional strategy, ensuring she can cope and pace her compassion over the long haul, and ultimately allowing her to help more patients.

Yet, while this kind of self-protective, coping strategy makes sense on an individual basis in challenging contexts, I argue that it should not be generalised into the sole model guiding the institutional support of compassion. Providing time-out for staff or seminars on self-care are good initiatives, but alone they end up placing the burden on the individual carer to ‘take better care of themselves’ in order to protect their ‘reserves’ and ensure they can weather the purportedly inevitable costs of compassion without becoming exhausted or burnt out. Again, while this may be an excellent remedial strategy for individuals in intense caring contexts, it won’t address the underlying motives that sustain compassion. Institutions must do more. The final section of this article considers one promising example for institutional support from the palliative care context.

How then does the motivational model help in explaining cases where compassion appears to fatigue and fail? I suggest the motivational model has three advantages: (i) it helps explain why interventions are possible and reminds us of the plasticity of compassion; (ii) it illuminates new avenues for motive-based interventions that strategically aim at underlying motives to feel compassion rather than remedial self-care; and (iii) it shows that sustaining compassion is not up to individuals alone but requires social and institutional support.

First, if motivational factors and choices to avoid compassion are playing a role in some apparent cases of compassion fatigue, then change is possible—interventions to sustain compassion are possible. On the fixed-capacity model, this seems much less certain. There is a concerning note of fatalism regarding the inevitability of the costs of compassion for individual carers. Remedies tend to focus on coping within one’s limits using a range of self-care and resilience strategies not directly related to compassion itself. But this emphasis on inherent limits conflicts with empirical research highlighting the plasticity of our emotions and the many variables that can affect the strength and scope of our compassion. It is also important given research has shown that our attitudes about the malleability of our emotions can influence the effort we put into cultivating them (Schumann et al. 2014). Narratives of compassion fatigue, which “suggest that being a healthcare provider inherently depletes a person’s innately limited capacity for compassion” can thus create a self-fulfilling prophecy, and “in some veins, has reinforced a culture of explicit distancing or suppression of compassion” (Simon-Thomas and Goetz 2017, p. 12). This is a rather different attitude to compassion than many traditions have relied upon, including palliative care. The motivational model provides an evidence-base for seeing compassion as a choice and something we can, with effort, intentionally cultivate and extend beyond its seeming limits (Cameron 2017).

Second, the motivational model illuminates the type of interventions likely to be particularly effective. On the motivational model, the key to overcoming failures of compassion lies in “changes in motivation rather than in underlying capacities or abilities to feel compassion” (Cameron and Rapier 2017, p. 396). Changing people’s motivations to feel and act compassionately can change compassion outcomes. Thus, rather than assuming failures of compassion are due to people reaching their limit, it directs us to investigate the goal-directed motives at play in specific contexts, such as in palliative care. Interventions work, according to the motivational model, by intervening in this ‘motivational mix’ and seeking to tip the cost–benefit analysis in the direction of choosing compassion. Motivational or motive-based interventions are thus vital for increasing and sustaining compassion (Cameron 2017; Weisz and Zaki 2017).

Third, the motivational model helps us see why failures of compassion cannot be rectified by individuals alone but requires institutional and social support. The motivational model pushes researchers to reckon with a broader range of factors shaping compassion-relevant situations than have typically been countenanced. Institutions play a crucial role here too in shaping people’s perceptions and fears about the costs of compassion in their specific context. In the next section, I consider some examples of social and institutional support for compassionate care.

The question of which motive-based interventions are most effective is an empirical one, and one that provides a promising direction for future research. Research in palliative care on sustaining compassion should investigate the specific ‘motivational mix’ palliative care HCPs face in their context, which will be distinct from that faced in everyday life as well as in other healthcare contexts. Much could be learnt from studying carers’ own perceptions of the costs and benefits of compassion: when and why they choose to approach or to avoid compassion. What motives and priorities are shaping how they balance these perceived costs and benefits, material, cognitive, emotional, moral, and spiritual? Such research would provide a welcome addition to the many studies employing the self-report measures of the ProQOL. It would provide a richer understanding of the kind of motive-based interventions that could help support HCP motivation to choose compassion.

This future research could also draw on the growing psychological literature on motive-based interventions. Several potential interventions, both short-term and long-term, have been suggested for further study in applied contexts (Cameron 2017; Weisz and Zaki 2017). Long-term interventions include, for example, mindfulness meditation practices, compassion cultivation training (CCT) and, in the next section, I explore spiritual care training. Among short-term interventions suggested are those designed to shift people’s perceptions about the costs of compassion. For example, one would be to “activate moral goals or the moral self-concept” prior to periods of intense compassionate caring (Cameron 2017, p. 268). This may strengthen motives to approach compassion in order to live up to one’s professional or moral self-identity, overriding competing motives to avoid its costs. Another short-term motive-based intervention would be attempts to change perceived group norms about compassion (Weisz and Zaki 2017). Research from the Common Group Identity Model (Dovidio and Gaertner 2014) suggests that if people learn an in-group to which they belong values compassion, this could shift their perception of the costs of compassion, strengthening intrinsic motivation to feel compassion and increased helping effort. Shifting perceived group norms in this way could be relevant with respect to both social and institutional norms about compassion.

5. The Example of Spiritual Care Education

To further explore the implications of a shift to motive-based interventions and the role of institutional support, in conclusion I offer some preliminary examples of social and institutional support for compassion, focussing on spiritual care education in palliative care.

One extraordinary recent example of the power of shared social norms can be seen in a study of nurses during the COVID-19 health crisis in Spain, which found higher ‘compassion satisfaction’ in nurses during this period (Ruiz-Fernández et al. 2020). What is interesting for our purposes is the motive-based reasons the authors hypothesize for why this might be the case and the link they posit between “[t]he greater visibility of nurses, their motivation to relieve suffering and social recognition” (Ruiz-Fernández et al. 2020, p. 4322). As the authors explain: “Somehow, in the face of the unique circumstances brought about by this pandemic, nurses have been able to connect strongly with their own intrinsic motivation for caregiving and have gained satisfaction from compassion through actively committing to patients” (Ruiz-Fernández et al. 2020, p. 4327). In contrast to the often-invisible status of caregiving work in society, during the pandemic the motivation to alleviate suffering felt by HCPs was echoed by a broader social recognition of the value of their professions and the importance of compassionate caregiving in situations of deep suffering. They authors note: “during the hardest weeks of the pandemic in Spain, as data were being collected for this study, a social movement of support and recognition for the work and effort of health professionals in general and nurses in particular was taking place, which was not the case prior to the pandemic” (Ruiz-Fernández et al. 2020, pp. 4327–28). Nurses’ work was visible, their values and motivation shared, and the daily risks and sacrifices they were making were honoured, as society reacted to their work with widespread public expressions of gratitude. Ruiz-Fernández and colleagues hypothesize that this broader social support may be key for sustaining HCP motivation to feel compassion in challenging contexts.

Other research has investigated how institutional support could play a role in shaping the ‘motivational mix’ of HCPs. For example, a study on critical care nurses found that meaningful recognition, especially in formal recognition programs, was a significant predictor of increased ‘compassion satisfaction’ and that its value for decreasing burnout and supporting HCP wellbeing “cannot be overstated” (Kelly and Lefton 2017, p. 443). Again, of interest is the link to motivation. It may be that like the case of social support above, formal institutional recognition of HCP’s contributions to care could reinforce their intrinsic motivation to feel and act compassionately. There are, of course, many other ways institutions could provide this recognition and valuing of compassionate care—both internally, for example, in what they reward financially, celebrate, model, and make salient, but also by advocating for changes in external structures, such as healthcare funding models.

In addition to shifting perceptions about the costs of compassion, institutions have the capacity to direct support towards long-term motivational interventions. One promising example can be found in palliative care’s turn to spiritual care education and calls for institutional support for all HCPs to undertake such training. From its beginnings, palliative care has led the way in recognising the importance of spirituality in healthcare and how emotional and spiritual distress can exacerbate patient suffering (Saunders 2006). There is broad consensus that spiritual care is integral to quality, patient-centred palliative care and improves quality of life for those facing life-threatening illness (Mesquita et al. 2017). Research has found that spiritual wellbeing is closely associated with a range of positive healthcare outcomes for patients (van de Geer et al. 2017).

It should be noted that the term ‘spiritual’ here is used in a broad and multi-dimensional sense, encompassing existential challenges (e.g., meaning, purpose, identity) and value-based issues (e.g., ethics, character, significant relationships), as well as religious concerns (e.g., faith, relation with God or the sacred) (Best et al. 2020). Spirituality captures a person’s sense of connectedness, for example, to self, others, nature or the divine. Following a bio-psycho-social-spiritual model (Sulmasy 2002), spiritual care emphasizes the importance of this spiritual dimension in caring for patients. It calls for assessment of patient needs to include spiritual needs and the provision of support to address those needs (Best et al. 2020).

While spiritual care provision has sometimes been the domain of trained experts such as chaplains or pastoral care staff, it is increasingly seen as something all HCPs on the multi-disciplinary team require training to develop. The interprofessional spiritual care model holds that “each clinician is responsible to approach the patient as a whole person and to provide relational, dignity-based, compassionate care” (Balboni et al. 2014, p. 1588), assessing in an integrated way the patient’s physical, emotional, social, and spiritual well-being. This, in turn, has led to calls for healthcare organisations and educational institutions to provide concrete support for the development of HCP’s spiritual care competencies.

The first key element identified as crucial for spiritual care education by the European Association for Palliative Care (EAPC) spiritual care reference group is the “development of the reflective capacity of staff to consider the importance of spiritual dimensions in their own lives” (Best et al. 2020, p. 3). While HCPs often cite a lack of time or training as a key barrier to the provision of spiritual care, research has found that “[s]eeing oneself as not spiritual or only slightly spiritual is a key factor demonstrably associated with not providing spiritual care” (Bar-Sela et al. 2019, p. 346). Thus, alongside a focus on screening tools and other communication prompts to assist HCPs in engaging with patient spiritual needs, there has also been a growing recognition of the importance of HCP’s self-reflection on their own holistic wellbeing. HCP’s personal awareness of their spiritual needs facilitates competence and confidence in recognising the spiritual needs of patients, as well as being associated with reduced burnout (Best et al. 2020). A key recommended feature of HCP spiritual care education is that it should help “participants to consider their own spirituality and the role it plays in their personal and professional lives” (Bar-Sela et al. 2019, p. 346).

Given the broad definition of ‘spiritual’ or existential wellbeing employed here, spiritual care training programs promote self-reflection across a range of areas related to HCP’s deepest values, goals, and priorities. In this sense, as outlined in Section 3, spiritual care education is an intervention targeted at the motivational level. Sensitivity to one’s spiritual wellbeing has the potential to shift people’s top-level motives and how they anticipate the costs of various courses of action before them, including choices to feel and act compassionately. As Best and colleagues note, regularly practicing self-awareness of spiritual issues in one’s own life “can help the healthcare practitioner to avoid being distracted by their own fears, prejudices, and restraints and attend to the patient and his/her family” (Best et al. 2020, p. 1). This analysis maps the discussion in Section 4 of the way anticipated costs shape our compassion. As Cameron hypothesises with respect to the success of other long-term interventions such as mindfulness meditation, they “may lead people to become more aware of anticipated emotions in mass suffering contexts and to be less afraid of the consequences of compassion” (Cameron 2017, p. 268). Likewise, with compassion cultivation training (CCT): “By encouraging focus on the present moment and the active cultivation of compassion, these programs are likely to clarify and dispel motivations to avoid compassion, and illuminate the benefits of extending compassion to others” (Cameron and Rapier 2017, p. 396). Spiritual care education programs which support staff to reflect on deeper aspects of their own lives could work in a similar way, sharpening awareness of how they are balancing the costs and benefits of compassion. They could help HCPs strengthen higher-order goals and motives they want to prioritise and mitigate fears about anticipated emotional costs, potentially increasing motivation to choose and sustain compassion in difficult caring contexts. Initial evidence supports a link between spiritual care education, staff spiritual wellbeing and compassion, finding significant and sustained improvements in compassion for the dying and compassion for oneself (Wasner et al. 2005).

A final essential feature of spiritual care education is that it must be institutionally supported. If spiritual care is vital to palliative care and increasingly seen as part of the role of each member of the interprofessional team, then both educational institutions and employers have a responsibility to foster spiritual care competencies. HCPs must be provided with time and space to undertake this training and integrate it with their professional practice. Staff need concrete institutional support to develop a “habit of practising personal spiritual reflection as a professional requirement” (Best et al. 2020, p. 4). This institutional support must be of a different kind than the optional time-out or self-care activities sometimes provided, where the remedy is typically individualised and left up to HCPs themselves to apply lessons about how to take better care of themselves, largely outside of work. By contrast, spiritual care education is intrinsically connected to the professional role and undertaken in the workplace and team settings. Institutions must also be open to two-way dialogue about emerging insights developed through spiritual care education. Facilitating awareness about what matters most to HCPs should extend to reflection on the broader organisational context, values, and pressures. As Paal and colleagues contend: “Implementation of spiritual care helps challenge the hidden values of healthcare institutions… spiritual care training involving the whole team may be helpful to reset the institutional agenda” (Paal et al. 2015, p. 28).

To the extent then that spiritual care education is institutionally supported and relates to palliative care HCPs’ professional identity and their own deeper priorities and motivations to deliver compassionate care, it presents a promising example of an institutional, motive-based intervention to sustain compassion and wellbeing—one which could also benefit other healthcare areas. By focussing on HCPs’ own spiritual needs and fostering their integrated wellbeing, spiritual care education targets underlying motives and can help uncover judgments and fears—often accurate—about the anticipated costs of compassion in challenging care contexts, opening them to broader institutional awareness, scrutiny, and responsibility. Combined with other forms of social and institutional recognition of compassionate care, such interventions could profoundly reshape the ‘motivational mix’ for palliative HCPs, tipping the cost–benefit analysis in the direction of compassion and thus supporting HCP motivation to feel and act compassionately.

6. Conclusions

This article contributes to the debate about moving beyond compassion fatigue in the healthcare literature by bringing to bear new psychological research on the motivational model of compassion and examining its implications for palliative care. The advantages of this conceptual model for illuminating motive-based interventions have been demonstrated, leading to the contention that sustaining compassion is not the responsibility of individuals alone and institutions must go further than supporting self-care activities. The article examined preliminary examples of broader social and institutional motive-based interventions, with a focus on the case of spiritual care education. The study of such interventions provides a promising direction for future research in palliative care on sustaining HCP compassion. Future studies should investigate the specific ‘motivational mix’ palliative HCPs face and their perceptions of the costs (material, cognitive, emotional, moral, spiritual) of compassion in this specific context; studying when and why they choose to approach or avoid compassion. Research examining how HCPs balance these perceived costs and benefits would provide a welcome addition to the many studies employing the self-report measures of the ProQOL and a richer understanding of how institutions can sustain compassion and wellbeing in palliative care.