INTRODUCTION
The true impact of empathy on the therapeutic relationship between clinician and pain patient is poorly understood. It is important to discuss this relationship and the potential role it plays in the widespread phenomenon of the undertreatment of pain. From a clinical and psychological perspective, there are multiple pragmatic and psychological factors that argue against the ability of the health care provider and the patient to relate to one another in a way that allows the clinician to truly understand the suffering of the patient. If clinicians were able to fully become in tune with this, it is hard to argue that the problem of undertreated pain would be of the magnitude that it has been and unfortunately continues to be. We feel that doctors, nurses, and others involved in patient care will ultimately be better served if they simply learn to accept the fact that they are unable to rely solely on their empathy. Even if we assume that these clinicians are good, caring people with genuine motivations for helping patients, this is not sufficient to make them competent in the treatment of pain. Ultimately, the use of rating scales and aids that facilitate the objective measure of pain, and thereby communication about it, are the only hope in allowing for better, more empathic pain care. Thankfully, a growing number of such instruments are becoming available, such as the Pain Assessment and Documentation Tool (PADT; Passik et al., 2004, 2005), CAGE (Beresford et al., 1990), Screening Tool for Addiction Risk (STAR; Friedman et al., 2003), Drug Abuse Screening Test (DAST; Gavin et al., 1989), Screener and Opioid Assessment for Patients with Pain (SOAPP; Butler et al., 2004), and the Opioid Risk Tool (ORT; Webster & Webster, 2005), among others. In addition to the question of empathy and its impact, we cannot forget that the souring of the regulatory and legal climates surrounding pain management creates fear, and fear widens the gulf between doctor and patient (Passik & Kirsh, 2006).
When trainees begin medical school, nursing school, or other allied health disciplines, one of their primary motivations for choosing their career path is the desire to help people. Indeed, a recent study showed that early education medical students scored high on measures of attitudes toward social issues in medicine and wanting to make an impact in this area; unfortunately, these attitudes were shown to diminish over the course of their training (Woloschuk et al., 2004). What aspect of intervening in the care of another human being pertains more to the desire to help and make a difference in patients' lives more readily than treating pain? When young medical trainees first enter their clinical rotations they are psychologically very close to patients and may actually overidentify with them. A study of the content of the nightmares experienced by trainees (Marcus, 2003) revealed that those at the beginning of training often find themselves in the patient role in a nightmare, such as being operated on without anesthesia. Medical education provides the necessary distance to allow for empathy and perspective taking, as opposed to actually feeling the pain of the patient. This distance is probably necessary to allow physicians to do what they must do to other human beings in situations where, if there was too much emotional investment, perhaps it would be impossible to perform painful procedures. Thus, distance begins to develop, and by the end of training, nightmares now more commonly put the trainees in the physician role, such as operating on someone without anesthesia. But is this distance a help or potential hindrance when they are called upon to treat pain?
The literature reports that medical treatment tends to be selective and preferential depending upon the age, race, and gender of the patient. For instance, older, non-White females have the highest likelihood of being undertreated for their cancer pain (Cleeland, 1998; Cleeland et al., 1997). In addition, Hispanics are half as likely to receive pain medications in emergency rooms when they have the same long bone fractures as Whites (Todd et al., 1993). Given these findings, are we to believe that medicine as a whole is ageist, racist, and sexist? Or, that in AIDS patients (Breitbart et al., 1998), being uneducated is a risk factor for poor pain care (along with a history of a substance abuse)? Or, is it possible that in fields historically dominated by younger, educated, White men, that being different from the physician works against the patient somehow and decreases the likelihood of their ability to empathize with suffering? It is feasible that more distance is created, along with a diminished ability to empathize with the plight of the patient. And what does a young, healthy person who has never suffered with severe chronic pain know about their patients' pain anyway? Would it not be better to publicize the fact that clinicians are human and prone to inherent automatic thoughts and feelings that might impact the delivery of care? Getting this issue out into the open could then be used as a rallying call to have clinicians pay even closer attention to these potential biases.
These issues have perhaps been best studied in cancer. That the undertreatment of pain is a problem in oncology, where no one disputes that there is a moral and ethical imperative to provide aggressive pain management and little disagreement that opioids are the cornerstone of care, only bespeaks the fact that the problem of undertreatment is even more profound in nonmalignant pain. In cancer clinics, studies have been done to examine how well oncologists and oncology nurses can intuit their patients' suffering. In studies of the agreement of patients' self-report with reports given by their professional caregivers about their estimations of the patients' pain (Grossman et al., 1991), depression (Passik et al., 1998; McDonald et al., 1999), and overall quality of life (Fisch et al., 2003), all of the agreement is always at the low end of the intensity of the symptom. In other words, as long as patients reports minimal levels of pain, depression, or dysfunction in quality of life, their physicians and nurses agree with them. But, when the problems become more intense, the agreement falls off dramatically. Thus, there is a marked tendency to underestimate the suffering of patients in many different facets.
This situation is clearly a breakdown in the empathic relationship of the provider and the patient. Therefore, we need to investigate and describe the factors that mitigate empathic pain care from the pragmatic to the psychological. Where the latter is concerned, we will examine both conscious and unconscious factors.
PRAGMATIC FACTORS
There are many pragmatic factors that are germane to the issue of detracting from the promotion of empathy for people in pain. They can be roughly categorized into system-related, patient-related, and professional-related barriers.
System-related barriers to the promotion of empathic understanding include the oft-cited time pressures created by the short duration in which the physician and patient are in the room together (Mechanic et al., 2001; Ohtaki et al., 2003). This also includes reimbursement issues that disenfranchise the physician from treating pain and thereby lower attention to and interest in pain management (Balkrishnan et al., 2002; Joranson, 1994). Finally, regulatory pressures have created a hostile and adversarial environment around pain issues, and many clinicians feel cornered into acting as more of a police detective than a caregiver to their patients (i.e., feel the need to keep a hypervigilant eye out for addicts and diverters who might get one over on the physician and subsequently cause him/her to lose his/her medical license or be otherwise punished for treating pain patients; Passik & Kirsh, 2006). All of these barriers impede communication and thereby empathy.
Patient-related barriers include the multiple fears that patients harbor that inhibit their aggressively and accurately reporting of pain and suffering to their physicians. Patients may fear becoming addicted to pain medication. They might also fear the perception of being seen as a bad patient/complainer or that discussing pain might distract the physician from the treatment of their primary disease. Finally, they may not want to acknowledge pain issues for fear that it may represent progression of disease (Ward et al., 1993). These barriers lead to inhibited communication about pain, which in turn fails to provide the physician with the building blocks for empathy and concern.
Physician-related barriers, such as the failure to acquire adequate knowledge of pain assessment and management and the fear of regulatory oversight, were discussed above. Below, we would like to explore unconscious mechanisms and content that has not been heretofore related to the problem of undertreated pain to the best of our knowledge. In the end, such nonrational and unconscious issues may have more detrimental impact than has previously been acknowledged.
UNCONSCIOUS PROCESSES: THE MECHANICS OF (UNEMPATHIC) JUDGMENT
Cognitive and social psychologists (Gati & Tversky, 1984; Kahneman, 2003; Kahneman & Tversky, 1996; Johnson-Laird et al., 2004; Redelmeier et al., 1995) have described numerous unconscious aspects of how humans make automatic judgments that are out of awareness. For example, when four items are randomly placed in four different positions, the item in the third position is preferred and chosen an inordinate number of times. This is not something people are generally aware of, yet it colors the perceptions of quality and preference. In a disturbing study on the effects of race, subjects played a video game in which they had to decide to “shoot or not shoot” a series of Caucasian and African-American characters that would appear either holding guns or other objects in complex backgrounds (Correll et al., 2002). They found that participants, regardless of their own race or statements regarding stereotypes or prejudice, were more likely to “shoot” an unarmed target when the target was African-American. The study concludes that these split-second decisions were based on automatic thoughts triggered by prior exposure to negative stereotypes in our culture.
To what extent do physicians' judgments of pain and suffering in their patients, so important to the ability to empathize, fall victim to such mechanical aspects of the way humans think and make judgments? Are patients thought to be in pain or not, to require attention or not, because of processes that go on out of awareness for the person making the assessment? When one hears that there is a consistent inability to match the patient's assessment of their pain, depression, or quality of life, one might come to believe that physicians might be using a prototype of “what an outpatient with cancer feels” rather than what the individual patient sitting in front of them actually feels. This would be a fruitful avenue for further research.
Social psychologists have identified one such unconscious process that would seem most relevant to this discussion, namely, the observer–subject bias (Haro et al., 2006). When one person looks at another's behavior and is asked to make a judgment about why that person is acting the way that he/she is, the observer is likely to posit an internal attribution for the behavior (often termed the fundamental attribution error). As an example, when you view someone get into a minor auto accident, you tend to view that person as a poor driver or unsafe. On the other hand, when people are asked why they themselves are behaving in the way that they are, they tend to posit external attributions (often termed the self-serving bias). Going back to the driving example, when people get into a minor accident, they will tend to first blame the faulty brakes on the car, the wet road conditions, the squirrel that ran in front of them, etc. Thus, when a physician observes a patient exhibiting pain behaviors, he/she may be more likely to posit an explanation based on character (Mr. Smith is a somatizer) than a situational one (Mr. Smith's disease is progressing and causing more pain). Again, such a bias goes on out of awareness, and physicians are unlikely to be aware of the fact that it is influencing the judgments that they are making about poor Mr. Smith's pain.
UNCONSCIOUS CONTENT
Is there unconscious sadism and hostility toward people in pain harbored by clinicians that impede pain management? In some of the earliest papers on the undertreatment of pain, the insightful and brutally honest Sam Perry (1984, 1985) believed this to be the case. Are the pain patients who are nonresponsive to our ministrations actually thwarting our desires to help? Does this lead to engendering anger and sadistic impulses? Are patients who are seen to be “bringing their problems onto themselves,” such as addicts and obese people, deserving of scorn? If not, how do we explain the callous treatment they sometimes receive (i.e., carrying out painful procedures without the provision of adequate analgesia)? In a classic paper on “hate in the countertransference” D.W. Winnicott (1960) examines how patients who are depressed and self-destructive (like so many people in chronic pain) engender unconscious hate in their caregivers that can drive the patient into deeper and deeper despair. Do people who have been the victims of abuse and neglect (like so many people in chronic pain) manage to unconsciously and unwittingly engage us in faulty caregiver scenarios that perpetuate more of the same? This process has been called projective identification by the psychoanalysts and is as germane to the care of people with pain as it is to psychoanalytic treatments of nonpain patients (Rizq, 2005; Waska, 2006; Yahav & Oz, 2006). Yet psychoanalysts continually involve themselves in introspection and their own psychotherapy and supervision to examine themselves for such tendencies. Would all of us who treat patients benefit from doing the same? Would the utilization of introspective reflection on the part of pain management clinicians bring about awareness of unconscious impediments to effective care and aid physicians in providing appropriate and individualized treatment for their patients? Consider the utility of self-assessment prior to the patient–physician interaction and again with a simple exercise midway during the interaction (see Table 1).
Reflection questions prior to interaction with a patient and suggested exercise during the interaction
A pain case example may help to elucidate the potential errors in utilizing these cognitive and psychodynamic processes in clinical practice. When clinicians encounter evidence of noncompliance in pain therapy, especially where opioids are concerned, a warning flag must be raised and the issue must be addressed. However, care must be taken when conceptualizing the problem. For many clinicians, the first assumption is that the patient might be an addict. However, impulsive drug use may indicate the existence of a psychiatric disorder, diagnosis of which may have therapeutic implications. Patients with borderline personality disorder can express fear and rage through aberrant drug taking and behave impulsively and self-destructively during pain therapy. Hay and Passik (2000) reported a case in which one of the more worrisome aberrant drug-related behaviors, forging of a prescription for a controlled substance, was an impulsive expression of fears of abandonment, having little to do with true substance abuse in a borderline patient. Such patients are challenging and often require firm limit setting and careful monitoring to avoid impulsive drug taking. A snap judgment of addiction might have forced the clinicians to act rashly (e.g., discharging the patient from the practice), but what good would this ultimately serve? The clinicians might feel empowered for a short while after they “caught a bad one,” but this feeling is usually short-lived and can eventually lead to burnout regarding engaging in pain management at all. And what of the patient? “Kicking them out” of a pain practice usually means they end up as someone else's problem, and the initial issue (in this case, borderline personality disorder) never gets addressed. Indeed, in this case, taking a strong-arm approach with the patient by assuming addiction would likely have just added to the ongoing replay of psychiatric trauma (i.e., cycle of approach–avoidance and abandonment by others) hallmarked by the disorder.
CONCLUSIONS
We truly believe that the problem of undertreated pain can be solved. To do so, we need to address the medical and legal climates and the realities of a clinical situation that detract from empathic care. As was discussed above, these realms are intricately tied to one another. In the end, professionals will need to accept the fact that, although they are caring people, there are so many barriers to the treatment of pain and the provision of empathic care that they simply cannot be overcome flying by the seats of our collective pants. We will have to accept our limitations and then work to overcome them with technological and educational initiatives that promote communication and empathy, such as screening tools and other aids.
