CHAPTER 104 Psychologic Strategies for Chronic Pain
“Pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage or described in terms of such damage.”
— International Association for the Study of Pain 1973 consensus definition of pain
The lifetime prevalence of back pain is essentially universal; excruciating, agonizing, unbearable, torturing back pain is not. This chapter attempts to elucidate those factors that help contribute to the transformation of back pain into a disabling condition.
Pain is not what occurs at the periphery; it is what the brain perceives, and it is indisputably modifiable by emotions and beliefs. Actual damage is neither necessary nor sufficient for the perception of pain. Anger, depression, anxiety, fear, and other psychologic variables can all increase the perception of both acute and chronic pain, as can believing it to be an indicator of a destructive process.
Back pain is an enormous problem for patients, health care providers, families, employers, and society.1–3 In 2006 an estimated 2.3% of all physician visits, which exceeds 20 million, were attributable to back pain. Most patients seeking care from a spine specialist do so because of pain.4 Because the natural history of most spine pain is self-limiting,5,6 almost anything done will lead to the patient reporting a decrease in symptomatology in a matter of days or weeks (with the exception of bed rest, which has now been well shown to do more harm than good).7 Most patients with back pain do not need to see a spine specialist; rather they are seen by their primary care physicians. When it becomes persistent, it is commonly not attributable to any specific pathology or disease process.8 Despite the fact that this has been well documented for more than a decade, the idea that nonspecific low back pain (LBP) must result from demonstrable pathology persists in the mind of patients and is often fostered by providers. This belief in and of itself may lead to the worsening of the pain. Patients who do go on to see a specialist often do so because the pain has persisted beyond the time of expected spontaneous resolution, and they are worried. Of those patients whose pain does not resolve within a relatively short time, some will go on to develop disabling chronic pain.
The societal and individual costs of chronic debilitating pain are staggering—2003 estimates show 149 million lost work days because of pain, at a cost of approximately $19.8 billion per year. It is likely that prompt recognition and intervention in cases with developing chronicity can lead to improved outcome with less need for extensive interventions.
Back pain is often ascribed to strains, sprains, annular tears, internal disc disruption, facet arthropathy, or bone pathology; however, it is often not explained by examination or imaging. Positive findings on an imaging test may be misleading because patients with severe symptoms often have normal imaging, and patients with herniations, degenerated discs, bulges, osteophytes, and facet arthropathy are often without symptoms.9–11
To some extent, over-reliance on imaging findings derives from the persistence of obsolete concepts concerning nociceptive pain. Essentially, these implied a more or less linear relationship between pain perception and peripheral stimulation (i.e., a nociceptor is activated, the signal is transmitted to the dorsal horn [DH] of the spinal cord, and from there via the thalamus to the cortex, where pain is appreciated). Pain was seen as an analog representation of some event (e.g., a child stepping on one’s toe produces minimal pain, whereas an adult or an automobile would produce correspondingly greater pains). As a result, when a patient complained of severe pain and no appropriate pathology was found, the validity of the complaints was challenged.
Pain Perception: Nervous System Attenuation and Amplification
More recent evidence shows that pain is a creation of the nervous system and not just a gauge of nociceptor activation. Nociceptive afferent signals are subject to marked attenuation and amplification by descending facilitatory and inhibitory tracts that have their action at the dorsal horn (DH).12 Further, the presence of prolonged nociceptive stimulation, inflammation, or nerve injury can lead to sensitization of the neurons that relay pain, death of inhibitory cells,13,14 loss of tonic inhibition, and structural neuroplastic changes. Perhaps even more interesting, activation of immune cells including glia,15 which were previously thought of as having only structural roles, produces exaggerated, widespread, and mirror image pains.16–18 Patients with idiopathic chronic low back pain (CLBP) subjected to quantified thumb pressure report more pain and show more functional magnetic resonance imaging (MRI) activation in brain areas likely to reflect pain perception than do controls, suggesting that at least some portion of CLBP is related to central sensitization. Evidence also implicates central sensitization as a significant factor in whiplash-associated pain.19 Thus spine pain can result from local tissue pathology, central sensitization, or both. It is therefore unrealistic to expect that reports of chronic spine-related pain will necessarily correlate with the presence of severity of spine pathology.
Guarding against the possibility of pain, as well as anticipation of its occurrence, activates cells in the rostroventral medulla that function to amplify incoming pain signals at the level of the DH. Animal models suggest that the simple facts of anticipating a pain and expecting it to be important are sufficient to trigger these “on cells,” in essence activating the “amplifiers” before the pain stimulus has begun.20
Increasing evidence points to genetic variability in pain appreciation and in responses to endogenous and exogenous opioids.21–23 Furthermore, there is compelling evidence that individuals reporting high/low pain in response to a standard stimulus demonstrates correspondingly high or low activation of somatosensory cortex, anterior cingulate gyrus (a likely index of affective components of pain), and frontal cortex.24 The conclusion is that those who report unusual pain actually experience unusual pain, at least in the absence of incentives for misrepresentation, and that “excessive” complaints of pain may relate more to neurologic and genetic traits than to characterologic ones.
Pain and the Psyche
Chronic pain syndrome (CPS) is a term (not a diagnosis) that has fallen into disfavor with pain specialists but is still often used by others. It describes a condition of severe intractable pain with marked functional impairment and other behavioral changes that have no clear relationship to organic disorder. (Poor concordance between chronic pain and structural pathology does not, as noted earlier, challenge the authenticity of the pain.) Typically, these patients have inordinate use of medications and health care services, which are largely nonproductive. Thus this is a nonspecific term for patients most typified by abnormal illness behaviors, primarily those of somatic preoccupation and regression into the sick role. The term is useful, in that it properly directs therapy toward the reversal of regression and away from an exclusive focus on nociception. It does not, however, substitute for a careful diagnosis of the physiologic, psychologic, and environmental factors that produce the syndrome.
Long25 studied more than 4000 patients with LBP and sciatica and more than 2000 with “chronic pain syndrome” and concluded that the primary determinant of vocational disability was the psychiatric status of the patient before the onset of the symptoms.
Carragee and colleagues26 followed 100 patients with mild CLBP and no prior spine-related disability for 5 years. Moderate or severe Modic changes (degenerative changes noted on spine MRI) of the vertebral endplate were the only structural variable that weakly predicted adverse outcome. Provocative discography and baseline MRI predicted no outcome variables but were weakly associated with pain episodes. Psychosocial variables strongly predicted long- and short-term disability and health care visits for LBP. A model based on scores on the Modified Zung Depression Test, Modified Somatic Pain Questionnaire, Fear Avoidance Beliefs Questionnaire (physical activity subscale), and smoking status identified 100% of long-term disability subjects, 88% of all disability subjects, and 75% of subjects having a remission.
It is reasonable to posit a stress-diathesis model in which the degree of disability from a given degree of organic pathology will vary with the psychologic reserves of the individual, the stresses of the workplace, and incentives/disincentives for recovery. Clearly these variables overlap—the person with poor coping skills and limited education is unlikely to obtain the most desirable work situation.
Depression, Anxiety, and Anger
The most frequent psychiatric illnesses (excluding somatoform disorders) in pain center patients are anxiety disorders, depression, and substance abuse. In 200 CLBP patients entering a functional restoration program, Polatin and colleagues27 found that 77% of patients met lifetime diagnostic criteria and 59% demonstrated current symptoms for at least one psychiatric diagnosis (excluding somatoform disorders). The most common were those listed. Fifty-one percent met criteria for personality disorder. Substance abuse and anxiety disorders appeared to precede CLBP, whereas major depression could either precede or follow it. Studies vary as to the prevalence of psychiatric disorder; however, they tend to agree about those that are most common.
Estimates of the prevalence of depression in chronic pain patients range from 10% to 83%. This extreme variance reflects variable settings, populations, and diagnostic criteria. In a Canadian general population survey of 118,533 people, CLBP was present in 9%. Major depression was present in 5.9% of those without pain and in 19.8% of those with CLBP. The rate of major depression increased in a linear fashion with pain severity.28 It is likely that the arrow of causality can point in either direction because there is evidence that pain predicts depression and depression predicts pain, and to similar degrees.29
In a probability sample of 5692 U.S. adults, 35% of those with CLBP had comorbid mental disorders. Major depression was present in 12.6%, dysthymia in 5.6%, any anxiety disorder in 26.5%, and any substance use disorder in 4.8%. There was no increased prevalence of (nonalcohol) drug abuse.30
Major affective disorder can present with pain, in which case treatment of the mood disorder often provides relief. More commonly, however, depression appears as a consequence of pain, though not necessarily a direct result of it. Rudy and colleagues29 showed that the link between pain and depression could be mediated by perceived life interference (loss of gratifying activities) and loss of self-control. Moreover, Strigo and colleagues,31 in a study of the association of major depressive disorder and experimental pain, observed that anticipation of pain was associated with increased activity in the amygdala, anterior insula, and anterior cingulate cortex in patients with major depressive disorder when compared with normals. This suggests that depressed patients experienced an affective response even before they experienced the painful stimulus. This was also associated with greater perceived helplessness. They posit that patients with a major depressive disorder have an altered functional response within specific neural networks during the anticipation of pain that may lead to an impaired ability to modulate the painful experience, as well as their emotional response to the pain.
There seems to be a vicious cycle in which pain behavior, loneliness, inactivity, helplessness, depression, withdrawal, loss of reinforcers and distractions, inactivity, and pain are mutually reinforcing. Improving one element in this series often benefits the others. These issues, of course, are not resolved by pharmacotherapy but do respond to successful rehabilitation.
Anxiety adversely affects pain through a number of mechanisms, and it can be the major reason for failure of rehabilitation. Phobic processes can promote a cycle of unnecessary self-protection, leading to deconditioning. When people became afraid to move, disability and dysfunction can result as much from unwarranted fear as from the pain itself. Anxiety can also lead to muscle guarding and tension that lead to muscle shortening and other physiologic responses that worsen pain. Nociceptors that are normally unaffected by norepinephrine become sensitive to it following injury, so neuropathic pains are often exacerbated by anxiety, as well as fear, anger, or excitement.
Anger is associated with exacerbation of both acute and chronic pain. A number of authors have found associations among anger regulation, both expression and suppression, and severity of chronic pain.32–34 The Ironic Process Model35 posits that attempts to suppress unwanted thoughts actually increases them. In a study examining the effects of anger suppression in pain severity, Burns and colleagues33 found that patients with chronic LBP who were told to suppress their anger toward a study confederate exhibited more pain behaviors and reported more pain than those who did not suppress their anger.
Psychogenic Pain/Somatization and Other Pain Amplifiers
Psychogenic pain (not a current diagnostic term) is a concept whose existence is disputed, yet pains of various sorts are clearly prominent features in somatization disorder. The terminology has changed multiple times, and the current term for what was called psychogenic pain is “pain disorder associated with psychologic factors.” The criteria require that pain causes significant distress or impairment in functioning; that psychologic factors be judged to have an important role in the onset, severity, exacerbation, or maintenance of the pain; and that the symptom or deficit not be intentionally produced or feigned.36 The method for determining that psychologic factors are causative is unspecified.
Psychogenic pain is akin to conversion disorders such as blindness and paralysis and is similarly typified by nonphysiologic findings on examination and behavioral inconsistencies. Patients may demonstrate behaviors that are incompatible with the degree of impairment they describe. A plethora of complaints and marked functional impairment may coexist with well-preserved muscle definition. It may be that the term is used for several unrelated conditions, given that some diagnosed with psychogenic pain appear euthymic, are animated, and sleep well, whereas others appear to suffer severely, cannot sleep, and even consider suicide.
One clue to the presence of somatization is apparent reluctance to discuss nonsomatic issues. If asked about family, work, or politics, the response inevitably and rapidly diverges to talk about doctors, symptoms, and treatments. This is not typically seen even in severe physical illness. Another clue is the sense of immediacy in the recounting of the traumatic event—a minor remote event is described as though it occurred yesterday.
There is evidence of a continuum among symptoms of post-traumatic stress disorder (PTSD), dissociation, somatization, and affect dysregulation. These interrelated symptoms commonly follow major trauma, and there seems to be a hierarchy of traumas, such that natural disasters lead to fewer symptoms than do adult interpersonal traumas, with childhood trauma causing the most severe symptoms.37–39 Rome and Rome40 hypothesized that a process akin to kindling follows psychic trauma, leading to symptom amplification, spontaneous symptoms, anatomic spreading, and cross-sensitization. These are processes that also characterize pain following neurologic trauma. They noted a melding of sensory and affective symptoms and a “polymodal allodynia” that rendered these people sensitized to both physical and emotional stressors.40 Most studies linking adult-onset chronic pain with childhood trauma have been retrospective. However, a recent study by Jones and colleagues37 looking prospectively at a 1958 British cohort of 7571 subjects found that, although adult onset of chronic pain was not associated with childhood surgery, it was associated with hospitalization for a motor vehicle accident, institutional care, maternal death, and familial financial hardship. Strengthening directionality of their findings, they also found that the association was not explained by adult psychologic distress or social class. Von Korff and colleagues38 also examined the effects of childhood psychosocial stressors and the onset of adult arthritis in a prospective study of 18,309 subjects from 10 countries participating in a World Mental Health Survey in the Americas, Europe, and Asia. They found that, controlling for age, sex, and early onset of psychologic disorders, subjects with significant childhood stressors had an increased risk of adult arthritis. Early-age onset of symptoms of depression and or anxiety were associated with an increased risk of adult arthritis even after controlling for childhood stressors.
Other psychiatric conditions that may present with pain include hypochondriasis, dementia, psychosis, and factitious disorder. Experience suggests that new onset of conversion/somatization in the elderly is rare and, when present, it may herald dementia. Malingering is by definition not a psychiatric illness. Although thought to be uncommon in chronic pain (on the basis of no data), it does occur.
Multiple psychologic factors affect both the perception of pain and ability to cope with it. Chronic stress increases both the perception of pain and disability. Distraction reduces pain awareness, whereas isolation and inactivity increase it and foster self-preoccupation. Perhaps the major psychologic factors that affect chronic pain are cognitions and incentives.
Cognitive theories of depression, anxiety, and pain hold that thoughts and beliefs are major determinant of affect (i.e., how a person feels is less determined by events than by his or her interpretation of them). The person who concludes from an unsuccessful job interview that the company has no openings reacts differently than the one who infers that he or she is undesirable and unlikely to find work. The terminal cancer patient who believes that “the surgeon got it all” will be more content than the healthy person who believes his intractable pain is due to severe but undetected pathology. Maladaptive cognitions tend to be automatic and habitual, so they are rarely examined for validity. They are simply accepted.
Cognitive factors have an impact on pain in several ways. First, the adverse quality of pain is modified by its interpretation. Such “catastrophic” interpretations of pain such as, “the nerves are being crushed” or “the exercises feel like they’re tearing something loose,” impede coping. The situation can be worsened by health care providers who attribute the pain to incidental findings on imaging that may bear only a modest relationship to the pain. Chronic back pain, which is the leading cause of disability and absenteeism from the workplace, lacks a specific structural explanation in more than 80% of cases. Additionally, it is often strongly driven by such psychosocial factors as fear of pain/reinjury, “catastrophizing,” depression, and anxiety. Several studies have looked at the effects of pain catastrophizing on pain perception and behavior.41–44 Thibault,44 in a 2008 study with 72 patients with musculoskeletal pain performing a lifting task, found that pain catastrophizing was associated with increased pain behaviors and overt signs of pain such as grimacing and such protective behaviors as decreased lifting. In 192 patients with chronic pain, Shelby and colleagues found that catastrophizing contributed to both pain and disability.45 Failure to address these issues in treatment of chronic back pain often leads to continued disability. Pain tolerance is reduced by thoughts emphasizing the averseness of the situation, the inadequacy of the person to bear it, or the physical harm that could occur. Such beliefs as “I’ll have a life again only after I’m cured,” “I can’t go out to dinner if I’m in pain,” and “I shouldn’t exercise if it hurts” have obvious impacts on adaptation.
Many patients with chronic pain develop a fear of movement, so-called “kinesiophobia.” Having become deconditioned due to rest following the onset of their pain, they hurt more whenever they attempt activities. This in turn leads to even more rest, etc. Breaking this cycle is critical for rehabilitation to take place. A number of studies have addressed kinesiophobia.46–48 Patients who rate high on scales of kinesiophobia report more pain and disability and engage in more self-protective behaviors.49
Self-appraisal may be as important as appraisal of the pain itself. Those who feel unable to influence events eventually give up. Belief in personal helplessness fosters pain and disability; on the other hand, a sense of self-efficacy promotes efforts to cope.50 Thus perceptions of helplessness lead to depression, resignation, and passivity, which in turn increase disability and pain. Self-efficacy, the opposite of helplessness, is correlated with successful rehabilitation in fibromyalgia, for instance.51
“Locus of control” is a psychologic construct that refers to one’s sense of the determinants of future events. The perception that events are a consequence of the individual’s own behavior (internal locus of control) is associated with better mood and function. Those with external locus of control tend to see future events as contingent on other people, or “fate.” People with chronic pain who have an external locus of control report depression and anxiety, feel helpless to deal with their pain, and they often rely on maladaptive coping strategies such as excessive rest and eating. Decreased perception of self-control may explain much of the relationship between depression and pain.