Pain and prejudice

• Managing pain has been a key goal for medicine over history. As our understanding has developed, theories have shifted away from an emphasis upon the nature of pain as a purely physical response to injury, towards an appreciation of the broader social and cultural factors that shape an individual’s experience of pain.
• This helps to account for the fact that, in certain circumstances, individuals with severe and traumatic injuries apparently experience surprisingly little pain, while in others, individuals with no apparent physical malady apparently experience a great deal of pain.
• One such theory is the biopsychosocial concept of pain. This integrates the key findings of the past 50 years of research, namely that the relationship between pain and injury is variable; pain may persist or occur in the absence of injury; pain is not a single sensation but has many dimensions; there is no adequate treatment for many types of pain; there are multiple ascending pathways that can carry pain information to the brain; and there are multiple areas of the brain that process pain information. But these apparent theoretical advances, while impressive, have not led to significant improvements in pain control.
• One reason for this is that today’s society is continually expanding definitions of sickness, which blur the distinctions between physical trauma, emotional upset, and social discomfort. The expansion of the ‘sick role’ means that people are tacitly encouraged to interpret their experiences as painful, at a time when the scope for changing those experiences is relatively limited.
• While appreciating that pain cannot be reduced to a simple physical response, and welcoming the advances in our understanding of this phenomenon, we perhaps need to take a more critical view of contemporary culture’s widespread adoption of the sick role, and how this is throwing up new barriers to our ability to treat and manage pain.

Interpreting pain – the tale of Bert Trautmann’s ‘painless’ broken neck

Bert Trautmann was born in Bremen, Germany, in 1923, and was a paratrooper on the losing side during the Second World War. He was captured by the British in 1945 and sent to a POW camp in Ashton, Makerfield, where football became a welcome distraction. Injury forced Bert to take up a place in goal, and there he shone. After the war Trautmann joined St Helens Town before moving to Manchester City in 1949. The move to City created outrage, season ticket holders threatened a boycott, and various groups in Manchester and around the country bombarded the club with letters denouncing the act. Trautmann had a lot to prove. He got his chance in 1956, during the FA Cup final.

In the seventy-fourth minute of the game against Birmingham City, Trautmann took an insane dive into the feet of opposing striker Peter Murphy and sustained a horrific neck injury. Details of exactly what happened next are murky, but this much is clear. Having been provided with the entire paraphernalia of on-the-pitch treatment (one bucket and one sponge), Trautmann remained on the pitch and continued in goal. He was involved in two further dramatic saves to help Manchester City beat Birmingham 3-1. After collecting his winner’s medal, Bert was seen to rub his ‘sore’ neck, and he complained of head pains until eventually seeing his doctor and being diagnosed with a broken neck. Bert became a hero and was awarded the footballer of the year for 1956.

The story of Bert Trautmann has not only entered into footballing legend. It has also taken a place within the modern interpretation of pain (see, for example, a recent booklet from the British Pain Society – Understanding and Managing Pain: Information for Patients (1)). Clearly Trautmann’s pain was not so severe as to have prevented him from continuing to play, and yet he had a catastrophic injury. There are many ways to square this: his desire to be accepted by the fans; his desire to not let down his teammates; his desire to win the FA Cup; the ‘rush’ of being in such a big game, and so on.

At that moment in Bert Trautmann’s life, other motives were apparently sufficient to overcome any experience of pain. Whatever the content of that motivation, the moral of the Trautmann Cup Final is that pain does not necessarily force entry into consciousness, and can be denied access according to certain conditions. For patients who suffer continual pain this can be a source of hope or inspiration that their pain may also be controlled (1).

The Trautmann story also highlights the Cartesian dualism that remains a marked feature of pain research, despite multiple attempts to eject it. On one hand, it is patently obvious that pain is something brought about by external events. A patient in pain may be engulfed by their experience, but they remain more than their pain. The pain is an internal feature of an external problem and is in the body not the mind. On the other hand, pain appears as something that very much involves the person; their intentions, thoughts, likes and hates and their complicated, messy past. This pain is very much ‘you’.

The tension between pain as a thing happening to you, and pain as you, has been boiling over for the past 50 years and is still to be resolved. The debate is more than merely philosophical; it has important implications for our clinical understanding of pain and treatment of patients.

Down with specificity theory

Most people probably fall spontaneously on the side of the Cartesian divide that sees pain as a response to injury. This view sees the relationship of pain to injury in a way similar to pulling a bell in a belfry. Someone pulling on the rope creates the injury or pain ‘stimulus’, and the bell rings to signal pain. Translated into the biological realm, the painful stimulus activates a pain pathway (the ‘rope’), which triggers a pain centre (the ‘bell’) somewhere in the brain.

This model of pain is known as ‘specificity theory’ and was first proposed by the French philosopher René Descartes almost 350 years ago. For Descartes, nerve filaments were activated by the fast moving particles of a noxious stimulus (such as a fire), creating a disturbance that passed directly to a central point in the brain. In the modern interpretation, nerve filaments have been replaced by dedicated pain fibre systems that pass into the spinal cord and ascend into the thalamus of the brain before being relayed to the primary sensory cortex, which is part of the outer rippled part of the brain.

The theory of pain based on a dedicated line system necessarily leads to the proposal of a predictable and consistent relationship between injury and pain. This fails to account for the mismatch of injury and pain that was experienced by Trautmann, and that has been reported elsewhere previously and since. During the Second World War, Henry Beecher observed that only one out of three soldiers wounded in battle complained of sufficient pain to receive morphine (3). Beecher was astonished to find that most of the soldiers denied having any pain, or claimed so little pain so as not to require pain relief. A 1978 study of Israeli soldiers with traumatic amputations following the Yom Kippur War provided similar findings (4).

Of course, there are reasons why a soldier extracted from battle might not complain of pain: he may feel his own wounds are insufficient to draw scarce medication from his fellow fallen comrades, for example. A 1982 study in a civilian setting, however, demonstrated similar findings (5). Researchers sat in the emergency department of a South London Hospital and asked patients to rate their level of pain as they entered. The researchers reported that in about 40 per cent of cases the patients complained of far less pain than might be expected based on the extent of their injuries, while about 40 per cent complained of rather more pain than might be expected. This left just 20 per cent of patients reporting the ‘correct’ amount of pain.

In addition to patients with injuries but no pain, there are a significant number of patients who have chronic unremitting pain in the absence of any identifiable injury or diagnostic marker. Such patients include those suffering from low back pain, fibromyalgia, irritable bowel syndrome, atypical facial pain and a number of other disorders that can be loosely grouped under the category of ‘functional disorders’ (6). These disorders are considered ‘functional’ because they can only be diagnosed following the report of symptoms rather than via an objective diagnostic test. Importantly, the disorders have no clinical effects, so far observed, aside from the experience of pain and a reported loss of normal function because of the pain.

According to specificity theory, an injury to tissue should invariably produce a painful experience that is commensurate with that injury, and there should not be pain without injury. Both cases of injury in the absence of pain, and the abundance of patients with pain in the apparent absence of injury, have undermined the belief that a neural system is indisputably linked to a single, specific psychological experience.

Replacing specificity

Beginning at the same time that Trautmann was performing his in-goal heroics, pain researchers began to report discoveries that would dramatically change our understanding of pain and account for the peculiarities of his experience. In 1955 Ronald Melzack and his colleagues reported striking observations that injury signals are not transmitted to the brain via a single dedicated pathway but by multiple ascending pathways, each with distinctive features including different speeds of transmission and terminations within the brain (7, 8). Modern neuroimaging techniques have decisively demonstrated that pain experience includes many regions of the brain rather than a single identifiable pain centre (9).

In 1965, Melzack teamed up with Patrick Wall to develop the gate control theory of pain (10). The theory proposed that pain is the result of the relative activity in small and large diameter nerve fibres. Large diameter fibres carry mechanical sensory information, whereas the small diameter fibres carry noxious information. Small fibre activity tends to facilitate the passage of information up the spinal cord (‘opening the gate’), whereas large fibre activity inhibits the flow of information (‘closing the gate’). This is one reason that rubbing a region of soreness helps to reduce pain.

Melzack and Wall reasoned that the cells of the spinal cord, which first receive incoming information, must select and compute outputs based on the combination of signals received. A descending influence upon these cells from the brain was also included, which in turn was influenced by input from the spinal cord cells, thus forming a spinal cord-brain loop. Gate control theory provided the first physiological mechanism for psychological interventions to minimise pain, such as distraction or relaxation, and shifted attention away from the peripheral source of injury and towards the spinal cord and brain.

As a reaction against the failures of specificity, and in an effort to incorporate these new facts about the biological system for processing noxious information, other theories have followed the gate control theory. These can be broadly characterised as ‘pattern theories’. The most popular of the pattern theories is the ‘biopsychosocial’ model of pain.

The biopsychosocial approach to pain is based on several propositions (11, 12), the central one being that an individual’s emotions and behavioural activity in response to an event is influenced by their appraisal of that event and their particular individual circumstances. Thus in addition to the biology of a noxious event, the biopsychosocial model introduces psychological and social factors that may mitigate or enhance the final experience of pain. Development of the biopsychosocial model has been an important part of the effort to undermine mechanistic accounts of pain that have done considerable harm to patients: for example, through removal of body or brain parts in a misguided attempt to eliminate the source or cause of pain, which is followed by frustration at patients’ persisting complaints of pain.

Crucially, the biopsychosocial model places an emphasis upon the content of pain experience rather than upon its source, which was inherent to the specificity model of pain. Specificity leads naturally to a definition of pain based on the stimulus. The question ‘what is pain?’ is defined in terms of a stimulus that is deemed to be painful because it causes pain. Thus pain, the response to a painful stimulus, is caused by a painful stimulus. The definition is circular, amounting to the suggestion that pain is pain – the mysterious and alien product of a painful stimulus.

In order to escape the inherent circularity of specificity theory, the International Association for the Study of Pain (IASP) drafted a definition of pain as ‘an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage’ (13). The clear advantage of this definition is that it places emphasis upon the content of pain, rather than the cause, and captures the multidimensional aspects of pain. The definition comments further that ‘pain is always subjective’, rounding off an understanding of pain as a conscious state rather than a physical process. The definition incorporates much of what researchers mean when they talk of pain as a ‘biopsychosocial’ phenomenon (14, 15).

This new understanding of pain has not been a merely academic exercise; it has also created pressure to treat pain as a symptom or disorder in its own right. The motivation to treat pain as a symptom or problem independently of any clinical diagnosis began with the publication of the gate control theory and the challenge to specificity. After all, if pain cannot be reliably judged upon an objective measure of injury or receptor activation, then assessment of pain must fall to subjective factors.

Considerable motivation was also provided by the very poor treatment of some patients’ pain, particularly those in the final stages of cancer (16). A basic humanism motivated early attempts to mitigate the pain and suffering of such patients, and the development of the Hospice and pain clinic soon followed.

Since the development of gate control theory, it has become much more common for pain to be treated as a psychological as much as a physical phenomenon. Recent high-profile investigations of pain have examined the placebo effect, which is the mere suggestion of pain relief accompanied by no other active intervention, and the neural commonalities of physical with mental pain (17, 18).

These reports gel with therapeutic interventions that emphasise the need to address the psychological and behavioural components that may be contributing to a patient’s painful condition. A standard therapy would include measures of pain and maladaptive behaviour such as guarding and activity avoidance. Regardless of the cause of the patient’s pain, the therapist works to provide more adaptive ways of thinking and behaving such that patients can return to normal function and suffer less discomfort. These interventions fall under the general rubric of cognitive behavioural therapy (CBT).

The biopsychosocial concept integrates the key findings of the past 50 years of research: namely that the relationship between pain and injury is variable; pain may persist or occur in the absence of injury; pain is not a single sensation but has many dimensions; there is no adequate treatment for many types of pain; there are multiple ascending pathways that can carry pain information to the brain; and there are multiple areas of the brain that process pain information. All this is impressive. Sadly, however, these apparent theoretical advances have not led to significant improvements in pain control.

The rise and rise of chronic pain

The period of ascendancy of the biopsychosocial model has coincided with an immense increase in the number of people laying claim to chronic pain. In the UK in 1955, for example, there were fewer than 20million working days lost to back pain but by 1995 the number had increased to over 100million (19). The figures are similar in the USA, where the rate of disability claims related to low back pain have increased by 14 times the rate of population growth (20).

Less than a third of these cases of back pain turn out to be the result of a demonstrable abnormality, such as a trapped nerve, and the vast majority are diagnosed as being non-specific in origin: in other words, without identifiable pathology or cause.

In general, the functional disorders, such as non-specific low back pain (NSLBP), represent a major and growing challenge to medicine. In UK primary care, these disorders account for 20 per cent of consultations and for 35 per cent of new referrals among medical outpatients. Treatment is expensive and largely without effect. In one study, 75 per cent of patients with functional chest pain still had symptoms 10 years after presentation (21).

Aggressive interventions, such as surgery to remove parts of the supposed pain system, largely fail to resolve the pain and often render the patient worse than before (16). More moderate intervention with medicines aimed either directly at the pain or at proposed underlying clinical problems has also been largely unsuccessful.

Consequently, patients with chronic pain are now increasingly referred to a specialised pain clinic where they will be enrolled in a therapeutic programme usually involving a variant of cognitive behavioural therapy. Patients enrolled in such programs fare little better than those just taking medications. On average, patients report a drop of 15 to 30 percent in their level of pain, between one month and a year following CBT treatment (22). These changes are at the border of what is considered to be clinically meaningful to patients, and there is little indication that patients give up their status as disabled and return to a normal mode of living.

A recent study has suggested that patients who learn to accept their status rather than search fruitlessly for a cause have a greater chance of returning to normal function (23). But while acceptance might be a reasonable aim, it hardly seems like a victory over the disorder, and is a somewhat lukewarm achievement for the biopsychosocial model.

The return of specificity

Working in part with the eminent pain and brain imaging group in Montreal, Bud Craig and colleagues published a series of reports in the mid-1990s that described a ‘specific’ pathway for the experience of pain and temperature in an effort to understand at least one form of chronic pain (24, 25).

Following stroke that involves damage to the thalamus (a major relay for all sensory information entering the brain), some patients experience a nearly constant deep aching or burning pain within the part of the body represented in the now damaged area of the thalamus (26). Occasionally, there is also heightened sensitivity to temperature, causing pain on contact with temperatures not usually experienced as painful, such as that of a hot shower. Paradoxically these sensations can be combined with decreased sensitivity to some or all noxious stimuli and a decreased ability to detect innocuous temperature sensations.

Craig and his colleagues suggested that the paradox could be solved by describing multiple ascending pathways that are usually integrated to produce a final experience. Specifically, they described two projection systems from the spinal cord to the thalamus. The first projection system codes for cool and noxious cold (temperatures less than approximately 35 degrees Celsius), and the second system codes for both noxious cold and heat, as well as noxious pinch.

Craig hypothesised that in stroke patients with central pain syndrome, the ascending system specific to cool and noxious cold has been damaged, and this disinhibits or unmasks the remaining heat, pinch and cold pathway, resulting in a constant aching burning pain combined with a loss of some noxious and non-noxious sensitivity. The proposal was powerful because it encompassed the known features of an extremely challenging disorder using the generally well-accepted concept of a plastic and integrative nervous system.

The ideas were also highly controversial, however, in part because the precise mechanistic details were mostly hypothetical and in need of experimental verification (27), but also because the work implied a return to specificity. Professor Pat Wall was most vociferous in his response, commenting that ‘the paper is a powerful message from a cult…[depending] on two unproven assumptions. The first is that each modality of sensation is dependent on its private nucleus in the thalamus. The second is that direct monosynaptic spinothalamic inputs are more important than indirect inputs.’ (28)

In their response to Wall, Craig and colleagues argued that, like it or not, the nervous system really is organised according to line systems that have been specialised by natural selection to carry certain types of information: ‘Our work is based on the principle that the mammalian central nervous system is functionally and anatomically well-organised…. This principle is the basis for much of what is known about the function of the brain. Wall’s comments are representative of an old argument that all can be explained only by acknowledging that everything is connected to everything else and that, in the extreme, the concept of “blueness” or “pain” is a mental property that cannot be localised in the brain.’ (29)

Craig and colleagues dismiss the tension between pain as a physical manifestation and pain as a mental property with a mighty harrumph. This has justification. Just because the experience of pain is mental does not mean that the cause is not physical. At least some disorders, perhaps certain cases of central pain among them, lend themselves towards an understanding based on dedicated line systems.

An excellent example is arthritis of the knee. This condition is severely painful and debilitating but is almost entirely curable following a total knee replacement. After knee replacement, 90 per cent of patients report no pain or only mild pain and their walking is no longer limited by functional impairment (30). According to the biopsychosocial model, such a clean outcome should be all but impossible, because patients should develop social relationships and psychological coping mechanisms likely to perpetuate the pain after surgery. This doesn’t happen, and the model that best seems to fit knee arthritis is in fact the much-maligned specificity theory.

At the same time, ‘blueness’ and ‘pain’ cannot be found inside neurons or pieces of brain. These are mental states that can only be experienced by socially conscious beings, and there are disorders that seem especially resistant to any explanation based on a dedicated line system. The rapidly rising numbers of people with low back pain, for example, implies a sociological rather than biological cause.

The IASP produced a report in 1995 entitled ‘Back Pain in the Workplace: Management of Disability in Non-Specific Conditions’ (31), which acknowledged as much. The report recommended a major overhaul of the way back pain was treated by medicine and society. Instead of diagnosing people with back pain of uncertain origin as having nonspecific low back pain (NSLBP), the report called for a diagnosis of activity intolerance with deliberate rejection of the condition as a medical problem.

Consistent with this stance, the report also argued to end the provision of disability payments and to reclassify as unemployed those who fail to return to work within two to four weeks of a NSLBP or activity intolerance diagnosis. The report generated considerable controversy including a large and vocal debate at the 1996 IASP World Congress in Vancouver and several exchanges within the pages of the IASP journal, Pain, including more vociferous comments from Patrick Wall (32, 33).

The major conclusion of the report, that NSLBP should essentially be recategorised as a non-disorder, jarred heavily with psychologists and pain specialists who emphasised the importance of recognising pain as a disorder in its own right. A major principle of cognitive behavioural therapy (CBT) is to recognise the pain as the problem and subsequently alter the patient’s perception so that he or she does not think in catastrophic terms, returns to normal behaviour as far as possible, and generally learns to accept and cope with the disorder.

This approach is certainly preferable to the use of aggressive surgeries, such as back fusions for NSLBP, which rarely result in a successful outcome and often leave the patient in more pain and distress than before. As noted, however, CBT in no way constitutes a cure for chronic pain of functional or nonspecific origin. It generally leads to quite modest improvements in pain experience and behaviour, and, perversely, may actually be adding to the problem.

Is pain really the problem?

The IASP back pain report noted that complaints of back pain ‘are not reliably indicative of injury’ and went on to note that pain behaviours, ‘including the verbal report of pain, should be seen as social communications and not merely as metrics of pain’. In other words, people might complain for lots of reasons and then adopt a role that produces the most ‘fitting’ outcome.

At a time in history when human beings are being increasingly treated as damaged goods and incapable of controlling their own lives, adoption of the sick role is likely to be perceived as ‘fitting’ to an increasingly large number of people (34). Under these circumstances, society can be said to be literally making people sick, aided and abetted by psychologists and pain specialists who indulge patients’ somatic description of their angst.

The lifetime incidence of low back pain is between 60 and 85 per cent, meaning that almost everybody will have some experience of back pain at some point in their life (31). In a period where dislocation and insecurity are widespread, the interpretation of pain, and other feelings of discomfort, as being biological, is attractive. Disease explains behaviour, provides a rational for withdrawal and even confers a sense of identity. The trigger for an initial bout of low back pain (or facial pain, or chest pain, or gut discomfort and so on) could be from a direct blow, an unfortunate bend, infection or other nefarious source.

Perpetuation and development of this acute experience into a disorder, however, appears as a social phenomenon. This is not to say that a patient’s pain is not ‘real’ but is rather to suggest that the nature of their distress and the causes are likely to lie outside the usual remit of medical science. Staying alive to the possibility of a currently unseen biological force is perhaps reasonable so long as the danger of chasing shadows remains apparent.

The major strength of the biopsychosocial model is the ability to capture the experience of pain, but the major weakness is the failure to provide causal mechanisms of pain. Pain is a mental event, expressed through a socially constructed awareness, but experience is not necessarily causal. The biopsychosocial model is an important descriptive tool, allowing us to get a grip on the mental nature of pain. However, it is not a ‘theory’ with which to understand the causes of pain. Theories require explanations of processes and, most of the time, that is critically absent.

The biopsychosocial model of pain has developed out of the tension between pain as a physical response to injury and pain as experience. This tension is real; pain does not fall neatly on to either side of the Cartesian divide. Nevertheless, at least some pains do conform better with a more deterministic model, and categorising people as having a chronic disorder based on subjective interpretation of their feelings has a definitive downside – it produces a medical diagnosis for a vastly expanding number of people without any hope of an effective treatment.

In the face of this compelling and depressing fact, to maintain that all pains are caused by an amalgam of biopsychosocial factors is as dogmatic as the earlier advocacy of specificity theory.

Stuart Derbyshire is assistant professor of anesthesiology and radiology at the University of Pittsburgh.

Acknowledgements: The author is supported by a grant from the Pittsburgh Foundation and the John F. and Nancy A. Emmerling Fund. Thanks to Dr. William Lariviere for comments on an earlier draft of this essay. The opinions expressed are those of the author only.
(1) Understanding and managing pain: information for patients, Pain Society, September 2003
(2) Derbyshire SWG, Trends in Neurosciences 2002; 25: 65-66
(3) HK Beecher, Measurement of Subjective Responses, Oxford University Press, New York, 1959
(4) PL Carlen, PD Wall, H Nadvorna, T Steinbach, ‘Phantom limbs and related phenomenon in recent traumatic amputations’, Neurology 1978; 28: 211-217
(5) R Melzack, PD Wall, TC Ty, ‘Acute pain in an emergency clinic: latency of onset and descriptor patterns’, Pain 1982; 14: 33-43
(6) S Wessely, C Nimnuan, M Sharpe, ‘Functional somatic syndromes: one or many?’ Lancet 1999; 354: 936-939
(7) DIB Kerr, FP Haugen, R Melzack, ‘Responses evoked in the brainstem by tooth stimulation’, American Journal of Physiology 1955; 183: 253-258
(8) D Bowsher, D Albe-Fassard, ‘The anatomophysiological basis of somatosensory discrimination’, International Review of Neurobiology 1965; 8: 35-75
(9) SWG Derbyshire, ‘Exploring the “neuromatrix”’, Current Review of Pain 2000; 6: 467-477
(10) R Melzack, PD Wall, ‘Pain mechanisms: a new theory’, Science 1965; 150: 971-979
(11) FJ Keefe, J Dunsmore, R Burnett, ‘Behavioral and cognitive-behavioral approaches to chronic pain: Recent advances and future directions’, Journal of Consulting and Clinical Psychology 1992; 60: 528-536
(12) DC Turk, TE Rudy, ‘Cognitive factors and persistent pain: A glimpse into Pandora’s Box’, Cognitive Therapy and Research 1992; 16: 99-122
(13) H Merskey, ‘The definition of pain’, European Journal of Psychiatry 1991; 6: 153-159
(14) SWG Derbyshire, ‘Sources of variation in assessing male and female responses to pain’, New Ideas in Psychology 15: 83-95, 1997
(15) G Waddell, ‘A new clinical model for the treatment of low-back pain’, Spine 12: 632-644, 1987
(16) R Melzack, PD Wall, The Challenge of Pain, London: Penguin, 1988
(17) P Petrovic, E Kalso, KM Petersson, ‘Ingvar M. Placebo and Opioid Analgesia – Imaging a Shared Neuronal Network’, Science 2002; 295: 1737-1740
(18) NI Eisenberger, MD Lieberman, KD Williams, ‘Does Rejection Hurt? An fMRI Study of Social Exclusion’, Science 2003; 302: 290-292
(19) G Waddell, ‘Low back pain: a twentieth century health care enigma’, Spine 21: 2820-2825, 1996
(20) JT Robertson, ‘The rape of the spine’, Surg Neurol 39: 5-12, 1993
(21) S Potts, C Bass, ‘Psychological morbidity in patients with chest pain and normal or near normal coronary arteries: a long term follow up study’, Psychological Medicine 1995; 24: 339-347
(22) S Morley, C Eccleston, AC Williams, ‘Systematic review and meta-analysis of randomized control trials of cognitive behavioral therapy for chronic pain in adults, excluding headache’, Pain 1999; 80: 1-13
(23) LM McCracken, C Eccleston, ‘Coping or acceptance: what to do about chronic pain?’, Pain 2003; 105: 197-204
(24) AD Craig, ET Zhang, MC Bushnell, A Blomqvist, ‘A thalamic nucleus specific for pain and temperature sensation’, Nature 1994; 372: 770-773
(25) AD Craig, EM Reiman, A Evans, MC Bushnell, ‘Functional imaging of an illusion of pain’, Nature 1996; 384: 258-260
(26) KL Casey (ed) Pain and Central Nervous System Disease, Raven: New York 1991
(27) WD Willis, X Zhang, CN Honda, GJ Giesler, ‘A critical review of the role of the proposed VMpo nucleus in pain’, The Journal of Pain 2002; 3: 79-94
(28) PD Wall, ‘Pain in the brain and lower parts of the anatomy’ [letter; comment], Pain 1995; 62: 389-391
(29) AD Craig, ET Zhang, MC Bushnell, A Blomqvist, ‘Reply to PD Wall’ [letter; comment], Pain 1995; 62: 391-393
(30) JA Garcia, B Bewley, JF Redden, ‘The St. Leger total knee replacement – a 7-year clinical assessment and survivorship analysis’, Knee 2003; 10:173-177
(31) WE Fordyce, Back Pain in the Workplace: Management of Disability in Nonspecific Conditions, IASP Press, Seattle, 1995
(32) E Thompson, H Merskey, RW Teasell, ‘Comments on the Back Pain in the Workplace Report’, Pain 1996; 65: 111-112
(33) PD Wall, ‘Editorial comment Back pain in the workplace’, Pain 1996; 65: 5
(34) F Furedi, Therapy Culture: Cultivating Vulnerability in an Uncertain Age Routledge, 2004