ath

srsityNeue Lon

e end

Received 1 July 2013Received in revised form 2 October 2013Accepted 18 October 2013

Keywords:Lay understandingNerve damage

Neuropathic pain (NP) is not easy to understand for those with the diagnosis. Even in specialist medicalservices, explanation may not be given or may not be integrated with patients existing beliefs about their

ute importantly to how they construct their problem [44]. In chronicpain, patients and professionals understandings tend to differ[21,38,46], and although patients concepts relate to medical expla-nations, they do not often adopt them andmay be quite critical [62].

greater thed on the bcriptions [

there are several well-established scales to help with dia[3,27], but it appears to be difcult to diagnose in prima[51], and underdiagnosis is common [25,55]. This, in turn, leadsto under-referral to specialist services, where mechanisms of NPare better recognised, often associated with specic diagnosesand to a certain extent with treatment [2].

It can be hard for patients to describe and understand NP andassociated symptoms, particularly when they are not identiedby medical personnel. Without explanation, pain is frequentlymisinterpreted as harm, prompting protective strategies that

Corresponding author. Address: Research Department of Clinical, Educationaland Health Psychology, University College London, Gower St., London WC1E 6BT,UK. Tel.: +44 2076791608; fax: +44 2079161989.

E-mail address: amanda.williams@ucl.ac.uk (A.C.de.C. Williams).

PAIN155 (2014) 34URL: http://www.ucl.ac.uk (A.C.de.C. Williams).with subjectivity and the construction of concepts of health and ill-ness through language, by all in a culture, not only those in healthcare [62]. For those receiving health care, their experiences contrib-

pact of NP on quality of life can be substantial andof other kinds of chronic pain [50]. NP is diagnospositive and negative symptom patterns and des0304-3959/$36.00 2013 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.pain.2013.10.021an thatasis of2], andgnosisry careGiven the impact of chronic pain [8,22], it is important that peo-ple with pain have a satisfactory understanding of their pain, butthere is relatively little research in the area. The common-sensemodel of understanding health problems [30] describes patientsbehaviours in response to their symptoms in relation to their impli-cit or explicit models of their illness. Use or nonuse of health care isone of those behaviours of interest. An alternative approach starts

jected by patients who see them as threats to the legitimacy of theirpain [29,40]. This means that they are not referred to, or do not takeup, psychologically based rehabilitation programmes for whichthere is good evidence of efcacy [61]. By contrast, health care staffsconcerns about pain can provide important validation.

Neuropathic pain (NP), resulting from lesions in or disease ofthe central or peripheral nervous system, affects 7% to 8% of thepopulation, increasing with age and chronic disease [8,55]. The im-1. IntroductionPsychological treatmentconditions. We were curious about how people with NP conceptualised the problem. Web sites relevantto NP were used to recruit 79 people with NP. They were sampled using Q-methodology, which requiressorting according to degree of agreement or disagreement with diverse statements about NP, derivedfrom the widest possible range of sources. The sets of sorted statements are analysed for factors whichrepresent shared constructions. The four factors that we found differed in important ways: (1) identi-cation of nerve damage as cause; (2) the necessity of identifying cause; (3) the acceptability of symptom-atic treatment; (4) the existence or not of psychological inuences; and (5) the usefulness ofpsychological treatment. The meaning of these factors was extended by participants free comments: cer-tain viewpoints showed associations with their medical and treatment histories and with the interfer-ence of pain with daily life. Overall, a biopsychosocial model of pain was only weakly represented, andno integrated model of pain emerged across the four different accounts. There was little reference toNP having been explained when the diagnosis was made. This study highlights the need for more acces-sible explanations of NP within and outside medical services if people with NP are to use their under-standing of NP to help them manage their pain and reduce its impact on their lives. 2013 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

In particular, the psychological components of anymodelmaybe re-Article history:How do people understand their neurop

Sally Martin a, Clare Daniel b, Amanda C de C Williama Previously, Research Department of Clinical, Educational and Health Psychology, Univeb Pain Management Centre and Facial Pain Service, National Hospital for Neurology andcResearch Department of Clinical, Educational and Health Psychology, University Colleg

Sponsorships or competing interests that may be relevant to content are disclosed at th

a r t i c l e i n f o a b s t r a c tic pain? A Q-studyc,College London, London, UKrosurgery, Queen Square, London, UKdon, London, UK

of this article.

9355

www.e l sev ie r . com/ loca t e / pa in

be broadly representative of the relevant opinion domain ([59] p.

175); it can never claim to be comprehensive. The concourse con-sisted of (1) the content of discussionwith 5 professionals (a clinicalpsychologist working in chronic pain, 2 experienced medical painspecialists, and 2 primary care doctors); (2) content from severalUKWeb sites for peoplewith NP (eg, www.patient.co.uk; www.bra-inandspine.org.uk); (3) social networking sites used by people withNP (such as Facebook groups); (4) textbook chapters and researchpapers on NP. We gathered 130 statements, satisfying the recom-mendation that 3 times the intended nal number of statementsbe obtained [52]; after removing repetition, overly complex propo-sitions and ambiguous statements, 45 items remained, adequate forthe recommended 40 to 80 statements for the nal Q-sort [59].

Q-methodology involves the presentation of these statementsexacerbate disability [6]. Early explanations that make sense ofpain can mitigate its impact [37,41]. The aim of enabling peoplewith chronic pain to become effective in self-management may re-quire more consistent investment in explaining pain. Pain that is amystery to the person with pain generates worry, which early inthe pain history may be functional [32], but in chronic pain is usu-ally part of a closed cycle of pain, disability and distress [15,20].

It is therefore helpful to study lay understanding of NP. The cur-rent study uses Q-methodology, which is well suited to the studyof subjectivity because it generates a distribution of participantssubjective viewpoints, opinions and attitudes about a given subjectmatter [10].

2. Methods

Ethical approval was given by the relevant university commit-tee. The study recruited online, with permission from the relevantWeb site administrators, via 10 Web sites used by people withneuropathic pain: for NP, for multiple chronic pain conditions,and for HIV, multiple sclerosis and diabetes. The Web sites werefound using 4 search engines: Google, Yahoo UK & Ireland, MSN/Bing, and Ask, in October 2011, using the terms neuropathic pain,neuropathy, diabetes, HIV, shingles, chronic pain, persistent pain, for-um, message board, and online support. An advertisement request-ing volunteers with NP to share how they understood it, andproviding a hyperlink, was distributed to these Web sites.

Free software (Q-assessor Beta [43]) was used to create neces-sary Web pages and a programme for the Q-sorting to be carriedout online. The hyperlink posted on theWeb sites led to the studysWeb pages, which explained the purpose, requested consent toproceed, and then presented the Q sort study and instructions.Recruitment of adult participants with NP aimed for a sample sizeof at least 50 to achieve stability in the factor structures derivedfrom analysis [34]. Two comparisons of computer-based and paperQ-sorts [45] found no difference in the reliability or validity be-tween the methods.

2.1. Q-methodology

Q-methodology uses a factor analytic paradigm but with theimportant difference that instead of using relationships amongitem scores as raw data, the factor analysis correlates relationshipsamong statements within respondents [9]. Q-methodology is not,however, concerned with the population but with subjectivityand shared viewpoints within that population [47].

Statements provided to respondents were drawn from a con-course that aimed to represent as wide as possible a range of waysinwhich neuropathic pain is understood [53]. The samplingmethodis not important provided that thenalQ set can justiably claim to

350 S. Martin et al. / PAIN(the Q-set) to a sample of participants (the P-set). In the rst phaseof the sorting process, participants arranged the statements fromthose with which they strongly disagreed, through those aboutwhich they felt neutral, to those with which they strongly agreed,by clicking on the designated button. Participants could amend theselection at any time. In the second phase, the participants werepresented with the statements that they had agreed with in phase1, then those they had disagreed with, and then those they hadbeen uncertain about. Using a drag-and-drop procedure, theysorted each statement into a quasinormal distribution on a gridprovided, putting 1 statement into each box. The horizontal gridprovided a distribution from strong disagreement to strong agree-ment, and the format was chosen after a pilot study showed a clearpreference (5 of 6 respondents) for more rather than fewer optionsat the extremes. At the extremes were 5 boxes vertically: stronglyagree (+3) through strongly disagree (3); adjacent to each were 6boxes vertically for agree and disagree; between these were 7boxes vertically for possibly agree and possibly disagree; and, cen-trally, 9 boxes vertically for uncertain.

Additionally, several open-ended questions requested partici-pants comments about the statements that they strongly agreedor disagreed with or that stood out for some other reason. Theywere also asked to list the treatments they had tried for their painand to state what might have triggered their pain. The open-endedquestions used an unstructured response format in which partici-pants provided their responses in text boxes without word limits.Lastly, participants were asked to provide their e-mail addressesif they wished to check factor interpretation following analysis[59].

After completing their sorts, the participants completed theBrief Pain Inventory (BPI; Short Form [12]), which assesses painand pain interference in people with chronic pain [54]. Using a010 numeric rating scale, from no pain to pain a bad as youcan imagine, this inventory asks respondents to rate their worst,least and average pain over the past 24 hours and their currentpain; the severity score is the mean of these 4 items. Interferenceis measured by 7 items with responses from 0, does not interfere,to 10, completely interferes, providing an overall mean of inter-ference. The items are general activity, mood, walking ability,work, relationships with others, sleep, and enjoyment of life. In astudy of people with NP, the BPI was reported to have a Cronbachalpha of 0.94 for both scales [64].

2.2. Data analysis

The distribution of participants subjective viewpoints [10]were subjected to factor analysis. Signicant factors representshared viewpoints within the group [45] or clusters of subjectivity[10,58]. The Q-assessor online programme (q-assessor.com), mod-elled on the PCQ Method programme [1], uses the centroid methodof factor analysis, rst grouping Q-sorts by similar item rankings[10,59] and then producing distinct factors by varimax rotation.Factors with an Eigenvalue greater than 1 and with at least 1 par-ticipant loading on it are assumed to be interpretable [53]. For afactor loading to be signicant at the P < .01 level, the cutoff wasgiven by 2.58 SE), where standard error (SE) was 1=

Np

, and Nwas the number of statements in the concourse [9]. For the currentstudy, with 45 statements, this formula identied factor loadings>0.39 as signicant at P < .01.

Interpretation of factors, that is, of the statements and theirloadings within each factor, used the method recommended byWatts and Stenner [59]. Q-assessor provided rank statement totalsfor each factor, indicating ranking of statements within factors andsignicant differences among factors in ranking of particular state-ments, described as distinguishing statements, whose normalised

55 (2014) 349355score differs from those of all other factors. Participants feedbackand other information (in this study, BPI scores and treatment his-tories of participants) enrich this interpretation [19].

N13. Results

We received 84 Q-sorts, but 5 respondents did not complete theadditional questions, leaving 79 complete responses. The studywas downloaded 416 times, but it is not possible to tell how manyof these were repeated attempts. The minimum response rate is,therefore, 20% but is likely to be somewhat higher. Of the 79 par-ticipants, 49 (62%) were male; 23% were between 50 and 80 yearsof age; 38% were between 40 and 50 years of age; 28% were be-tween 30 and 40 years of age; and 11% were younger than 30 yearsof age.

Some items were effectively redundant: all respondents dis-agreed that NP could be caused by personality or attention seekingor that diet affected pain level; all agreed that NP could be felt else-where than at the injury site and that sleep problems exacerbatedit.

3.1. Factor interpretation

Q-assessor analysed 79 Q-sorts and produced 4 factors with anEigenvalue >1 and clear exemplars (participants) loading on them.There were 56 exemplars who loaded signicantly on 1 of 4 factorsand 23 who did not load on a single factor but on diverse combina-tions, with none of the combinations favoured over others. Wetherefore considered all viewpoints in the 79 Q-sorts to be ade-quately represented by the factors below [19,59]. Factors were dis-tinguished from one another by a focus on cause, on nerve damageas cause, and on rejection or acceptance of psychological inuenceor treatment. Factors are described in order of decreasing numberof exemplars, showing item rankings in brackets: rankings of +3and 3 represent strong agreement and strong disagreement,respectively (Table 1).

3.1.1. Factor 1NP is a nervous system problem; psychology inuences pain

experience and acceptance; NP is open to psychological treatments.This viewpoint was endorsed by about one third of the group. It

had 2 related themes and many distinguishing statements (signif-icant at P < 0.05) characteristic of factors which represent a sub-stantial proportion of the sample. It identied NP as a physicalproblem but accepted that psychological methods could help itsmanagement (Appendix A). One theme concerned the multifacto-rial causes of NP and its maintenance, with a predominant modelof altered processing of pain signals by the nervous system. Psy-chological inuences on pain were admitted, but a psychosomaticmodel was clearly rejected.

3 Neuropathic pain signals that there is something seriouslywrong with the part of the body where the pain is felt.

2 The most important treatments for neuropathic pain areprovided by doctors.

+2 Neuropathic pain can be caused by changes in the spinal cordand/or the brain.

+2 Neuropathic pain can be made worse by stress, anxiety anddepression.

2 Neuropathic pain is sometimes not pain but expression ofemotion.

The second theme described acceptance of pain (as opposed toseeking a cure) and learning to live with pain as the goal, towardswhich psychological methods could provide an additional (ratherthan an alternative) resource.

+3 Psychological therapy such as cognitive behavioural therapy

S. Martin et al. / PAIcan change a persons experience of neuropathic pain such aspain intensity. +3 It is important that people learn to live with neuropathicpain.

+3 More support is needed for people who are willing tomove away from focusing solely on medical treatment fortheir pain.

+2 People who cope better with neuropathic pain have acceptedthe situation.

3 Psychologists and psychological therapies are not useful forpeople with neuropathic pain.

The open-ended comments showed that participants thoughtthat mood and stress affect pain, that accepting NP was helpful,and that they endorsed or were open-minded about psychologicaltreatment:

I think psychological therapies have an important role to play in pain manage-ment and helping improve ones mood.

Self-help literature can provide more relief than medication at times!

How people behave and think affects their pain. I have completed a mindful-ness course and this has helped a huge amount with controlling my reactionto pain.

Six participants who loaded on factor 1 had used cognitivebehavioural treatment for pain, whereas none of the participantsloading on the other factors listed this specically. Nine partici-pants loading on factor 1 also made reference to having tried othernonmedical treatments, such as yoga and meditation, and therewere several comments expressing dissatisfaction with medicalprovision.

It is interesting that in neutral positions were statements con-cerning the direct inuence of psychological processes on pain. Itmay be that acceptance of the place of psychology arose more fromtrying treatments in open-minded ways or in the face of no furtherbiomedical options, rather than from subscribing to an integratedbiopsychosocial model of neuropathic pain. There were also com-ments seeking to distinguish these psychological models from psy-chosomatic accounts. With reference to the statement that if thepain/illness/injury was not visible, then the pain must all be inthe patients head, one respondent wrote, Unfortunately this isthe outlook of a lot, if not the majority, of people that if they can-not see a visible wound then how can relatively normal-lookinglegs cause a person so much pain with no exterior signs? Anothercommented, It is really important to be reassured that it is not inour heads. It is hard to accept that it cannot be cured, but it doeshelp if the doctors are honest.

Treatments tried in this group (24 of the 26 answered) were, indescending order of numbers of patients using them, and in vari-ous combinations: anticonvulsants; opioids; antidepressants;mainstream psychological treatments such as cognitive behav-ioural therapy; nonsteroidal anti-inammatory drugs (NSAIDs);physiotherapy and other physical therapies; meditation/mindful-ness or yoga; acupuncture; and one each for lidocaine patches,nerve blocks and facet joint injections. Only 3 were using no drugs.

3.1.2. Factor 2NP is nerve damage; psychology is irrelevant in pain experi-

ence; neutral about psychological treatments.This factor was exemplied by 15 participants and had a single

coherent theme: that of a neurophysiological model of neuropathicpain, with explicit rejection of psychological inuence of any sort.Cause was important, and symptomatic treatment was stronglyrejected.

+3 Neuropathic pain is caused by damaged nerves sending dis-torted signals to the brain.

55 (2014) 349355 351 +3 Modern medicine cant identify all the causes of NP, so testresults can come back negative.

1 +3 Doctors need to treat the cause of NP, not the symptoms. +3 Neuropathic pain seems to have a will of its own. 3 Areas in the brain that are responsible for neuropathic painare also connected to emotion. Therefore, neuropathic pain is inpart an emotional experience.

3 Negative thoughts can increase neuropathic pain.

This factor is dominated by biomedical formulations of NP andby nerve damage as the cause of NP; brain processes are describedin neurological but not psychological terms, with strong rejectionof an association with psychosocial processes, such as cognitions,emotion and personality. This group was the only one to rejectstrongly the statement that negative thoughts can increase neuro-pathic pain, making it a distinguishing statement (factors 1 and 4scored this neutrally). Additional comments further repudiatedany place for psychological inuence:

I disagreed mostly with statements suggesting emotional causes. Thats justnonsense. Assertion that neuropathic pain is psychological is a ridiculousassumption and could lead to a medical professional determining that pain doesnot exist when it does.

Despite the clear rejection of psychological processes in neuro-pathic pain, psychological treatment methods were not rejectedbut ranked neutrally; none of the participants listed any formalpsychological interventions among the treatments tried, and mosttreatments were by conventional medicine. In descending order ofnumbers using them, and in various combinations: unspeciedanalgesics, anticonvulsants, antidepressants, physiotherapy andphysical methods, opioids, and one each of NSAIDs, acupuncture,botox, occipital nerve stimulation, and surgery.

3.1.3. Factor 3NP is (irreparable) nerve damage, for which symptom manage-

ment is needed; psychological inuences play a part in pain per-ception, but psychological treatments have no role.

Eleven participants loaded on this factor, which shared much ofthe biomedical model endorsement of factor 2 but not its preoccu-pation with cause. Participants found symptomatic treatmentacceptable. Several had NP associated with incurable conditions(cancer, HIV, multiple sclerosis), so symptomatic treatment wastheir best option. There was more tolerance for psychological fac-tors as contributors to the pain experience but not as a route forintervention.

+3 Neuropathic pain is caused by damaged nerves sending dis-torted signals to the brain.

+2 Modern medicine cant identify all the causes of NP, so testresults can come back negative.

+3 Neuropathic pain seems to have a will of its own. +2 Neuropathic pain can be made worse by stress, anxiety anddepression.

3 Doctors need to treat the cause of neuropathic pain not thesymptoms.

2 Accepting neuropathic pain can reduce distress.

Several comments emphasised the uncontrollability andunpredictability of NP: Neuropathic pain seems to have a willof its own stood out to me. There is often no specic event thatcauses it to act up. Often it is in the middle of the night. AndAm I going mad? Is it all in my head? That question stood outas thats how I feel most of the time. I dont think the causecan be treated: its irreversible.

A few comments associated this characteristic with the accept-ability of symptomatic treatment: The expectation that doctors

352 S. Martin et al. / PAINneed to understand or x the cause is unrealistic at thistime. . .treating the symptoms is important. Treatments triedwere, in descending order of numbers using them, and in variouscombinations, were: unspecied analgesics, antidepressants, anti-convulsants, opioids, physiotherapy, and 1 each of NSAIDs, topicallidocaine, medical marijuana, vitamins, diet, and deep breathingand positive thoughts.

3.1.4. Factor 4NP (of recent onset) cause should be identied; psychological

inuences play a role; treatment may be medical and/or psycho-logical. A distinct factor emerged from 4 participants Q-sorts,resembling factor 2 in the focus on cause, yet neutral about state-ments endorsed in factor 2, which described neurophysiologicalmechanisms. It was also neutral about symptomatic treatmentand accepting of psychological inuence and of psychologicaltreatment. These participants had pain for a notably shorter timethan those exemplifying the other 3 factors, caused by chemother-apy, HIV, surgery, and an unknown event.

+3 It is essential for the cause of neuropathic pain to be foundto be able to treat it.

+3 Because biological, psychological and social factors all inu-ence perception of neuropathic pain, treatment should there-fore include psychosocial interventions as well as medical ones.

+2 There is a clear relationship between underlying causes ofpain, the physical signs and the neuropathic pain that is felt.

2 It is important that people learn to live with neuropathicpain.

3 Medication is the best way to manage symptoms of neuro-pathic pain.

Those scoring on this factor were relatively neutral about somebiological explanations, and there was some expression of disap-pointment or disillusionment with orthodox medicine in the freetext comments: Doctors have few tricks in their book to managethe pain. This is perhaps surprising given the relatively shortduration of pain and therefore of attempts at treatment. Pain inter-ference was scored higher than in the other 3 factors, suggesting agroup of people unable to nd a satisfactory way to live with thepain; they were relatively neutral about the statements on adjust-ment and acceptance.

No treatment was mentioned by more than 1 respondent, ex-cept surgery by 2, but the following were used in various combina-tions and in addition to surgery: unspecied analgesics,physiotherapy and physical methods, opioids, anticonvulsants,meditation and yoga, and alternative methods.

3.2. Participants responses to the factors

An effort was made to engage the participants via e-mail tocheck that their views had been accurately represented by theinterpretation of the factor on which they loaded. The 45 partici-pants with e-mail addresses who loaded on any factor were senta summary of that factor. Of those, 13, including at least 1 repre-sentative of each factor, replied: all agreed with the interpretationof their factor and stated that their views and attitudes about neu-ropathic pain had been accurately represented.

4. Discussion

We found that 4 main distinct accounts of neuropathic painemerged, summarised in terms of the key concepts shared or dif-fering. We focus particularly on the psychological content, becausein contrast with the musculoskeletal literature, in which psycho-

55 (2014) 349355logical models are becoming integrated in assessment and treat-ment, the NP literature recognises the distress experienced by

N1people with NP, but there is little work on how people understandand attribute meaning to their pain.

Factor 2 (F2) represents the extreme of commitment to causethat is expressed in biological-mechanical terms, with completerejection of psychological inuence, as if any admission of psycho-logical inuence undermined the authenticity of NP [21,46]. Evenso, those represented by F2 were agnostic about the possible ben-ets of psychologically-based treatments. F2 was the only factor tobe distinguished from others by its rejection of particularstatements.

Factor 1 (F1) was less concerned with cause, although thenerve-damage model was clearly held. This is unsurprising, giventhe respondents diagnoses (Table 1), but it is also common forpeople with NP to be provided with explanations which emphasisenerve damage [17]. An even clearer contrast between F1 and F2 layin the agreement with statements concerning the role of psychol-ogy in adjustment and in treatment. Nearly half of those repre-sented by F1 had tried or were considering cognitive behaviouraltherapy; none of those represented by other factors mentioned it.

Factor 3 (F3) and Factor 4 (F4) seemed to be in part determinedby particular features of NP. For F3, NP arose mainly from incurableconditions, so that questions of cause were less important thanthose of adjustment, inuenced by psychological factors, and ofsymptom management by medical or psychological means,whereas for F4, NP was of relatively recent onset and those repre-sented by it were not yet satised with diagnostic attempts, soquestions of cause were paramount. This may have representednot only hopes of resolving the cause, abolishing pain and restoringhealth or, if not, of validating the complaint of pain and legitimat-ing associated disability [46,47]. It may also reect the predomi-nant focus on treatment of life-threatening disease at theexpense of pain, as noted in cancer [7], or possibly claricationby health care staff that treatment options had been exhausted,fostering disillusionment with medical interventions, which wasevident in some statements.

It is interesting that agreement with psychological inuencewas specic to adjustment, rather than to incorporating a moreintegrated model of pain experience [56]. It was hard to nd anysophisticated statements of such an integrated pain model in liter-ature accessible to the public; instead, there were plenty of implicitand explicit statements of a psychosomatic version of pain, how-ever incompatible with the science of NP [5,14,31,63]. Patientswith chronic pain conditions are sensitive to constructions of painas being controllable, evoking a tendency to blame or hold respon-sible the affected individual [60], so they distance themselves frompsychosomatic attributions of pain [21], a common, albeit inaccu-rate, attribution when investigations are negative [49]. Across allaccounts, participants comments indicated that they had receivedpsychosomatic explanations of their pain and had been distressedand offended, consistent with other studies which use open-endedmethods to sample patients experiences [11,21].

Limited access to evidence-based and accessible informationabout neuropathic pain, except perhaps within pain clinics, meansthat people with NP are often exposed only to biomedical models[17] not written for the lay public [eg, [26,42,57]] and which makelittle or no mention of descending pathways in the spinal cord andbrain. NP mechanisms are complex and only partly understood[57], so it is unsurprising that clinicians struggle to convey an accu-rate and integrated model of NP comprehensible to the layperson.Further, where there is a denitive diagnosis, such as diabetic neu-ropathy or cancer, it is easier to regard pain as an inevitable part ofthat disease rather than to enquire further about how pain is gen-erated [4]. It is, however, of some concern that the information

S. Martin et al. / PAIavailable to patients outside written material produced by clinicsis likely unwittingly to exacerbate suspicion of psychology in painrather than to correct these misapprehensions.None of the 4 factors completely rejected the possibility of psy-chological treatment of pain or related problems, although freetext comments, in particular, suggested that trying them was morein the spirit of trying anything or of disappointment with orthodoxtreatment approaches than out of conviction that psychology wasof particular value. It is also possible that people with NP whoare hostile to psychologys involvement in pain treatment andwho saw the invitation might have decided not to respond to aninvitation from a psychologist. Only F1 endorsed the notion ofacceptance [33] or learning to live with pain. F3 and F4 strongly re-jected the idea, citing the unpredictability of pain and its constantinterruption and interference [16,17,35]. (Note that we use partic-ipants language: acceptance was a term provided in statementsand which respondents understood in their own ways and usedin free text. We are not asserting that what they understood by itis what Risdon and colleagues [47] and others have described.)

All respondents appeared to have consulted mainstream medi-cal services, although we do not have details. The term doctor ap-pears to stand for an oversimplied and somewhat demonisedclinician whose help is sought but who, for various reasons, oftendisappoints and becomes a gure of blame [21,41]. Medicationsof all types were listed by participants, except those in F4, who ap-peared to view medication use as impeding the search for thecause of pain [47]. This is in sharp contrast with guidelines (eg,[39]) which recommend neuroleptic and antidepressant drugs asrst-line NP treatment, although limited by restricted efcacy[24] and adverse effects [13]. There may also have been a self-selection process by which people obtaining good relief from NPmedication had little interest in taking part in the study and mayeven have been less likely to participate in the patient forumsthrough which many were recruited.

4.1. Strengths and limitations

As described above, sampling could be neither representativenor purposive and therefore cannot claim to describe the full rangeof viewpoints concerning NP. However, it is likely that the factorsprovide a snapshot of shared viewpoints and understandings thatare constructed by people with NP, as intended by the method[19]. We sampled beyond clinical populations because only a smallminority with persistent pain reach pain clinics. However, peoplewho complete online research tend to be better educated, youngerand middle class [23], so we may have seriously undersampledolder people among whom some NP conditions such as postherp-etic neuralgia are more prevalent than in younger people. Addi-tionally, the sorting process was time-consuming, and 3participants e-mailed to say that they gave up because of this;we do not doubt that many more did so but did not inform us. De-spite our efforts to reduce demand, it remains a concern and abarrier.

4.2. Implications and future research

There is no evidence from this study of a shared integrated bio-psychosocial model of neuropathic pain, but at least an outline ofone, used by clinicians and patients, would be helpful. Participantstended to emphasise biological aspects of pain and were ambiva-lent about or rejected the role of psychological factors. All 4 factorstook a position on cause, usually with physical and psychologicalcauses clearly segregated [29]. There is material here for more de-tailed investigation of lay understanding of the causes of NP, butthis probably requires visual and dynamic materials, rather thanwords alone. Further, it is unknown what explanations of pain, if

55 (2014) 349355 353any, are given to patients with NP as part of their health care, orhow they are incorporated or not into their existing frameworkof understanding. In general, people with chronic pain value a

es od

es: Mdicaoth

es: datient)

Treatment: medicaes: thrit

dicaoughes: s

dicaple

ator

1coherent explanation for their pain [36] and it may resolve worryand enable them to tackle pain differently [20]. Patients lookmainly to their doctors for explanation of symptoms [48], but doc-tors and other health care professionals may nd it hard to articu-late an integrated model of pain, may apply it only when there isno identied disease, or may apply it only to certain symptoms;many have poor understandings of chronic pain [18]. Results alsosuggest a wide-ranging approach to treatment modalities by peo-ple with NP and a willingness to consider psychological treat-ments, but availability of the latter is restricted [18,28], andthere are no trials of the predominant treatment, cognitive behav-ioural therapy, specic to NP [61].

4.3. Conclusion

The current study found, in a self-selected group of people withNP, constructions of their pain with some similarities but distinctand interesting differences, some of which may be associated withparticular features of the pain and the individualsmedical histories.It was clear, however, that no viewpoint integrated physical and

Factor 3: NP is (irreparable) nerve damage, forwhich symptom management is needed;psychological inuences play a part in painperception but psychological treatmentshave no role

Diagnoses/causrelated (MS, artchemotherapy.Treatments: meand positive th

Factor 4: NP (of recent onset) cause should beidentied; psychological inuences play arole; treatment may be medical and/orpsychological

Diagnoses/causunknownTreatments: memeditation, com

BPI, Brief Pain Inventory; MS, multiple sclerosis; NSAIDs, nonsteroidal anti-inammTable 1Factor themes and participant details for each .

Factor number and description Diagnoses/causTreatments trie

Factor 1: NP is a nervous system problem;psychology inuences pain experience andacceptance; open to psychologicaltreatments

Diagnoses/causTreatments: mebehavioural and

Factor 2: NP is nerve damage; psychology isirrelevant in pain experience; neutral aboutpsychological treatments

Diagnoses/causdiabetes); traum(nerve entrapm

354 S. Martin et al. / PAINpsychological factors to make sense of the conditions. Participantsdescribed receiving little explanation in health care, and this is likelyto inuence both the content and the complexity of their models.This is an important area for investigation and improvement.

Conict of interest statement

None of the authors have any conicts of interest with respectto this work.

Appendix A

Example of Q-sort distribution for Factor 1.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.pain.2013.10.021.

References

[1] Atkinson J. PQ Method 2.0 Manual; 1992. Available at: http://www.rz.unibwmuenchen.de/p41bsmk/qmethod/.[2] Baron R, Binder A, Wasner G. Neuropathic pain: diagnosis, pathophysiologicalmechanisms, and treatment. Lancet Neurol 2010;9:80719.

[3] Bennett MI, Attal N, Backonja MM, Baron R, Bouhassira D, Freynhagen R, ScholzJ, Tlle TR, Wittchen H-U, Jensen TS. Using screening tools to identifyneuropathic pain. PAIN 2007;127:199203.

[4] Bennett MI, Rayment C, Hjermstad M, Aass N, Caraceni A, Kaasa S. Prevalenceand aetiology of neuropathic pain in cancer patients: a systematic, review.PAIN 2012;153:35965.

[5] Blackbeard J, Wallace VCJ, ODea KP, Hasnie F, Segerdahl A, Pheby T, Field MJ,Takata M, Rice ASC. The correlation between pain-related behaviour and spinalmicrogliosis in four distinct models of peripheral neuropathy. Eur J Pain2012;16:135767.

[6] Blyth FM, March LM, Nicholas MK, Cousins MJ. Self-management of chronicpain: a population-based study. PAIN 2005;113:28592.

[7] Breivik H, Cherny N, Collett B, de Conno F, Filbet M, Foubert AJ, Cohen R, Dow L.Cancer-related pain: a pan-European survey of prevalence, treatment, andpatient attitudes. Ann Oncol 2009;20:142033.

[8] Breivik H, Collett B, Ventafridda V, Cohen R, Gallacherd D. Survey of chronicpain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain2006;10:28797.

[9] Brown S. Political subjectivity: applications of Q methodology in politicalscience. New Haven, CT: Yale University Press; 1980.

[10] Brown SR. A primer on Q methodology. Operant Subjectivity 1993;16:91138.

[11] Burbaum C, Stresing A-M, Fritzsche K, Auer P, Wirsching M, Lucius-Hoene G.Medically unexplained symptoms as a threat to patients identity? A

f pain NMedian duration of painBPI scores: mean (SD)

S, trauma, surgery, PHN.tion, physical methods, cognitiveer psychological treatments

n = 26Duration median 67 yPain 5.5 (1.8)Interference 5.8 (2.4)

isease-related (cancer, HIV,c (injury), hereditary, mechanical, environmental (toxins).tion, surgery, physical methods.

n = 15Duration median 6yPain 5.6 (2.2)Interference 6.0 (1.6)

raumatic (surgery, injury), disease-is, shingles), hereditary,

tions, physical treatments, breathingts.

n = 11Duration median 10 yPain 4.3 (2.0)Interference 5.1 (2.7)

urgery, HIV, chemotherapy,

tion, physical methods, yoga andmentary therapies

n = 4Duration median 12 yPain 5.4 (2.0)Interference 7.3 (2.8)

ies; PHN, post-herpetic neuralgia.

55 (2014) 349355conversation analysis of patients reactions to psychosomatic attributions.Patient Educ Couns 2010;79:20717.

[12] Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory.Ann Acad Med Singapore 1994;23:12938.

[13] Closs SJ, Staples V, Reid I, Bennett MI, Briggs M. Managing the symptoms ofneuropathic pain: an exploration of patients experiences. J Pain SymptomManage 2007;34:42233.

[14] Crombez G, Beirens K, Van Damme S, Eccleston C, Fontaine J. The unbearablelightness of somatisation: a systematic review of the concept of somatisationin empirical studies of pain. PAIN 2009;145:315.

[15] Crombez G, Eccleston C, Van Damme S, Vlaeyen JWS, Karoly P. Fear-avoidancemodel of chronic pain: the next generation. Clin J Pain 2012;28:47583.

[16] Daniel C, Narewska J, Serpell M, Hoggart B, Johnson R, Rice A. Comparison ofpsychological and physical function in neuropathic pain and nociceptive pain:implications for cognitive behavioural management programs. Eur J Pain2008;12:73141.

[17] Daniel HC, Van der Merwe JD. Cognitive behavioural approaches andneuropathic pain. In: Cervero F, Jensen TS, editors. Handbook of clinicalneurology, vol. 81. Amsterdam, The Netherlands: Elsevier; 2006. p.85568.

[18] Donaldson L. 150 Years of the Annual Report of the Chief Medical Ofcer: Onthe State of Public Health 2008. London, UK: DH; 2009.

[19] Donner JC. Using q-sorts in participatory processes: an introduction to themethodology. Soc. Develop Papers 2001;36:2449 [Available at: http://siteresources.worldbank.org/SOCIALANALYSIS/1104890-1120158652972/20566697/SDP-36.pdf#page=30].

[20] Eccleston C, Crombez G. Worry and chronic pain: a misdirected problem-solving model. PAIN 2007;132:2336.

[21] Eccleston C, Williams A, Stainton Rogers W. Patients and professionalsunderstandings of the causes of chronic pain: blame, responsibility andidentity protection. Soc Sci Med 1997;45:699709.

[22] Elliott AM, Smith BH, Penny KI, Smith WC, Chambers WA. The epidemiology ofchronic pain in the community. Lancet 1999;354:124852.

[23] Eysenbach G, Khler C. How do consumers search for and appraise healthinformation on the world wide web? Qualitative study using focus groups,usability tests, and in-depth interviews. BMJ 2002;324:5737.

[24] Finnerup NB, Sindrup SH, Jensen TS. The evidence for pharmacologicaltreatment of neuropathic pain. PAIN 2010;150:57381.

[25] Freynhagen R, Bennett MI. Diagnosis and management of neuropathic pain.BMJ 2009;339:b3002.

[26] Geber C, Baumgrtner U, Schwab R, Mller H, Stoeter P, Dieterich M, SommerC. Revised denition of neuropathic pain and its grading system: an open caseseries illustrating its use in clinical practice. Am J Med 2009;122:S3S12.

[27] Haanp M, Attal N, Backonja M, Baron R, Bennett M, Bouhassira D, Cruccu G,Hansson P, Haythornthwaite JA, Ianetti GD, Jensen TS, Kauppila T, Nurmikko TJ,Rice ASC, Rowbotham M, Serra J, Sommer C, Smith BH, Treede R-D. NeuPSIGguidelines on neuropathic pain, assessment. PAIN 2011;152:1427.

[28] Haythornthwaite JA, Benrud-Larson LM. Psychological aspects of neuropathicpain. Clin J Pain 2000;16:15.

[29] Kenny D. Constructions of chronic pain in doctor-patient relationships:bridging the communication chasm. Patient Educ Couns 2004;52:297305.

[30] Leventhal H, Benyamini Y, Shafer C. Lay beliefs about health and illness. In:Ayers S, Baum A, McManus C, Newman S, Wallston K, Weinman J, West R,editors. Cambridge handbook of psychology, health and medicine. Cambridge,UK: Cambridge University Press; 2007. p. 1248.

[31] Maratou K, Wallace VCJ, Hasnie FS, Okuse K, Hosseini R, Jina N, Blackbeard J,Pheby T, Orengo C, Dickenson AH, McMahon SB, Rice ASC. Comparison of

[43] Q-Assessor. Available at: .[44] Radley A. Health psychology, embodiment and the question of vulnerability. J

Health Psychol 2000;5:297304.[45] Reber B, Kaufman S, Cropp F. Assessing Q-Assessor: a validation study of

computer-based Q-sorts versus paper sorts. Operant Subjectivity2000;23:192209.

[46] Rhodes LA, McPhillips-Tangum CA, Markham C, Klenk R. The power of thevisible: the meaning of diagnostic tests in chronic back pain. Soc Sci Med1999;48:1189203.

[47] Risdon A, Eccleston C, Crombez G, McCracken L. How can we learn to live withpain? A Q-methodological analysis of the diverse understandings ofacceptance of chronic pain. Soc Sci Med 2003;56:386.

[48] Salmon P, Peters S, Stanley I. Patients perceptions of medical explanations forsomatisation disorders: qualitative analysis. BMJ 1999;318:3726.

[49] Sharpe M, Mayou R. Somatoform disorders: a help or hindrance to goodpatient care? Brit J Psychiatry 2004;184:4657.

[50] Smith BH, Torrance M, Bennett MI, Lee AJ. Health and quality of life associatedwith chronic pain of predominantly neuropathic origin in the community. ClinJ Pain 2007;23:1439.

[51] Smith BH, Torrance N, Johnson M. Assessment and management ofneuropathic pain in primary care. Pain Manag 2012;2:5539.

[52] Stainton Rogers R. Q Methodology. In: Smith JA, Harre J, Langenhove LV,editors. Rethinking methods in psychology. London, UK: Sage Publications;1995. p. 17892.

[53] Stenner PH, Dancey C, Watts S. The understanding of their illness amongstpeople with irritable bowel syndrome: a Q methodological study. Soc Sci Med2000;51:43952.

[54] Tan G, Jensen MP, Thornby JI, Shanti BF. Validation of the Brief Pain Inventory

S. Martin et al. / PAIN155 (2014) 349355 355dorsal root ganglion gene expression in rat models of traumatic and HIV-associated neuropathic pain. Eur J Pain 2009;13:38798.

[32] Matthews A. Why worry? The cognitive function of anxiety. Behav Res Ther1990;28:45568.

[33] McCracken L, Zhao OBrien J. General psychological acceptance and chronicpain: there is more to accept than the pain itself. Eur J Pain 201;14:1705.

[34] McNaught A, Howard C. Q-methodology: pragmatic considerations andepistemological concerns. Health Psychol Update 2001;10:248.

[35] Morley S, Eccleston C. The object of fear in chronic pain. In: Asmundson G,Vlaeyen J, Crombez G, editors. Understanding and treating fear of pain. Oxford,UK: Oxford University Press; 2004. p. 16388.

[36] Morris DB. The culture of pain. Los Angeles, CA: University of California Press;1991.

[37] Moseley GL, Nicholas MK, Hodges PW. A randomized controlled trial ofintensive neurophysiology education in chronic low back pain. Clin J Pain2004;20:2430.

[38] Newton-John TRO. Psychological effects of chronic pain and their assessmentin adults. In: Jensen T, Wilson PR, Rice ASC, editors. Clinical Pain Management.Chronic Pain, vol. 3. London, UK: Arnold Press; 2002. p. 10111.

[39] NICE. Neuropathic pain: pharmacological management. NICE. Guidance/Clinical Guidelines; 2010. Available at: .

[40] Peters S, Stanley I, Rose M, Salmon P. Patients with medically unexplainedsymptoms: sources of patients authority and implications for demands onmedical care. Soc Sci Med 1998;46:55965.

[41] Petrie KJ, Frampton T, Large RG, Moss-Morris R, Johnson M, Meechan G. Whatdo patients expect from their rst visit to a pain clinic? Clin J Pain2005;21:297301.

[42] Portenoy R. Neuropathic pain. In: Portenoy R, Kanner R, editors. Painmanagement: theory and practice. Philadelphia, PA: FA Davis; 1996. p.83125.for chronic nonmalignant pain. J Pain 2004;5:1337.[55] Torrance N, Ferguson JA, Afolabi E, Bennett MI, Serpell MG, Dunn KM, Smith

BH. Neuropathic pain in the community: More under-treated than refractory?PAIN 2013;154:6909.

[56] Tracey I, Mantyh PW. The cerebral signature for pain perception and itsmodulation. Neuron Rev 2007;55:37791.

[57] Treede RD, Jensen TS, Campbell JN, Cruccu G, Dostrovsky JO, Grifn JW,Hansson P, Hughes R, Nurmikko T, Serra J. Neuropathic pain: redenition and agrading system for clinical and research purposes. Neurol 2008;70:16305.

[58] Van Exel NJA, de Graaf G. Q methodology: a sneak preview; 2005. Available at:.

[59] Watts S, Stenner P. Doing Q methodology: theory, method and interpretation.Qual Res Psychol 2005;2:6791.

[60] Weiner B, Perry RP, Magnusson J. An attributional analysis of reactions tostigmas. J Pers Soc Psychol 1988;55:73848.

[61] Williams ACDC, Eccleston C, Morley S. Psychological therapies for themanagement of chronic pain (excluding headache) in adults. CochraneDatabase System Rev 2012;11:CD007407.

[62] Willig C. A discourse-dynamic approach to the study of subjectivity in healthpsychology. Theor Psychol 2000;10:54770.

[63] Woolf C, Mannion R. Neuropathic pain: aetiology, symptoms, mechanisms andmanagement. Lancet 1999;353:195964.

[64] Zelman DC, Gore M, Dukes E, Tai K-S, Brandenburg N. Validation of a modiedversion of the Brief Pain Inventory for painful diabetic peripheral neuropathy. JPain Symptom Manage 2005;29:40110.

How do people understand their neuropathic pain? A Q-study1 Introduction2 Methods2.1 Q-methodology2.2 Data analysis

3 Results3.1 Factor interpretation3.1.1 Factor 13.1.2 Factor 23.1.3 Factor 33.1.4 Factor 4

3.2 Participants responses to the factors

4 Discussion4.1 Strengths and limitations4.2 Implications and future research4.3 Conclusion

Conflict of interest statementAppendix AAppendix A Supplementary dataReferences