Rheumatology 2001; 40: 95-101
© 2001 British Society for Rheumatology
Risk factors for persistent chronic widespread pain: a community-based study
1 Arthritis Research Campaign (ARC) Epidemiology Unit, and
2 Unit of Chronic Disease Epidemiology, School of Epidemiology and Health Sciences, Stopford Building, University of Manchester, Oxford Road, Manchester M13 9PT, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Chronic widespread pain is the cardinal clinical feature of the fibromyalgia syndrome, which, in the majority of clinic patients, is persistent. By contrast, in community-derived patients, pain is persistent in only half of the affected individuals, particularly those with psychological distress. Whether such distress is a consequence of the pain or a manifestation of a wider process of somatization which is associated with the persistence of pain is unclear.
Objectives. We tested in a large, prospective, population-based study the hypothesis that features of somatization predict the persistence of chronic widespread pain.
Methods. In all, 252 (13%) of 1953 adult subjects selected from a population register were classified as having chronic widespread pain based on a detailed questionnaire which included a pain drawing. The patients also completed a number of psychosocial instruments which measure features known to be associated with somatization. Two hundred and twenty-five (91%) of the patients were successfully followed up after 12 months and provided data on pain status using the same instruments.
Results. In all, 126 (56%) patients reported chronic widespread pain at follow-up, 74 (33%) reported other pain and 25 (11%) reported no pain. Persistent chronic widespread pain was strongly associated with baseline test scores for high psychological distress and fatigue. In addition, these subjects were more likely to display a pattern of illness behaviour characterized by frequent visits to medical practitioners for symptoms which disrupt daily activities. The prevalence of persistent pain increased with the number of risk factors the subjects were exposed to.
Conclusions. Although almost half of the cases of chronic widespread pain resolved within 1 yr, this study has demonstrated for the first time that those subjects who display features of somatization are more likely to have widespread pain which persists. These findings have implications for the identification and treatment of persons with persistent chronic widespread pain.
KEY WORDS: Chronic widespread pain, Persistence, Risk factors, Somatization.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Fibromyalgia has been described as a non-articular rheumatic syndrome whose cardinal feature is chronic widespread pain in the presence of widespread tenderness (as measured by multiple tender points) [1]. Population studies have suggested that chronic widespread pain is common [2]. We have shown that chronic widespread pain in community subjects is associated with psychological disturbance, reporting other somatic symptoms, fatigue, low levels of self-care, and measures of hypochondriasis and bodily preoccupation [2, 3]. Persons with such symptoms are also more likely to have a mental disorder such as depressive or anxiety disorders [4]. These findings suggest that chronic widespread pain is associated with a constellation of symptoms indicative of the process of somatization [5], and we have shown recently that these factors predict the onset of new cases of chronic widespread pain [6]. The process of somatization is a common form of illness behaviour in which persons communicate psychological distress through physical symptoms for which medical attention may be sought [7]. This form of presentation to primary care is distinct from the psychiatric diagnostic criteria for somatization disorder [8, 9], although the two phenomena share features such as the reporting of multiple symptoms and the seeking of medical help. Somatization disorder is a more specific syndrome and represents only a small proportion of true somatizers [7].
Clinic studies of persons with widespread pain, which are dominated by persons who consult with fibromyalgia, suggest that such pain symptoms tend to persist in the majority of patients [10]. However, there is little information on the natural history of these symptoms in the general population. The only population-based study of the natural history of chronic widespread pain reported that community subjects generally had a good prognosis when compared with clinic patients [11]. In that study, of 34 subjects with chronic widespread pain only 12 (35%) still had symptoms after a median follow-up period of 27 months. Subjects with persistent symptoms were more likely to be female, to be older and to be psychologically distressed. Whether these associations between psychological distress and chronic widespread pain varied with age and sex was unknown. However, since only a small number of subjects with persistent pain were identified, the relevance of these observations to outcome and its predictors requires confirmation [12]. More importantly, whether persistent episodes of chronic widespread pain are associated with the process of somatization is unknown. To assess the association between the persistence of pain and the process of somatization, it is necessary to assess the features of somatization rigorously in a larger sample of persons with chronic widespread pain. We hypothesized that those features of the process of somatization which predict new cases of chronic widespread pain are associated with the persistence of pain.
The aims of the present study were to examine rigorously the hypothesis that features of the process of somatization, including a history of reporting somatic symptoms, medical help-seeking behaviour, an increased level of psychological distress and an increased level of fatigue, a symptom often associated with somatization [7], predict the persistence of such pain symptoms.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Design outline
The study was a population-based prospective outcome study. At baseline, subjects who had chronic widespread pain were identified and various indicators of somatization were measured. Twelve months after the baseline survey, subjects were followed up and those with persistent chronic widespread pain were identified.
Baseline
A random sample of 3004 subjects aged 18–65 yr was selected from a population-based primary care register, covering a mixed sociodemographic suburban area in south Manchester.
Each subject was sent a questionnaire by post which enquired whether the subject had experienced any pain during the previous month that had persisted for at least 24 h and, if so, whether the pain had lasted for more than 3 months. Four line drawings of the body were included (front, back and sides) on which subjects were asked to indicate the site(s) of pain. These methods have been used previously to determine the location and duration of pain [2, 11]. On the basis of this information, subjects were categorized into three pain groups. Chronic widespread pain was defined using the definition of the American College of Rheumatology (ACR) in their criteria for fibromyalgia [1]. To satisfy these criteria, subjects must have pain that is present both in two contralateral quadrants of the body and in the axial skeleton, and this pain must be present for at least 3 months. All subjects who had experienced pain during the previous month which had lasted for longer than 24 h but which did not satisfy criteria for chronic widespread pain were placed in the category ‘other pain’. The ‘no pain’ group comprised the subjects who did not report any pain in the past month lasting for more than 24 h. Of the 3004 questionnaires mailed, 1953 were returned completed (75% response rate after adjustment for persons not resident at their listed address; n=402). Of the subjects who responded, 252 (13%) satisfied the criteria for chronic widespread pain and were therefore eligible for follow-up.
Features of the process of somatization
The baseline pain questionnaire also included sections to assess aspects of physical complaints, illness attitudes and behaviours, and levels of psychological distress which have been associated with the process of somatization. These were as follows:
- Somatic Symptom Checklist. This was originally validated as a screening test for somatization disorder [13]. The scale includes six items: trouble breathing, frequent vomiting (when not pregnant), loss of voice for more than 30 min, being unable to remember what you have been doing for hours or days (without the influence of alcohol or drugs), difficulty swallowing, and frequent pain in the fingers or toes. There is an additional item for females—frequent trouble with menstrual cramps. These symptoms are included in the American Psychiatric Association criteria for somatization disorder [14]. In that study a threshold of between 3 and 4 symptoms reported resulted in a sensitivity of 73% and a specificity of 94% for identifying cases of somatization disorder [10]. To avoid spurious associations with persistent cases of chronic widespread pain, only non-pain somatic symptoms were examined (i.e. frequent trouble with menstrual cramps and frequent pain in the fingers and toes were excluded). The total score was therefore between 0 and 5 for both males and females.
- Illness Attitude Scales. The nine Illness Attitude Scales (IAS) [15] assess attitudes and concerns about illness and health. Each scale includes three items, each scored from 0 to 4, providing a total score between 0 and 12. Individual scales assess worry about health, concern about pain, health habits, hypochondriacal beliefs, thanatophobia (fear of death), disease phobia, bodily preoccupation, treatment experience and the effect of symptoms. A recent study based on a principal components analysis demonstrated that the IAS measures two dimensions reflecting ‘health anxiety’ and ‘illness behaviour’ [16]. The Health Anxiety subscale consists of 11 items (such as ‘Are you worried that you may get a serious illness in the future?’) and has a total score between 0 and 44 and a general population mean score of 9.1 (S.D.=6.9). The Illness Behaviour subscale consists of six items (such as ‘Do your bodily symptoms stop you from working?’); it has a total score between 0 and 24 and a general population mean score of 4.7 (S.D.=4.2).
- General Health Questionnaire. The 12-item version of the General Health Questionnaire (GHQ) [17] was included as a measure of psychological distress. Each item has four possible responses, but for scoring these were dichotomized at the midpoint. Scores for individual items are summed to give a total score ranging between 0 and 12, high scores indicating higher levels of psychological distress.
- Fatigue Questionnaire. The Fatigue Questionnaire [18] is an 11-item instrument developed for use in population studies to measure physical and mental aspects of fatigue, a feature commonly observed in many somatizing patients. Each item has four response options. In a similar manner to the GHQ, scores for individual items were dichotomized at the midpoint and these individual scores were summed to provide a total score between 0 and 11, high scores corresponding to high levels of fatigue.
Follow-up
The 252 subjects with chronic widespread pain at baseline were mailed an identical questionnaire after 12 months. The methods for categorizing pain were the same as those used for the baseline survey. The observer categorizing pain status was blind to the subjects' baseline pain status, measures of somatization and all other information included in the baseline questionnaire. Chronic widespread pain at follow-up was defined using the ACR definition as discussed previously.
Statistical analysis
Those subjects who provided complete data at baseline and follow-up were included in the analysis. The distribution of the baseline psychosocial scale scores was not Gaussian. Thus, subjects scoring zero on the GHQ and Fatigue scales were classified as one group (the referent group), and the remaining subjects were split into thirds of the distribution of their scale scores to produce three approximately equally sized groups. For the IAS somatization subscales and the Somatic Symptom Checklist, subjects' scores were divided into thirds. The association between persistent chronic widespread pain and the scale scores of those subjects scoring in the middle and highest thirds were compared with that of patients scoring in the lowest third. If a significant association was evident, the subjects' scale scores were then dichotomized at the point where this was greater. The dichotomized scales were used to construct a multivariate model of the associated features of persistent chronic widespread pain. Each scale was added to the model one at a time. After each addition, the goodness of fit of that model was tested using likelihood ratio tests. Those scales which significantly improved the fit of the model (P<0.05) were retained. Since no significant heterogeneity by age or sex was evident [Mantel–Haenszel (log-rank) test for heterogeneity [19], P>0.05], the results are presented as summary odds ratios (OR) with 95% confidence intervals (CI). The risk of persistent pain symptoms was then examined for those subjects exposed to none, one, two or three of the risk factors which were included in the final model. All analyses were conducted using the Stata statistical software [20].
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Response rates and study subjects
In all, 252 subjects (88 males, 164 females) satisfied the definition of chronic widespread pain at baseline. On examination of the electoral register at the time of the follow-up survey, it was found that six of these subjects (three males, three females) were no longer registered as living at the address of record at their general practice. It was therefore assumed that these subjects did not receive the follow-up questionnaire and they were removed from the denominator, leaving a study population of 246. Of these, 225 [82 males (96%), 143 females (87%)] responded at follow-up, giving an adjusted participation rate of 91%. No difference in the age structure was observed when those subjects who participated were compared with the 21 who did not (Table 1
). Non-participants were more likely to be female and to have a high score on the GHQ. However, they did not differ significantly on any of the other baseline measures of the process of somatization.
|
Prevalence of persistent chronic widespread pain
Of the 225 subjects at baseline who satisfied the criteria for chronic widespread pain, 25 (11%) reported no pain at follow-up and 74 (33%) reported other pain, while just over half [126 (56%)] again reported chronic widespread pain. Women were not significantly more likely than men to report persistent chronic widespread pain [56.6% (95% CI 48.1–64.9%) and 54.9% (95% CI 43.5–65.9%) respectively]. However, for both sexes the prevalence of persistent pain increased with age (Table 2).
|
Features of the process of somatization as predictors of persistent chronic widespread pain
Table 3 shows the age- and sex-adjusted univariate associations of the baseline scales with persistent chronic widespread pain. Subjects scoring in the middle and highest thirds of the Illness Behaviour scale were three times more likely to report persistent chronic widespread pain than those in the lowest third. Scoring in the highest range of the fatigue scale was associated with a fourfold increased risk of persistence. However, the greatest risk of persistent chronic widespread pain was associated with scoring in the highest range of the GHQ. Although it was not significant, a history of reporting two or more non-pain somatic symptoms was associated with a 60% increased risk of having persistent chronic widespread pain. Interestingly, a high score on the Health Anxiety scale was not associated with the persistence of pain.
|
Multivariate analysis demonstrated that the three scales which were significantly associated with persistent pain in the univariate analysis remained the best predictors of persistent pain when adjusted for the effects of age, gender and the effects of the other scales. Thus, high scores on the Illness Behaviour scale, the GHQ and the Fatigue scale were all associated with a two-fold increased risk of persistent chronic widespread pain (Table 4
).
|
The prevalence of persistent chronic widespread pain was observed to increase with the number of those factors identified in the multivariate analysis which subjects were exposed to. Thus, the prevalence of persistent chronic widespread pain was 35% in subjects exposed to none of the factors, 48% in those exposed to one, 66% in those exposed to two and 87% in those exposed to all three (Table 5
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study examined the natural history of chronic widespread pain in community subjects. Of those persons with chronic widespread pain at baseline, just over half reported symptoms at follow-up. The prevalence of persistent pain did not differ by gender, although it did increase with age. We have shown for the first time new associations with features of the process of somatization, including psychological distress, a pattern of illness behaviour which is typified by medical care-seeking behaviour and high levels of fatigue. These findings support the hypothesis that the persistence of chronic widespread pain can be predicted by simple measures of somatization.
These findings support our previous study [11], which found that those persons with chronic widespread pain who in addition had high levels of psychological distress were more likely to have symptoms that persisted. In that study, 35% of subjects had pain which persisted at a median follow-up interval of 2 yr, compared with 56% of subjects in the present study. This difference may be explained by the difference in the length of follow-up because, over time, a proportion of persons with chronic widespread pain are likely to have pain which is less widespread or to recover completely and report no pain.
In considering these results, the following methodological issues must be highlighted. First, of those subjects who reported chronic widespread pain at baseline, a number did not participate at follow-up. Although the participation rate at follow-up was high, we were concerned about the possible effects of non-participation bias. We therefore compared the baseline responses of those subjects who participated with those who did not. Non-participants were more likely to be female and had higher baseline GHQ scores. However, there were no other significant differences between the two groups. Nevertheless, these differences would only affect the internal comparisons in the present study if the relationship between female sex and GHQ score and the persistence of chronic widespread pain were different in those subjects who participated compared with those who did not. This seems unlikely.
Secondly, the subjects in the present study were assessed at two time points 12 months apart, and prevalent cases of chronic widespread pain were identified at those two time points. We do not know what happened in the intervening months. It is possible that subjects who were classified as having chronic widespread pain at baseline may have had symptoms which resolved during that period and who then developed symptoms which again satisfied the criteria at follow-up. Although this seems unlikely, since by definition widespread pain symptoms must be present for at least 3 months before the time of survey, it must remain a possibility. We could have attempted to determine what had happened in the intervening months; however, recall of pain symptoms over a 12-month period is notoriously inaccurate [21]. In the light of this, we believe that more accurate information would be provided by enquiring whether a person's pain symptoms had been present over the shorter time-span of the previous 3 months.
Thirdly, the pain symptoms of subjects were gathered using a manikin included in a postal questionnaire. It has been argued that the location and extent of chronic pain are best determined by physical examination by a qualified practitioner. However, we would argue that since pain is a wholly subjective experience, and that even during physical examination a qualified practitioner relies on patient self-report, this claim appears unjustified. The construct validity of this approach to determining pain has been demonstrated. We [3] and others [22, 23] have used manikin-derived pain symptoms previously to demonstrate associations between psychological and physical factors and the presence of regional and widespread pain symptoms in community subjects.
Fourthly, the pain status of subjects and the features of the process of somatization were measured at the same time during the baseline survey. These data, therefore, do not allow us to say anything about the temporal relationship between chronic widespread pain and these measures. However, this study has demonstrated that, for those persons with chronic widespread pain and who have additional features associated with the process of somatization, their prognosis after 12 months is worse than that of persons without such additional features, irrespective of whether these features precede or are a consequence of pain.
Fifthly, in the present study we did not assess whether subjects had been diagnosed with a physical illness at the time of the baseline survey. It is conceivable that at baseline a proportion of subjects may have reported chronic widespread pain symptoms which were associated with physical illness, and this may be particularly true in older subjects. Pain symptoms at follow-up may also have been associated with that illness. Although we have no data to allow us to assess this possibility, it is still apparent from the present study that psychological factors, even if they are a consequence of physical illness, are associated with the persistence of pain in community subjects.
In summary, we have found that, of those persons selected from the community who satisfy criteria for chronic widespread pain, just under half will have symptoms which will resolve over 12 months. Persons who display psychological features associated with the process of somatization are more likely to have pain symptoms which persist. What, therefore, are the implications of these findings? By administering three simple self-completion questionnaires, medical practitioners are able to identify a group of persons whose symptoms are likely to resolve and a group whose symptoms are likely to persist. In this way these simple measures may be used to target health-care provision more usefully. These findings may also be useful in developing therapeutic interventions. Persistent chronic widespread pain may be alleviated if appropriate cognitive behavioural therapies, which aim to alter specific illness behaviour patterns, were targeted to persons with such pain. This avenue of possible therapeutic benefit to persons with persistent chronic widespread pain merits further study.
|
Acknowledgments |
|---|
The authors are grateful for the participation and help of the doctors, staff and patients of the general practice in greater Manchester, to Professor Peter Croft, Dr Sidney Benjamin and Dr Stella Morris for their help in conceiving and conducting the study, and to Ann Papageorgiou for administration of the survey. The study was supported by the Arthritis Research Campaign, Chesterfield, UK
|
Notes |
|---|
Correspondence to: J. McBeth
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Wolfe F, Smythe HA, Yunus MB et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Arthritis Rheum1990;33:160–72.[ISI][Medline]
- Croft P, Rigby AS, Boswell R, Schollum J, Silman A. The prevalence of chronic widespread pain in the general population. J Rheumatol1993;20:710–3.[ISI][Medline]
- Hunt IM, Silman AJ, Benjamin S, McBeth J, Macfarlane GJ. The prevalence and associated features of chronic widespread pain in the community using the ‘Manchester’ definition of chronic widespread pain. Rheumatology1999;38:275–9.
[Abstract/Free Full Text] - Benjamin S, Morris S, McBeth J, Macfarlane GJ, Silman GJ. The association between chronic widespread pain and mental disorder. Arthritis Rheum2000; 43:561–7.[ISI][Medline]
- Kleinman A, Kleinman J. Somatisation: the interconnectedness in Chinese society among culture depressive experiences and meaning of pain. In: Kleinman A, Good B, eds. Culture and depression. Studies in the anthropology of cross culture psychiatry of affect and disorder. London: University of California Press, 1985: 429–90.
- Silman AJ, McBeth J, Papageorgiou AC, Hunt IM, Benjamin S, Macfarlane GJ. Features of the process of somatisation predict the onset of chronic widespread pain: Results from a large population based study. Arthritis Rheum1998;41(Suppl.):S358.
- Bhui K, Hotopf M. Somatization disorder. Br J Hosp Med1997;58:145–9.[Medline]
- World Health Organization (1992) ICD-10 Classification of Mental and Behavioural Disorders. Clinical Descriptions and Diagnostic Guidelines. Geneva: World Health Organization, 1992.
- American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders. DSM-IV. 4th ed. Washington DC: American Psychiatric Association, 1994.
- Ledingham J, Doherty S, Doherty M. Primary fibromyalgia syndrome—an outcome study. Br J Rheumatol1993;32:139–42.
[Abstract/Free Full Text] - Macfarlane GJ, Thomas E, Papageorgiou AC, Scollum J, Croft PR, Silman AJ. The natural history of chronic pain in the community: A better prognosis than in the clinic? J Rheumatol1996;23:1617–20.[ISI][Medline]
- White KP, Harth M. The occurrence and impact of generalized pain. In: Croft P, Silman AJ, eds. Baillière's Clinical Rheumatology. London: Ballière Tindall, 1999: 379–89.
- Othmer E, DeSouza C. A screening test for somatisation disorder. Am J Psychiatry1985;142:1146–9.
[Abstract/Free Full Text] - American Psychiatric Association. American Psychiatric Association diagnostic and statistical manual. Washington DC: American Psychiatric Association, 1980.
- Kellner R, Abbott P, Winslow WW, Pathak D. Fear, beliefs and attitudes in DSM-III hypochondriasis. J Nerv Ment Dis1987;175:20–4.[ISI][Medline]
- Speckens AE, Spinhoven P, Sloekers PPA, Bolk JH, Van Hemert AM. A validation study of the Whitely Index, the Illness Attitude Scales, and the Somatosensory Amplification Scale in general medical and general practice patients. J Psychosom Res1996;40:95–104.[ISI][Medline]
- Goldberg DP, Williams P. The user's guide to the General Health Questionnaire. Windsor: NEFR-Nelson, 1988.
- Chalder T, Berelowitz G, Pawilkowska T, Watts L, Wessely S, Wright D et al. Development of a fatigue scale. J Psychosom Res1993;37:147–53.[ISI][Medline]
- Mantel N, Haenszel WH. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst1959;22:719–48.[ISI][Medline]
- Stata reference manual, release 3.1, 6th ed. College Station, TX: Stata Corporation, 1993.
- Linton SJ, Gotestam GK. A clinical comparison of two pain scales: correlation, remembering chronic pain, and a measure of compliance. Pain1983;17:57–65.[ISI][Medline]
- Croft PR, Papageorgiou AC, Ferry S, Thomas E, Jayson MI, Silman AJ. Psychological distress and low back pain. Evidence from a prospective study in the general population. Spine1995;20:2731–7.[ISI][Medline]
- White KP, Speechley M. The London Fibromyalgia Epidemiology Study: The prevalence of fibromyalgia syndrome in London, Ontario. J Rheumatol1999; 26:1570–6.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Carnes, S. Parsons, D. Ashby, A. Breen, N. E. Foster, T. Pincus, S. Vogel, and M. Underwood Chronic musculoskeletal pain rarely presents in a single body site: results from a UK population study Rheumatology, July 1, 2007; 46(7): 1168 - 1170. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Staud, C. J. Vierck, M. E. Robinson, and D. D. Price Overall fibromyalgia pain is predicted by ratings of local pain and pain-related negative affect--possible role of peripheral tissues Rheumatology, November 1, 2006; 45(11): 1409 - 1415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. H Smith, A. M Elliott, and P. C Hannaford Is chronic pain a distinct diagnosis in primary care?: Evidence arising from the Royal College of General Practitioners' Oral Contraception study Fam. Pract., February 1, 2004; 21(1): 66 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M Valencia-Flores, M H Cardiel, V Santiago, M Resendiz, V A Castano, O Negrete, C Rosenberg, G Garcia-Ramos, J Alcocer, and D Alarcon-Segovia Prevalence and factors associated with fibromyalgia in Mexican patients with systemic lupus erythematosus Lupus, January 1, 2004; 13(1): 4 - 10. [Abstract] [PDF]
|
|
|
|
 |
|
 A C Papageorgiou, A J Silman, and G J Macfarlane Chronic widespread pain in the population: a seven year follow up study Ann Rheum Dis, December 1, 2002; 61(12): 1071 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J Macfarlane, J. McBeth, A. J Silman, and I K Crombie Widespread body pain and mortality: prospective population based study Commentary: An interesting finding, but what does it mean? BMJ, September 22, 2001; 323(7314): 662 - 662. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||