Family Practice Vol. 21, No. 1, 66-74
© Oxford University Press 2004, all rights reserved.
Article |
Is chronic pain a distinct diagnosis in primary care?
Evidence arising from the Royal College of General Practitioners' Oral Contraception study
Department of General Practice and Primary Care, University of Aberdeen, Foresterhill Health Centre, Westburn Road, Aberdeen AB25 2AY, UK
E-mail: blairsmith{at}abdn.ac.uk
Received 19 March 2003; Revised 20 August 2003; Accepted 8 September 2003.
Smith BH, Elliott AM and Hannaford PC. Is chronic pain a distinct diagnosis in primary care? Evidence arising from the Royal College of General Practitioners' Oral Contraception study. Family Practice 2004; 21: 66–74.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsDiscussionReferences |
|---|
Background and Objectives. There is a case for considering chronic pain as a distinct diagnosis in primary care, independent of its cause or body site, to allow development of common management strategies. This study examined the prevalence and factors associated with chronic pain among women still in the Royal College of General Practitioners' Oral Contraception Study, and compared the factors associated with chronic pain arising at different body sites, to examine the support for this case.
Methods. A survey of 11 797 women examined the presence and site(s) of chronic pain, and other health factors. Socio-demographic factors, attendance at the family doctor for the pain and self-rated health were examined for their association with any chronic pain and site-specific chronic pain.
Results. Thirty-eight percent of women reported any chronic pain, for which most had attended their family doctor. Manual social class, former oral contraceptive use and area of residence were independently associated with any chronic pain. Chronic pain at different sites was associated with different socio-demographic factors, and different proportions of women consulting their family doctor or reporting unfavourable health. These proportions rose with the number of reported sites of chronic pain.
Conclusions. The findings confirm the high prevalence of chronic pain and its importance to primary care. They suggest that chronic pain occurrence at different body sites varies across population subgroups, with differences in its impact on health. These differences call into question the independence of a chronic pain diagnosis. However, other factors common to any chronic pain support the development of common management and prevention strategies in primary care, and the variations highlighted will inform the targeting of these. The number of sites of chronic pain may be more important than the actual site(s) in determining its impact.
Keywords. Chronic pain, chronic pain syndrome, cross-sectional study, epidemiology, survey.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsDiscussionReferences |
|---|
Pain of at least 3 months' duration has been shown to affect up to half of the general population,1,2 with 5–15% having severe chronic pain.3,4 The symptom is associated with widespread physical, psychological and social distress and disadvantage,4,5 and has a low rate of recovery.6–8 Regional and widespread pain may also be associated with higher mortality.9 While specialist, pain clinic management may be effective,10 this can only reach a small proportion of sufferers. Population-based approaches to prevention and management are therefore required, probably through primary care where most chronic pain presents.11 The development of such management strategies requires an understanding of the distribution of chronic pain within the population, and of the factors associated with its occurrence.
Many previous studies have reported the epidemiology of chronic pain occurring at specific sites, such as the back, neck or shoulder.12 It has been argued, however, that chronic pain is a distinct diagnostic entity, encompassing all site-specific or diagnostic subgroups.13–15 This concept circumvents problems with symptoms that are vague or difficult to categorize, and with pain occurring at multiple sites. It is based on the hypothesis that the severity or extent of chronic pain is more important than its cause in determining its long-term impact and management needs. This would be important to GPs and other primary care professionals, who could apply common assessment and treatment strategies to any patient with chronic pain, irrespective of the site or cause of pain. The development of management and prevention strategies for a distinct chronic pain condition may facilitate the targeting of care more efficiently than separate strategies developed for site- or cause-specific painful problems. Such developments imply the presence of important aetiological and impact factors common to chronic pain occurring at each site. However, it is not clear whether such common factors exist, and there are also good reasons to suppose that differences may be more important.
Since its establishment in 1968, the Royal College of General Practitioners (RCGP) Oral Contraception Study (OCS)16 has made a major contribution to our understanding of the health effects of combined oral contraceptives. In 1994/95, a postal health survey of women still under observation was conducted. This survey included questions about chronic pain and its primary care treatment, enabling us to explore the distribution, risk factors and associated health and treatment seeking for all chronic pain, and chronic pain occurring at different sites and numbers of sites. If chronic pain is a distinct diagnosis, there should be no important differences in associated aetiological and impact factors between chronic pain at specific sites, and the extent of chronic pain (number of sites at which it occurs) should be more important than its site in determining the impact.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsDiscussionReferences |
|---|
This was a cross-sectional study conducted within the RCGP oral contraception cohort study.16 The main RCGP study began in 1968, when, during a 14-month period,
47 000 women were recruited by 1400 GPs throughout the UK. All of the women were married or living as married, most were white, and many had completed their family. The cohort has been regularly followed-up. By 1994, 25% of the original cohort were still under observation, most of the other women being lost to follow-up because they moved from the practice area of the recruiting doctor. Between November 1994 and July 1995, nearly all (11 797, 95%) of the women still under observation were sent a postal questionnaire via participating practices. Up to two reminders were sent to non-responders. The questionnaire asked about a number of health issues, including pain occurrence, current smoking habits, a global question on general health and three questions from the General Health Questionnaire, about sleep, overcoming difficulties and unhappiness/depression (see Box 1).17,18 From the questionnaire responses, an individual was defined as having chronic pain if she gave positive responses to each of two questions:19
- during the past month, have you had an ache or pain which has lasted for 1 day or longer?
- did the pain start more than 3 months ago?
| BOX 1 Questions about different aspects of health included in the postal survey How would you describe your health at present? Have you recently lost much sleep over worry?
Much more than usual Have you recently felt you couldn't overcome your difficulties?
Much more than usual Have you recently been feeling unhappy and depressed?
Much more than usual
|
In addition, respondents were asked to identify as many sites of the pain as appropriate from a list of nine [head, neck, shoulder(s), arm(s), chest, back, abdomen/ stomach, hip(s), leg(s)], and to say whether they had seen their family doctor as a result of the pain.
Socio-demographic risk factors were defined from: information collected at recruitment to the main RCGP OCS [date of birth (to calculate age at December 1995), number of cigarettes smoked daily, social class based on husband's occupation, area of residence]; information provided during the follow-up period and pertaining at December 1995 (parity, use of contraceptives); and information from the postal survey (smoking habits). Impact factors were defined from information provided in the postal survey (global assessment of health, responses to the three general health questions).
Data were analysed in SPSS for Windows. Unadjusted odds ratios (ORs) for the presence of any chronic pain were calculated for the socio-demographic factors. To allow for possible confounding, adjusted odds ratios (AORs) were calculated by entering each variable (apart from smoking at survey) into a backward stepwise multiple logistic regression model. Smoking at survey was not entered into the model as well as smoking at recruitment to avoid co-linearity in the analysis. Similar models were then constructed for chronic pain at each site, chronic pain occurring at 
5 sites compared with 
4 (as an indicator of the extent of chronic pain), and chronic pain resulting in consultation with a doctor compared with chronic pain not resulting in a consultation (as an indicator of the severity of chronic pain). Associations with impact factors were explored by cross-tabulations between chronic pain occurrence, site and number of sites, and responses to the four health questions, each collapsed into a binary variable (‘favourable’ and ‘unfavourable’; see Box 1). Differences were tested by chi-square tests.
Results
A total of 10 073 questionnaires were returned (response rate 85.4%). A total of 9084 respondents had completed both pain questions and formed the study population; a further 255 individuals responded negatively to the presence of pain, but positively to the duration, and were excluded.
Any chronic pain
Overall, 3490 women [38.4%, 95% confidence interval (CI) 37.4–39.4%] were identified as having any chronic pain. Smoking and manual social class at recruitment, former oral contraceptive use and living in Wales were associated with any chronic pain (Table 1). After multiple logistic regression, manual social class, former oral contraceptive use and living in Wales were independently associated with any chronic pain. Among those reporting any chronic pain, 45.4% (1579/3479) reported their health to be only fair or poor, compared with 17.1% (954/5579) of those without any chronic pain (chi-square = 851.1, P < 0.001) (Table 2). A higher proportion of those with any chronic pain also reported, to varying degrees, unfavourable health in each of the other three aspects of health assessed. The family doctor had reportedly been consulted by 80.4% (95% CI 79.1–81.7%) of women as a result of the chronic pain (Table 3). Consultation with the doctor was independently associated with older age (AOR in those > 60 years = 1.80, 95% CI 1.39–2.34, P < 0.001, compared with those <50 years), smoking at recruitment (AOR = 1.24, 95% CI 1.04–1.48, P = 0.019, compared with non-smokers), manual social class at recruitment (AOR = 1.29, 95% CI 1.08–1.55, P = 0.006, compared with non-manual social class) and former oral contraceptive use (AOR = 1.41, 95% CI 1.18–1.69, P < 0.001 compared with women never using oral contraception).
|
|
|
The chronic pain occurred in one site in 23.2% of women with chronic pain, while 18.2% reported chronic pain at
5 sites (Table 3). The proportion consulting their family doctor ranged from 75.6% among those with pain at one site to 93.1% of those reporting chronic pain at all nine sites (chi-squareTREND = 58.84, P < 0.001). Smoking at recruitment was independently associated with a greater risk of reporting chronic pain at 5 sites compared with at 1–4 sites (AOR = 1.22, 95% CI 1.02–1.46, P = 0.026), as was manual social class (AOR = 1.27, 95% CI 1.04–1.56, P = 0.020). No other factors were associated with pain at five or more sites. The proportion of respondents with unfavourable aspects of health increased significantly with the number of sites at which the chronic pain occurred (Table 2). For example, the proportion reporting poor or fair overall health ranged from 25.1% (95% CI 22.2–28.1%) among those with chronic pain at one site to 84.4% (95% CI 73.8–95.0%) among those with pain at nine sites.
Site-specific chronic pain
The most common site of chronic pain was the back (54.0%), followed by the legs (Table 4). The proportion of respondents consulting a doctor ranged from 79.7% of those with head pain to 89.6% with chest pain (chi-square = 33.98, 8 df, P < 0.001). Of the 802 respondents reporting chronic pain at a single site only, the proportion consulting a doctor ranged from 60.4% in those with chronic shoulder pain to 93.8% of those with chronic chest pain (chi-square = 24.6, 8 df, P < 0.003). Although there was no association between age and any chronic pain, the distribution of site-specific chronic pain varied with age. Chronic pain in the head, and in the abdomen, was significantly associated with younger age, while chronic pain in the hip, and in the leg was associated with older age (Table 5). Chronic pain in the arm, the chest, the back and the leg was associated with manual social class, and there was some evidence of geographical variation in chronic arm, back and leg pain. The proportion of women with unfavourable aspects of health also varied according to the site of the pain (Table 2).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsDiscussionReferences |
|---|
This study confirms the high prevalence of chronic pain, and the associated poor levels of health and high rates of primary care consultation. This reinforces the need for the development of effective primary care management and further research. We found some reasons for addressing these to site-specific chronic pain, rather than to all chronic pain. There were several statistically significant differences in factors associated with chronic pain at different body sites: (i) the proportion of women consulting the doctor; (ii) associated socio-demographic factors; and (iii) associated general health. These suggest that treatment needs and/or the perceived importance of the pain may be different, and that different population subgroups may be vulnerable to chronic pain at different sites. This may be important for prevention. The contrasting associations with age may be particularly important, with younger women tending to experience chronic pain at the head and abdomen, and older women experiencing it at the hip and leg. Similar findings arose in the Nuprin study,20 and the variations are consistent with clinical observation. For example, the association between head and abdominal pain and younger age may reflect factors associated with the menstrual cycle, while that between hip and leg pain and older age may represent age-related pathology such as osteoarthritis. Similarly, the consistent associations between chronic chest pain and poor socio-demographic, psychological and general health may reflect pain resulting from ischaemic heart disease. There may well be different patterns in a cohort of men.
This said, any chronic pain was associated with high rates of medical consultation, several independent socio-demographic factors and more unfavourable general health than among women without chronic pain. These common factors lend support to the existence of a distinct diagnosis, or syndrome, as defined by Last.21 This is consistent with previous research, which has found common psychological and behavioural factors predictive of chronic widespread pain.22–24 Some differences in socio-demographic factors were impossible to explore in this cross-sectional study. For example, age differences could be the result of a cohort effect, or a preponderance for women with chronic pain to localize this to different sites at different ages. This phenomenon has been suggested in children with chronic abdominal pain, who have been found more likely to describe chronic head pain in adulthood, though this association disappeared after controlling for psychiatric disorders.25 Further longitudinal studies are required to address this issue.
The number of sites at which chronic pain was experienced appeared to be at least as important as the actual location of the pain. Indeed, there was greater variation in the proportion of women either reporting unfavourable aspects of health or consulting a doctor with the number of sites than with the actual site of chronic pain. The American College of Rheumatology considers pain at multiple sites as a distinct subgroup of pain conditions.19 Studies such as that by Thomas et al. have found such widespread pain to be associated with poor health outcomes.26
Although the RCGP OCS study was not designed specifically for studying chronic pain, the size of the cohort and the availability of several general health-related questions have provided useful information. The prevalence of any chronic pain in this sample was high, but broadly similar to that found in other general population studies; the relative frequency of chronic pain at different sites was also compatible with other studies, as was our finding of associations between any chronic pain and lower social class, poor general and psychological health, and smoking.1–7,27 Some previous work, based in the UK, has found geographical variation in the prevalence of chronic back pain,28 though with high-risk locations different from this study, which found the highest risk for any chronic pain, and for chronic back and leg pain among women living in Wales. Pain and its response are perceived differently across different countries and cultures,29 and, although we cannot apply our findings beyond the UK, some variation is therefore not surprising. To our knowledge, no study has previously found a link between former oral contraceptive use and chronic pain, though a possible link with low back pain (not necessarily chronic) was found in Sweden.30
There are several limitations of this study. The sample was restricted to UK women, aged between 42 and 81 years at the time of follow-up. However, it included a wide social and geographic spread within this setting, and there was a good response to the survey. The results are based primarily on a cross-sectional survey conducted within a cohort study, which had been subjected to a high loss of subjects. Our results might have been affected if women lost from the cohort were at different risk of developing chronic pain than women remaining in the study. Most of the attrition of subjects was due to women moving from the practice area of the recruiting doctor, or to practices withdrawing from the study; women who remained under observation tended to be older, of lower social class and of higher parity than those who had left.18 These factors were associated with chronic pain at some sites in our study, and so bias from loss to follow-up may have affected our estimates of associations with some risk factors. However, the similarity between our prevalence rates and those in other studies suggests that this loss to follow-up was not a major factor. Data on the participants were collected at different points in time, and some of the variables could have changed between collection and measurement of pain status in 1994/95. Of the data collected at recruitment, we have no information on any change of social class during the study period, and it is unclear how this potential source of bias might have affected our results. Smoking status was collected both at recruitment and in 1994/95, and both were considered in our analysis. All of the other information contributing to the analysis was collected either during the follow-up period or at the time of survey, and was therefore current.
The site of chronic pain was reported using a list that covered most of the body. Although it may not have been comprehensive, only 27 women with chronic pain were unable to select at least one site. Clustering of sites has not been analysed in this study. It may have been useful to ask women to identify the most important site of chronic pain in those with pain at multiple sites, in order to allow deeper exploration of site-specific chronic pain. The questionnaire did not provide data on the duration of pain, or on associated symptoms, and we could not distinguish between persistent and intermittent chronic pain. Finally, as a cross-sectional study, we were unable to determine whether the associations were the result of cause or effect, though risk factors that were present at recruitment preceded the information collected on chronic pain by up to 27 years.
In summary, while recognizing the limitations of the study, we have confirmed the high prevalence of chronic pain in the community, and some of its socio-demographic and health-related associations. We found that there were important differences in some of the associations with chronic pain at different sites. However, we also found associations that were consistent across all sites of chronic pain, and that the importance of the number of sites at which chronic pain arises may be greater than the actual site. If chronic pain is a distinct diagnosis, subgroups of the condition should also be described, based on socio-demographic factors and on the extent of the pain. This is broadly the approach of the International Association for the Study of Pain,31 which considers both a general phenomenon and a site-specific classification. This approach will harness the benefits of common treatment and prevention strategies in primary care, and optimize the targeting of these. It will mirror a suggested common approach to other ‘functional somatic syndromes’, such as irritable bowel syndrome and chronic fatigue syndrome, whose similarities may outweigh differences in primary care.32
|
Acknowledgments |
|---|
We are grateful to the thousands of women who completed the questionnaire, the hundreds of doctors who contributed to the RCGP OCS, and Professor Peter Croft and Dr Martyn Lewis of Keele University for their comments on the manuscript. The postal questionnaire was funded by an unconditional grant from Wyeth-Ayerst Inc. Additional support for the study has been received from the RCGP, Schering AG and Schering Health Care. BS is supported by an NHS R&D Primary Care Career Scientist Award, funded by the Scottish Executive Health Department.
|
Notes |
|---|
* Categorized as ‘favourable’.
categorized as ‘unfavourable’.
|
References |
|---|
|
TopAbstractIntroductionMethodsDiscussionReferences |
|---|
1 Andersson HI, Ejlertsson G, Leden I, Rosenberg C. Chronic pain in a geographically defined general population: study of differences in age, gender, social class, and pain localisation. Clin J Pain 1993; 9: 174–182.[ISI][Medline]
2 Elliott AM, Smith BH, Penny KI, Smith WC, Chambers WA. The epidemiology of chronic pain in the community. Lancet 1999; 354: 1248–1252.[CrossRef][ISI][Medline]
3 Verhaak PF, Kerssens JJ, Dekker J, Sorbi MJ, Bensing JM. Prevalence of chronic benign pain disorder among adults: a review of the literature. Pain 1998; 37: 215–222.
4 Smith BH, Elliott AM, Chambers WA, Smith WC, Hannaford PC, Penny KI. The impact of chronic pain in the community. Fam Pract 2001; 18: 292–299.
5 Latham J, Davis BD. The socioeconomic impact of chronic pain. Disabil Rehabil 1994; 16: 39–44.[Medline]
6 Macfarlane GJ, Thomas E, Papageorgiou AC, Schollum J, Croft PR, Silman AJ. The natural history of chronic pain in the community: a better diagnosis than in the clinic. J Rheumatol 1996; 23: 1617–1620.[ISI][Medline]
7 Croft PR, Macfarlane GJ, Papageorgiou AC, Thomas E, Silman AJ. Outcome of low back pain in general practice: a prospective study. Br Med J 1998; 316: 1356–1359.
8 Elliott AM, Smith BH, Hannaford PC, Smith WC, Chambers WA. The course of chronic pain in the community: results of a four-year follow-up study. Pain 2002; 99: 297–307.
9 Macfarlane GJ, McBeth J, Silman AJ. Widespread body pain and mortality: prospective population based study. Br Med J 2001; 323: 662–665.
10 Williams ACD, Nicholas MK, Richardson PH et al. Evaluation of a cognitive behavioural programme for rehabilitating patients with chronic pain. Br J Gen Pract 1993; 43: 513–518.[ISI][Medline]
11 Haetzman M, Elliott AM, Smith BH, Hannaford P, Chambers WA. Chronic pain and the use of conventional and alternative therapy. Fam Pract 2003; 20: 147–154.
12 Crombie IK, Croft PR, Linton SJ, LeResche L, Von Korff M (eds). Epidemiology of Pain. Seattle: International Association for the Study of Pain Press; 1999.
13 Addison RG. Chronic pain syndrome. Am J Med 1984; 77: 54–58.[CrossRef][ISI][Medline]
14 Sanders SH. Chronic pain: conceptualization and epidemiology. Ann Behav Med 1985; 7: 3–5.
15 Smith BH. Chronic pain: a challenge for primary care. Br J Gen Pract 2001; 51: 524–526.[ISI][Medline]
16 Royal College of General Practitioners. Oral Contraceptives and Health. London: Pitman Medical; 1974.
17 Goldberg D, Williams P. A User's Guide to the General Health Questionnaire. Windsor (UK): NFER-Nelson; 1988.
18 Owen-Smith V, Hannaford PC, Warskyj M, Ferry S, Kay CR. Effects of changes in smoking status on risk estimates for myocardial infarction among women recruited for the Royal College of General Practitioners' Oral Contraception Study in the UK. J Epidemiol Community Health 1998; 52: 420–424.[Abstract]
19 Wolfe F, Smythe HA, Yunus MB et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia: report of the multi-center committee. Arthritis Rheum 1990; 33: 160–172.[ISI][Medline]
20 Sternbach RA. Survey of pain in the United States: the Nuprin pain report. Clin J Pain 1986; 2: 49–53.
21 Last RJ. A Dictionary of Epidemiology, 4th edn. Oxford: International Epidemiological Association; 2001.
22 Macfarlane GJ, Morris S, Hunt IM et al. Chronic widespread pain in the community: the influence of psychological symptoms and mental disorder in healthcare seeking behaviour. J Rheumatol 1999; 26: 413–419.[ISI][Medline]
23 McBeth J, Macfarlane GJ, Benjamin S, Silman AJ. Features of somatization predict the onset of chronic widespread pain: results of a large population-based study. Arthritis Rheum 2001; 44: 940–946.[CrossRef][ISI][Medline]
24 McBeth J, Macfarlane GJ, Hunt IM, Silman AJ. Risk factors for persistent widespread pain: a community-based study. Rheumatology 2001; 40: 95–101.
25 Hotopf M, Carr S, Mayou R, Wadsworth M, Wessely S. Why do children have chronic abdominal pain, and what happens to them when they grow up? Population based cohort study. Br Med J 1998; 316: 1196–2000.
26 Thomas E, Silman AJ, Croft PR, Papageorgiou AC, Jayson MIV, Macfarlane GJ. Predicting who develops chronic low back pain in primary care: a prospective study. Br Med J 1999; 318: 1662–1667.
27 Palmer KT, Syddall H, Cooper C, Coggon D. Smoking and musculoskeletal disorders: findings from a British national survey. Ann Rheum Dis 2003; 62: 33–36.
28 Walsh K, Cruddas M, Coggon D. Low back pain in eight areas of Britain. J Epidemiol Community Health 1992; 46: 227–230.[Abstract]
29 Lasch KE. Culture and pain. Pain: Clin Updates 2002; 10: 1–4.
30 Wreje U, Isacsson D, Aberg H. Oral contraceptives and back pain in women in a Swedish community. Int J Epidemiol 1997; 26: 71–74.
31 International Association for the Study of Pain. Classification of chronic pain. Pain 1986; Suppl 3: S1–S225.
32 Wessely S, Nimnuan C, Sharpe M. Functional somatic syndromes: one or many? Lancet 1999; 354: 936–939.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S Patel, S. Peacock, R. McKinley, D C. Carter, and P. Watson GPs' experience of managing chronic pain in a South Asian community--a qualitative study of the consultation process Fam. Pract., April 15, 2008; (2008) cmn012v2. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Clarke, A Judge, A Herbert, K McPherson, S Bridgman, M Maresh, C Overton, and D Altman Readmission to hospital 5 years after hysterectomy or endometrial resection in a national cohort study Qual. Saf. Health Care, February 1, 2005; 14(1): 41 - 47. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||