THE ASSOCIATION BETWEEN CHRONIC WIDESPREAD PAIN AND MENTAL DISORDER: A Population-Based Study

Objective. Patients with chronic widespread pain (CWP) have been reported to have a greater prevalence of mental disorders and somatization than that found in the general population, but the true association between CWP and mental disorders is unknown. In this study, we investigated whether there is an increased prevalence of mental disorder in people with CWP from the general population. We also describe the psychiatric diagnoses associated with CWP. Methods. In a population-based case–control study, 1,953 subjects (75% of a random sample of individuals age 18–65 years) completed a questionnaire that included a pain assessment and the 12-item General Health Questionnaire (GHQ-12). Of 710 subjects scoring >1 on the GHQ-12, 301 were assessed further using a structured psychiatric interview and detailed assessment of medical records to identify cases of mental disorder, in accordance with criteria of the 10th edition of the International Classification of Diseases. The association between CWP and mental disorder was modeled using logistic regression, adjusting for possible confounders including age, sex, and nonresponders. Results. We estimated the overall population prevalence of mental illness to be 11.9%. The odds of having a mental disorder for subjects with versus those without CWP were 3.18 (95% confidence interval 1.97–5.11). Most subjects with mental disorders were diagnosed as having mood and anxiety disorders. Only 3 cases of somatoform disorders were identified, and all were associated with pain. Conclusion. This study, although unable to demonstrate a cause-and-effect relationship, showed that 16.9% of those with CWP were estimated to have a psychiatric diagnosis, suggesting that these disorders should be identified and treated. Although there are many accounts of a relationship between chronic pain and mental disorders, these have been based mainly on investigations of consulting populations. Thus, compared with general population prevalence studies, patients attending pain (1), rheumatology (2), and fibromyalgia (3,4) clinics all have an excess of psychiatric disorders, although this may not be a consistent finding (5). Clinical samples are, however, liable to be influenced by consultation behavior and may be unrepresentative (6). In support of this, we recently showed that in a general population sample, people with chronic widespread pain (CWP, defined below) who sought primary care had greater psychological distress and more mental disorders than those who did not (7). However, even those who do not seek health care may have an increased prevalence of mental disorders. A Finnish study (8) found a 56% prevalence of mental disorder in people with fibromyalgia syndrome (FMS) compared with 17% in those without FMS, but these results are difficult to interpret since psychological symptoms were included as part of the classification process of fibromyalgia. Many psychiatric diagnoses have been described in association with chronic pain, but 2 diagnostic groups have predominated. Mood (depressive) disorders have been conspicuous in some studies of clinical samples (1,9,10), and depressive symptoms have been associated, in population studies, with pains in bodily regions (11,12). These findings must be interpreted with caution, since assessments of depression may fail to differentiate depressed mood, symptoms, and diagnosis, do not always use standardized diagnostic criteria, and may be influenced by the inclusion of symptoms (particularly related to appetite, sleep, and sexual function) or social dysfunction, which can be directly related to organic status or other mental disorders. Thus, depressive disorders may be overdiagnosed in people with chronic pain. Conversely, depressive disorders may be underdiagnosed due to the process of somatization (13), whereby patients focus on physical complaints and deny psychosocial problems, and physicians fail to identify these mental disorders (14,15). The diagnosis of somatoform disorders (particularly somatization disorder) is uncommon in the general population but has been prominent in some reports of chronic pain patients attending clinics (1,4). Published population surveys of the prevalence of mental disorders have not assessed pain routinely, nor have they assessed organic state or consultation for physical symptoms or disability, all of which contribute to the diagnosis of somatoform disorders. Assessments of somatization disorder based on physical symptom counts (16) are of limited value in this context, because higher somatic (including pain) symptom counts are also associated with depression and anxiety (17,18). In the general population, the relationship between chronic pain and somatoform disorders, identified in accordance with standardized diagnostic criteria, remains uncertain. To address these issues, surveys require the use of standardized criteria and assessments for both chronic pain and mental disorders. The American College of Rheumatology (ACR) has defined classification criteria for fibromyalgia in terms of CWP and multiple tender points (19). The evidence that mental disorders are associated with chronic pain, including fibromyalgia, in both clinic patients and the general population is inconclusive, and the relationship between separate diagnostic groups and CWP remains to be examined. In this study, we tested the hypothesis that subjects with CWP in the general population had higher rates of mental disorder than subjects without CWP. In addition, we determined the prevalence of psychiatric diagnostic groups including mood, anxiety, and somatoform disorders and examined the association of these diagnostic groups by comparing persons with CWP, other pain, and no pain. SUBJECTS AND METHODS Design and population. A population-based case–control study involving prevalent cases of CWP was performed. A population sample of 3,004 subjects age 18–65 years was randomly selected from a general practice register in the south Manchester, UK, area. This is a mixed but relatively affluent sociodemographic area with high employment levels compared with other parts of the region. The general practice register provides a convenient sampling frame for community subjects, since >95% of the population in the UK is registered with a general practice (20). Each subject was sent a postal questionnaire and, if necessary, a reminder postcard and then a further copy of the questionnaire.