The power of longitudinal population-based studies for investigating the etiology of chronic widespread pain.

In this issue of Pain, Gale et al. show that lower childhood intelligence is associated with increased risk of chronic widespread pain (CWP) later in life [2]. Although an association between chronic pain and cognitive ability has been shown before, the Gale et al. report is the first longitudinal study in a large general population cohort. Two important characteristics stand out in their study. First, its longitudinal design enabled the researchers to study causal direction as well as interesting explanatory mechanisms in the association between intelligence and pain. A previous crosssectional study interpreted this association as an effect of pain on cognitive function [8]. Gale et al., however, show that the reverse might also occur. Specifically, the authors found that lower intelligence predicted the development of CWP many years later. They suggest several interesting mechanisms to explain their findings, including inappropriate coping styles and suboptimal health literacy. These suggestions imply important roles for health seeking behavior and dealing with symptoms or symptom related information, and will inspire further research in this area. The second key feature of the Gale et al. study is that it used a population-based design, which contrasts with many previous etiological studies that compared patients to healthy controls. This approach is especially relevant because of the nature of CWP, which is the core symptom of fibromyalgia, one of the most prevalent functional somatic disorders (FSD). CWP was defined according to the American College of Rheumatology criteria for fibromyalgia. The definition is based on questions on symptom diversity and chronicity, which is preferable to identifying cases based on self-report of fibromyalgia. Not all persons fulfilling the criteria for an FSD present with a formal diagnosis [4,10]; one of the factors increasing the chance of an FSD diagnosis in persons with persistent functional somatic symptoms (FSS) was high intelligence [5]. This increased chance of an FSD diagnosis in highly intelligent people might mask the fact that the FSS on which the FSD diagnosis is based are less prevalent in this group [7]. The strategy used by Gale et al., which is based on the assessment of the core symptom CWP and not on the self-reported diagnosis of fibromyalgia, avoids this problem. This study is consistent with previous studies that associated lower intelligence with a large variety of somatic and psychiatric health problems as well as mortality. These health problems include various types of FSS in both adults [1,3,7] and adolescents [6]. The results suggest that lower intelligence may be a generic risk factor for morbidity and mortality that deserves further study,