Arthritis and Pain Management: The Epidemiology of Pain

Time Required: 
10 minutes

The fundamental concern of rheumatologists and their patients is pain. In frequency, musculoskeletal pain ranks second only to respiratory conditions in the list of symptoms presented to general practitioners. Yet the emphasis has been on the delineation of distinct clinicopathological syndromes rather than of pain per se.

The consistent message from studies of low back and soft tissue pain is that the strongest risk for developing new symptoms is having had a previous episode [1]. Such pain appears to track through an individual's life, each episode increasing the probability of having another one. This life history of pain in any one person is the outcome of a continuing and changing interplay between predisposition and environment. Any attempt to explain in a simple fashion that occupational activity, for example, is a risk factor for low back pain is bedevilled by the fact that earlier pain experience might well have dictated the type of work that people do or their attitude towards their job. In a study from Norway, early pain experience was charted in a cohort of children [2]. The most common pain complaint at the age of 5 years was abdominal pain, but by 10 years headache dominated. Those children who complained of abdominal pain at the age of 5 were more likely than other children to be complaining of headache at the age of 10. This serves to highlight the fact that, the further we look into the way in which pain affects people throughout their lives, the more complex the picture becomes.

Another important feature of musculoskeletal syndromes is their propensity to occur together. This concurrence is reported consistently across studies from different sources, including neck and low back pain, and upper limb and lower limb syndromes. The more areas of pain that are involved at any one time, the more likely that unaffected areas will become involved later. There are a number of potential explanations for this. Multiple joint involvement in osteoarthritis (OA), for example, may reflect the strong age-related incidence of the condition at all sites; older people are increasingly likely to develop problems at more than one site. It may reflect common pathogenesis; long years of heavy lifting is implicated in the aetiology of both hip and knee OA, as well as low back pain and shoulder syndromes. Other links, such as the influence of psychosocial factors on symptom occurrence, seem not to be specific to any one syndrome but a feature of persistent musculoskeletal pain in general. The pattern seems to be that local factors (mechanical stress and injury) may determine the initial location, but the reasons for persistence and recurrence are more general. Fibromyalgia represents the extreme example of this, where the peripheral origins of the pain have become lost in the widespread nature of the problem.

The neuroscientists have provided theories and experiments to support the idea that pain expression is not simply a result of unhealed peripheral injury. Although warning against swinging the pendulum too far in the direction of all persistent pain being a creation of central mechanisms, Wall and others have contributed to the breakdown of the Cartesian separation of mind and body [2]. Fibromyalgia fits the new model of what, in Wall's words, is a "virtual reality world", in which central influences of emotion and cognition join with the consequences of peripheral injury to give painful symptoms a life of their own.

How much chronic musculoskeletal pain is related to such central influences, as summed up in the term 'somatisation'? The core notion of somatisation is that psychological distress can be expressed as physical symptoms. In general population studies, there is a consistent relation between measures of psychological distress and the extent of pain complaints. In studies carried out among Boeing workers in America [3], in a population-based cohort from a Manchester population [4], and among factory workers in Finland [5], measures of distress at baseline predicted the likelihood of later episodes of low back or other musculoskeletal pain. There is a growing body of evidence that shows levels of distress in people presenting with low back pain in primary care are the strongest predictors of the outcome of the episode. These were all shorter studies, and prior experience of pain might have influenced the way in which psychosocial distress became embedded in a painful context. We need to know whether such relations persist over a long period of time, and whether they represent longer-term risks for persistent pain. There must still be a suspicion that the link between pain and psychological distress can never be neatly packaged into a linear process of cause and effect, and that childhood experiences, or parental experiences, for example, colour these relationships from an early age.

The evidence from epidemiology suggests that pain and its persistence needs to be released from the strait-jacket of our obsession with localisation and peripheral mechanisms, and studied as a phenomenon in its own right.

For a systematic review on the economic burden of rheumatoid arthritis, please consult Rosery et al (2005) [6].


Knowledge Level: 
Professional Issues: