Toward a set of reliable and valid measures for chronic pain assessment and outcome research.

During the last 10-15 years, a number of reliable and valid measures for assessing chronic pain syndromes and evaluating treatment have been proposed and tested. Because of the multidisciplinary nature of the field, these measures have been drawn from physiological, psychological (emotional, cognitive, and behavioral), sociocultural, and economic studies. Together they form the definition of a chronic pain data base which can assist in classifying chronic pain syndromes, assessing patients for treatment, and evaluating the results of treatment. Since all measurement is indirect, reliability, validity, and appropriateness for use in multivariate analysis have been the major selection criteria. Along with these assessment measures, a set of risk factors, judged as affecting both the incidence of chronic pain and response to treatment, have been reviewed. It has become clear that to further the scientific knowledge and clinical understanding of the chronic pain syndrome and its treatment, data using such a set of reliable measures must be collected. Furthermore, these data should be analyzed using current multivariate methodology which gives attention to measurement error as well as to causal interactions and control issues. Fortunately, the tools are now becoming available, just as consumers and certifiers of chronic pain treatment are demanding careful program evaluation and an account of results.