Assessing change in chronic pain severity: the chronic pain grade compared with retrospective perceptions

BACKGROUND: There is no standard method of measuring change in chronic pain severity. Clinical trials commonly use serial assessment scales, completed at two points in time, to estimate change in pain severity, while clinicians usually ask patients to make a retrospective assessment of change. How the two methods compare is not known. AIM: To assess different methods of measuring change in chronic pain severity, by comparing changes in scores on a serial measure of chronic pain severity using the Chronic Pain Grade (CPG) questionnaire and responders' retrospective perception of change in pain severity. DESIGN OF STUDY: Postal self-completion questionnaires. SETTING: The Grampian region of Scotland. METHOD: Postal questionnaires were sent in March and September 1998 to a random sample of 535 adults with chronic pain, drawn from responders to a postal survey of the region conducted in 1996. RESULTS: Corrected response rates of 87.5% and 90.7% were obtained. Over a six-month period poor levels of agreement were found, with responders' retrospective perceptions mirroring recorded changes in 41.8% of individuals (kappa = 0.081). A low partial correlation coefficient between the two measures (-0.209) was also found. Over a two-year period there were again poor levels of agreement, with responders' retrospective perceptions mirronng recorded changes in 35.2% of individuals (kappa = 0.071). A low partial correlation coefficient (-0.401) was again found. CONCLUSION: There was poor agreement and low correlation between two commonly used methods for assessing change in pain severity over time. This finding has important