Does cognitive change predict the outcome of cognitive-behavioural pain management?

Abstract

Does cognitive change predict the outcome of cognitive-behavioural pain management? The psychometric properties of the Pain Cognitions Questionnaire (PCQ: Boston et al. , 1990), a brief self-rated measure of spontaneously occurring pain-related cognitions, were examined in a sample of 833 severely impaired chronic pain patients who completed it prior to and at several points following discharge from an inpatient cognitive-behavioural pain management programme. Principal components analysis allowed the identification of a two-factor structure which remained stable in this population across all points of assessment. The first factor (11 items) was labelled 'positive self statements'; the second (12 items), 'lack of mastery over pain'. Both factors showed adequate reliability (test-retest and internal consistency) and sensitivity to treatment-induced change; and preliminary support was obtained for criterion-related validity. The PCQ was then used to examine an underlying assumption of cognitive-behavioural therapies (CBT), viz that cognitive change mediates successful treatment outcome. Regression modelling was used to examine the extent to which change in PCQ scores from pre- to post-treatment was predictive of a range of outcomes including pain, depression, medication use and general functioning (Sickness Impact Profile scores) at one-month follow-up after controlling for pre-treatment levels of relevant variables. Results were broadly, though weakly, supportive of the cognitive hypothesis, with cognitive change accounting for up to 15% of outcome variance.