A meta-analysis to establish the construct validity and normative values of the Pain Catastrophising Scale

Introduction: Outcome measures in clinical psychology tend to be developed in an ad-hoc way, with psychological constructs added to theoretical understanding without formal evaluation of their validity and relationship with existing constructs. Pain catastrophising is an example of a construct with no proven differentiation from other pain-related cognitions. The Pain Catastrophising Scale (PCS) is widely used and several theories exist regarding its theoretical basis and causal relationship with pain outcomes. Aims: This thesis aims to establish psychometric properties for the PCS from a wide and varied population; to assess the sensitivity of the scale and create norms for pain types; and assess the construct validity of pain catastrophising. Method: A systematic review was conducted to collect baseline PCS scores from research studies since its development in 1995 to the present day. Meta-analysis including multivariate regression explored variables influencing pain catastrophising. Correlations between the PCS and other measures were used to evaluate the construct validity of pain catastrophising. Results: Good internal reliability (α=.92, 95% CI .91-.93) and test-retest reliability scores (Spearman correlation coefficient=.88, 95% CI .83-.93) were found for the PCS. Participants’ pain type was highly related to PCS scores, with those with generalized pain scoring highest. No significant effects of age or gender were found. Language of the PCS affected PCS scores. Study type influenced PCS scores, but was confounded with pain diagnosis, with controlled trials more likely than quasi-experimental studies to recruit clinical samples. Divergent validity of the construct of pain catastrophising was tentatively supported by limited data. Discussion: Within the limits of available data, the use of the PCS is supported as a valid and reliable measure. Pain catastrophising varies depending on the pain type and intensity experienced. Further research is recommended to clarify the construct validity of pain catastrophising through consistent use of outcome measures