Selective processing of threat-related cues in day surgery patients and prediction of post-operative pain

Objective. To investigate the use of a measure of selective processing bias associated with anxiety as a predictor of post-operative pain independently of self-report measures of anxiety. Methods. Forty-seven women admitted for minor gynaecological surgical procedures completed a selective processing task (modified Stroop) and the State-Trait Anxiety Inventory immediately prior to surgery. Following surgery they completed the McGill Short-Form Pain Questionnaire. Intraoperative analgesia consumption was also recorded. Results. Participants demonstrated significantly slower colour-naming times for physical threat cues than control cues. This was not due to an emotionality effect, as colour-naming times for neutral and positive cues were not significantly different. This bias was congruent with the participants' current concerns, as colour-naming times were significantly slower for physical threat words than for social threat words. This index of selective processing bias significantly predicted post-operative pain independently of self-reported state and trait anxiety. Conclusions. The advantages of measures of psychological constructs that are not reliant on self-reporting are discussed.