BACKGROUND Presently, it is difficult to predict which patients are at increased risk of ongoing pain problems postoperatively. This study followed a group of patients from the week before their operation until 3 months after it, to identify potential risk variables. METHODS Fifty-four patients undergoing moderate-major gynaecological surgery at Christchurch Women's Hospital were recruited and assessed preoperatively over an 11-week period. At this initial assessment, participants were subjected to a cold pressor test (CPT). Telephonic follow-up was conducted at 6 weeks and 3 months postoperatively, to determine pain status. Information regarding the type of operation and surgical approach was collected from hospital records. RESULTS Pain threshold (time taken to report the onset of pain), as measured by the CPT, was significantly predictive of prolonged pain outcomes (area under the curve = 0.80, 95 % CI 0.66, 0.95). Pain tolerance (total time taken to end the CPT voluntarily) was similarly predictive but non-significant (area under the curve = 0.69, 95 % CI 0.47, 0.90). CONCLUSIONS The preoperative cold pressor test shows some promise for predicting ongoing postoperative pain. However, more research is needed to determine the clinical significance of these findings in larger samples and how they could be incorporated into clinical practice.