Objectives: Patients undergoing major surgery often require several days of postoperative analgesia. However, few data exist on the longitudinal course of postoperative pain and the psychometric properties of pain assessment tools used in this setting. Our objective was to validate use of the modified Brief Pain Inventory through reanalysis of pain data from a multiple-dose, placebo-controlled, randomized trial of analgesia after coronary artery bypass graft surgery.
Methods: Four hundred sixty-two patients who underwent coronary artery bypass graft surgery via median sternotomy were administered a shortened form of the original Brief Pain Inventory that contained 3 severity and 5 interference items. Additionally, patients were presented with a single-item measure of procedure-specific pain. Daily pain and interference ratings were available from days 4 to 14 postoperatively. We performed factor analysis to evaluate the consistency with which the modified Brief Pain Inventory items loaded on 2 separate factors corresponding to the original Brief Pain Inventory's pain severity and pain interference subscales. We calculated 2 reliability measures, internal consistency and test-retest reliability, for each subscale.
Results: The modified Brief Pain Inventory consistently measured 2 underlying constructs, severity and interference, with Cronbach alphas of 0.85 or greater for the 2 Brief Pain Inventory scales, and test-retest stability coefficients ranging from 0.58 to 0.95 for each pair of consecutive assessment periods. The procedure-specific pain question showed substantial overlap with a general measure of pain severity, suggesting concurrent validity.
Discussion: The modified Brief Pain Inventory was stable and valid over the assessment period, suggesting that it can be used during the subacute postoperative period to assess postoperative pain among patients with coronary artery bypass graft surgery.