A comparison of the validity of two indirect utility instruments as measures of postoperative recovery.

Cost-effectiveness analyses of surgical interventions require valid measures of postoperative recovery. The objective of this study was to compare the validity of two indirect utility instruments, the Short Form 6D (SF-6D) and EuroQol 5D (EQ-5D), as measures of postoperative recovery.A prospective cohort of patients undergoing elective colorectal resection at two university-affiliated institutions from October 2012-October 2013 completed the SF-6D and EQ-5D (including the EQ-visual analog scale [EQ-VAS]) at baseline (before surgery), and at 4 and 8 wk after surgery. Responsiveness and construct validity were assessed through a priori hypotheses.A total of 165 patients were included. The SF-6D was the most responsive to the expected postoperative changes at 4 and 8 wk compared with the EQ-5D and the EQ-VAS. The 4-wk SF-6D, EQ-5D, and EQ-VAS discriminated between patients with and without complications after controlling for confounders with adjusted mean differences of -0.070 (95% confidence interval [CI] -0.126 to -0.015), -0.133 (95% CI -0.231, -0.030), and -7.91 (95% CI -14.77, -1.04), respectively. Mean SF-6D and EQ-5D values were significantly different from the US population norms at all time points, but the magnitude of change was highest for the SF-6D. The strength of correlation between all three instruments was moderate at all time points (r=0.550-0.684, all P<0.05).The SF-6D preference-based health index appears to be a more valid measure of postoperative recovery than the EQ-5D and EQ-VAS in surgical cost-effectiveness analyses.