The Dutch language version of the Pain Disability Index (PDI-DLV): psychometric properties in breast cancer patients

Abstract

ABSTRACT Background Pain after breast cancer surgery is a common and disabling problem. A reliable and valid questionnaire to assess pain-related disability is the Pain Disability Index (0–70). However, properties of the Dutch version (PDI-DLV) have never been investigated in this population. Objective To assess psychometric properties of the PDI-DLV after breast cancer surgery. Methods For reliability, relative and absolute reliability were calculated with a one-week test–retest interval, as well as internal consistency. Moreover, content and construct validity were examined to evaluate validity. Results One hundred twenty-three women were included. Relative reliability was good (intraclass correlation coefficient = 0.80). Standard error of measurement and minimal detectable change (absolute reliability) were 5.57 and 15.45 points, respectively. The mean difference between two measurements was −1.98 points, with 95% limits of agreement equal to 13.19 and −17.15. The within-subjects coefficient of variation was 59%. Internal consistency was confirmed (α = 0.87). The PDI-DLV was scored as understandable and complete (content validity). Construct validity was supported by confirmation of more than 75% of the tested hypotheses and of the one-factor model. Conclusion The PDI-DLV is a valid questionnaire to assess pain-related disability 1 year after breast cancer surgery. Although absolute reliability is disputable, its good relative reliability allows evaluating changes between subjects.