Imke Boekel1, Alisa L Dutmer2, Henrica R Schiphorst Preuper2, Michiel F Reneman3. 1. Faculty of Medicine, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 2. Department of Rehabilitation Medicine, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 3. Department of Rehabilitation Medicine, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. m.f.reneman@umcg.nl.
Abstract
PURPOSE: A cross-sectional and longitudinal study was conducted to analyse construct validity, responsiveness, and Minimal Clinically Important Change (MCIC) in the Work Ability Score (WAS) and Pain Disability Index Work item (PDI-W) in patients with Chronic Low Back Pain (CLBP). METHOD: Construct validity was assessed by testing predefined hypotheses. Responsiveness and MCIC were measured with an anchor-based method. The area under the receiver Operating Characteristic Curve (AUC) and the optimal cut-off point were calculated. Smallest Detectable Change (SDC) was calculated to determine measurement error. RESULTS: In total, 1502 patients (age 18-65 years) with CLBP were included. For validity of the WAS and PDI-W, respectively, seven and six out of 10 hypotheses were not rejected. The WAS (n = 355) was responsive to change with an AUC of 0.70. MCIC was 1.5 point, SDCindividual 4.9, and SDCgroup 0.3. MCICs were 4.5, 1.5, and - 0.5 points for, respectively, low, middle, and high scoring baseline groups. The PDI-W (n = 297) was responsive to change with an AUC of 0.80. MCIC was - 2.5 points, SDCindividual 5.2, and SDCgroup 0.3. MCICs were - 0.5, - 2.5, and - 4.5 points for, respectively, low, middle, and high scoring baseline groups. CONCLUSION: Construct validity of the WAS and PDI-W was insufficient in this patient sample. The WAS and PDI-W are responsive to change. On average, improvements of 1.5 point (WAS) and - 2.5 points (PDI-W) were interpreted as clinically important. However, MCICs are also baseline dependent. Due to a risk of measurement error, at the individual level change scores should be interpreted with caution.
PURPOSE: A cross-sectional and longitudinal study was conducted to analyse construct validity, responsiveness, and Minimal Clinically Important Change (MCIC) in the Work Ability Score (WAS) and Pain Disability Index Work item (PDI-W) in patients with Chronic Low Back Pain (CLBP). METHOD: Construct validity was assessed by testing predefined hypotheses. Responsiveness and MCIC were measured with an anchor-based method. The area under the receiver Operating Characteristic Curve (AUC) and the optimal cut-off point were calculated. Smallest Detectable Change (SDC) was calculated to determine measurement error. RESULTS: In total, 1502 patients (age 18-65 years) with CLBP were included. For validity of the WAS and PDI-W, respectively, seven and six out of 10 hypotheses were not rejected. The WAS (n = 355) was responsive to change with an AUC of 0.70. MCIC was 1.5 point, SDCindividual 4.9, and SDCgroup 0.3. MCICs were 4.5, 1.5, and - 0.5 points for, respectively, low, middle, and high scoring baseline groups. The PDI-W (n = 297) was responsive to change with an AUC of 0.80. MCIC was - 2.5 points, SDCindividual 5.2, and SDCgroup 0.3. MCICs were - 0.5, - 2.5, and - 4.5 points for, respectively, low, middle, and high scoring baseline groups. CONCLUSION: Construct validity of the WAS and PDI-W was insufficient in this patient sample. The WAS and PDI-W are responsive to change. On average, improvements of 1.5 point (WAS) and - 2.5 points (PDI-W) were interpreted as clinically important. However, MCICs are also baseline dependent. Due to a risk of measurement error, at the individual level change scores should be interpreted with caution.
Authors: Lidwine B Mokkink; Caroline B Terwee; Dirk L Knol; Paul W Stratford; Jordi Alonso; Donald L Patrick; Lex M Bouter; Henrica Cw de Vet Journal: BMC Med Res Methodol Date: 2010-03-18 Impact factor: 4.615
Authors: Remko Soer; Michiel F Reneman; Bert L G N Speijer; Maarten H Coppes; Patrick C A J Vroomen Journal: Spine J Date: 2012-11 Impact factor: 4.166