Jonathan David Comins1,2,3, Volkert Dirk Siersma4, Martin Lind5, Bent Wulff Jakobsen6, Michael Rindom Krogsgaard7. 1. Section for Sports Traumatology M51, Bispebjerg-Frederiksberg Hospital, IOC Sports Medicine Copenhagen, Copenhagen, Denmark. jodc@regionsjaelland.dk. 2. Section of Physical and Occupational Therapy, Department of Rheumatology, Zealand University Hospital, Koege, Denmark. jodc@regionsjaelland.dk. 3. Research Unit for General Practice, Department of Public Health, University of Copenhagen, Copenhagen, Denmark. jodc@regionsjaelland.dk. 4. Research Unit for General Practice, Department of Public Health, University of Copenhagen, Copenhagen, Denmark. 5. Department of Orthopedic Surgery, Aarhus University, Aarhus, Denmark. 6. Aleris-Hamlet Hospitals, Aarhus, Denmark. 7. Section for Sports Traumatology M51, Bispebjerg-Frederiksberg Hospital, IOC Sports Medicine Copenhagen, Copenhagen, Denmark.
Abstract
PURPOSE: For clinical trials, it is essential that measures are sensitive to change. The aim of this study was to conduct a head-to-head comparison of responsiveness of four PROMs used to measure outcome after anterior cruciate ligament (ACL) reconstruction. The PROMs compared were the knee injury osteoarthritis outcome score (KOOS), the international knee documentation committee subjective form (IKDC), the Lysholm score, and the knee numeric-entity evaluation score (KNEES-ACL). We hypothesized that KNEES-ACL would be more responsive than the other PROMs, as KNEES-ACL was created based on patient interviews and validated using Rasch analysis. METHODS: One-hundred and sixty-six consecutive adults completed the four PROMs before and 3, 6, and 12 months after ACL-reconstructive surgery. Responsiveness was calculated as Cohen's Effect Size and Standardized Response Means. Bootstrapping was used to generate 95% confidence intervals for comparisons of responsiveness across PROMs. Repeated-measures ANOVA was also computed for each PROM. RESULTS: The largest effect sizes at 12 months were seen for KNEES-ACLSports-Behaviour (1.35, p < 0.001) and KNEES-ACLSports-Physical (1.19, p < 0.001), the smallest for KOOSADL (0.35, p < 0.001) and KOOSSymptoms (0.39, p < 0.001). IKDC and Lysholm lay between these with IKDC slightly more responsive. Head-to-head comparisons of similar subscales of KOOS and KNEES-ACL showed substantial differences in effect size in the domains of symptoms (0.69, p < 0.001), daily activities (0.31, p = 0.005), and Sports activity (0.63, p = 0.013) all in favour of KNEES-ACL. CONCLUSION: These results demonstrate superior responsiveness for KNEES-ACL, which is the only PROM that has been constructed through exhaustive patient feedback and validated for patients with ACL deficiency using the most stringent psychometric methods (Rasch analysis). KNEES-ACL is the most precise and accurate PROM for patients with ACL injury and the most trustworthy instrument for clinicians and clinical researchers. LEVEL OF EVIDENCE: Prospective cohort study, Level II.
PURPOSE: For clinical trials, it is essential that measures are sensitive to change. The aim of this study was to conduct a head-to-head comparison of responsiveness of four PROMs used to measure outcome after anterior cruciate ligament (ACL) reconstruction. The PROMs compared were the knee injury osteoarthritis outcome score (KOOS), the international knee documentation committee subjective form (IKDC), the Lysholm score, and the knee numeric-entity evaluation score (KNEES-ACL). We hypothesized that KNEES-ACL would be more responsive than the other PROMs, as KNEES-ACL was created based on patient interviews and validated using Rasch analysis. METHODS: One-hundred and sixty-six consecutive adults completed the four PROMs before and 3, 6, and 12 months after ACL-reconstructive surgery. Responsiveness was calculated as Cohen's Effect Size and Standardized Response Means. Bootstrapping was used to generate 95% confidence intervals for comparisons of responsiveness across PROMs. Repeated-measures ANOVA was also computed for each PROM. RESULTS: The largest effect sizes at 12 months were seen for KNEES-ACLSports-Behaviour (1.35, p < 0.001) and KNEES-ACLSports-Physical (1.19, p < 0.001), the smallest for KOOSADL (0.35, p < 0.001) and KOOSSymptoms (0.39, p < 0.001). IKDC and Lysholm lay between these with IKDC slightly more responsive. Head-to-head comparisons of similar subscales of KOOS and KNEES-ACL showed substantial differences in effect size in the domains of symptoms (0.69, p < 0.001), daily activities (0.31, p = 0.005), and Sports activity (0.63, p = 0.013) all in favour of KNEES-ACL. CONCLUSION: These results demonstrate superior responsiveness for KNEES-ACL, which is the only PROM that has been constructed through exhaustive patient feedback and validated for patients with ACL deficiency using the most stringent psychometric methods (Rasch analysis). KNEES-ACL is the most precise and accurate PROM for patients with ACL injury and the most trustworthy instrument for clinicians and clinical researchers. LEVEL OF EVIDENCE: Prospective cohort study, Level II.
Authors: Christine M McDonough; Eva Stoiber; Ivan M Tomek; Pengsheng Ni; Young-Jo Kim; Feng Tian; Alan M Jette Journal: J Orthop Sports Phys Ther Date: 2016-08-05 Impact factor: 4.751