What is the Responsiveness and Respondent Burden of the New Knee Society Score?
Although the new Knee Society score (NKSS) has been validated by a task force, a longitudinal study of the same cohort of patients to evaluate the score’s responsiveness and respondent burden has not been reported, to our knowledge.
We analyzed the NKSS for (1) responsiveness; (2) respondent burden; and (3) convergent validity in 148 patients studied longitudinally during more than 1 year.
During an 8-month period, 165 patients underwent TKA by the same surgeon at our institution, of whom 148 (90%) completed this study; the others were excluded because of distance to travel or loss to followup at the specified time. The NKSS, WOMAC, and SF-12 were completed by each patient 1 day before surgery and at 3 and 12 months postoperatively. At the same times, the original KSS (OKSS) which is designed as an observer’s assessment, was completed by the same orthopaedic fellow for all patients. Responsiveness of the NKSS was assessed by determining effect size, standardized response mean (SRM), and ceiling and floor effects. Respondent burden was assessed through time to completion recorded in minutes and ease of completion which was measured objectively on a Likert scale of 1 to 5 by the patients. Convergent validity was assessed by correlating the NKSS with the WOMAC, SF-12, and OKSS (current, widely used scales) by Pearson’s correlation coefficient.
Effect size was largest (2.83 and 3.38) and SRM was highest (2.29 and 2.68) for the NKSS at 3 and 12 months respectively, indicating the NKSS to be the most-responsive score followed by the OKSS, WOMAC, and SF-12. The NKSS exhibited no ceiling and floor effects. The NKSS took a longer time to complete (5.49 ± 3.56 minutes) compared with the WOMAC (4.64 ± 3.19 minutes) and SF-12 (4.35 ± 3.27 minutes). The mean difference in time taken for the NKSS versus the WOMAC was 0.85 minutes (95% CI, 0.54–1.17 minutes; p < 0.001) and the mean difference for the NKSS versus the SF-12 was 1.14 minutes (95% CI, 0.76–1.15 minutes; p < 0.001). Its ease of completion generally was comparable to that of the WOMAC and SF-12. Convergent validity showed a strong correlation (r > 0.6; p < 0.001) of the NKSS with the WOMAC at all times and moderate to strong correlation (r = 0.4–0.6; p < 0.001) with the SF-12 and OKSS at the first two assessments, which became strong (r > 0.6; p < 0.001) at 12 months.
The NKSS exhibited greater responsiveness than the WOMAC, SF-12, and OKSS scales and showed no ceiling effect, indicating adequate potential for recording future improvement. The NKSS also showed reliable convergent validity when correlated with these other scores. However, it posed a greater respondent burden in terms of time to completion.
As independent nondevelopers of the NKSS, we found it to be a responsive tool for assessment of TKA outcomes. We have confirmed that the NKSS can be used interchangeably for this purpose with the WOMAC scale and that it correlates positively with other established scales of the SF-12 and OKSS. Further study of the short-form version will establish whether it also can be used effectively while reducing the respondent burden.