Short-Form New Knee Society Score: What Is Its Responsiveness and Convergent Validity with Other Scores?

Abstract

The short-form version of New Knee Society Score (SF-NKSS) was designed with a purpose to reduce respondent burden. Literature review revealed only one report by Scuderi et al on responsiveness of derived SF-NKSS, but it was evaluated in two separate patient cohorts pre- and postsurgery. Our study had evaluated responsiveness and convergent validity of derived SF-NKSS in a single, large patient cohort followed longitudinally from preoperative status to over 1 year. Our database of 148 knee arthroplasty patients operated by the same surgeon, whose NKSS, Western Ontario and McMaster Universities Arthritis Index (WOMAC), and 12-item Short-Form Survey (SF-12) scores were prospectively collected preoperatively and postoperatively at 3 and 12 months for a study published earlier, was analyzed for derived SF-NKSS. Responsiveness was evaluated by determining the effect size, standardized response mean (SRM), and ceiling and floor effects. For convergent validity, Pearson's correlation coefficient was used. SF-NKSS was found to be most responsive with the largest effect size and SRM at 3 months (3.02 and 2.50, respectively) and at 12 months (3.58 and 2.92, respectively) with no ceiling or floor effect. SF-NKSS was followed in responsiveness by original NKSS, WOMAC, and SF-12 in a descending order. Convergent validity showed a strong correlation (<i>r</i> = 0.8-1.0; <i>p</i> < 0.001) of SF-NKSS with NKSS and a moderate to strong correlation (<i>r</i> = 0.5-0.6; <i>p</i> < 0.001) with WOMAC and SF-12. We concluded that SF-NKSS is a reliable, highly responsive tool for post-total knee replacement evaluation. It also has the ability to register improvement in the patient's recovery, which can continue even after 1 year. We found that the SF-NKSS can be used interchangeably with the original NKSS.