Background: A digital adaptation of the nine-hole peg test (9HPT) was developed with the potential to provide novel disability features for patients with multiple sclerosis (PwMS).
Objectives: The objectives were to evaluate the 9HPT features based on reliability, prognosis, and discrimination between treatment groups.
Methods: The MS partners Advancing Technology and Health Solutions (MS PATHS) cohort data were used to derive new features including completion time and speed. Association and reliability between features and clinical outcomes were tested by intraclass correlation coefficients (ICCs) with repeated measures. The added prognostic value of the features for a clinically meaningful decline was assessed by time-to-event analyses with likelihood ratio tests. The estimated effect size between treatment efficacy groups was acquired from linear mixed-effects models. Sample size was calculated for a hypothetical randomized clinical trial.
Results: For the 10,843 PwMS, speed and completion time were associated with MS disability. Compared with time, speed showed higher reliability (ICC = 0.78 vs 0.74), added benefits in predicting disability worsening (p < 0.001), better discrimination between high- and low-efficacy groups (effect size: 0.035 vs 0.015), and an 18% reduction in required sample size for a 1-year clinical trial.
Conclusion: Integrating horizontal hand distances traveled over the 9HPT pegboard can be a more reliable measure of hand function.
Keywords: Digital assessment; disability prognosis; disease progression; feature engineering; manual dexterity test; nine-hole peg test; treatment efficacy; upper-limb function.