Psychometric measurement properties and reference values of the six-spot step test, the six-minute walk test, the 25-foot walk test, and the 12-item multiple sclerosis walking scale in people with multiple sclerosis.

Unlabelled: This review investigated the psychometric properties of the most commonly used short-, long-, complex- and patient-reported walking outcome measures in multiple sclerosis(MS): the timed-25-foot walk test (T25FW), the six-minute walk test (6MWT), the six-spot step-test (SSST), and the 12-item MS walking scale (MSWS-12), along with reported reference data of these tests.

Methods: Based on PubMed and Embase searches, psychometric as well as descriptive data of T25FW, 6MWT, SSST, and MSWS-12 were extracted from studies evaluating persons with MS (pwMS). Descriptive data was also extracted from healthy controls (HC), if reported. Data was displayed as median [IQR].

Results: A total of n=84 studies (N=36.929 pwMS, 64% females, age 49.6 [43;51] yrs; N=3.093 HC, 40.0 [37.5;47.5] yrs) were included. In pwMS, expanded disability status scale (EDSS) was 4.0 [3.0;4.8], patient determined disease scale (PDDS) 2.6 [2.0;3.0], time since diagnosis 11.6 [9.9;13.2] yrs, and MS-phenotypes (RR/SP/PP/unknown) were 68/19/10/3%. PwMS performed substantially worse than HC across all walking capacity outcomes. Weak to strong associations were found for construct validity (r=0.29-0.88, EDSS or PDDS and walking capacity or ability outcomes) and concurrent validity (r=0.16-0.88 between walking capacity and ability, r=0.73-0.95 between walking capacity outcomes). Ecological validity showed weak to moderate associations between daily steps and walking outcomes (r=0.42-0.68). Good to excellent test-retest, intrarater, and interrater reliability were reported across outcomes (ICC=0.71-1.00). Responsiveness was most frequently reported as the minimal-clinically-important-difference, minimal-detectable-change, or as the minimal-important-change (data not shown).

Conclusion: Overall the T25FW, 6MWT, SSST, and MSWS-12 demonstrate moderate to excellent psychometric properties (i.e., valid, reliable, and responsive to changes), which make them clinically useful and applicable to research.