Psychometric properties of a core set of measures of balance for people with cerebellar ataxia secondary to multiple sclerosis

Objective: To examine the reliability, validity and interpretability of four clinical measures in assessing the severity of balance dysfunction among people with cerebellar ataxia (CA) secondary to multiple sclerosis (MS). Design: Cross sectional observation study. Setting: Data collected across four outpatient clinics in New Zealand and United States of America. Participants: Sixty consecutive participants with CA secondary to MS. Main outcome measures Balance was assessed and video-recorded using the Berg Balance Scale (BBS), Timed Up and Go (TUG) test, the posture and gait sub-component of the International Co-operative Ataxia Rating Scale (PG-ICARS) and gait, stance and sit sub-components of the Scale for the Assessment and Rating of Ataxia (SARABal). The videos were later used to estimate reliability. The Barthel Index, Expanded Disability Status Scale (EDSS), ICARS and SARA were assessed and disease duration recorded. Results: Reliability was good for all four measures (range between ICC 0.95 and 0.99). Internal consistency was moderate to good for all four measures (α range 0.72-0.94), moderate to good correlation between the measures of balance (ρS range 0.72-0.85) and poor to moderate correlation with disease severity (EDSS), functional independence (Barthel Index) and disease duration (ρS range -0.37 to 0.76). Minimal Detectable Change (MDC) was derived for BBS (3), PG-ICARS (2) and SARABal (2). Measures were able to discriminate between assistive walking device users and non-users. Conclusions: All four measures showed good reliability and acceptable validity; however, owing to the item repetition in scoring of the PG-ICARS and moderate construct, criterion and convergent validity of the TUG, the BBS and SARABal are recommended for balance assessment in clinical practice for people with CA secondary to MS.