BackgroundBalance confidence is an essential component of fall risk assessment in persons with multiple sclerosis (pwMS).AimsThe aims of this cross-sectional study were to 1) investigate the ability of the 16-item Activities-specific Balance Confidence scale (ABC-16), 6-item Activities-specific Balance Confidence scale (ABC-6), and each item of the ABC-16 for distinguishing fallers and 2) determine cutoff scores for these scales to discriminate fallers and non-fallers in pwMS.MethodsOne hundred and fifty-six participants [fallers/non-fallers: 60 (38.5%)/96 (61.5%), median EDSS: 1.5] were enrolled. Balance confidence was assessed using the ABC-16 and ABC-6. The self-reported number of falls in the past three months was recorded. Descriptive assessments, including walking, balance, and cognition were performed. Logistic regression and receiver operating characteristic analyses were conducted to estimate the sensitivities and specificities of the ABC-16 and ABC-6.ResultsBoth the ABC-16 (AUC: 0.85) and ABC-6 (AUC: 0.84) had the discriminative ability for falls. Each item of the ABC-16 scale was a significantly related to falls [odds ratio (OR) range: 1.38 to 1.89]. Items 8 and 10 had the highest odds ratio (OR: 1.85; 95%CI: 1.47–2.33, OR: 1.89; 95%CI: 1.49–2.40; respectively). We found cutoff scores of ≤ 70 of 100 (sensitivity: 71.67, specificity: 86.46) and ≤ 65/100 (sensitivity: 76.67, specificity: 79.17) in discrimination between fallers and non-fallers for the ABC-16 and ABC-6, respectively.ConclusionBoth original and short forms of the ABC scale are an efficient tool for discriminating fallers and non-fallers in pwMS. Although all items are related to falls, outdoor walking activities have the strongest associations with falls than other items.