A domain-specific approach to characterizing balance confidence post-stroke.

This was a cross-sectional questionnaire analysis, not a treatment study, so it cannot show that targeting one balance domain improves balance confidence or walking performance. The reactive-balance items were worded as not losing balance or becoming unsteady, which is ambiguous because reactive tasks often involve loss of balance or unsteadiness. All reactive-balance items were placed in the context of walking or stepping, so responses may also reflect walking or anticipatory-control confidence. The item-to-domain assignments were made by a single expert team, not by formal large-scale consensus. The ABC scale did not capture sensory orientation as a separate domain, and walking items may not include biomechanically challenging circumstances such as uneven surfaces or narrow paths. The analysis tested a pre-specified model rather than identifying the best-fitting model.

The authors declared no potential conflicts of interest.

Do not use the three domain scores as a validated treatment-targeting tool. The study only grouped ABC items into three factors, and the reactive-balance items were worded ambiguously and included walking or stepping, so they may not measure reactive balance alone.