Abstract

The Community Dependency Index (CDI) was developed due to concerns that the Barthel Index (BI) was limited as a measure of physical function in community settings. However, no studies have compared the two rating scales within multiple sclerosis (MS). The aim of this study was to determine whether, in a community-based sample of people with MS, the CDI is a better measure than the BI. BI and CDI data were collected from 90 people with MS. Four measurement properties were compared: scaling assumptions (item mean scores, corrected item-total correlations), acceptability (score distributions, floor/ceiling effects), reliability (Cronbach's alpha) and validity (concurrent, discriminant, group differences, relative validity). Both scales satisfied recommended criteria for scaling assumptions (indicating it was legitimate to report a summed score) and internal consistency reliability (alpha >/0.85). The scales were highly correlated (r = 0.96), indicating they measured the same construct. Both scales demonstrated good group differences validity, but the BI was marginally superior. Notable ceiling effects (BI > CDI) were demonstrated for both scales, particularly in those less disabled. This study sample had relatively minor levels of disability, with over 70% still being independently mobile. In this sample of people with MS, the measurement properties of the BI and CDI examined were very similar, suggesting the CDI does not appear to have achieved its goal of better measurement.