S.P.5 Trial readiness: Clinical interpretability of change scores of the North Star Ambulatory Assessment in Duchenne muscular dystropy

Abstract The North Star Ambulatory Assessment (NSAA) is a clinician-rated measure of ambulation in Duchenne muscular dystrophy (DMD). A key role of the NSAA is to monitor disease progression and treatment impact (e.g. steroids). However, to date, the responsiveness (i.e. ability to detect change) of the NSAA has not been determined. The aim of this study was to address this issue. Data was collected on boys with DMD from the UK North Star clinical database ( n = 804 measurements). Two sub-groups of patients were defined: daily versus intermittent prednisolone steroid regimes. The NSAA was transformed to 0–100 range based on Rasch measurement methods. The responsiveness of the NSAA was examined through mean change scores over time, pair-wise squared t -values from paired samples t -tests, and effect size (ES) calculations. Minimal Important Difference (MID) statistics were derived based on distribution-based indicators. Data were analysed using RUMM2030 and SPSS-19. Generally, mean NSAA scores were higher in the daily group (mean change (SD) range: 52.9 (24.4) to 73.6 (13.6)) compared to the intermittent group (mean change (SD) range: 46.2 (14.5) to 68.9 (12.9)). The associated responsiveness statistics for the daily group suggested consistent low/moderate changes (mean change (SD) range: −7.5 (12.2) to 13.8 (17.6); ES range: 0.15 to 0.67) over time. For the intermittent group this fluctuated between low, moderate and large changes (mean (SD): −7.2 (8.5) to 13.8 (11.9), ES: 0.02 to 1.01). The mean MID was 8.8 and 6.9, for PF Regime_1 and PF Regime_2, respectively. In this first study to explore the responsiveness of the NSAA in DMD, the findings support the ability of the NSAA to detect important change over time. In addition our initial proposed MIDs provide a start point to provide clinical meaning of changes scores over time. This study also further demonstrates the importance of using Rasch measurement methods to better understand the development and use of rating scales.