Psychometric assessment of a modified part I of the unified multiple system atrophy rating scale

Abstract

Background

Despite widespread use in MSA clinical trials, the UMSARS has recognized limitations in responsiveness.

Objectives

To evaluate the psychometric properties of a modified UMSARS Part I incorporating collapsed response categories (0 and 1) and excluding item 11 (sexual function).

Methods

Psychometric properties and responsiveness of the modified UMSARS Part I were assessed and compared with other UMSARS scores using data from a randomized, double-blind, placebo-controlled trial.

Results

For the mUMSARS Part I, internal consistency (Cronbach's alpha) was 0.80 at Week 48 and 0.76 at end-of-treatment, and test-retest reliability (intraclass correlation coefficient) was 0.61. Construct validity for mUMSARS Part I ranged from r = 0.51–0.72, comparable to the UMSARS Total score (r = 0.49–0.69) and higher than UMSARS Part I (r = 0.42–0.60). At Week 48, mUMSARS Part I effect sizes across global impression anchors were large in worsening groups (1.64–2.92) and lower in no-change groups (0.97–1.62).

Conclusions

The modified scale demonstrates acceptable psychometric properties, supporting its use in trials assessing progression of MSA. Both the original UMSARS Part I and mUMSARS Part I performed well across psychometric assessments; however, the mUMSARS Part I showed consistently stronger construct validity and greater responsiveness to change.