Background

The Expanded Disability Status Scale (EDSS) is a commonly used measure of disability based on a clinician administered neurological exam in patients with Multiple Sclerosis (pwMS). Ordinal scores range from 0 (no disability) to 10 (death) in 0.5-point increments. The Patient Determined Disease Steps (PDDS) is a patient-reported measure of disability in pwMS, with ordinal scores ranging from 0 (no disability) to 8 (bedridden) in 1-point increments. Few studies have characterized the association between EDSS and PDDS. Those that have are limited by small sample sizes (n=96 to 103) and inconsistent and sometimes unrealistic correlations for patients with mild disability. For example, one of the studies reported a PDDS of 0 corresponds to an EDDS of 2.9.

Objectives

To characterize the association between EDSS and PDDS in a large sample of pwMS, which can help in the better application of PDDS in practice.

Methods

A total of 406 subjects participating in a US-based prospective cohort study were used for the analyses. All subjects had EDSS and PDDS assessments at baseline and approximately 12-months post-baseline, providing a total of 812 assessments. Mixed effect regression models using cubic splines, growth curve models and quadratic polynomials were employed to characterize the association between EDSS and PDDS and compared with models available in literature.

Results

The mean (standard deviation) age was 48.6 (10.35) years and 72.9% were female. Based on the Akaike’s Information Criterion, the quadratic polynomial regression was found to be the best fitting model. The equation predicting EDSS had the following form: EDSS = 1.4359 + 0.0830 * PDDS + 0.0999 * PDDS², which was modestly convex in shape. Patients with a PDDS of zero were predicted to have an EDSS of 1.4 and patients with a PDDS of 8 were predicted to have an EDSS of 8.4.

Conclusions

The fitted relationship between EDSS and PDDS in this large sample of pwMS showed generally similar scores across the ranges of the scales. The equation predicting EDSS as a function of PDDS revealed a more realistic fit when compared against other published equations. This study successfully developed a user-friendly crosswalk between EDSS and PDDS scores, using repeat measurements over a large sample. This crosswalk can aid in the better application and interpretation of PDDS.

Sponsored by: Biogen