Concordance between patient- and physician-reported Myasthenia Gravis Activities of Daily Living (MG-ADL) scores

Abstract

Introduction/aims: Myasthenia gravis (MG) is a neuromuscular disease characterized by abnormal skeletal muscle fatiguability. The MG Activities of Daily Living (MG-ADL) scale assesses eight symptoms and is often used as primary endpoint in MG clinical trials where it is completed by neurologists. However, in observational studies, patients frequently complete the MG-ADL scale independently of their neurologist. In this study we aimed to assess the concordance between self- and physician-reported MG-ADL scores.

Methods: An international observational study was conducted among adult patients with MG scheduled for a routine visit or who entered the hospital via emergency services. Consenting patients and physicians completed the MG-ADL. Concordance between assessments was calculated using Gwet's agreement coefficient (Gwet's AC) for the MG-ADL individual items and the intraclass correlation coefficient (ICC) for the MG-ADL total score.

Results: Data were collected from 137 patients (63% female; mean age, 57.7 years). Physicians assessed the patient's symptoms as slightly more severe (8.1 vs 7.5 MG-ADL total score, respectively), corresponding to a difference of 0.6 on a range from 0 to 24. The ICC for the MG-ADL total score between the patient and the physician assessment was 0.94 (95% confidence interval, 0.89 to 0.95), showing excellent concordance. Gwet's AC showed substantial to almost perfect agreement for all items, except eyelid droop, for which the agreement was moderate.

Discussion: Our results demonstrate that patients and neurologists have a concordant assessment of the patient's MG symptoms when using the MG-ADL scale. This evidence supports patient self-administration of the MG-ADL in clinical practice and research.

Keywords: MG-ADL; concordance; myasthenia gravis; observational study; proxy assessment.