Measuring physical function in Amyotrophic Lateral Sclerosis: Novel solutions to an old problem

Abstract

Sensitive, objective outcome measures are essential for tracking clinical severity in neurodegenerative conditions such as Amyotrophic Lateral Sclerosis or Parkinson’s Disease. The current ‘gold standard’ multi-item rating scales and performance-based tests lack sensitivity and contain subjective biases. The goal of this thesis was to develop sensitive and clinically meaningful measurement tools in two areas: hand dexterity and respiratory function. To achieve this, current methods and recent trends in novel solutions were investigated. Unmet needs were found for: a simple, sensitive and clinically meaningful test of hand dexterity and a wearable respiratory volume sensor. A novel device for objective assessment of hand dexterity was designed and mechanically validated. Clinical validation of the device then took place in two large cohorts: healthy people and people with ALS, along with secondary validation in a small cohort of people with Parkinson’s Disease. A dexterity performance score was developed using a logistic regression machine learning algorithm which distinguished between healthy people and people with ALS. The device had a much lower floor effect and consequently a higher completion rate than the performance tests. In addition, investigation into an initial design of a novel wearable respiratory sensor demonstrated its mechanical validity. It showed potential as continuous measure of respiratory volume, however the current calibration models used to convert the sensor signals to lung volumes were unsuitable for this new, more sensitive measurement tool.