Quantifying the Association between Objectively Measured Physical Activity and Multiple Sclerosis in the UK Biobank

Area of Science:

• Neurology
• Epidemiology
• Biostatistics

Background:

• Utilized UK Biobank accelerometry data for objective physical activity (PA) measurement.
• Linked PA data with multiple sclerosis (MS) diagnosis information.
• Investigated PA differences in prevalent MS cases and predictive value for incident MS.

Purpose of the Study:

• Quantify PA differences between MS cases and healthy controls.
• Assess the predictive performance of objective PA measures for incident MS.
• Identify key predictors of MS development.

Main Methods:

• Compared accelerometer-derived PA summaries between 316 MS patients and matched controls.
• Employed Cox proportional hazards models to predict time to MS diagnosis in 19 predictors.
• Analyzed 585,900 person-years of follow-up data for incident MS cases.

Main Results:

• Statistically significant differences in all PA summaries between MS cases and controls (P < 0.001).
• Lower relative amplitude (RA) was the most predictive PA variable for incident MS (C = 0.594).
• A Cox regression model including age, stroke, and RA predicted MS risk with concordance C = 0.691.

Conclusions:

• Objective PA measures reveal significantly lower activity levels in individuals with MS.
• Younger age, history of stroke, and lower relative amplitude are associated with increased future MS risk.
• Objective PA data offers valuable insights into MS pathophysiology and prediction.