Andy
Senior Member (Voting rights)
Background:
Fatigue affects 50% to 90% of patients with multiple sclerosis, Parkinson’s disease, stroke, ME/CFS, and long COVID and represents a primary barrier to rehabilitation participation yet remains poorly understood despite decades of research.The Problem:
Systematic examination of recent comprehensive reviews reveals a dual diversity failure. Brain diversity is underexamined: fatigue, fundamentally a central nervous system symptom, is assessed predominantly through questionnaires (95%-100% of studies) and peripheral biomarkers, while functional neuroimaging appears in fewer than 15% of studies. Population diversity is neglected: reviews consistently document a predominance of Western European and North American cohorts, with racial and ethnic composition inconsistently reported and demographic stratification virtually absent.The Solution:
Identical fatigue scores reflect heterogeneous brain mechanisms, basal ganglia hypometabolism, frontal dysfunction, inflammatory network disruption, and differentially distributed across populations. Five testable predictions distinguish this framework: (1) neuroimaging reveals distinct subtypes; (2) subtypes show different clinical phenotypes; (3) treatment responses vary by subtype; (4) populations differ in subtype distribution; and (5) stratification resolves apparently inconsistent findings.Recommendations:
Fatigue rehabilitation research should (1) incorporate multimodal neuroimaging as standard, with a tiered protocol accommodating participant burden; (2) intentionally recruit demographically diverse populations; (3) employ data-driven clustering to identify brain-based subtypes; and (4) conduct mechanism-stratified treatment trials. An illustrative study design incorporating multimodal neuroimaging, comprehensive clinical and biological assessment, and diverse population sampling can test core predictions within existing research infrastructure.Impact:
Mechanism-based patient stratification enables precision rehabilitation, reducing required trial sample sizes while identifying effective interventions and preventing harm in vulnerable subgroups.Open access