Exercise may ease Parkinson's-related fatigue, but comparison treatments matter
Randomized trials favored exercise over usual care for fatigue. Comparisons involving stretching, sleep guidance or other active approaches were less clear.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
Exercise reduced fatigue scores compared with usual care in the pooled analysis, and the authors rated the evidence as moderately certain. Other comparisons suggested that sleep hygiene, relaxation, education or stretching might have effects comparable to some exercise programs. The review does not establish that a particular exercise type or intensity is best.
The main usual-care comparison used only part of the included evidence. The abstract does not provide follow-up durations, participant totals or enough direct comparisons to identify the most effective exercise approach.
What researchers found
Fatigue scores
Exercise versus nonexercise usual care across 7 studies; standardized mean difference. The negative value means lower fatigue scores with exercise. Standard deviations measure how much participants' scores vary, not scale points or percentages. Assessment timing is unspecified.
How fatigue was measured
Most studies used the Parkinson's Fatigue Scale or Fatigue Severity Scale. Pooling results as standardized differences allows scores from different scales to be compared; the outcome remains measured fatigue, not progression of Parkinson's disease.
Why the comparator matters
The trials tested walking, strengthening, dance and combined training. Some comparison groups received stretching or other potentially helpful activities. Those studies answer a different question from trials comparing exercise with usual care alone.
The original publication
Physical Exercise for Managing Fatigue in Parkinson's Disease: Clinical and Research Recommendations From a Systematic Review and Meta-Analysis.
Costa V, Zambetta ML, Gianlorenço AC
J Geriatr Phys Ther · 2026
- PubMed ID
- 41538284
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.