Systematic review/meta-analysis of five RCTs (n=270) found exercise interventions reduced fatigue in Parkinson’s disease (SMD −1.34) but with substantial heterogeneity and low certainty, while acupuncture showed no benefit.

The paper highlights exercise as a potentially useful symptomatic, non‑pharmacological strategy and pinpoints methodological gaps (small, heterogeneous trials, lack of PD‑specific fatigue measures) that should guide future mechanistic and sufficiently powered clinical studies, but it offers limited…

BACKGROUND: Fatigue is a prevalent and disabling non-motor symptom in Parkinson's disease (PD), yet evidence for effective management strategies, particularly non-pharmacological approaches, remains limited. OBJECTIVE: To systematically review and meta-analyse the efficacy of non-pharmacological interventions aimed at reducing fatigue in people with PD (PwP). METHODS: MEDLINE, CINAHL and AMED were searched from inception to January 2026 for randomised controlled trials (RCTs) in which fatigue was a primary outcome. Risk of bias was assessed using the Cochrane RoB2 tool. Standardised mean differences (SMDs) were pooled using random-effects models, and certainty of evidence was assessed using GRADE. The protocol was registered with PROSPERO (CRD42023394180). RESULTS: Five RCTs (n = 270 participants) met the inclusion criteria, evaluating exercise-based interventions (3 trials, n = 142) and acupuncture (2 trials, n = 128). Exercise interventions were associated with a statistically significant reduction in fatigue compared with controls (SMD =  - 1.34, 95% CI - 2.24 to - 0.44, p = 0.003), although heterogeneity was substantial (I2 = 89%). Acupuncture showed no significant effect compared with sham (SMD = 0.17, 95% CI - 0.13 to 0.51, p = 0.26), with low heterogeneity (I2 = 0%). Risk of bias was low in the acupuncture trials but rated as 'some concerns' for all exercise trials. GRADE assessment indicated low certainty evidence for exercise and moderate certainty evidence for acupuncture. Secondary outcomes such as quality of life and sleep were inconsistently reported and showed minimal change. CONCLUSIONS: Exercise-based interventions show preliminary promise for reducing fatigue severity in PwP, but the evidence is limited by heterogeneity and methodological concerns. Acupuncture appears ineffective compared to sham. High-quality, adequately powered RCTs evaluating diverse non-pharmacological approaches, including psychological and self-management strategies, are urgently needed, using standardised PD-specific fatigue measures.