Effects of Upper Limb Exercise or Training on Hand Dexterity and Function in People With Parkinson Disease: A Systematic Review and Meta-analysis.
Featured on Physle Daily · Saturday, 3 October 2026
Parkinson disease often affects the hands, making everyday tasks like buttoning clothes or writing harder, so this review asked whether exercise or hand-training programs actually help. Researchers pooled results from eighteen randomized controlled trials involving 704 people, comparing various upper limb exercise or training approaches, including a few focused specifically on handwriting. Across studies, there was moderate-quality evidence of a small improvement in hand dexterity, meaning coordinated finger and hand movement, after training. Evidence for whether people felt their hand function improved in daily life was far less certain, showing no statistically significant effect and very low confidence in that result.
Handwriting-specific training showed some improvement too, but the evidence was described as low quality. The interventions varied widely in approach and dosage, which limits conclusions about what specifically works best or how much training is needed.
If you live with Parkinson disease, you may already know how much it can affect fine hand movements, like doing buttons, using cutlery, or writing. This review looked at eighteen studies, together involving about seven hundred people, testing different kinds of hand and arm exercise or training programs. The combined results showed a small improvement in hand dexterity, the ability to coordinate precise finger and hand movements, with moderate confidence in that finding.
However, when it came to how people rated their own hand function in daily life, the evidence did not show a clear benefit, and the confidence in that finding was very low. A few studies focused specifically on handwriting training and showed some improvement there too, but the quality of that evidence was low.
Because the programs tested varied so much in type and amount of training, it is not yet clear exactly what kind of exercise, or how much, works best.
This is a summary of research, not medical advice. Talk to your own healthcare provider about anything affecting your care.
For clinicians weighing how to address hand and upper limb impairment in Parkinson disease, this systematic review and meta-analysis synthesizes evidence from eighteen randomized controlled trials (n=704, low to unclear risk of bias) testing varied exercise or training interventions against outcomes of dexterity, self-reported function, and handwriting. The pooled analysis found moderate-quality evidence of a small positive effect on within-hand dexterity (SMD=0.26; 95% CI 0.07–0.44), supporting training as a plausible avenue for addressing manual dexterity deficits.
In contrast, self-reported hand function showed a nonsignificant effect (SMD=0.67; 95% CI -0.40–1.75) based on very low-quality evidence, meaning confidence in that null result is minimal rather than reassuring. A narrative synthesis of handwriting-focused interventions suggested improved performance, but this rests on low-quality evidence from only three trials. The marked heterogeneity in intervention type, dose, and delivery across studies limits any conclusion about optimal treatment parameters or which task-related elements drive benefit.
This review indicates that exercise and training may support dexterity outcomes with moderate certainty, but it does not establish effective approaches for self-reported function or handwriting, and future trials examining dose and active ingredients are needed before firmer practice conclusions can be drawn.
Meta-analysis of randomized controlled trials found moderate-quality evidence of a small improvement in hand dexterity after upper limb exercise or training (SMD=0.26; 95% CI 0.07, 0.44).
Evidence for improvement in self-reported hand function was very low quality and the effect was not statistically significant (SMD=0.67; 95% CI -0.40, 1.75), so no clear benefit could be established.
Handwriting-specific training showed improved performance in a narrative review of three studies, but this was based on low-quality evidence, and interventions overall varied widely in type and dose, limiting conclusions about optimal approaches.
Source
Archives of physical medicine and rehabilitation
Proud EL, Miller KJ, Morris ME et al. · 2024
doi: 10.1016/j.apmr.2023.11.009