Boxing vs Sensory Exercise for Parkinson's Disease: A Double-Blinded Randomized Controlled Trial.
Featured on Physle Daily · Sunday, 16 August 2026
This 20-week randomized controlled trial compared boxing training with sensory exercise for people with Parkinson's disease, asking whether motor benefits from either program would persist after the intervention ended. Forty participants with idiopathic Parkinson's disease were randomized to 10 weeks of boxing (n=20) or sensory exercise (n=20), with motor symptoms assessed using the UPDRS-III at baseline, 10 weeks, and 20 weeks (10 weeks after training stopped). A statistically significant interaction between group and time was found, with the sensory exercise group showing greater improvement than the boxing group. At the week-20 follow-up, the sensory group's gains had not diminished, unlike the boxing group.
The abstract does not report whether within-group changes over time were themselves significant, and the specific mechanisms behind the difference are not described, so these findings should be seen as a starting point for further research rather than a settled conclusion.
Researchers wanted to know whether two types of exercise programs for Parkinson's disease, boxing and sensory training, would keep helping motor symptoms even after people stopped doing them. Forty people with Parkinson's disease were split into two groups: 20 did boxing and 20 did sensory exercise, each for 10 weeks.
Motor symptoms were measured with a standard rating scale at the start, at 10 weeks, and again at 20 weeks, ten weeks after the exercise programs ended. The sensory exercise group showed a statistically significant improvement in motor symptoms compared to the boxing group over this period. When researchers checked back at 20 weeks, the benefits in the sensory group had not faded, while the same could not be said for the boxing group.
This is a fairly small study of 40 people, and it does not explain why one type of exercise seemed to hold up better than the other. It offers useful early evidence but is not a final answer about which exercise approach works better long-term for Parkinson's disease.
This is a summary of research, not medical advice. Talk to your own healthcare provider about anything affecting your care.
This double-blinded RCT compared 10 weeks of boxing versus sensory exercise training in 40 people with idiopathic Parkinson's disease, using UPDRS-III scores at baseline, week 10, and week 20 (a 10-week washout period) as the outcome measure. Repeated-measures ANOVA revealed a statistically significant group-by-time interaction (F(1,39)=4.566, P=.036), with post hoc analysis indicating the sensory exercise group's improvement was retained at washout compared to boxing (P<.006). The design allows for a reasonable between-group comparison and includes a genuine follow-up assessment after program cessation, which is a notable strength given how little is known about durability of exercise effects in Parkinson's disease.
However, the sample size is modest (20 per arm), and the abstract does not report within-group significance testing, effect sizes, or details on blinding integrity, adherence, or adverse events, limiting how far these findings can be generalized. The authors themselves frame this as a signal for future research to include washout-period assessments, rather than as confirmation that one exercise modality is clinically superior. Replication with larger samples and mechanistic explanation for the group difference remains needed.
In a 40-person RCT, sensory exercise training showed a statistically significant improvement in motor symptoms compared to boxing over the study period.
At a 10-week washout follow-up, the sensory exercise group's motor symptom improvements persisted, whereas the boxing group's did not show the same retention.
This is a single small trial (20 per group) without reported within-group significance or mechanistic explanation, so findings should be treated as preliminary rather than conclusive.
Source
Neurorehabilitation and neural repair
Sangarapillai K, Norman BM, Almeida QJ · 2021
doi: 10.1177/15459683211023197