Effects of exercise modality-dose combinations on cognitive function in Parkinson's Disease: A bayesian dose-response...

This study aimed to identify the most effective exercise modalities and optimal dose ranges for improving cognitive function in individuals with Parkinson's disease (PD).

We conducted a systematic search of five databases up to June 1, 2025, and included 33 RCTs involving 1,256 PD patients. Exercise doses were standardized using MET·min/week. A Bayesian dose-response network meta-analysis was performed in R to model nonlinear cognitive effects across nine exercise modalities. Certainty of evidence was assessed using the GRADE framework.

The pooled analysis revealed an inverted U-shaped relationship between exercise dose and cognitive outcomes, with the greatest benefit observed around 890 MET·min/week. The optimal range for cognitive improvement was identified between 500 and 1000 MET·min/week, beyond which the effects plateaued or declined, particularly above 1200 MET·min/week. Among the evaluated modalities, only multicomponent training and mind-body exercise demonstrated consistent and significant cognitive improvements at moderate doses. While virtual reality and aerobic walking showed beneficial effects at specific dose points, their overall efficacy was less stable. No consistent effects were observed for cycling, dance, or resistance training within the primary dose range.

This study is the first to construct a comprehensive dose-modality response model for cognitive enhancement in PD, providing empirical evidence for the optimal dose range and preferred exercise strategies. The findings offer a robust basis for developing tailored, precision-based non-pharmacological interventions for individuals with PD, with potential applicability to broader populations affected by neurodegenerative conditions.