Objective: To evaluate the effects of a combined motor–cognitive exercise intervention in individuals with Parkinson’s disease (PD) and to explore latent response subgroups to identify variables associated with intervention responsiveness.

Background: PD is a progressive neurodegenerative disorder characterized by worsening motor and non-motor symptoms. Among non-pharmacological treatment approaches, combined motor-cognitive training has emerged as an important component of PD management [1]. However, exercise trials have often reported inconsistent findings. Characterizing heterogeneity in treatment responsiveness may improve understanding of intervention effects and inform personalized rehabilitation strategies [2].

Method: Thirty-eight individuals with PD completed a specialist-led combined motor–cognitive training program. The intervention consisted of supervised 60-minute sessions conducted twice weekly over 20 weeks, including an initial 8-week adaptation phase (T1) followed by a 12-week main training phase (T2). All participants underwent physical function and clinical assessments at baseline (T0), T1, and T2. Linear mixed-effects models were used to evaluate time effects across assessment points. A multivariate latent class linear mixed model was applied using UPDRS Part III and cognitive function scores to identify latent response subgroups not apparent at the group level. Logistic regression with elastic net regularization was performed to determine predictors of subgroup.

Results: Significant improvements were observed in UPDRS Part II and depression at both T1 and T2 compared with T0, and Mini-BESTest scores improved at T1. In contrast, UPDRS Part III and the 6MWT showed worsening at both T1 and T2 (Table 1). Latent class analysis identified two distinct response subgroups. Both subgroups demonstrated overall maintenance or improvement in cognitive function; however, they were primarily distinguished by trajectories of UPDRS Part III scores (Figure 1). Key predictors of subgroup included anxiety, grip strength, and the cognitive domain of quality of life (QoL), with an explanatory power of 54.9% (Table 2).

Conclusion: This study demonstrates heterogeneity in responsiveness to combined motor–cognitive exercise interventions in individuals with PD. Baseline factors including anxiety, grip strength, and the cognitive domain of QoL may help guide personalized rehabilitation strategies.

Table 1

Figure 1

Table 2

References: [1] Ferrazzoli, D., Ortelli, P., Cucca, A., Bakdounes, L., Canesi, M., & Volpe, D. (2020). Motor-cognitive approach and aerobic training: a synergism for rehabilitative intervention in Parkinson’s disease. Neurodegenerative disease management, 10(1), 41-55.
[2] Baudendistel, S. T., & Earhart, G. M. (2025). Characteristics of responders to interventions for Parkinson disease: a scoping systematic review. Neurodegenerative disease management, 15(4), 173-186.

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/heterogeneous-responsiveness-to-combined-motor-cognitive-exercise-in-parkinsons-disease-a-latent-class-analysis/