Directional Symptom Dependencies in MS & Parkinson's Disease Dashboard

ORIGINAL RESEARCH article
J. Multidiscip. Healthc., 23 June 2026
Volume 2026:19, pp. 1–19 | https://doi.org/10.2147/JMDH.S607760

Directional Symptom Dependencies in Multiple Sclerosis and Parkinson’s Disease: A Comparative Bayesian Network Analysis

Alham Al-Sharman, Hanan Khalil, Dhafer Malouche, Saddam Kanaan, Meeyoung Kim, Nabil Saad, Marah Abdelrazeq

---

---

Abstract

Multiple sclerosis (MS) and Parkinson’s disease (PD) are progressive neurological disorders with complex interactions among motor symptoms, psychological disturbances, sleep problems, fatigue, and pain. Because correlation-based approaches cannot establish effect direction, this study used Bayesian Network (BN) analysis (Hill-Climbing algorithm, Gaussian BIC, 500 bootstrap resamples) to identify and compare directional dependency pathways in MS (n=104) and PD (n=54).

Key Findings

MS network — 18 nodes, 27 directed edges; 21/27 edges (77.8%) bootstrap-stable. Anxiety is the central hub:

• Anxiety → Sleep quality: β = 0.51 (p < 0.001)
• Anxiety → Physical activity: β = 0.37 (p < 0.001)
• Anxiety → Pain severity: β = 0.30 (p < 0.001)
• Anxiety → Physical fatigue: β = 0.24 (p = 0.016)
• Pain interference → Physical fatigue: β = 0.40 (p < 0.001)
• Pain interference → Cognitive fatigue: β = 0.23 (p = 0.005)

PD network — 14 nodes, 15 directed edges; 14/15 edges (93.3%) bootstrap-stable. Anxiety is again the central hub:

• Depression → Anxiety: β = 0.78 (p < 0.001)
• Anxiety → Pain interference: β = 0.59 (p < 0.001)
• Age → Physical activity: β = −0.50 (p < 0.001)
• Anxiety → Balance: β = −0.48 (p < 0.001)
• Age → Balance: β = −0.40 (p < 0.001)

Anxiety is a shared central hub in both diseases, but the downstream pathways differ. Findings are cross-sectional directional dependency structures (hypothesis-generating), not confirmed causal pathways.

Citation

Al-Sharman A, Khalil H, Malouche D, Kanaan S, Kim M, Saad N, Abdelrazeq M. Directional Symptom Dependencies in Multiple Sclerosis and Parkinson’s Disease: A Comparative Bayesian Network Analysis. J Multidiscip Healthc. 2026;19:1–19. doi: 10.2147/JMDH.S607760