Gu Yuan, Gong Yishu
School of Medicine, Stanford University, Palo Alto, CA, United States of America.
The Harvard T.H. Chan School of Public Health, Harvard University, Boston, MA, United States of America.
PeerJ. 2025 May 29;13:e19519. doi: 10.7717/peerj.19519. eCollection 2025.
Parkinson's disease (PD) is a chronic neurological disorder that affects millions of people worldwide. A common motor symptom associated with PD is gait impairment, leading to reduced step count and mobility.
Monitoring and analyzing step count data can provide valuable insights into the progression of the disease and the effectiveness of various treatments. In our study, the generalized additive model (GAM) was used to identify statistically significant variables for step counts. Additionally, a web application was developed as an interactive visualization tool.
The GAM model shows that the following variables are statistically significant for daily step counts: sex ( = 0.03), handedness ( = 0.015), PD status of father ( = 0.056), COVID-19 status (Yes No, = 0.008), cohort (PD Healthy, < 0.0001), the cubic regression spline with three basis functions of age by cohorts ( < 0.0001), and the random effect of individual age trajectories ( = 0.0001).
Based on the PPMI data, we find that sex, handedness, PD status of father, COVID-19 status, cohort, and the smoothing functions of age are all statistically significant for step counts. Additionally, a web application tailored specifically for step count analysis in PD patients was developed. This tool provides a user-friendly interface for patients, caregivers, and healthcare professionals to track and analyze step count data, facilitating personalized treatment plans and enhancing the management of PD.
帕金森病（PD）是一种慢性神经疾病，影响着全球数百万人。与帕金森病相关的一种常见运动症状是步态障碍，导致步数减少和行动能力下降。
监测和分析步数数据可以为疾病进展和各种治疗方法的有效性提供有价值的见解。在我们的研究中，广义相加模型（GAM）用于识别与步数具有统计学显著意义的变量。此外，还开发了一个网络应用程序作为交互式可视化工具。
GAM模型显示，以下变量对每日步数具有统计学显著意义：性别（P = 0.03）、用手习惯（P = 0.015）、父亲的帕金森病状态（P = 0.056）、新冠病毒感染状态（是/否，P = 0.008）、队列（帕金森病患者/健康人，P < 0.0001）、按队列划分的具有三个基函数的年龄三次回归样条（P < 0.0001）以及个体年龄轨迹的随机效应（P = 0.0001）。
基于PPMI数据，我们发现性别、用手习惯、父亲的帕金森病状态、新冠病毒感染状态、队列以及年龄平滑函数对步数均具有统计学显著意义。此外，还开发了一个专门针对帕金森病患者步数分析的网络应用程序。该工具为患者、护理人员和医疗保健专业人员提供了一个用户友好的界面，用于跟踪和分析步数数据，有助于制定个性化治疗方案并加强帕金森病的管理。
