Evaluating multivariable prediction models for Parkinson’s disease prognosis: a scoping review protocol

Comprehensive Summary

Eickholt et al explore the strengths and limitations of multivariable prediction models that gauge the progression of Parkinson’s disease (PD) in individuals over time. The study presents a scoping review, an approach to assess the breadth of information, to evaluate published prognostic models that aim to predict symptoms of PD, such as motor decline, non-motor symptoms, and dementia. Eickholt et al will collect data from studies across PubMed, EMBASE, Web of Science, and Scopus, evaluate each model using TRIPOD+AI and PROBAST reporting guidelines to determine the risk and applicability of the prediction models to a larger clinical setting. By organizing results according to specific clinical outcomes, the authors aim to clarify the current state of PD prognostic modeling and identify key gaps that must be addressed to improve prediction accuracy and guide future research.

Outcomes and Implications

Parkinson’s disease (PD) is a neurodegenerative disorder that impacts more than 10 million people around the world, yet there are currently no disease-modifying therapies or cures to slow its progression. As a result, clinicians cannot reliably predict how quickly the disease will advance or when specific motor and non-motor symptoms will emerge in individual patients. Prognostic models are prediction models to estimate the probability of certain symptoms occurring. Prognostic models aim to address this uncertainty by estimating the likelihood of future clinical outcomes based on multiple patient characteristics. In this study, Eickholt et al systematically evaluate existing PD prognostic models to identify weaknesses in validation, reporting, and clinical applicability. The authors emphasize that substantial additional research is needed and position this review as a foundation for strengthening future models, ultimately supporting more informed decision-making for both patients and clinicians.