A simple-to-use web-based calculator for survival prediction in Parkinson's disease.

Research overview

The text below is the source article’s original abstract under a verified reuse license. cc by

BackgroundTo establish and validate a nomogram and corresponding web-based calculator to predict the survival of patients with Parkinson's disease (PD).MethodsIn this cohort study, we retrospectively evaluated patients (n=497) with PD using a two-stage design, from March 2004 to November 2007 and from July 2005 to July 2015. Predictive variables included in the model were identified by univariate and multiple Cox proportional hazard analyses in the training set.ResultsIndependent prognostic factors including age, PD duration, and Hoehn and Yahr stage were determined and included in the model. The model showed good discrimination power with the area under the curve (AUC) values generated to predict 4-, 6-, and 8-year survival in the training set being 0.716, 0.783, and 0.814, respectively. In the validation set, the AUCs of 4- and 6-year survival predictions were 0.85 and 0.924, respectively. Calibration plots and decision curve analysis showed good model performance both in the training and validation sets. For convenient application, we established a web-based calculator (https://tangyl.shinyapps.io/PDprognosis/).ConclusionsWe developed a satisfactory, simple-to-use nomogram and corresponding web-based calculator based on three relevant factors to predict prognosis and survival of patients with PD. This model can aid personalized treatment and clinical decision-making.

Related topics