Distributional Regression for Lungfunction of Cystic Fibrosis Patients with a Special Focus on Spatial and Random Effects

Marisa Lange, University of Goettingen

Co-authors: Colin Griesbach, University of Goettingen; Elisabeth Bergherr, University of Goettingen

Abstract: We present a modelling approach for lung function in cystic fibrosis patients using data from the German Cystic Fibrosis Registry. Applying Gaussian distributional regression via the gamboostLSS framework, we incorporate spatial effects through city-level triangulation with Voronoi polygons and Markov random fields. The model reveals clinically meaningful covariate effects and regional disparities.