Prediction of High Risk of Deviations in Home Care Deliveries

Anders L. Madsen, Kristian G. Olesen, Heidi Lynge Løvschall, Nicolaj Søndberg-Jeppesen, Frank Jensen, Morten Lindblad, Mads Lause Mogensen, Trine Søby Christensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review


This paper presents a real-world application of Bayesian networks to support existing home care quality supervision. In Denmark home care is delivered by municipalities, where the individual citizen is free to select the service provider, private or public. The aim of our work is to support the home care control process by identifying significant deviations automatically, pointing to reasons for a significant deviation and identifying future home care deliveries where there is a high probability of deviation between granted and delivered care to the individual citizen. Home care is granted as packages of time measured in minutes and we define a too high delivery rate as larger than 150. In the municipality under study in this work (municipality of Hj{ø}rring), the supervision of home care delivery is a manual and time consuming process prone to human error. This paper presents the results of efforts to automate parts of the supervision using Bayesian network modelling and data analysis. The results of the pilot study shows significant potential in applying Bayesian network modelling and data analysis to this challenge for the benefit of the municipality, the employees and the citizens.
Original languageEnglish
Title of host publicationProceedings of the 10th International Conference on Probabilistic Graphical Models, PMLR
EditorsManfred Jaeger, Thomas Dyhre Nielsen
Number of pages12
PublisherML Research Press
Publication date2020
Publication statusPublished - 2020
Event10th International Conference on Probabilistic Graphical Models - Hotel Comwell Rebild Bakker, Skørping, Denmark
Duration: 23 Sept 202025 Sept 2020


Conference10th International Conference on Probabilistic Graphical Models
LocationHotel Comwell Rebild Bakker
Internet address
SeriesThe Proceedings of Machine Learning Research


Dive into the research topics of 'Prediction of High Risk of Deviations in Home Care Deliveries'. Together they form a unique fingerprint.

Cite this