Bayesian analysis and prediction of patients' demands for visits in home care

Home care (HC) providers are complex structures which include medical, paramedical, and social services delivered to patients at their domicile. High randomness affects the service delivery, mainly in terms of unplanned changes in patients' conditions, which make the amount of required visits highly uncertain. In this paper, we propose a Bayesian model to represent the HC patient's demand evolution over time and to predict the demand in future periods. Results from the application to a relevant real case validate the approach, since low prediction errors are found.