Healthcare mobility across Italian Local Health Authorities (LHAs) generates substantial interregional financial flows, raising questions of quality, equity, and resource allocation. Existing analyses treat flows as static origin-destination tables, ignoring network structure, sparsity, and overdispersion. We propose a Bayesian latent space model with Hurdle Negative Binomial specification for weighted patient flows. Embedding LHAs in Euclidean space, closer positions predict higher flows, capturing unobserved affinity.
Modeling Healthcare Mobility in Italy: a Dynamic Latent Space Network Approach / Manente, C., Alfò, M., D'Angelo, S.. - (2026), pp. 591-596. (SIS-FENStatS 2026 Rome; Italy ) [10.1007/978-3-032-30881-8_95].
Modeling Healthcare Mobility in Italy: a Dynamic Latent Space Network Approach
Manente, Cecilia
Primo
;Alfò, MarcoSecondo
;D'Angelo, SilviaUltimo
2026
Abstract
Healthcare mobility across Italian Local Health Authorities (LHAs) generates substantial interregional financial flows, raising questions of quality, equity, and resource allocation. Existing analyses treat flows as static origin-destination tables, ignoring network structure, sparsity, and overdispersion. We propose a Bayesian latent space model with Hurdle Negative Binomial specification for weighted patient flows. Embedding LHAs in Euclidean space, closer positions predict higher flows, capturing unobserved affinity.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
