| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 110.71 KB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Characterizing the profile of frequent users of an emergency service is
essential for the definition of health management policies, looking for an efficient
meeting of patients’ needs and offering a better quality of services to users of the
National Health Service. Therefore, it is reasonable that several research studies de veloped in this area focus on the very frequent users of hospital emergency services,
usually referred to as hyper users. Their definition varies according to the studies,
but in general, a hype user is a patient with 4 or more visits to the emergency
department per year. This study analyzes the inflow of users to the emergency de partment of the Hospital do Esp´ırito Santo de Evora (HESE), by type of user, in a ´
period between January 2018 and May 2021. Based on Generalized Linear Models
(GLM) and Generalized Additive Models (GAM) it is attempted to identify some
factors that allow to outline the profile of the hyper user of the HESE emergency
service. The results obtained with several models are compared and some consid erations are made about the modelling limitations taking into account the type of
data and the context of the collection period. The profile of a frequent user of the
HESE emergency services includes the triage code indicator, the user residence, age
and gender, the admission shift, the waiting time and the physician's diagnosis.
Description
Keywords
Emergency department exploratory data analysis genealized additive linear models generalized linear models