An analysis of all hospitalization causes in mainland Portugal

: evolution and the clinical coding

Abstract

Analysing the evolution of all cause-specific hospitalizations at a national level provides important information that comprehends an overall representation of the country healthcare supporting researchers, healthcare managers and decision-makers to efficiently plan and allocate resources (Hoeymans et al., 2012; Rossetto et al., 2019; Swe et al., 2020). In order to delineate policies to address needs and priorities in specific population groups, it is important to assess trends and patterns throughout time by specific characteristics, such as gender and age (Krumholz et al., 2015). Additionally, in Portugal on January 1st 2017 was implemented a new classification system of diseases, International Classification of Diseases, Tenth Revision, Clinical Modification and Procedure Coding System (ICD-10-CM/PCS), which implied a wide adaptation at several levels and which may have affected the hospitalization causes trends (Alonso et al., 2019, 2020). This thesis aims to provide a characterization of how hospitalizations causes have been evolving from 2000 until 2016 by highlighting the main trends and patterns as well as the main differences that have occurred during this period, throughout a retrospective descriptive study. Afterwards, an assessment of the ICD-10-CM/PCS implementation implications on hospitalizations causes trends will be performed with an interrupted time series approach. The results of this study point out evidence of declining trends among allcause hospitalizations in mainland Portugal with some differences according to age and gender among each specific cause. After the characterization of how hospitalizations evolved, we observed that ICD-10-CM/PCS caused variations in some cause-specific hospitalizations, yet, when considering all-cause hospitalizations, no variation due to the transition was found. Therefore, future studies on epidemiological profiles and trends should consider the variations vi ii caused by the codification transition. However, we must be cautious with results as 2017 hospital data completeness still needs to be assured and it would be preferable to include a symmetric distribution before and after the transition, by thus including 2018 as well.

Date of Award	13 Jul 2021
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	António Andrade (Supervisor) & João Vasco Nunes dos Santos (Co-Supervisor)