Authors:
Bernardo Marques
;
Eliana Sousa
;
Tiago Silva-Costa
;
Ricardo Correia
and
Alberto Freitas
Affiliation:
Faculty of Medicine, University of Porto and University of Porto, Portugal
Keyword(s):
Data quality problems, Administrative data, Hospital information systems.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Enterprise Information Systems
;
Health Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
This paper is a preliminary study over the problems resulting from the integration of a departmental information system database over a central database. This work will allow the comparison between the quality of the data collected for clinical purposes by a medical department, and the data collected for administrative and epidemiological purposes in a central hospital database. It is expected that the different purposes for these two data collections can have an impact on data consistency, namely on it completeness and detail of information, among other data quality problems. We expect to detect the type of variables that are better recorded in each information system, by calculating and comparing the quality of similar variables. We also expect to detect differences between both systems in the registries of the same patients. This paper can play an important role for better understanding the quality of the integration of departmental systems in the general hospital information syst
em, pointing out some limitations about consistency and information flow. It is also our goal to suggest some recommendations and strategies to prevent data quality problems and to improve communication between central and departmental databases.
(More)