Authors:
Thomas Ostermann
1
and
Reinhard Schuster
2
Affiliations:
1
Witten/Herdecke University, Germany
;
2
University of Luebeck, Germany
Keyword(s):
Entropy, Diagnostic Diversity, Hospital Comparison, Classification.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Clinical Problems and Applications
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Enterprise Information Systems
;
Health Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
In Germany hospital comparisons are part of health status reporting. This article presents the application of
Shannon’s entropy measure for hospital comparisons using reported diagnostic data. We used Shannon’s
entropy to measure the diagnostic diversity of a hospital department by means of reported ICD–9–codes.
Entropy values were compared both with respect to the hospital status (i.e. primary, secondary, tertiary or
specialized hospital) and specialisations (e.g. surgery, gynaecology). There were relevant differences in
entropy values between the different types of hospitals. Primary hospitals differed from specialized
hospitals (0.535 ± 0.09 vs. 0.504 ± 0.07). Furthermore, specialized departments like obstetrics or
ophthalmology did generate lower entropy values than area-spanning departments like paediatrics or general
internal medicine, having significantly higher values. In conclusion, we showed how entropy can be used as
a measure for classifying hospitals. Besides of hospital
comparisons, this approach can be implemented in
all fields of health services research for measuring variability in nominal or ordinal data. The use of entropy
as a measure for health services research and classification algorithms should be encouraged to learn more
about this measure, which unreasonably has fallen into oblivion in health services research.
(More)