Authors:
Javier Fernández-Sánchez
1
;
Cristina Soguero-Ruiz
1
;
Pablo de Miguel-Bohoyo
2
;
Francisco Javier Rivas-Flores
2
;
Ángel Gómez-Delgado
3
;
Francisco Javier Gutiérrez-Expósito
1
and
Inmaculada Mora-Jiménez
1
Affiliations:
1
Universidad Rey Juan Carlos, Spain
;
2
University Hospital of Fuenlabrada, Spain
;
3
University Hospital of Sueste and Arganda del Rey, Spain
Keyword(s):
Hypertension, Chronic Condition, Health Status, Clinical Risk Groups, ICD9-CM Diagnosis Codes
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Data Manipulation
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Methodologies and Methods
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Observation, Modeling and Prediction of User Behavior
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Soft Computing
Abstract:
Hypertension is a chronic condition that has a considerable prevalence in the elderly. Furthermore, hypertensive patients double cost of normotensive individuals. The budget reduction and the increasing concern about the sustainability of the healthcare system have caused that improving the efficiency and use of resources are a priority in developed countries. Identification of chronic hypertensive patients, i.e., patients with high blood pressure, can be performed by means of population classification systems such as Clinical Risk Groups (CRGs). CRGs classify individuals in health status categories taking both demographic and clinical information of the encounters that individuals have with the healthcare system during a defined period of time. In this work, we determine the characteristic profile and the evolution of diagnosis codes according to the International Classification of Diseases 9 revision, Clinical Modification (ICD9-CM), focusing on healthy and chronic hypertensive pat
ients at different chronic statuses (CRG). Our data correspond to the population associated to the University Hospital of Fuenlabrada (Madrid, Spain) during the year 2012, providing about 46000/16000 healthy/hypertensive individuals. We found that profiles associated to different health statuses have different patterns in terms of ICD-9 diagnosis codes. Furthermore, a prediction method is proposed to determine the health status of a new patient according to demographic (age and gender) and clinical (diagnosis codes) data. We conclude that gender is the less informative characteristic, though the combination of age and diagnosis codes have a great potential when they are non linearly combined.
(More)