Study of the dynamics of hospital services through the use of multivariate statistical techniques: the case of the Clinical Hospital of the University of Chile

Objective: To obtain useful hospital management indicators, based on the technique of principal components analysis. Material and method: We analyzed information on the activity of medical-surgical services in the Hospital Clinico de la Universidad de Chile from January 2003 to December 2004. The following variables were analyzed: number of discharges, lethality rate, readmissions rate, number of outpatient consultations, number of hospital stays, and the surgical complexity index. The principal components analysis (PCA) technique was applied and the R correlation matrix was used. Results: The first two principal components for each year were selected, accounting cumulatively for 75.92% of the variability in the data: 47.02% for the first and 28.90% the second. The first component may be interpreted as a new index related to clinical complexity and the second related to healthcare demand. Conclusions: The use of these new indicators would aid classification of hospital services and performance monitoring, allowing appropriate corrective measures to be introduced.