ABSTRACT
BACKGROUND: Utilization of Diagnosis Related Groups (DRGs') for hospital payments in France. OBJECTIVES: To search the variations and the explicative factors of length of stay in a pneumologic unit, to analyse the relevance of the indicators by using the DRG's and the supplementary variables in a medical and economical perspective. SETTING: Pneumologic unit of Limoges' teaching hospital. METHODS: From 1-4-93 to 31-5-94, a prospective study analysed 1081 stays (one-day-longer) by the medical unit summary and by some variables describing the stay (planned admission, appropriateness), the patient (physical mobility at entrance, severity of disease), the procedures (physiotherapy, social needs). Statistical tests associated univariate analysis, linear and logistic regressions, hierarchic segmentation. RESULTS: DRG's explain 23% of logarithm length of stay. The mathematical model has separated five variables, which are associated with an increase of length of stay (age, number of procedures, physical mobility at entrance, social complexity, intensity of physiotherapy), one associated with a decrease of length of stay (planned admission). CONCLUSION: DRG's, alone, do not describe the dispersion of the length of stay, which is usually used to assess the consumption of resources during an hospitalization. The record of supplementary variables needs further studies, to improve the patients' classification.
