Using length of stay and inactive days in the hospital to assess appropriateness of utilisation in Barcelona, Spain

Study objective - To compare the level of inappropriate utilisation of a teaching hospital in two different calendar years and to analyse the relationship between changes in appropriateness of utilisation and changes in average length of stay. Design - Retrospective evaluation of the medical records of patients admitted to a hospital in two periods: 1988 and 1990. Setting - Hospital Universitari del Mar, a teaching hospital with 440 beds located in Barcelona, Catalonia, Spain. Participants - Medical records were randomly selected from records of adults discharged from hospital with a principal diagnosis other than normal delivery or any psychiatric condition. Altogether 750 records with complete data were reviewed for 1988 and 633 for 1990. Measurements and main results - The appropriateness evaluation protocol (AEP) was used to assess whether or not the admission and each day of the hospital stay were appropriate. For data analysis we used the extension of logistic regression that allows quantification of within-patient clustering of inactive days. The average length of stay (LOS) was 11.7 days in 1988 and 9.5 in 1990 (p < 0.001). In contrast, the proportion of inappropriate admissions was 12% in 1988 and 19% in 1990 (p < 0.001). Among those patients who were appropriately admitted, the average odds of a given day being inactive in 1990 was no lower (OR = 1.09) than the odds in 1988, but the clustering of inactive days was significantly (p < 0.001) lower in 1990. Conclusions - In the hospital studied, a lower average length of stay was not associated with an improved level of appropriate utilisation. Policies exclusively focused on lowering LOS may not directly lead to a reduction in inappropriate hospital utilisation.