Changes in bed resources and admission patterns in acute public hospitals in Victoria, 1987-1995

Objective: To describe changes in admission patterns, bed resources and hospital use in acute public hospitals and their relationship with early readmissions and interhospital transfers in Victoria between 1987 and 1995. Design: Descriptive study of longitudinal trends using data from the Victorian Inpatient Minimum Database and the Acute Health Services Branch of the Department of Human Services, Victoria. Setting: State of Victoria. Main outcome measures: Acute public hospital beds and hospital separations per 1000 population; separation type (same-day or longer); mean length of stay; interhospital transfers; and readmissions to the same hospital within 28 days. Results: Between 1987-88 and 1994-95, public hospital beds in Victoria decreased from 3.2 to 2.8/1000 population, and mean length of hospital stay decreased from 6.4 to 4.2 days. There was a significant direct correlation between number of beds/1000 and length of stay (r=0.90; 95% confidence interval [CI], 0.52-0.98). Bed occupancy remained constant at 80%. Over the same period, same-day admissions increased from 22% to 42% of hospital separations, interhospital transfers increased from 2.7% to 4% of separations, and readmissions to the same hospital within 28 days for any reason increased from 12.4% to 15% of separations (21% increase). Beds/1000 were inversely correlated with interhospital transfers (r=-0.83; 95% CI, -0.31 to -0.97), while readmission rates were inversely correlated with beds/1000 (r=-0.89; 95% CI, -0.98 to -0.50) and length of hospital stay (r=-0.95; 95% CI, -0.99 to -0.74). Conclusions: There were significant changes in the patterns of use of public hospitals between 1987 and 1995, possibly reflecting technological advances and changes in clinical practice, as well as policy to improve efficiency. Early readmission rates may be a useful proxy measure of potentially avoidable adverse outcomes.