A prospective cohort study of 6-month mortality in a community hospital experiencing a gradual reduction in critical care services

Objective: To analyse the effect of reduction of critical care sen ices on admissions, resource consumption and outcome. Design: Observation outcome study with analysis of patient data collected p rospectively during 1993, 1995 and 1997. Setting: High dependency and intensive care unit (HDU/ICU) of a community h ospital serving a population of 168,000. The number of beds decreased from 12 (1993), to 10 (1995) and to 8 (1997) with concomitant decrease in staff. Patients and participants: Three patient cohorts admitted to the HDU/ ICU d uring 1993, 1995 and 1997. Measurements and results: Admissions were classified into recovery room car e or critical care admissions and stratified according to workload (Levels I-IV), Illness severity scores of critical care admissions were recorded ac cording to the APACHE II system. Mortality data were acquired from a nation al database. The total number of admissions to the unit did not change over the years Length of step decreased significantly over the pears, Standardi sed mortality rates based on mortality within 30 days of discharge from the HDU/ICU were 1.17 (95% confidence interval 0.96-1.43) for critical care ad missions during 1993, 0.86 (0.70-1.06) for 1995 and 0.98 (0.79-1.22) for 19 97. Survival 180 days after discharge from the HDU/ICU did not differ signi ficantly over the years. Conclusions: The results suggest that an excess of resources were used in c ritical care services during 1993 and 1995. Reduction of HDU/ ICU beds by 3 0% from 7.1 to 4.8 beds/100,000 was not associated with increased 6-month m ortality of the patients admitted.