Effects of Organizational Change in the Medical Intensive Care Unit of a Teaching Hospital

OBJECTIVE: To compare the effects of change from an open to a closed intensive care unit (ICU) format on clinical outcomes, resource utilization, teaching, and perceptions regarding quality of care. DESIGN: Prospective cohort study; prospective economic evaluation. SETTING: Medical ICU at a university-based tertiary care center. For the open ICU, primary admitting physicians direct care of patients with input from critical care specialists via consultation. For the closed ICU, critical care specialists direct patient care. PATIENTS: Consecutive samples of 124 patients admitted under an open ICU format and 121 patients admitted after changing to a closed ICU format. Readmissions were excluded. MAIN OUTCOME MEASURES: Comparison of hospital mortality with mortality predicted by the Acute Physiology and Chronic Health Evaluation II (APACHE II) system; duration of mechanical ventilation; length of stay; patient charges for radiology, laboratory, and pharmacy departments; vascular catheter use; number of interruptions of formal teaching rounds; and perceptions of patients, families, physicians, and nurses regarding quality of care and ICU function. RESULTS: Mean +/- SD APACHE II scores were 15.4 +/- 8.3 in the open ICU and 20.6 +/- 8.6 in the closed ICU (P=.001). In the closed ICU, the ratio of actual mortality (31.4 percent) to predicted mortality (40.1 percent) was 0.78. In the open ICU, the ratio of actual mortality (22.6 percent) to predicted mortality (25.2 percent) was 0.90. Mean length of stay for survivors in the open ICU was 3.9 days, and mean length of stay for survivors in the closed ICU was 3.7 days (P=.79). There were no significant differences between periods in patient charges for radiology, laboratory, or pharmacy resources. Nurses were more likely to say that they were very confident in the clinical judgment of the physician primarily responsible for patient care in the closed ICU compared with the open ICU (41 percent vs 7 percent; P<.Ol), and nurses were the group most supportive of changing to a closed ICU format before and after the study. CONCLUSIONS: Based on comparison of actual to predicted mortality, changing from an open to a closed ICU format improved clinical outcome. Although patients in the closed ICU had greater severity of illness, resource utilization did not increase.