OBJECTIVE: To detect possible reasons for mortality of critical patients transferred from the ICU to the hospital wards and to analyze the possible attributable causes for such mortality. DESIGN: An observational study of prospectively collected data, analyzed retrospectively. POPULATION: Cohort analysis of 5328 with consecutive admissions to our ICU, whose evolution was followed up to hospital discharge or death. PERIOD: From January 2006 to December 2009. METHOD: An analysis was made of differential significance of epidemiological, clinical-care, death risk estimate, coincidence between ICU admissions reasons and causes of death after ICU discharge, as well as limitation of health care effort incidence. Inappropriate ICU discharge was considered to exist if the death occurred during the first 48 hours after ICU transfer, without limitation of care effort. RESULTS: A total of 907 patients died (SMR=0.9; 95% CI, 0.87-0.93), 202 of whom died after ICU discharge (3.8% of total sample and 22.3% of all deceased patients), ward length of stay being 12.4±17.9 days. No significant differences were found between deaths in the ICU or post-ICU deaths regarding infective complications appearing after admission to the ICU. Greater mortality was also not found in those re-admitted to the ICU after having been transferred to the ward. It was verified that the cause of death in the ward did not significantly coincide with the cause of admission to the ICU. DISCUSSION: Some mortality after ICU discharge is to be expected. Our data do not allow us to attribute this mortality rate to care deficiencies (inappropriate ICU discharges or deceased care in the wards). The reasons for this mortality have a varied and variable explanation. It mostly corresponds to an evolution of the patients differing from that expected when they were discharged from ICU.
OBJECTIVE: To detect possible reasons for mortality of critical patients transferred from the ICU to the hospital wards and to analyze the possible attributable causes for such mortality. DESIGN: An observational study of prospectively collected data, analyzed retrospectively. POPULATION: Cohort analysis of 5328 with consecutive admissions to our ICU, whose evolution was followed up to hospital discharge or death. PERIOD: From January 2006 to December 2009. METHOD: An analysis was made of differential significance of epidemiological, clinical-care, death risk estimate, coincidence between ICU admissions reasons and causes of death after ICU discharge, as well as limitation of health care effort incidence. Inappropriate ICU discharge was considered to exist if the death occurred during the first 48 hours after ICU transfer, without limitation of care effort. RESULTS: A total of 907 patients died (SMR=0.9; 95% CI, 0.87-0.93), 202 of whom died after ICU discharge (3.8% of total sample and 22.3% of all deceased patients), ward length of stay being 12.4±17.9 days. No significant differences were found between deaths in the ICU or post-ICU deaths regarding infective complications appearing after admission to the ICU. Greater mortality was also not found in those re-admitted to the ICU after having been transferred to the ward. It was verified that the cause of death in the ward did not significantly coincide with the cause of admission to the ICU. DISCUSSION: Some mortality after ICU discharge is to be expected. Our data do not allow us to attribute this mortality rate to care deficiencies (inappropriate ICU discharges or deceased care in the wards). The reasons for this mortality have a varied and variable explanation. It mostly corresponds to an evolution of the patients differing from that expected when they were discharged from ICU.