Abstract & Commentary

Factors that Increase Risk for ICU Readmission: Implications for ICU Discharge Practices

By Linda L. Chlan, RN, PhD, School of Nursing, University of Minnesota, is Associate Editor for Critical Care Alert.

Dr. Chlan reports that she receives grant/research support from the National Institutes of Health.

Synopsis: A significant risk factor for ICU readmission is the illness severity (acute physiology score) associated with persistent physiological abnormalities at ICU discharge, regardless of initial ICU admission illness severity.

Source: Kramer A, et al. Intensive care unit readmissions in U.S. hospitals: Patient characteristics, risk factors, and outcomes. Crit Care Med 2012;40:3-10.

The purpose of this study was to use data from the multi-institutional APACHE IV database (Cerner Corporation) to compare characteristics of and outcomes for patients who were readmitted to the ICU and those who were not readmitted after ICU discharge; to identify risk factors for ICU readmission; and to examine case-mix adjusted outcomes for patients with and without ICU readmission. The APACHE IV database accessed for this study represented information on consecutive ICU admissions over a period of 8 years (2001-2009). These data included in the sample consisted of 97 units and 35 hospitals from across the United States, which represented a cross-section of facilities by bed size, teaching and non-teaching hospitals, and geographic regions (except for the Northeast region of the United States, which was not represented). Demographic, clinical, and physiological data were collected on each patient on day 1 of ICU admission and again on the day of discharge associated with the initial ICU admission. Outcomes included mortality, length of ICU stay, and length of hospital stay; ICU stay and length of hospital stay were truncated at 30 days and 50 days, respectively, to limit extreme outliers.

The researchers used various statistical procedures to compare patients with ICU readmission and those without ICU readmission. The researchers focused their analysis at the patient level, meaning that they did not make comparisons among unit types or adjust for hospital or ICU characteristics. Patients were not included in the study if they died while in the ICU or were discharged from the ICU and were not candidates for readmission (e.g., discharged to another facility or home).

There were 229,961 patients who met the researchers' predefined inclusion criteria. Overall, 6.1% of patients from the hospitals included in the APACHE IV database were readmitted to the ICU. Of those patients who had an ICU readmission, 88.4% had one readmission whereas 11.6% had multiple readmissions. The median time between ICU discharge and readmission was 3.14 days. There were no differences among diagnoses (non-surgical indications). Patients who were readmitted were older, had more comorbidities, were receiving dialysis, and had a higher illness severity (Acute Physiology Score, APS) at discharge from the ICU. Not surprisingly, those patients readmitted to ICUs had longer ICU and hospital stays as well as higher mortality. Contrary to findings from other studies, there was no difference in mortality in those patients discharged from the ICU during the night or on the weekend. The researchers emphasized the importance of considering patient characteristics, particularly illness severity, when assessing ICU readmission rates.


The findings from this large, nationally representative sample of ICU patients provide a comprehensive assessment of patient characteristics and risk factors for ICU readmission, including patient outcomes. Many of the findings from this study are consistent with those risk factors for ICU readmission previously reported in the literature, most prominently older age. A major finding from this study is that ICU readmission is associated, not surprisingly, with severe illness, specifically as measured by APS at ICU discharge and not at ICU admission. Further, readmission to the ICU is associated with an increase in mortality rate, longer ICU stay, and longer periods of hospitalization.

These findings have numerous implications for ICU clinicians in this day and age of cost-containment. Given the immense pressure in some hospitals to discharge patients "quicker and sicker," these findings may give pause to this practice. Given that the major risk factor for ICU readmission was a continued high level of illness severity at discharge, in the end it may be a cost-savings or cost-neutral practice to keep a very ill patient in the ICU for a day or two longer, especially given that the median time to readmission was around 3 days! Given that those patients included in this sample with ICU readmission had much longer ICU and hospital stays, a careful examination of ICU discharge practices may be warranted. This paper calls attention to the importance of examining for patient differences and illness severity when contemplating discharge.

There also are implications of these findings for care processes, such as for clinician interactions with patients and their family members as to expectations after ICU discharge. For those patients who are the most ill and have higher APS scores, there may be a rocky recovery that may warrant readmission to the ICU. Family members may want to be prepared for this possibility.