ICU Bouncebacks: Rates, Reasons, and Risk Factors
By Karen L. Johnson, PhD, RN
Which of the following patients is most likely to be readmitted to the ICU from a general ward?
- Patient A is 56 years old and 5 days postesophagogastrectomy, with a hematocrit of 34, a 5-kg weight gain since admission, and now-positive blood cultures;
- Patient B is 72 years old and 7 days postradical neck dissection, with a heart rate of 120 beats/min, BP 110/60 mm Hg, respirations 32/min, and temperature 38°C;
- Patient C is 65 years old and was transferred 24 hours ago from the ICU with a recent upper gastrointestinal bleed and hepatic encephalopathy;
- Patient D is 42 years old and several days post-motor vehicle crash and surgical repair of a grade IV liver laceration, with a new-onset right lower lobe pneumonia.
If you guessed that Patient A was most likely to return to the ICU from a general ward you were correct. In fact, if you guessed Patient B, C, or D you were also correct. All 4 of these patients are at high risk to be readmitted, or to "bounce back" to the ICU. What is the incidence of ICU bounce-backs? What are the major reasons for ICU readmission? What are the risk factors? This special feature will address these questions. For the purpose of this essay, ICU bounce-backs are defined as patients who are discharged from and readmitted to the ICU during the same hospitalization.
Comparing ICU readmission rates is complicated by many factors including national health care policy, institutional policies, varying patient populations, ratios of ICU beds to hospital beds, and the availability of intermediate care beds.1 The incidence of ICU readmissions is difficult to state because studies have used different denominators to calculate it. The reported incidence of readmission to the ICU ranges from 4.6% to 16% of the total number of ICU admissions,2,4-7 12% of total ICU discharges,3 and 4.3-9.6% of total live ICU discharges.1,8 Since only patients who are discharged alive from the ICU are at risk for readmission, it would make sense to use the total number of live ICU discharges as the denominator. Rosenberg and Watts pooled data from 8 large, multi-institutional studies of ICU outcomes and reported a mean ICU readmission rate of 7% (range, 4-10%).9
Patients who are readmitted to the ICU within 24 hours after ICU discharge are of particular interest because of the possibility of premature discharge. In studies that reported these data, readmission to the ICU within 24 hours occurred in 17-30% of readmitted patients.1,7,8
Hospital mortality appears to be higher for patients readmitted to the ICU, with reported mortality ranging from 21%2 to 58%.3 Even when adjusting for severity of illness, one study reported that readmitted patients were 11 times more likely to die in the hospital, and their hospital stays were almost twice as long, as nonreadmitted patients.8 Rosenberg and Watts9 offer 3 plausible explanations for the higher death rates among patients readmitted to the ICU: 1) readmitted patients may respond less adequately to ICU treatment, resulting in subsequent readmission or death; 2) patients readmitted to an ICU may reflect potentially ineffective care; and 3) increased mortality rates among readmitted patients may reflect poor quality care such as premature ICU discharge.
Major Reasons for Readmission
Studies that have examined ICU readmission categorize the causes of readmission as either a recurrence of the initial disease (if the readmission diagnosis involved the same organ system as the initial admission) or a new complication (if the readmission diagnosis involved a different organ system from the initial admission). Studies that have examined reasons for readmissions have reported that 30-65% of patients were readmitted to the ICU for a recurrence of the initial disease.1,3,4,6,8 The initial ICU admission diagnoses most frequently associated with readmission included upper GI bleeding, respiratory failure, and neurologic impairment. Patients with the diagnoses of postoperative elective abdominal aortic aneurysm, drug overdose, and chest pain to rule out myocardial infarction have been reported to have very low risk of ICU readmission.1,5
Studies that have examined the incidence of patients who return to the ICU for a new complication report that 30-39% of patients return as a result of a new complication and the leading cause of readmission in these studies was consistently a pulmonary disorder (pulmonary edema, respiratory arrest, aspiration, or bacterial pneumonia).1,3,4 Nosocomial pneumonias that developed subsequent to ICU discharge, poor ventilatory reserves, and ventilatory failure from inability to clear pulmonary secretions were frequently reported reasons for these pulmonary disorders.9
In the 8 studies conducted between 1983-2001 that were reviewed for this essay,1-8 it is striking to note that pulmonary problems—either as a recurrence of the initial disease or as a new complication—account for an overwhelming majority of the causes for ICU readmissions.
Risk Factors Associated with Readmission
A few studies have identified warning signs, or variables present upon initial ICU discharge, that statistically predicted ICU readmission.4-6,8 These variables included increased heart rate,5,6 increased respiratory rate,5,6 decreased hematocrit,5,6 positive fluid balance,6 positive blood cultures,6 and discharge Acute Physiology Score greater than 408. In a retrospective study of ICU readmissions, Snow and colleagues reported that while 78% of the discharges were deemed appropriate, 62% of the patients manifested several retrospectively selected warning signs that might have altered the clinicians to re-evaluate the ICU discharge plans.4
ICU Readmission Rates as a Quality Indicator?
In this essay, I have tried to briefly summarize the evidence to date on what we know about the incidence, mortality, reasons for, and risk factors associated with ICU readmission. What we do not know, and can’t seem to agree on, is whether we should even pay attention to ICU readmission rates. The Society of Critical Care Medicine (SCCM) supports ICU readmission as a quality indicator. SCCM’s Quality Indicators Committee ranked ICU readmission within 48 hours as the top indicator for judging ICU quality.10 SCCM’s Guidelines for ICU Admission, Discharge, and Triage11 state "readmission to the ICU for a similar problem should be closely monitored as it may directly relate to quality of the discharge process."
There are others who debate that patient-related factors cannot be separated from process-of-care issues.8 Readmission to the ICU may reflect a patient’s specific disease process and the inherent physiologic instability of a severely ill patient.9 Chen and colleagues showed that different patient groups had different likelihoods of being readmitted.1 If ICU readmission rates reflect the quality of process-of-care issues, then I question what process of care are we evaluating: care in the ICU? The ICU discharge decision-making process? Post-ICU care? Cooper and colleagues argued hospitals that are more clinically aggressive may have a lower threshold for readmission to the ICU and thus have higher ICU readmission rates.7 On the other hand, hospitals with a very low ICU readmission rate may be an indicator of patients who have inappropriately long ICU stays.9
While the debate continues, it seems intuitive that readmissions to the ICU are not a good thing. It seems as though we should try to keep bounce-backs to a minimum. But what is that number? Currently, based on the evidence we have, ICU readmission rate may reflect some crude measurement of quality9 and we should, as critical care clinicians, review these cases one by one, because it is at least grounded in good patient care.10
1. Chen LM, et al. Patients readmitted to the intensive care unit during the same hospitalization. Crit Care Med. 1998;26:1834-1841.
2. Baigleman W, et al. Patient readmission to critical care units during the same hospitalization at a community teaching hospital. Intensive Care Med. 1983;9:253-256.
3. Franklin C, Jackson D. Discharge decision making in a medical ICU: Characteristics of unexpected readmissions. Crit Care Med. 1983;11:61-66.
4. Snow N, et al. Readmission of patients to the surgical intensive care unit: Patient profiles and possibilities for prevention. Crit Care Med. 1985;13:961-964.
5. Rubins HB, Moskowitz MA. Discharge decision making in a medical intensive care unit: Identifying patients at high risk of unexpected death or unit readmission. Am J Med. 1988;84:863-869.
6. Durbin CG, Kopel RF. A case control study of patients readmitted to the intensive care unit. Crit Care Med. 1993;21:1547-1553.
7. Cooper GS, et al. Are readmissions to the intensive care unit a useful measure of hospital performance? Med Care. 1999;37:399-408.
8. Rosenberg AL. Who bounces back? Physiologic and other predictors of intensive care unit readmission. Crit Care Med. 2001;29:511-518.
9. Rosenberg AL, Watts C. Patients readmitted to ICUs: A systematic review of risk factors and outcomes. Chest. 2000;118:492-502.
10. Angus DC. Grappling with intensive care unit quality—Does the readmission rate tell us anything? Crit Care Med. 1998;26:1779-1780.
11. Society of Critical Care Medicine. Guidelines for ICU Admission, discharge, and triage. Crit Care Med. 1999;27:633-638.