KEY POINTS
Patients admitted to modern hospitals may develop serious adverse events in up to 20% of admissions. In addition, hospitalized patients can deteriorate unexpectedly due to the development of a new problem.
In a high percentage of cases, deterioration is gradual in onset and is associated with the development of derangement in the patient’s vital signs.
Many hospitals have introduced Rapid Response Teams (RRTs) to review deteriorating patients when they develop derangements in vital signs that fulfill predefined criteria.
Evidence for the effectiveness of RRTs is conflicting, and the optimal team composition and thresholds for activation remain undetermined.
INTRODUCTION
Modern hospitals treat patients with increasingly complex medical conditions. Despite advances in medical technology and the advent of new medicines and interventions, many patients admitted to hospitals suffer adverse events. The most studied of these events are unplanned admissions to the intensive care unit (ICU), unexpected hospital deaths, and cardiac arrests. Other studies have shown that such events are preceded by the development of new problems or derangements in vital signs for several hours, and that the response to these problems by ward staff may be suboptimal.
Traditional code teams are activated when a patient suffers a cardiac arrest manifesting as a loss of circulation or respiration. In this chapter, we discuss the concept of the Rapid Response Team (RRT), which is activated when a patient develops less severe and earlier signs of instability. We also describe the Rapid Response System (RRS), which is the entire system used to support the team.
SERIOUS ADVERSE EVENTS ARE COMMON IN HOSPITALIZED PATIENTS
Modern hospitals in developed countries care for patients of increasing age, acuity, and complexity.1 Studies conducted in North America, Australia, New Zealand, and the United Kingdom suggest that such patients suffer adverse events in up to 20% of cases depending on the definition used and population assessed (Table 12-1).
Summary of Studies Reporting Adverse Events in Hospitalized Patients
| Reference(s) and Year of Inception | Study Population | Definition of Adverse Events | Major Findings |
|---|---|---|---|
| Schimmel2 1960-1961 | 1014 patients admitted over 8 months to a university-affiliated hospital | Every noxious response to medical care occurring among patients…resulting from acceptable diagnostic and therapeutic measures deliberately instituted at the hospital |
|
| Leape et al,3 1984 | 30,195 patients in 51 hospitals in New York | Unintended injury that was caused by medical management that resulted in measurable disability |
|
| Thomas et al,4 1992 | 14,700 patients in 28 hospitals in Utah and Colorado | Injury caused by medical management rather than by the disease process and resulted in prolonged LOS or disability at discharge |
|
| Andrews et al,5 1989-1990 | 1047 patients from 3 units of a university teaching hospital in the United States | Situations in which an inappropriate decision was made when, at the time, an appropriate alternative could have been chosen |
|
| Wilson et al,6 1992 | 14,179 patients in 28 hospitals in New South Wales and South Australia | Unintended injury or complication that resulted in disability, death, or prolonged hospital stay and was caused by the health care management rather than by the underlying disease process |
|
| Davis et al,7 1998 | 6579 patients in 13 New Zealand hospitals with more than 100 beds | Same as the study by Wilson et al |
|
| Vincent et al,8 1999-2000 | 1014 patients in 2 London hospitals | Unintended injury that was caused by medical management rather than the disease process |
|
| Baker et al,9 2000 | 3745 patients in 20 Canadian hospitals | Same as the study by Wilson et al |
|
| Bellomo et al,10 1998-1999 | 1125 patients undergoing major surgery in a university teaching hospital | Specific criteria for 11 predefined adverse events |
|
In 1964, Schimmel2 reported on the incidence of adverse events in a cohort of 1014 patients admitted over an 8-month period to a university teaching hospital in the United States. Participating house officers reported “every noxious response to medical care occurring among their patients.” The study found that 20% suffered iatrogenic injury, 6.7% of which were fatal. Subsequently, two large studies, one in New York3 and the other in Utah and Colorado,4 estimated a much lower incidence of adverse events of 2.9% to 3.7%. However, both of these studies defined adverse events from a medicolegal perspective in an attempt to estimate the incidence of medical negligence. In a different study assessing a broader definition of medical error, Andrews and coworkers5 found a 17.7% incidence of adverse events.
Four subsequent studies defined adverse events as “unintended injury or complication resulting from medical management rather than the underlying disease process.” These studies were conducted in multiple countries worldwide including Australia,6 New Zealand,7 England,8 and Canada9 and enrolled more than 25,500 hospitalized patients (Table 12-1). These studies reported an incidence of adverse events ranging from 7.5%9 to 16.6%6 and suggested that between 36.9%9 and 51%6 were preventable.
The above studies assess adverse events from the perspective of iatrogenesis and negligence. Patients may suffer an adverse event that does not fall into these categories. Bellomo and coworkers10 conducted a 4-month study of serious adverse events (SAEs) in 1125 patients undergoing major surgery (defined as surgery requiring admission for more than 48 hours) at the Austin hospital. A dedicated research coordinator assessed patient records for the presence of 11 predefined SAEs: acute myocardial infarction, pulmonary embolism, acute pulmonary edema, unscheduled tracheostomy, respiratory failure, cardiac arrest, cerebrovascular accident, severe sepsis, acute renal failure, emergency ICU admission, and death. The study reported that 16.9% of patients suffered postoperative SAEs, and that 7.1% of patients died. Further, in those older than 75 years who underwent unscheduled surgery, the mortality was 20%.10
SERIOUS ADVERSE EVENTS ARE PRECEDED BY SIGNS OF INSTABILITY
A number of studies have assessed the clinical course and management of patients in the hours leading up to SAEs and cardiac arrests (Table 12-2). Some of these studies11,12 have used an expert panel to determine whether the cardiac arrest or iatrogenic arrest was avoidable and whether it was associated with medical error. Such studies suggest that approximately 60% of cardiac arrests were avoidable. Similarly, an assessment of 100 consecutive emergency ICU admissions suggested that 54% of patients received suboptimal care, and that suboptimal care was associated with increased mortality.13
The major limitation of these studies is their retrospective design and lack of objective criteria for assigning preventability. Consistent with this notion, Hayward and Hofer14 reported an analysis of 111 deaths in 7 Veteran hospitals in the United States, which suggested that previous studies had overestimated the incidence of death due to medical error. In addition, the authors demonstrated considerable interobserver variability in estimation of preventability, suggesting that “preventability was in the eye of the reviewer.”14
Other investigators have retrospectively assessed patients’ case histories for objective signs of physiological or biochemical instability in the hours leading up to the cardiac arrest or unplanned ICU admission. At least five studies15-19 have demonstrated that patients develop new complaints or deterioration in commonly measured vital signs or laboratory investigations in up to 84% of cases in the 24 hours prior to the event (Table 12-2). Such perturbations are not only objective, but they are routinely measured and assessed by treating medical and nursing staff (Figs. 12-1 to 12-3). However, the limitation of these studies is that they fail to demonstrate whether intervention during the course of deterioration would have altered the patient outcome. In addition, they do not assess a control group to document the frequency of such perturbations in patients not suffering cardiac arrest and unplanned ICU admission.
Summary of Studies Reporting Antecedents to Serious Adverse Events and In-Hospital Cardiopulmonary Arrests
| Reference and Year of Inception | Study Population and Setting | Method of Assessment | Major Findings |
|---|---|---|---|
| Bedell et al,11 1981 |
|
|
|
| Schein et al,15 Jul-Oct 1987 |
|
|
|
| McQuillan et al,13 Winter 1992 | 100 consecutive emergency admissions to adult ICU in England (Portsmouth and Southampton) | Opinions of two external assessors on quality of care before admission ↓ especially recognition, investigation, monitoring and management of abnormalities of airway, breathing, and circulation. | Assessors agreed that 20% received optimal care and 54% suboptimal care. ICU mortality of these patients was 25% and 48%, respectively. Suboptimal care resulted from lack of organization and knowledge, failure to appreciate urgency, failure to seek advice |
| Buist et al,17 Jan-Dec 1997 |
| Retrospective assessment of medical records for abnormalities in vital signs and blood tests |
|
| Hodgetts et al,12 1999 |
|
|
|
| Hodgetts et al,16 1999 |
|
| Risk factors for arrest included abnormalities in respiratory rate, breathing, pulse rate, systolic blood pressure, or temperature, as well as chest pain, hypoxia, or concern by the doctor or nurse |
| Buist et al,22 May-Dec 1999 | 6303 patients admitted over 7 months to 320-bed hospital in Dandenong Australia | Prospective assessment of patients identified by predefined abnormal observations | 8.9% of admissions fulfilled criteria. Oxygen desaturation and hypotension comprised 68% of all events. The presence of any abnormality was associated with a 6.8-fold increased risk of mortality |
