Improving rapid response system performance in a Chinese Joint Commission International Hospital.

Introduction

Serious adverse events (SAEs)[1] are an inseparable part of medical care. SAEs common to both inpatient and outpatient settings include cardiac arrest, death, and unplanned admission to the intensive care unit (ICU). Implementation of a rapid response system (RRS) in adult patient populations is associated with an overall reduction in cardiopulmonary arrests and in-hospital mortality.[2] Many vital dysfunctions are easily observable and typically include alterations in the respiratory rate, SpO2 levels, heart rate, blood pressure, and level of consciousness.[3-5]

RRSs have been widely introduced throughout hospital health systems to identify and treat patients with deteriorating conditions in general hospital units.[6] The RRS has changed the way staff working in acute hospital wards recognize and respond to patients at risk of clinical deterioration[7] before they become critically ill. The primary aim of the RRS is to decrease in-hospital mortality, principally through the prevention of cardiac arrest.[8] In an RRS, adequate efferent limb activation criteria and the corresponding actions of health care staff (monitoring of vital signs, early detection of patient deterioration, and medical emergency team (MET) activation) are key factors for reducing the incidence of SAEs.[8] Several countries have RRSs in place that are aligned to their medical system. Therefore, we sought to develop the optimal RRS plan for our health care system.

Our hospital, the Second Affiliated School Hospital Zhejiang University School of Medicine (SAHZU) is a comprehensive network with 2200 beds across two metropolitan teaching hospitals in the city of Hangzhou, China. We adopted a systems approach to improving the management and outcomes of patients with SAEs at our hospital, including multidisciplinary collaboration, operational planning, training, competency criteria for different MET members, staff requirements, modified early warning system (EWS) criteria, and related management functions. We developed a regional MET and RRS for all hospital staff in both inpatient and outpatient areas.

In the present study, we aimed to assess the effectiveness of the developed RRS, to improve its utilization. We sought to ensure the cooperation of involved health care staff, to reduce the time needed to obtain the best response for each patient.

Methods

Ethical approval

The SAHZU research and ethics committee exempted this study from the requirement for approval needed in studies involving human participants because this was a quality assurance project. All patients admitted to the hospital were considered participants in this assessment.

Setting

The SAHZU is a comprehensive, nonprofit, tertiary university health care provider in Hangzhou. Over 2000 clinical health professionals serve a catchment population of 8,000,000. Services include acute care services at two hospital sites. Annual outpatient visits exceed 3,000,000; there are over 100,000 hospital admissions annually, with over 7000 critical patients. The SAHZU is an enthusiastic adopter of the RRS, with establishment of METs at two of our hospitals in 2013.

Intervention areas

We covered all wards and other public areas, such as the outpatient areas, cafeteria, garden, consulting rooms, hallways, restrooms, and auxiliary units.

Patients in critical care units (emergency department [ED], ICU, neurological intensive care unit [NICU], surgical intensive care unit (SICU), and anesthesia department[AD]) were not included in the RRS.

Study design

In this section, we introduce the RRS structure and protocol and describe the interdepartmental/cross-institutional interface of the RRS. Finally, we present a brief analysis of RRS utilization and present process indicators.

RRS structure and protocol

The RRS comprises four key elements, often referred to as the limbs of the RRS. These include an afferent limb (identification of patient deterioration by the ward staff and triggering of a response), the efferent limb (MET), and the feedback and administrative components.[9]

Organizational structure of the RRS

The hospital vice president holds the overall responsibility for the technical, administrative, and operational aspects of the RRS. METs are designated by a committee comprising the president, medical director, nursing director, anesthesia director, emergency director, and ICU director. The committee is responsible for establishing related policies as well as the supervision and control mechanisms of RRS operation. The committee is also responsible for establishment of the RRS management information system, data analysis, and monitoring and evaluation functions.

RRS personnel structure and requirements

Implementation of an RRS requires additional resources for the efferent limb; the workload of the MET is substantially increased as they also respond to other medical emergencies.[9,10] The RRS includes the MET, broadcast system, and dedicated RRS staff. Based on the available infrastructure, catchment area population, and shortest approach to the hospital, we established four METs. These teams comprised personnel from the ED, ICU, NICU, SICU, and AD.

MET staff aptitude requirements are laid down by our hospital, as below:

Doctor: a) ICU and emergency physicians; b) registered cardiology, respiratory, or anesthesiology physicians who hold an intermediate license; resident doctors.

Nurse: N2-level nurses in the departments of internal, cardiac, or respiratory medicine or in the ICU, ED, or AD.

N2 nurse requirements: a) Minimum 3 years’ clinical experience; b) competent in critical care; c) score of > 85 in both practical skills and theory; d) ability to assist in four types of operations; e) minimum course requirements completed (26 credits in 1 year); clear, continuous quality improvement case reports and published journal articles; f) Basic Life Support (BLS) certification.

RRS training content and skill requirements

MET member training and skill requirements

Training methods include lectures on the treatment of SAEs. The training content for MET members includes first aid skills, critical disease monitoring equipment, and critical care. MET members are required to pass BLS and ACLS certification.

The essential competencies for physicians on the METs are: (1) skilled in providing airway support including oral or nasal endotracheal intubation, assisted ventilation, tracheal intubation skills, and use of a laryngeal mask; (2) skilled in use of a portable multifunction ventilator (ICU); (4) proficient in cardiopulmonary resuscitation (CPR), external cardiac pacing, temporary cardiac pacing, and defibrillation; (5) well-versed in drug usage, indications, and contraindications according to 2010 American Heart Association (AHA) CPR guidelines.[11]

Auxiliary staff training and skill requirements

Three levels of training courses were developed according to Joint Commission International (JCI) and AHA curriculum requirements. CPR skills are imparted to all hospital medical staff to inculcate a sense of emergency consciousness and promote their rescue capabilities.

(1) Basic CPR: administrative and logistics staff; (2) BLS: doctors, medical technicians and nurses; (3) ACLS: emergency unit and ICU; anesthesiology, cardiology, respiratory doctors who provide emergency, intensive care; N2 nurses in intensive care units (emergency ICU, ICU, NICU, cardiac care unit, cardiovascular monitoring, cardiothoracic surgery monitoring); and the Postanesthesia Care Unit.

Testing and evaluation

All medical personnel are required to pass theoretical and practical tests and their clinical skills are monitored on an ongoing basis. Annual performance reviews of MET members are compulsory; the performance of other medical staff is reviewed once every 2 years.

Operational functioning of the RRS

Early warning system (EWS)

According to the EWS (also known as the aggregate weighted scoring system or aggregate weighted track and trigger system), scores are assigned for each vital sign, from 0 (normal) to 3 (extreme derangement), based on deviation from the “normal” reference range.[12] This EWS determines the threshold levels of the trigger criteria, based on our clinical practice and widely accepted values reported in previous studies.[13,14]

RRS activation

METs are on call around the clock for both inpatient and outpatient areas. In the event of an SAE, the RRS can be activated by calling 665555. The broadcast system provides details of the SAE to the METs and activates the event. After completion of the intervention, the rescue nurse is required to complete the SAE response sheet.

The target time for arrival of the MET at the location of the call is “less than 5 minutes”. Each of the four established METs is responsible for its assigned catchment area: (1) ED: responsible for medical buildings 3, 8, 9, 10, 11 and the garden; (2) ICU: responsible for medical buildings 1, 2, 3, 12 and the canteen; (3) NICU and SICU: responsible for medical buildings 5, 6, 7.

MET trigger criteria

Based on their assessment of the patient, the MET decides on the next step. The trigger criteria are: (1) Airway: respiratory distress requiring airway support; (2) Breathing: respiratory rate < 9/minute or > 36/minute; SpO2 < 90% with oxygen supplementation; (3) Circulation: systolic blood pressure < 80 mmHg or ≥200 mmHg; Heart rate < 40/minute or ≥130/minute; (4) Neurological status: impaired consciousness; (5) discretion of MET: any other serious concerns regarding the patient.

Statistical methods

All statistical analyses were performed using IBM SPSS version 20 (IBM Corp., Armonk, NY, USA). Descriptive statistics were generated and analyzed. In addition, we analyzed data on the incidence of SAEs, sex distribution of patients, activation areas, activation status, reasons for activation, MET arrival time, calling periods, CPR, reasons for over-activation, and patient outcomes.

Results

Between May 2013 and December 2015, RRS activation occurred for 198 SAEs involving 208 patients: 109 (55.1%) male and 89 (44.9%) female patients; mean age ± standard deviation (SD: 54.2 ± 1.6 years. The RRS was activated by 192 (97.0%) medical professionals and 6 (3.0%) auxiliary staff. Overall, among 385 activations, the main cause was unconsciousness in 133 (34.5%). The two second most common reasons for RRS activation were airway obstruction and weak or absent carotid pulse (total, 98; 25.4%) (Table 1). The mean time elapsed from RRS activation and MET arrival at the calling site was 2.4 ± 0.1 minutes. The corresponding elapsed time in case of the ED team was 2.3 ± 1.5 minutes and that for the NICU, ICU, and SICU teams was 2.9 ± 1.0 minute, 3.1 ± 1.9 minutes, and 3.5 ± 1.9 minutes, respectively (data not shown).

Table 1.

Clinical reasons for RRS activation.

Reasons for activation (N=385)	n	%
Unconsciousness	133	34.5
Respiratory distress	34	8.8
Airway obstruction	49	12.7
Unconsciousness	31	8.1
Drop in blood pressure	48	12.5
Weak or absent carotid pulse	49	12.7
Other	41	10.6

RRS, rapid response system.

A total 123 (62.1%) calls were made during the period between 8:00 and 17:00 (Table 2). A total of 86 (43.4%) patients required CPR. A total of 17 calls (8.6%) were deemed to be over-activation for which the RRS was not applicable (Table 3). After the intervention, 82 (41.4%) patients survived, and 61 (30.8%) patients were transferred to ICU units for further treatment (Table 4).

Table 2.

RRS activation during work shifts.

Activation period	n	%	Activation time (s)
8:00–17:00	123	62.1	2.6 ± 1.8
17:01–1:00	35	17.7	2.8 ± 1.3
1:01–7:59	40	20.2	3.2 ± 1.7

RRS, rapid response system.

Table 3.

Applicability of RRS activation.

Reasons for activation	n	%
Valid activation	181	91.4
No additional treatment required	13	6.6
Patient left hospital on their own	3	1.5
RRS activation for patient with do-not-resuscitate status	1	0.5

RRS, rapid response system.

Table 4.

Outcomes of RRS intervention.

Patient outcomes	n	%
Survived	82	41.4
Died	15	7.6
Admitted to medical unit	61	30.8
Other	40	20.2

RRS, rapid response system.

Discussion

Regional RRS administration

The rationale for the establishment of a MET is that early recognition and timely response to imminent in-hospital complications will reduce the risk of morbidity and mortality.[15]Multidisciplinary and interdepartmental collaboration is critical to the success of an RRS, as is efficient administrative and logistics support for paging, broadcasting, and maintenance of equipment. The hospital information system is a key resource in planning an RRS.

Including staff from ICU-related units has been found to be an efficient use of human resources and time, to ensure that the MET arrives at the calling site within the target of 5 minutes, as rescue performance metrics for CPR are limited to 5 minutes.[16] Some studies have evaluated the performance of general ward staff with respect to activating an RRS when the triggering criteria are fulfilled; study findings have revealed a wide variation in the rate of delayed activation (21%–81%).[17-19] We should mention that not all patients need to be treated within 5 minutes, for example, a patient at risk of a suicide attempt. Greater attention is needed for patients who have experienced an SAE.

Training and simulation

Patients tend to display abnormal vital signs for hours or days preceding any serious emergency. The failure to activate an RRS could be owing to the current trigger criteria being observed too late in the progression of the patient 's clinical deterioration.[20] Therefore, in our study, we tended to call an RRS using a critical point criterion. In this way, we could provide treatment for patients as soon as possible.

Standardized aptitude and skills of MET members are key to the quality and efficiency of the RRS. On evaluating the initial experience of the RRS, some areas for improvement were identified and incorporated into the training program. We identified training needs among personnel from the departments of anesthesia (skills in endotracheal intubation) and pharmacy (refresher training in emergency drugs, dosages, indications, and contraindications) and we regularly addressed these in the BLS and ACLS training.

Scenario building and simulation exercises were particularly helpful in identifying potential bottlenecks and operational challenges in responding to calls from nonclinical areas like canteens. All simulation exercises were conducted using a medical patient simulator. This training method is widely used in allied health care and the aviation industry. Simulation exercises for commonly occurring scenarios are more cost-effective than establishing special training centers. Moreover, participation in these exercises helps improve teamwork and facilitates cooperation between multidisciplinary medical personnel.

EWS early recognition

At our hospital, all personnel can activate the RRS. According to surveys, 48% to 56% of general ward nurses activated the RRS. In general wards, routine vital sign measurement is undertaken manually and intermittently, with or without automated equipment, at intervals that are based on the patient's severity of illness.[21,22]

Of the 198 SAEs in our study, 6 (3.0%) were activated by auxiliary staff. Activation was based on the EWS criteria, e.g., unconsciousness or another serious situation. We found that nonmedical personnel can help to improve the efficiency and speed of the RRS response. In patients with vital signs that fall within the EWS criteria, the staff activate the RRS earlier. We found that more than 40% of patients required CPR and these patients received timely treatment.

The SAHZU is a general hospital for adult patients; we only included SAEs that occurred among the intended target population of the EWS, i.e., adult patients aged ≥18 years who were admitted to the hospital. Therefore, the EWS criteria are not applicable to children.

The EWS is a recent development as part of health care reform and improved quality of care. Although robust evaluations of EWSs are lacking, use of an EWS helps in the efficient early recognition and treatment of adverse events. Even in resource-constrained settings, an EWS can help to supplement medical staff and improve overall efficiency. Application of an EWS is an effective tool for the administration SAEs.

Retrospective analysis

Implementation of an RRS may involve a multitude of social, political, and hierarchical barriers, which must be considered before and during the implementation phase.[9,23]Regular engagement of all stakeholders to share information, feedback, and to participate in decision-making is important.[24] From the outset, it is important to emphasize a management philosophy with two key elements: data-driven continuous improvement of processes focused on the needs of the end user, and respect for the people delivering the service. Previous studies have shown that every RRS must endeavor to adapt to the clinical environment and needs. Therefore, continuous monitoring and formative evaluation is an important aspect of managing the RRS. Use of feedback to institute corrective measures can help in resolving problems.

Owing to our relatively short experience with the RRS, we focused largely on qualitative aspects that will help to avert MET failures and to fine tune the system.[25-28] The direction of future work is to continue to track the data, to build an evidence base using data of a large sample. The dates of SAEs are collected for analysis, including the analysis of reasons for activating the RRS.

Current RRS issues and questions

With an increased number of RRS activations, shortcomings and challenges were encountered during the development and implementation of the RRS. The most important is the definition of MET call criteria because the 8.6% RRS activation was categorized as an over-response. To some extent, very sensitive activation criteria may over-trigger the MET, causing system fatigue with no tangible benefit.[2] In particular, patients who are assigned do-not-resuscitate (DNR) status or are potentially suicidal (e.g., threatening to jump from a building) do not constitute a valid indication for activation of the RRS. Therefore, it is necessary to modify specific thresholds of the trigger criteria for SAEs.

Despite these issues regarding RRS activation, recent systematic reviews and meta-analyses of RRSs have concluded that increasing evidence indicates an RRS can be effective in reducing admissions to the ICU and reducing hospital mortality.[2]

Conclusion

We revealed the progressive improvement in SAEs after the introduction of regional RRS management, with relatively more frequent MET calls.

At present, few domestic hospitals have established an RRS. There is a need to improve standards and operating protocols. Our RRS draws on national and international experience, to ensure the qualification of MET members and optimal use of human resources and related equipment. The target response time for arrival of the MET at the calling location is < 5 minutes. Future steps include a further focus on execution, data-based reviews, feedback, and continuous quality improvement.

Last, the findings from this study are limited in their generalizability to all populations, for example patients with end-stage disease or suicidal patients; therefore, further research into the RRS is needed. Regular feedback, data analysis, and staff training can potentially decrease over-activation of the RRS, as can using the EWS in conjunction with the RRS.