Article Text
PDF
Abstract
Background Rapid response teams (RRTs) help in the early recognition of deteriorating patients in hospital wards and provide the needed management at the bedside by a qualified team. RRT implementation is still questionable because there is insufficient evidence regarding its effects. To date, according to our knowledge, no published studies have addressed the effectiveness of RRT implementation on inpatient care outcomes in Egypt.
Objective We aimed to assess the impact of an RRT on the rates of inpatient mortality, cardiopulmonary arrest calls and unplanned intensive care unit (ICU) admission in an Egyptian tertiary hospital.
Methods An interventional study was conducted at a university hospital. Data was evaluated for 24 months before the intervention (January 2018 till December 2019, which included 4242 admissions). The intervention was implemented for 12 months (January 2021 till December 2021), ending with postintervention evaluation of 2338 admissions.
Results RRT implementation was associated with a significant reduction in inpatient mortality rate from 88.93 to 46.44 deaths per 1000 discharges (relative risk reduction (RRR)=0.48; 95% CI, 0.36 to 0.58). Inpatient cardiopulmonary arrest rate decreased from 7.41 to 1.77 calls per 1000 discharges (RRR, 0.76; 95% CI, 0.32 to 0.92), while unplanned ICU admissions decreased from 5.98 to 4.87 per 1000 discharges (RRR, 0.19; 95% CI, −0.65 to 0.60).
Conclusions RRT implementation was associated with a significantly reduced hospital inpatient mortality rate, cardiopulmonary arrest call rate as well as reduced unplanned ICU admission rate. Our results reveal that RRT can contribute to improving the quality of care in similar settings in developing countries.
- Hospital Mortality
- Cardiopulmonary Resuscitation
- Patient safety
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
The current evidence on rapid response teams (RRTs) effects is contradictory, their effectiveness is still up for debate.
To date, according to our knowledge, no published studies have addressed the effectiveness of RRT implementation on rates of mortality, cardiopulmonary arrests and intensive care unit (ICU)-unplanned admission in Egypt.
WHAT THIS STUDY ADDS
This quasiexperimental pre–post-interventional study helps in the evaluation of the effectiveness of RRT implementation on the rates of mortality, cardiopulmonary arrest and unplanned ICU admission, additionally, it aims to provide evidence for the generalisability of the results.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
RRTs play a crucial role in early detection of deteriorating patients within hospital wards, ensuring patient safety and enhancing the quality of care. This function of recognising and promptly responding to early signs of patient deterioration will be monitored using key performance indicators (KPIs). Metrics such as the rate of cardiopulmonary arrests outside critical care areas will be tracked.
By monitoring these KPIs, healthcare facilities can assess the effectiveness of their interventions, including the processes for recognising and responding to early signs of patient deterioration. This allows them to identify areas for improvement in their clinical practices and patient safety protocols.
Background
In December 2004, the Institute for Healthcare Improvement (IHI) launched a challenge to the medical community with its ‘Saving 100 000 Lives’ campaign, to promote a culture of safety and ensure the best healthcare outcomes for patients. Rapid Response Teams (RRTs) were one of the six recommendations of the campaign. The concept is simple: any healthcare worker can call the RRT to quickly assess a patient and intervene when lifesaving care may be required, bypassing normal hospital procedures.1 In December 2018, the General Authority for Healthcare Accreditation and Regulation (GAHAR), as part of Egypt Universal Health Insurance System, made a similar challenge where the recognition of and response to a patient’s clinical deterioration became one of the national safety requirements (NSRs) mandated for the accreditation of hospitals in Egypt.2 This new NSR underscores the importance the Egyptian government placed on patient safety and proper clinical monitoring as it worked to expand healthcare coverage throughout the country. An estimated 60%–85% of in-hospital cardiac arrests are preceded by earlier clinical signs, which indicated an increased risk of deterioration. However, these clinical signs are usually missed or ignored by the healthcare team. Based on these reports, researchers speculated that deploying an early alert and an RRT could be able to avert cardiac arrests and improve patient outcomes.3 RRTs play a crucial role in preventing cardiac arrests rather than responding to them. They also work to prevent unplanned admissions to the critical care unit and unexpected/preventable fatalities.4 Studies showed that RRT deployment was linked to a reduction in mortality, with an estimated one-and-half lives saved per week.5 Furthermore, other studies reported that in-hospital cardiac arrest significantly decreased.6 The outcomes’ results including overall mortality or non-intensive care unit (non-ICU) mortality were not consistent among the different studies, which is a gap of evidence for such intervention.7 8
To date, according to our knowledge, no Egyptian studies have described the impact of the implementation of RRT intervention in Egyptian hospitals on healthcare outcomes. The current study aimed to evaluate the impact of the implementation of a rapid response system (RRS) in a tertiary hospital in Egypt on mortality, arrest calls and unexpected ICU admission.
Methods
Design
A quasiexperimental pre–post-interventional study was conducted to fulfil the study objectives. Historical pre-implementation data were abstracted for 24 months (January 2018 to December 2020), then compared with post implementation data collected for 12 months (January 2021 to December 2021).
Setting
The study was conducted at Cardiothoracic Hospital, Ain Shams University; an Egyptian University tertiary hospital that has a capacity of 145 beds, including 56 ICU beds. The scope of the hospital includes cardiac as well cardio surgical cases of adults and of paediatric age groups. The average annual admission rate is around 6000 patients per year.
The study included all patients who met the inclusion criteria and were admitted to the hospital throughout the study period.
Inclusion criteria for study group: patients who are admitted to the Cardiothoracic hospital inpatient department.
Exclusion criteria for study group: outpatient department and cardiac catheterisation laboratory patients with short stay. This study team developed forms specifically for the study and conducted training rounds that first targeted the inpatient department as per the hospital policy.
Implementation of RRT
RRT implementation consisted of the following:
Development of RRT policy: a related hospital policy was developed by the different stakeholders identifying the criteria for calling the RRT, RRT composition, time to respond, mechanism to call the team and required documentation.
Criteria for calling RRT: a scoring system was adopted using the Modified Early Warning Score (MEWS) for adults and Pediatric Early Warning Score (PEWS) for paediatric age group.9 10 The scoring system depends on recording some vital signs such as heart rate, blood pressure, respiration rate and consciousness level, together with the added criterion of any staff concern or worry regarding the patient’s status.
Components of RRS:
The afferent arm (activators).
The efferent arm (responders) was led by an ICU physician and consisted of the primary physician, the acting chief nurse at the time of call and the primary nurse. A Situation, Background, Assessment & Recommendation documentation form was designated for handover between the afferent and efferent arms.
The quality improvement arm (hospital quality team) monitored the newly developed process.
The administrative arm consisted of the RRT committee overseeing the intervention’s function and reporting to the hospital’s management.
During the 12- month post-RRT period (January 2021 to December 2021), all patients admitted to the inpatient departments (2338 patients) were subjected to our study where RRT tools; the MEWS or the PEWS was implemented as part of their care plan according to the patient’s age.
The required interventions for each case according to the MEWS/PEWS included one of the following:
Informing the most responsible physician and/or nurse supervisor.
Assessing the patient more frequently every 1 or 2 hours according to MEWS/PEWS scoring guideline.
If the score exceeds certain point, RRT is called.
Training and monitoring:
The first round of RRT training programme started in August 2020 targeting medical and nursing staff working in the inpatient departments. The training consisted of a theoretical part done using presentations and handouts and a practical part using hands-on training on the RRT forms. A second round of training was conducted between December 2020 and March 2021. A third round started in June 2021 till September 2021, to reinforce the implementation of RRT service and improve the documentation through attending lectures and on-job training. In 2020, only training took place, while the actual implementation of the RRT commenced in January 2021.
Evaluation
Study outcomes included:
Inpatient mortality rate: calculated as the number of inpatient deaths per month divided by the number of inpatients discharges during the same month then multiplied by 1000. Inpatients who died during the study were counted, regardless of whether they received cardiopulmonary resuscitation (CPR) or not. Deaths occurring outside the hospital or after transfer to another facility were not included.
Inpatient cardiopulmonary arrest rate: calculated as the number of inpatient code blue calls occurring outside the ICU and the cardiac catheterisation laboratory per month divided by the number of inpatient discharges during the same month then multiplied by 1000.
ICU-unplanned admission rate: calculated as the number of unplanned ICU admissions per month divided by the number of inpatient discharges during the same month then multiplied by 1000.
Statistical analysis
Data analysis was done using SPSS V.24. Categorical variables were described as frequencies and percentages. Skewed quantitative variables were described using median and IQRs. Categorical and quantitative variables were compared using the χ2 and Mann-Whitney U, respectively. Statistical significance was considered at p≤0.05.
We used the ‘tsmodel’ package in R V.4.2.0 for time series analysis as described by Bernal et al, 2017 to evaluate the effect of the implementation of RRT intervention on the mortality/1000 discharges, number of code blue calls/1000 discharges and unplanned ICU admissions/1000 discharges.11 12 A Poisson distribution with a log link function was specified.
We used Minitab V.20 to produce U-controlled charts. The following tests were performed: 1 point>3 SD from the centre line 6 points in a row all increasing or decreasing, 9 points in a row on the same side of the centre line, 14 points in a row alternating up and down. No violations were detected. Absolute risk reduction and relative risk for each of the study indicators were calculated.
Results
Description of admitted patients before and after RRT implementation
As shown in (table 1), the characteristics of the patients’ gender slightly differ between the pre and post intervention. However, the large sample size is likely to facilitate spotting small statistical differences regardless of their clinical importance. One of the clinically important differences was the proportion of inpatient deaths in the pre-implementation period being double the one observed post the intervention.
Characteristics of the study population
Description of the RRT cases and RRT response details
During the post implementation period, 34 RRT activations were initiated, representing 1.5% of admission (online supplemental figure 1). RRT cases’ demographic characteristics and RRT response details are summarised in table 2.
Supplemental material
Supplemental material
Description of RRT cases and the RRT response details (N=34)
Out of RRT cases with valid data, 22 cases (84.6%) were adults and about half of them (57.7%) were males. The most common diagnosis was valve lesions (27.2%). Most of the calls were at daytime (63%), nurses were the main activators (90%). MEWS/ PEWS scoring was the main cause of RRT activation (66.7%). The majority of RRT cases remained in wards. Only one RRT case developed a cardiac arrest after the arrival of the RRT team who subsequently performed CPR. All 34 RRT cases recovered, and none of them died in the hospital. The median time of RRT arrival was 4 min and the median of RRT call duration was 11 min (online supplemental table 1). The most common interventions performed for RRT cases were medication administration and ordering blood tests (5) times, followed by ECG and Echocardiography (online supplemental figure 2).
Supplemental material
Supplemental material
Effect of RRT implementation on study indicators
By the end of the preintervention period, 4183 discharges were recorded, of which 372 were inpatient deaths. There were 31 code blue calls and 25 unplanned ICU admissions. By the end of the postintervention period, 2261 discharges were recorded, of which 105 were inpatient deaths and there were 4 code blue calls and 11 unplanned ICU admissions.
Table 3A shows that the introduction of the RRT intervention was associated with a significant reduction in the inpatient death rate and in the inpatient code blue calls rate. The decrease continued over the postintervention period as indicated by the statistically significant RRT intervention term for both outcomes. After the introduction of the RRT intervention, the unplanned ICU admission rate also decreased and kept the same level over the postintervention period.
(A) Unadjusted time series modelling of the effect of RRT on inpatient death rates, code blue activation rates and unplanned ICU admission rates. (B) Adjusted*† time series modelling of the effect of RRT on inpatient death rates, code blue activation rates and unplanned ICU admission rates
Supplemental material
Supplemental material
Supplemental material
The same results were observed after accounting for overdispersion and adjusting for seasonality in the time series analysis (table 3B). There was no significant time trend in any of the three rates over time, independent of the intervention, and no significant cyclical effects were detected.
Table 4 shows comparisons between the outcome measures during pre-RRT and post-RRT periods. The inpatient mortality rate significantly decreased from 88.93 to 46.44 deaths per 1000 discharges after RRT implementation, (absolute risk reduction (ARR), 42.49; relative risk reduction (RRR), 0.48; 95% CI, 0.36 to 0.58). Inpatient cardiopulmonary arrest (code blue) rate, significantly decreased from 7.41 to 1.77 inpatient code blue calls per 1000 discharges after RRT implementation, (ARR, 5.64; RRR, 0.76; 95% CI, 0.32 to 0.92). Unplanned ICU admissions rate decreased from 5.98 to 4.87 per 1000 discharges after RRT implementation, (ARR, 1.11; RRR, 0.19; 95% CI, −0.65 to 0.60). Figure 1A–C shows ‘U control charts’ for the three outcomes measured before and after the intervention.
Comparisons between outcome measures during pre-RRT and post-RRT periods
U control charts for mortality, cardiopulmonary arrest and unplanned ICU admission per 1000 discharges plots (A–C). (A) U control chart for mortality per 1000 discharges. (B) U control chart for cardiopulmonary arrest (code blue) per 1000 discharges. (C) U control chart for unplanned ICU admission per 1000 discharges. ICU, intensive care unit; LCL, lower confidence limit; RRT, rapid response team; UCL, upper confidence limit.
Disscussion
This study was conducted at Cardiothoracic Hospital, an Egyptian University Hospital, and the study duration took 36 months.
The aim of our study was to determine some selected indicators before the establishment of the RRT, then, to establish RRT service and to evaluate the outcomes of RRT establishment by measuring the selected indicators after the implementation of the RRT.
Our study findings aligned with Barocas et al, for benchmarking the use of RRT in tertiary hospitals. RRT was activated 34 times during the postintervention period of 2338 admission with a rate of 1.5%. Barocas study showed that of 45 651 admissions, 728 calls were made which represents 1.6%. This could be due to the similarity between our setting (a tertiary surgical hospital) and that in Barocas’s study.13
Our results regarding the RRT activators showed that nurses were the main activators for RRT calls (90%), while physicians activated only 10% of the calls. The study by Al-Omari et al supports our results, which reported that nurses activate RRT more often than physicians, as the study reported 1503 RRT activations of which nurses helped activate 69% while physicians activated 31%. This could be referred to the fact that nurses are typically responsible for the routine follow-up of patients.14
The present study showed that the median time of RRT arrival was 4 min and the median of RRT call duration was 11 min. The response time was within the acceptable range predefined in the RRT policy being less than 5 min.15
Our results were in line with the study of Segon et al, as they reported that the average time to respond to an RRT call by at least one member of the RRT was 3.1 min.7
Whereas in the study of Lee et al, the mean response time to RRS activation was 7.6 min.16
This can be due to the robust culture of safety in the Cardiothoracic hospital which was accredited in 2017 and registered by GAHAR in 2021.
In this study, the most common interventions performed by the RRT were giving medications (61.5%), performing ECG/echocardiogram (38.5%) and arterial blood gas (ABG) (30.8%) and chest radiograph (15.4%).
Chan et al largely supported this, as the most common RRT interventions they reported were electrocardiogram (41.0%), additional peripheral intravenous line access (40.0%), ABG was (32.2%) and chest radiograph (31.1%).17
In our study, on RRT calls, a small number of patients needed admission to ICU (24%), while the majority of patients stayed in ward (76%). This was aligned with Bannard-Smith et al, as the transfer to ICU occurred in 24% of cases.18
While in the Rashid et al’s study, a large majority of patients who needed RRT support were moved to a higher level of care (75.68%) and a comparatively small number of patients remained in the room (19.51%).19
RRT effect on mortality
The current study showed that RRT implementation was associated with a significant reduction in mortality rate. Inpatient mortality rate significantly decreased from 88.93 to 46.44 deaths per 1000 discharges after RRT implementation, (RRR, 0.48; 95% CI, 0.36 to 0.58).
The study of Al-Omari et al showed that hospital mortality decreased, and the mortality rate was also lowered from 7.89 to 2.8 per 1000 hospital admissions (RR 0.62; 95% CI 0.59 to 0.65; p<0.0001).14
In addition, Rashid et al observed that an analysis of 41 RRT calls showed a decrease in mortality by 4.88%.19
Furthermore, in a meta-analysis done by Solomon et al, RRT’s application was associated with a notable decline in hospital mortality the analysis included (relative risk=0.88, 95% CI=0.83–0.93).20
In contrary to our results, a study by Segon et al, as they reported that after the implementation of RRTs, the inpatient mortality rate did not improve. There was an inpatient mortality of 3.13% in the preintervention period and decreased to 2.91% in the postintervention period (p =0 .27).7
RRT effect on cardiopulmonary arrest
Our results showed that RRT implementation resulted also in a significant reduction of inpatient cardiopulmonary arrest (code blue) rate, as it significantly decreased from 7.41 to 1.77 inpatient code blue calls per 1000 discharges after RRT implementation, (RRR, 0.76; 95% CI, 0.32 to 0.92).
Our results were supported by Al-Omari et al who reported a reduction in non-ICU cardiopulmonary arrests from 10.53 to 2.58 per 1000 hospital admissions (RR 0.72; 95% CI 0.69 to 0.75; p<0.0001).14
Furthermore, The RRT was also associated with a reduction in cardiopulmonary arrests in adults (RR 0.65, 95 % CI 0.61 to 0.70, p=0.001) and paediatric patients (RR=0.64, 95 % CI 0.55 to 0.74).21
Additionally, a study by Beitler et al reported that non-ICU cardiopulmonary arrest codes significantly decreased from 3.28 to 1.62 codes per 1000 discharges after RRT implementation (RR, 0.493; 95% CI, 0.399 to 0.610; p<0.001).22
RRT effect on ICU-unplanned admission
In this study, the unplanned ICU admission rate decreased from 5.98 to 4.87 per 1000 discharges after the introduction of RRT and kept the same level over the postintervention period.
Many of the earlier trials that were analysed were unable to demonstrate how the RRT activation affected the number of ICU admissions. Sharek et al study’s findings showed that the number of ICU admissions decreased significantly from 44.65 to 20.7 per 1000 hospital admissions after the RRT system was set in place (RR 0.53; 95% CI 0.42 to 0.62; p=0.0001).23
Also, Al-Omari et al’s study showed that implementation of the RRT resulted in a significant reduction in the total ICU admissions, from 44.65 to 20.70 per 1000 hospital admissions (RR 0.53; 95% CI 0.42 to 0.62; p<0.0001).
Similarly, Ludikhuize et al stated that unplanned-ICU admissions showed a declining trend (OR, 0.878; 95% CI, 0.755 to 1.021; p=0.092).24
However, Segon et al revealed that there was no statistically significant change in the number of unplanned-ICU transfers after the introduction of RRT. The number of unexpected ICU transfers declined from 15.8% of ICU admissions to 15.5%, (258 of 1663, p= 0.80).7
The causes of the contradictory results of RRTs are complicated, and in certain situations, they may be connected to local practice and cultural factors that lead to the underutilization of the team. RRTs are extremely well-liked by nursing staff and can help identify underlying patient safety concerns in hospitals.25
Strengths and limitations of the study
This study is one of the first studies to evaluate the effectiveness of implementing an RRT in the Middle East. Quasiexperiments usually are more generalisable and have higher external validity as they can use real-world interventions instead of an artificial laboratory setting. In terms of limitations, for this analysis, we used secondary aggregate data derived from multiple sources. Although the rotation of surgery residents represented a big challenge, we tried to overcome it by repeating the training and doing on-job training for new residents to overcome this barrier. The financial burden of hiring a dedicated staff for RRT was overcome by using a single team for both the RRT and the code blue team.
Conclusions
In conclusion, the implementation of RRT at the Cardiothoracic hospital has been associated with a significant reduction in hospital inpatient mortality, cardiopulmonary arrest (code blue) rates and unplanned ICU admissions. Continuous monitoring is essential to ensure the sustainability of the RRT and maintain the gains. By monitoring key performance indicators such as the rate of cardiopulmonary arrests outside critical care areas, healthcare facilities can assess the effectiveness of their interventions including recognising and responding to early signs of patients’ deterioration and identify areas for improvement. Furthermore, we are contemplating the incorporation of patient-activated RRT to achieve a focus on patient-centred care and engagement.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Study protocol has been approved by the Research Ethics Committee (REC) at Ain Shams university, approval number: FMASU 110/2020. Data has been extracted without any patients’ personal information.
Acknowledgments
The authors would like to express their great appreciation to the Hospital’s management and staff, who generously contributed with their time and efforts in this study.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors RH: Conceptualisation, methodology, data collection and draft writing. RSH supervised and mentored all the research steps and manuscript writing and editing. WMH: Statistical analysis. SAH supervised and mentored all the research steps and manuscript editing. ISH supervised and mentored all the research steps and manuscript editing; The authors read and approved the final manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.