Background: Chemotherapy-induced febrile neutropenia (FN) is a major cause of morbidity and mortality in cancer patients if not treated promptly. As we were facing considerable delays in the management of chemotherapy-induced febrile neutropenic patients in the Emergency Department (ED), we initiated an improvement project aiming for “door-to-antibiotic time” of 60 minutes or less for all patients with fever and suspected chemotherapy-induced neutropenia. Methods: A multidisciplinary team was established to work on the project. We used quality improvement tools for mapping the existing patient flow processes of patients with FN in the ED. Several proposed change ideas have been tested using the Model for Improvement. These change ideas include improving the triaging process, creating an electronic “chemotherapy alert caution” and order sets for physicians, and using the hot-line by nurses to call the pharmacy to expedite the process of preparation of antibiotics. Outcome and process measures were collected weekly and they were discussed thoroughly and analyzed by the team. Run charts were used to monitor the progress. Results: After six Plan-Do-Study-Act cycles, all process measures improved and ultimately the “door-to-antibiotic time” was achieved by reducing it from 255 minutes to 49 minutes. During project testing and implementation, the nursing staff skills improvement and education were taken into consideration as a balancing measure. Conclusion: In a six-month period, the project led to a timely administration of antibiotics for patients with FN in the ED. This improvement was sustained for more than two years after the project initiation.
Febrile neutropenia (FN) is a well-known complication of chemotherapy treatment, and it can lead to life-threatening sepsis. It has been reported that about 10–15% of patients with solid tumors and more than 80% of those with hematologic malignancies will develop fever during chemotherapy course. FN is defined by the presence of fever (single oral temperature of more than or equal to 38.3°C or a temperature of more than or equal to 38°C sustained over a one-hour period) and the presence of neutropenia with absolute neutrophil count (ANC) of less than 500 cells/mm3 or an ANC that is expected to decrease to less than 500 cells/mm3 during the next 48 hours. Among cancer patients presenting with an episode of FN, one series found that 30% of patients had microbiologically confirmed bacteremia and another group identified a 10% prevalence of gram-negative bacterial infection.
An episode of FN is a potentially life-threatening condition, and therefore prompt evaluation and timely antibiotic administration are considered crucial. Achieving the goal of timely antibiotic administration in a pediatric ED can be difficult. In a recent study by Lynn et al. on the management of FN patients in the emergency room (ER), they found that latency of the first dose of antibiotic, pneumonia, and platelets count less than or equal to 50,000/mm3 were identified as independent factors associated with serious complications. A study conducted among patients admitted to intensive care unit (ICU) with severe sepsis showed that 39.8% had the risk of death, and for each hour delay in antibiotic administration, mortality increases by 7.6%. In another study of non-neutropenic patients with severe sepsis admitted to ICU, antibiotic administration within one hour was associated with a mortality rate of 25% compared with 67% if the antibiotic was administered after four hours.
In 2010, Infectious Diseases Society of America published clinical practice guidelines for the use of antimicrobial agents in neutropenic patients with cancer. They recommend that antibiotics should be given immediately (within two hours) to patients presenting with FN. The Surviving Sepsis Campaign in 2008 recommends that appropriate antimicrobial therapy should be administered within one hour in cases of severe sepsis. A retrospective study on 1628 FN hospitalization from 653 pediatric patients found 60-minute intervals (time to antibiotic) to be the most strongly associated factor with the composite adverse events outcome.
In our hospital, the available data revealed that there are considerable delays in the management of patients with chemotherapy-induced FN in the Emergency Department (ED). Thus, we initiated an improvement project aiming to make “door-to-antibiotic time” equal to or less than 60 minutes for all patients presenting to ER with fever and suspected chemotherapy-induced neutropenia.
Our staff reported considerable delays in the management of patients with chemotherapy-induced FN in the ED. A sample of 10 medical records of patients admitted with FN were retrospectively reviewed, and the mean “door-to-antibiotic” administration time in ED was found to be 255 minutes with a range of 150–450 minutes, which put our patients at risk and also led to the reluctance of patients and their families to visit ED due to major delays in receiving their treatment.
In an effort to improve the quality and safety of care rendered to those patients, a multidisciplinary team was assembled to work on an improvement project aiming for “door-to-antibiotic time” of 60 minutes or less for all patients (adults and pediatrics) presenting to the ER with fever and suspected chemotherapy-induced neutropenia within a six-month period; the project started in April 2013 and ended in November 2015. The primary outcome measure was the time interval from arrival to ED triage to antibiotics administration, which should be equal to or less than 60 minutes.
The improvement team used a number of quality tools including a driver diagram to identify the main drivers affecting the desired outcome [Figure 1]. In addition, the whole process of patient flow and management of FN in ED was mapped and the flaws in the process were identified. Improving these points of flaws within the process were considered as process measures because improving these measures will lead to overall improvement in the outcome measure.
As a result of this exercise and in aligning with the project driver diagram, we came up with a bundle of interventions and change ideas to be tested and implemented, including the following:
Securing support and commitment from ED chairman and the executive of nursing, followed by inviting ED physicians and nurses to attend a building capacity workshop.
Using technology to create a “Chemotherapy Caution Alert” in the electronic medical records (EMRs) for patients receiving chemotherapy so they can be identified upon arrival to ER. This chemotherapy alert should be flagged by the nurse who administered chemotherapy to the patient during the previous clinic visit or admission, and this caution will be in the system for a period of six months.
Securing the availability of needed resources, including making sure there is a thermometer in triage reception to check temperature upon arrival so any patient with fever and history of recently receiving chemotherapy will be triaged as level II.
Developing order sets that include initial blood works and evidence-based antibiotics orders for physicians.
Developing an Alert Card for patients to be carried by them always for easier identification in ED.
Placing awareness posters in the ER and clinics to alert the staff about the importance of this project.
Then, we created a data collection form to collect the following data: time of arrival to ED, triage level, time to assign to a bed in ED, time of orders by physician-blood works and first antibiotic dose and time of first antibiotic dose administration.
Several Plan-Do-Study-Act (PDSA) cycles have been used to test the changes using the Model for Improvement. The data were collected from all patients who came to ED using the data collection form after validating it. The team met regularly (weekly at the beginning) to discuss the data.
PDSA cycle 1
Plan: To check the compliance of nurses who administered chemotherapy in the clinic or during admission to activate the chemotherapy caution alert on the system.
Do: A preliminary review of data showed that 90% of the patients did not have the chemotherapy caution alert.
Study: From this PDSA cycle, we have learned that many nurses were not aware of the importance of flagging patients who are receiving chemotherapy on the EMR system.
Act: To enhance the nursing engagement in the project.
PDSA cycle 2
Plan: To give nurses more training about chemotherapy caution alert.
Do: When discussing the data with chemo nurses, they said that sometimes they do not do it due to time constraints or because they simply forgot; for that, the project team reinforced the idea behind this project and its importance.
The second round of data collection was conducted to measure the percentage of compliance study.
Study: The result of this PDSA Cycle showed that chemotherapy caution alert was still missing in 80% of patients.
Act: Conduct a meeting for the project team member to overcome the challenges reported by nurses in a more technical way.
PDSA cycle 3
Plan: To check with the Information Technology (IT) department about the possibility of automating the flagging process by having it as a forced function step in the EMR, where the chemotherapy caution alert will appear automatically once the chemo nurse completes his or her chemo checklist on the system.
Do: The IT did the needed modification in the EMR. Another round of data collection was done to measure the percentage of compliance.
Study: The result showed that 100% of patients had the chemotherapy caution alert.
Act: The desired outcome was reached for this phase of the process. Since this phase has turned out to be an automated process, there was no need to keep monitoring the compliance unless there is a technical breakdown. We decided to move to the next phase by making sure that the ED physicians are using the preprepared order sets to manage FN patients.
PDSA cycle 4
Plan: Access compliance of ED physicians in using the preprepared order sets that were developed to manage a patient with FN.
Do: We reviewed all medical records of the patients with FN in ED from April to July 2013 to check compliance of ED physicians in using order sets and collected data about the physicians' compliance in using these order sets.
Study: The analysis for the collected data showed that only 30% of physicians used the preprepared order sets, and most of them were pediatricians. Therefore, it was obvious that physicians form ED need more education and reinforcement to use order sets in the system. In addition to these findings, physicians reported that the antibiotics order set is time-consuming because it requires them to calculate the weight-dependent dose.
Act: Physicians were educated to use the order set in the system during the departmental meeting and the findings were discussed with ED physicians to get their feedback and comments on the process.
PDSA cycle 5
Plan: Making the order sets more user-friendly by solving the challenges that were reported by the physicians in the previous cycle.
Do: We discussed this issue with the clinical pharmacist, and the doses of antibiotics for children were changed to a fixed weight-based dose. Those fixed weight-based doses were uploaded in to the EMR system so that it appears for the physicians immediately when they are about to prescribe an antibiotic for the patient with FN. Another meeting with pediatric emergency physicians was arranged to discuss again the purpose of this improvement project and the importance of using the order sets that would shorten the time required for ordering various blood tests and antibiotics. The changes that we made in the system to ease their workload were highlighted. Then, another round of data collection was performed.
Study: In this PDSA cycle, the compliance with the antibiotic order sets for pediatrics was tested again, and this time, it was improved and reached 85% compliance.
Act: Keep monitoring the process to ensure high compliance, check whether the patients are using the chemotherapy alert card so that they would show it to the ED staff upon their arrival to expedite their management process.
PDSA cycle 6
Plan: To check if patients on chemotherapy are carrying with them the chemotherapy alert card that was given to them during their outpatient clinic visit or during their admission.
Do: Collect data to assess the patients' compliance in having the chemotherapy alert card with them when visiting the ED.
Study: Only 25% of the patients carried with them the chemotherapy alert card.
Act: The team of this project must work on improving the patients' awareness about the importance of having the chemotherapy alert card. The team also worked on increasing the awareness of patients about the importance of the alert card by the patient educators in the clinic during the chemotherapy session and for inpatients as well.
Nurses in clinics and in the inpatient setting were asked to make sure that every patient has the alert card before being discharged. Moreover, a boosting intervention was done by placing large awareness posters in clinics and inpatient wards empowering patients to request the chemotherapy alert card from physicians and nurses, and by making the chemotherapy alert card available in clinics and wards. After all these interventions, a round of data collection showed that the percentage of patients who carried the alert card increased to 75%.
A total of 358 patients with FN were followed during the project. In the first cycle of PDSA, the average time from presenting to ER until starting the administration of the antibiotics was 255 minutes, and the highest average was recorded in May 2013 which was 301 minutes. The effect of the interventions started to positively affect the administration time in mid-June 2013 after the third PDSA cycle. The significant drop in the waiting time was observed initially in August and September 2013, but it did not reach the target of 60 minutes or less. In mid-November 2013, The average time to antibiotic administration approached 60 minutes as shown in Figure 2. Although the monthly average time of administration of antibiotics was less than the 60-minute goal, some patients still did not receive their treatment within the desired time [Figure 3].
All in all, average the “door-to-antibiotic time” improved from 255 minutes to 49 minutes, which will assure more safety and better care quality for patients who present to ED with fever and suspected chemotherapy-induced neutropenia in our hospital.
The percentage of patients who got antibiotic within 60 minutes increased significantly from an average of 0–100% in April–June 2013 to 83–100% in the period November 2014 till March 2015.
In addition to the improvement, the competencies of the ED nurses in relation to chemotherapy and central line management were improved as we trained the majority of the ED nurses in these topics. Figure 4 depicts the percentages of nurses who were trained in these two areas before and after the start of our project.
The results showed that the oncology patients who come to the ED with fever and suspected neutropenia follow a standard process to receive a timely antibiotic, and this process is sustainable for more than two years.
In this quality project, we were able to reduce the average time of antibiotics administration for oncology patients coming to ED with fever and suspected chemotherapy-induced neutropenia to less than 60 minutes. This was sustained for a long time by standardization of the process and utilization of the available EMR, which did not require any extra financial cost. Achieving this 60-minute goal was also reported by using other quality initiatives,[4,10–13] and it is also clear that following a standardized set of procedures or protocol will significantly reduce the time needed to start antibiotics for these patients.[11,12]
The most important lesson from this improvement project was demonstrating that improvement science and quality tools are very helpful and practical in improving the system thinking and improving specific interlined processes to achieve the ultimate aim of the improvement project, and it is consistent with what was reported in the literature. In our project, we learned that improving the triaging process in ED and linking various processes with automated order sets had led to a positive impact on managing oncology patients with fever and suspected chemotherapy-induced neutropenia aiming to avoid complications due to delay in the management of those patients.
Limitations are expected in any quality improvement project. As quality projects most of the time are aiming to change practices or process, some staff tend to resist any type of change. Thus, keeping the staff involved and updated about the project and taking their feedback into consideration will not only minimize their resistance to change but would also encourage them to embrace the new change. Using average time as a measure has its advantages and disadvantages. It shows the overall trend for all patients but it does not address each individual patient's care. In our study, the percentage of patients who did get the antibiotic in 60 minutes increased significantly. Reaching 100% compliance in administering the antibiotic to all patients in 60 minutes or less is a challenging goal as the processes in the ED is affected by many variables including the workload and the urgency of cases. We believe that staff involvement goes beyond just informing them, and it may require upskilling or reskilling the staff to be able to accomplish the desired improvement. Furthermore, an essential component for the success of any QI project is the involvement of all stakeholders in the project. In our project, we involved the hospital leadership, physicians in the ED and Oncology Department, nursing staff in both departments, pharmacy staff, information technology, and others.
There was a dire need to improve the safety and the quality of care provided to patients with fever and suspected chemotherapy-induced neutropenia in the ED. Applying quality improvement methodology, using various quality tools, automated evidence-based order sets, working with committed team members, and getting the support from the institute's leadership allowed for the success of this project. The average administration of antibiotic in 60 minutes or less to patients with FN was attained and sustained for more than two years. Continuous monitoring of compliance is essential to make sure that we detect and overcome all the avoidable situations that may delay the timely management of cancer patients.
We would like to acknowledge all administrative, clinical, and quality improvement/patient safety leaders, physicians, nurses, pharmacists, health educators, IT specialists, and administrative assistants for their support throughout this project. Specifically, we would like to thank Faith Vabaza, RN, Corina Meissenheimer, RN, Ms. Amal Abu Shaqaif, Dr. Abdulllah Alromi, and Ms. Areej ALkhesaifi for their valuable contribution to this project.
The ultimate goal of this project was to test and implement evidence-based practice rather than testing hypotheses to generate generalizable findings. Thus, the ethical approval for the quality improvement project is not needed. Moreover, the hospital management and oncology department have approved and supported this project. However, all collected information did not include any patient identifiers and it was kept private and confidential.
Financial support and sponsorship
The authors disclosed no funding related to this article.
Conflicts of interest
The authors disclosed no conflicts of interest related to this article.
For reprints contact:email@example.com