Code Blue: A closer look
A closer look
STACY HOLCOMB, RN, MS, CCRN, and PAM GARLAND, RN, MN, CCRN, CNRN, are clinical managers and clinical nurse specialists; SANDY NEMETH, RN, is a performance improvement coordinator; JOANNE CULVERN, RN, BSN, CCRN, is a unit shift supervisor; FLORENCE KAMRADT, RN, MSN, is a clinical nurse specialist; KIM STEWART, RN, is a unit shift supervisor; and LISA CULVER, RRT, is a clinical specialist for respiratory care. All the authors work at Gwinnett Hospital System in Lawrenceville, Ga.
This emergency team changed the way it collects Code Blue data and found new ways to improve its resuscitation efforts.
Code Blue! Code Blue! Few pages trigger adrenaline the way this one does when there's a medical emergency like a cardiac arrest. In the United States, there are about 300,000 cardiac arrests each year.1 More than a third of patients who arrest do so while in a hospital.2
Cardiac arrests commonly call for techniques of basic life support (BLS) and advanced cardiac life support (ACLS). Cardiopulmonary resuscitation (CPR) is administered to any patient in cardiac arrest, except for those with a DNR order in place.3
Since ventricular fibrillation (VF) is the most common cause of sudden cardiac death, defibrillation is often warranted. If initiated within one minute of the start of VF, defibrillation is 98% effective; however, each minute defibrillation is delayed decreases the patient's chance of survival by 7% to 10%.2
As a first step toward improving outcomes among patients who suffer cardiac arrest at Gwinnett Hospital System, a nonprofit community hospital north of Atlanta, our emergency response team began to develop a database of Code Blue statistics. At our facility, a Code Blue is defined as a patient who is not breathing and/or does not have a pulse.
Our goal was to use this database to identify areas where we could improve our response time to these codes and thus, increase the number of positive clinical outcomes. Early on, our efforts led to several changes, including how we record the events of a code and how we train specific team members.
Updating forms to aid process improvement
In accordance with Joint Commission on Accreditation of Healthcare Organizations (JCAHO) standards, staff members at our institution collect Code Blue data, including patient demographics (age, sex, race), number and length of codes, and survival rates. We also routinely gather information and descriptions of problems encountered during resuscitative events.
Although our data collection met JCAHO standards, weas members of the emergency response teamwanted to collect more detailed data that could help us improve our Code Blue response. We reviewed JCAHO's standards as well as the Utstein standards and recommendations for reporting Code Blue information, which are approved by the American Heart Association (AHA).4 According to the Utstein guidelines, in-hospital resuscitation data that should be collected includes patient information, event documentation, and outcome variables such as survival rates and quality of life after resuscitation.4 The Utstein guidelines also provide a standard reporting form for gathering this information.
Based on the Utstein guidelines, we modified our documentation forms to include information from the Utstein standard reporting form. Our data collection process includes four steps to aid us in quality improvement. First, we record the events of a code on a Code Blue Documentation Form, which we call a "flow sheet." Second, the nurse who leads the ACLS team reviews and evaluates these events and documents her review on a Code Blue Evaluation Form. This, in turn, is reviewed by the clinical manager of that unit, who completes the Code Blue Management Review Form, which is on the reverse side of the Evaluation Form. (Among other things, the manager uses the Review Form to list problems that occurred during the code and steps that can be taken to prevent them in the future.) In the fourth step, all of this information is entered into a computer database. The original documentation formor flow sheetis kept in the patient's record, and a copy is sent to the clinical manager of the unit, along with a copy of the Management Review Form.
Helping recorders to improve their skill
In our hospital, the emergency response team designates a specific team member as the recorder to document the events of a code as they occur. In our review of codes, we found that a team member would be assigned this role at random. Most often, the recorder was a med/surg staff nurse who was typically involved in one to two codes per year.
We found that our nurses' lack of familiarity with the process led to a high rate of errors on the Documentation Form. Almost two-thirds of all codes63%had documentation problems, mostly related to where and how the information was entered on the form.
We determined that the best way to reduce the number of documentation errors was to provide these RNs with more in-depth training in the specific role of recorder. To do this, we videotaped the events of a mock code and asked nurses in each unit to watch the video and record the events of the code on the Documentation Form. This allowed them to practice recording in a non-emergent environment.
Nurses were then asked to compare their record of events with a completed "master" copy of the form, and we helped them identify where their recording could have been more accurate or uniform. Three months after the first workshop, we reviewed our code data and found that only 38% of the documentation forms contained errors, such as data entered in the wrong field, or fields left blank. One year after initiating training, the error rate was 29%. Nurses in the med/surg unit have committed to take part in this training exercise four times each year.
Four "gold standard" time intervals
Another change we made was to collect data on four critical time intervals. We modified our forms to include four time intervals that the Utstein guidelines identify as gold standard variables:4
Event onset to start of CPRthe time lapsed between reaching the patient and starting CPR. We expect this time to be less than 30 seconds. (Our form lists the word "minutes" after this interval and the other three for consistency's sake. Nurses write in the word "seconds" when applicable.)
Event onset to first defibrillationthe time lapsed between reaching the patient and first defibrillation, if defibrillation is applicable. A time is entered only when the patient is found to be in VF or pulseless ventricular tachycardia.
Event onset to advanced airway managementthe time lapsed between reaching the patient and the start of airway management. Advanced airway management includes not just intubation, but effective bag-valve-mask ventilation.
Event onset to first administration of resuscitation medicationthe time lapsed between reaching the patient and administering the first drug.
If any of the intervals do not apply to a specific patient, "N/A" is entered in the corresponding space on the form to make it clear that the data is truly not applicable, and not just omitted when it should have been recorded. Many factors, including clock accuracy, affect the recording of time intervals. The Utstein guidelines recommend using the defibrillator time.4 Because the model of defibrillator used in our facility does not display the time on its screen, our team opted to use the wall clock located in the patient area.
Clarifying the duties of the ACLS team leader
In reviewing our Code data, we also recognized a need to more clearly define the role of the ACLS team leader and to provide more education for all nurses who serve in that role. For consistency, the critical care nurse responding to the Code is always designated as the ACLS team leader. That role is a validated competency for charge nurses in the critical care unit. While working closely with the ED physician who responds to the Code, the team leader's responsibilities include:
Evaluating BLS on arrival to the code
Identifying cardiac rhythm
Initiating ACLS protocol
Evaluating ACLS protocol during the code
Reviewing the Code Blue Documentation Form for accuracy
Reviewing and documenting EKG rhythm strips
Completing the Code Blue Evaluation Form
One critical function of the ACLS team leader is to review the events of a code with the recorder. After doing so, the ACLS team leader has to exercise judgment in her evaluation of the events of a code.
For example, if a code is called at 0700 and the patient is found to be in sinus bradycardia but goes into a rhythm that requires CPR at 0705, the interval from the critical event onset to CPR is recorded as immediate. The five minutes from the time the patient was reached until CPR is started doesn't reflect the true interval; "immediate" is more accurate because CPR was started immediately at the onset of a rhythm that required CPR.
The same holds true for the interval between event onset and first defibrillation. At the onset of a code, the patient may not be in a rhythm that requires defibrillation. The time recorded on the evaluation form should reflect the time between the actual onset of the event that requires defibrillation and the initiation of defibrillation. The time intervals for advanced airway management and administration of resuscitation medication are also subject to the judgment of the team leader.
A year later, it's time to evaluate
By documenting the four critical time intervals, we discovered that we were already meeting or exceeding the time frames outlined in the AHA guidelines, and improvements in this area weren't needed after all.
We did, however, set up system-wide education that emphasized early recognition and intervention. This training included teaching the staff how to recognize a patient in whom a deteriorating condition is impending, emphasizing the need to call the ICU or nursing supervisor to assess such patients, and reminding the staff to promptly notify the physician of any changes in the status of a patient.
We suspect that this training, coupled with our mock codes, may have been responsible for the improvements we saw with our Code Blue outcomes. Prior to implementing these changes, 51% of patients who coded were successfully resuscitated. One year later, 67% of patients who coded were successfully resuscitated.
In our evaluation, we also identified several instances in which patients who probably should have had a DNR orderthey had a very poor prognosis and multiple comorbiditiesdidn't. Future plans include a more in-depth look at why such patients don't have a DNR, and how to help both families and clinicians understand when a DNR is appropriate.
By changing our data collection process and providing better training to nurses who respond to codes, our staffand more specifically, our med/surg nursessaid they were able to increase the rates of successful resuscitations. Indeed, as a result of these initial improvements, our confidence in responding to Code Blue! Code Blue! has grown tremendously.
1. Thel, M. C., & O'Connor, C. M. (1999). Cardiopulmonary resuscitation: Historical perspective to recent investigations. Am Heart J, 137(1), 39.
2. Nelson, M. E., & Zena, C. S. (1998). Losing the race to resuscitate. Nurs Manage, 29(4), 36D.
3. Marik, P. E., & Craft, M. (1997). An outcomes analysis of in-hospital cardiopulmonary resuscitation: The futility rationale for do not resuscitate orders. J Crit Care, 12(3), 142.
4. Cummins, R. O., Chamberlain, D., et al. (1997). Recommended guidelines for reviewing, reporting, and conducting research on in-hospital resuscitation: The in-hospital "Utstein style." Ann Emerg Med, 29(5), 650.
Emil Vernarec, ed. Stacy Holcomb, Lisa Culver, Pamela Padula Garland, Sandra Nemeth, Joanne Culvern, Florence Kamradt, Katherine Stewart. Code Blue: A closer look. RN 2002;8:36.
Published in RN Magazine.