(This article also appears in the October issue of EMS Insider. EMS Insider, the premier publication for EMS managers, supervisors, chiefs and medical directors, is a must-have resource for the critical, accurate information EMS leaders need. The monthly publication offers quality investigative reporting, exclusive articles, management tips and the very latest news on legislative issues, grants, current trends and controversies. For more about how to become an Insider, go to
Fourteen years ago, the medical community was stunned when a paper published by the National Institute of Medicine reported a remarkably high medical error rate in hospitals. “To Err is Human: Building a Safer Health System” stated that “at least 44,000 people, and perhaps as many as 98,000 people, die in hospitals each year as a result of medical errors that could have been prevented, according to estimates from two major studies.” A significant portion of those errors were attributed to mistakes involving the administration of medication.
Although there were no related studies involving out-of-hospital medication errors, one could extrapolate that the error rate was at least as bad, if not worse, according to Sedgwick County (Kan.) EMS Director Scott Had-ley. He points to the additional risks of practicing medicine in the field:
- Emergency situation;
- No external crosscheck;
- No electronic decision support;
- High-risk medications; and
- Drug shortage issues and substitutions.
As a result, Hadley and his Sedgwick County colleagues began to look for ways to decrease medication errors. They started by trying to understand the frequency of medication errors in their system. In an internal survey, 100% of the 107 EMS providers who responded said they always verify for accuracy before giving medication. Yet, 60% admitted to making a medication error, and 40% said they didn’t make an error that they know of—“that’s critical,” Hadley says.
Survey results were supported by evidence collected during the systems credentialing process: Some providers were unknowingly making medication errors. Significantly, they found that 31% of those committing a medication error had no idea that they had made a mistake. Even though they verbalized the correct medication and dose, they administered the incorrect dose.
Identifying the scope of the problem was a good start, but how to address it? Sedgwick County’s EMS Quality Improvement Manager, Paul Misasi, reached out to other systems, conducting a survey to determine if other EMS agencies had found the same problem. Of the 178 respondents, he learned that 96% use a verification process before administering medications. The majority use the “Five Rights”—the right patient, dose, medication, route and time. Interestingly, only 30% of the agencies said the verification process is done verbally. The majority of the time, the EMS provider ticks off the five rights mentally. Seventy percent said that’s how they were taught to perform the task.
Of those who said they used the Five Rights, only 60% said they follow it exactly. When asked if a one-person, mental checklist is adequate to prevent medication errors, slightly better than half said it was. Eight out of ten agreed or somewhat agreed that a two-person verbal process would be more successful.
After reviewing the results of the two surveys and other information, Sedgwick County EMS, in collaboration with its medical director, Sabina Braithwaite, MD, MPH, FACEP, developed a process to help identify and correct for system errors. The Medication Administration Cross-Check (MACC) is a one-page, easily remembered, standardized method for administering medications every time for every medication. Error traps, written into the process, help to create “pause points” that ensure safety. The MACC requires two providers to verbalize the procedure in a feedback loop.
It’s similar to asking someone else to proofread your own work: The second person often catches errors inadvertently caused by what was intended versus what was actually accomplished. “The second person is actually authorizing the medication,” Hadley says. “For the majority, it does a good job.”
The MACC doesn’t require a paramedic partner for it to work. “Even though an EMT cannot deliver the dose, he or she can read,” Hadley says.
There were some initial objections to using the MACC. Some providers were concerned that it would delay treatment. “Is it better to give the wrong dose faster or the right medication slower?” Hadley asks. Internal studies demonstrated that the two-person process takes approximately 20–25 seconds. “Not a lot of time,” he says.
Other concerns identified during the beta testing included a perception of a lack of professionalism or competency on the part of the paramedic, if the MACC is used in front of the patient. To combat this, EMS providers were instructed to explain to the patient what they were doing, telling them they are going to converse with their partner on a safety check regarding the medication they are about to administer. “How many people intervene and tell a pilot not to do a safety check?” Hadley says. They found that patients were receptive and didn’t think less of the provider’s abilities, he reports.
Success helped breed compliance. “Those who use the verbal verification process report 50% fewer errors than those who verify mentally,” Hadley says.
Identifying errors that do not occur is a challenge. Sedgwick County began tracking medication errors prior to implementing the MACC in order to establish a baseline. By comparing historical data to data collected after the MACC was implemented, Hadley says they estimate the MACC has prevented about 15 potential errors so far. Since March 2012, when the program was implemented, the average number of errors pre-MACC was 1.63 per month. Post-MACC errors dropped to 1.33 per month. “The power is in finding the events that didn’t cause harm,” Hadley says. He admits that determining the exact number of errors that were avoided is difficult because the medication errors prior to using the MACC were self-reported and could be under-represented.
Using the category index developed by National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP), Sedgwick County can see that the majority of the errors that occurred (45%) fell into Category C, meaning the wrong medication reached the patient, but no harm was caused. Three percent fell into a more serious category, contributing to temporary harm or requiring intervention. None of the errors resulted in permanent harm or death of the patient.
To help track medication errors both internally and nationally, Sedgwick County enters its data into the EMS Voluntary Event Notification Tool (E.V.E.N.T.) at http://event.clirems.org. The online tool is a program of the Center for Leadership, Innovation, and Research in EMS with sponsorship provided by the North Central EMS Institute, the National EMS Management Association, the Emergency Medical Services Chiefs of Canada, the National Association of Emergency Medical Technicians and the National Association of State EMS Officials.
The purpose of E.V.E.N.T. is to improve the safety, quality and consistent delivery of EMS, through the collection of data submitted anonymously by EMS practitioners. The data is used to develop policies, procedures and training programs to improve the safe delivery of EMS. The developers say that a similar system used by airline pilots has led to important airline system improvements based upon pilot-reported “near miss” situations and errors.
In the past, identifying an error usually resulted in some form of punishment for the provider. Typically the severity of the punishment was based on the degree of harm to the patient. As a result, few providers were willing to report an error and few systems could identify areas of improvement.
“We cannot keep blaming and shaming people who make an error. We have to fix it,” Hadley says. That’s why Sedgwick County subscribes to the Just Culture philosophy. “People must feel confident reporting errors,” he says. “[Self-reporting is] critical to making system-level improvements.” If the mistake could happen again tomorrow, it is probably not the fault of the individual.
Just Culture uses a system of peer review. However, Hadley notes that peer review protection laws differ from state to state, and need to be taken into account when developing a medication error reporting system. “Some [peer review programs] are discoverable,” he says.
Since implementing Just Culture at Sedgwick County, Hadley says that more and more providers are coming forward to report medication errors. “Employees don’t come to work to make mistakes. Mistakes will happen. We want to know what happens in our system so we can improve,” he says.
For those interested in the MACC and supporting documents, please contact Hadley at firstname.lastname@example.org or Misasi at email@example.com.