CEO SUMMARY: In this unusual collaboration, the participating Pennsylvania hospitals dramatically reduced blood specimen labeling errors. This initiative to share best practices incorporated techniques that were refined in other projects designed to reduce medical errors and improve patient care. Another interesting feature of this multi-hospital quality initiative is that the participating institutions agreed to publish their rates of errors involving mislabeled patient blood specimens.
HERE IS AN UNUSUAL STORY for the clinical laboratory industry. It involves nine participating hospitals, the State of Pennsylvania, and the willingness to publicize error rates on the handling of laboratory test specimens.
“Working with the Pennsylvania Patient Safety Authority (PPSA), these nine hospitals reduced the rate of mislabeled blood specimens in the nine facilities by 37%,” observed Barbara Booth, MT (ASCP), Laboratory Service Improvement Coordinator at Geisinger Wyoming Valley Medical Center in Wilkes-Barre.
“This project was initiated in 2009,” she continued. “The goal was to improve patient safety by reducing one common source of errors—blood specimen labeling errors.”
This program was organized by the PPSA. Hospitals that participated had to agree to:
1) Report blood specimen labeling errors through the authority’s Pennsylvania Patient Safety Reporting System (PA-PSRS);
2) Submit monthly laboratory reports to an authority analyst; and
3) Investigate mislabeling events using a standardized event investigation tool.
This initiative is noteworthy because it shows how transparency in quality outcomes is advancing both for hospitals and for clinical laboratory services. It also demonstrates how government health agencies will step up and take a lead role in fostering cooperation across multiple healthcare organizations to pursue improved quality in patient care.
Further, this is an example of how cooperating institutions will increasingly be willing to share best practices that directly contribute to a reduction of medical and laboratory errors. Data from these initiatives will be available to the public and other health care providers.
For Booth, this project targeted an ongoing source of laboratory test errors familiar to every laboratory administrator and pathologist: blood specimen identification errors. “In 2009, PPSA (at http://patientsafetyauthority.org) recognized that blood specimen identification errors occur at rates of 0.1% to 6.5%,” stated Booth.
“Laboratory professionals know that misidentified specimens can lead to delays in diagnosis and cause providers to treat the wrong patient or the wrong disease, and contribute to severe transfusion reactions as well,” added Booth. “Thus, to reduce blood specimen labeling errors, in 2009, the authority asked hospitals to address this issue in a collaborative project.”
Booth was the leader of this effort at her hospital. She reported on the outcomes and lessons learned from this unique nine-hospital collaboration at the Lab Quality Confab that took place in San Antonio, Texas, in November.
“Nine facilities agreed to collaborate, even though most organizations do not like to air their dirty laundry,” stated Booth. “All participating institutions recognized that this program offered a chance to review internal procedures and improve patient outcomes. The potential for improving patient care and patient safety clearly outweighed any disadvantages.
Goal: Cut Errors in Half
“The primary goal of the collaborative was to decrease blood specimen mislabeling events by 50% in 12 months,” noted Booth. “We defined an event as unlabeled, partially labeled, illegible labels, or use of the wrong patient name or label. Point-of-care testing (POCT) was excluded from this project.
“At Geisinger Wyoming Valley, our initial thought was to make improvements hospital-wide,” she said. “But this became our first lesson, as it was suggested that we focus on one department. This is consistent with the principle that, by starting small and achieving some success, we would build energy and momentum to take on larger quality- improvement efforts later.
“We launched our improvement efforts at Geisinger Wyoming Valley Medical Center with the medical intensive care unit,” explained Booth. “As a 20-bed unit, it was relatively small and it allowed us to engage respiratory, laboratory, and nursing.
“At my hospital, our improvement team included staff from laboratory, nursing, respiratory, risk management, regulatory PI, and the medical director,” she stated. “We launched a system-wide electronic educational course for all specimen collectors. Next, we observed the processes collectors used when obtaining blood specimens from a patient.
“To ensure that we were consistent in our observations, we developed an observation tool,” she added. “We wanted to ensure that each observer knew what to look for and would gather similar information.
“The tool asked: ‘Did the patient have an identification band?’ ‘Where was it?’ ‘What information was on it?’ ‘What identifiers were checked?’” commented Booth.
“Another lesson: observe procedures before and after process changes and include a variety of observers,” she continued. “For our observations, we used staff crossover, meaning labs looked at respiratory, respiratory looked at nursing, and nursing looked at labs.
“Crossover is important,” added Booth. “When your staff looks at its own procedures, the tendency is to see only the good aspects.
Differences in Collection
“These observations uncovered plenty of information: every department has slightly different procedures for collecting blood and identifying patients,” she said. “Different processes existed because we had different staff (including respiratory, nurses, and sometimes physicians) collecting blood.
“When we observed the processes, we saw that laboratory tended to be more consistent and that staff from other departments had more variation in their collection processes,” recalled Booth.
“Our findings may not surprise experienced laboratory professionals,” noted Booth. “The observations showed that collectors were not doing all things we
thought they should be doing to collect blood properly and identify the patient!
“Seeking to standardize blood collection, we listed the steps that a collector would follow to conduct the best process each time,” said Booth. “Then we taught this process by having team members go out to each clinical service, including the lab, respiratory, nursing, and the ICU.
“Initially, our efforts to standardize here’s another lesson,” she added. “Interactions always included more than one team member, and the focus was always on patient safety. Whenever someone complained, it was explained that the program was designed to improve patient care. At that point, there was no further discussion because staff recognized the value of eliminating errors and thus improving patient care.
“Despite our efforts, we continued to have some errors in blood specimen identification,” she said. “Therefore, we conducted post-error interviews to improve the process. The nine teams developed a standardized event investigation tool that is available online from the Patient Safety Authority (http://tinyurl.com/7nrcf8s).
“Using this form, we interviewed collectors in a non-punitive manner and encouraged them to be honest,” Booth explained. “At first, it was a struggle, particularly with nurses. This was another lesson. The interviews gave our laboratory staff an opportunity to break down barriers and build relationships with the nurses.
“The data from this tool showed that the most significant factors contributing to errors were: a) procedures not followed; b) interruptions or distractions; c) emergencies; and d) significant workload increases,” she said. “This data helped us identify reasons why collectors did not use the standardized process we developed.
Procedures Not Followed
“The number one source of errors for misidentified blood specimens was that procedures were not followed,” recalled Booth. “We next began to dig into the reasons why procedures were not followed.
“Three causes jumped to the top of the list,” she said. “There was a need to improve the lab test requisition. Issues with the printing of labels required a fix. There was also the failure of collectors to properly identify patients when drawing blood.
“Over the course of this improvement program, the nine facilities decreased identification errors by 37%!” noted Booth. “Error rates at our facility did increase after the project was completed, however, they did not rise to the previous levels. Most importantly, we now had tools and a structure to address this increase.
“The authority published an article about the collaborative on its website and the article details all of the steps we took to solve the various problems we encountered,” Booth concluded. “Its title is: ‘Reducing Errors in Blood Specimen Labeling: A Multihospital Initiative’ and the URL is http://tinyurl.com/7nrcf8s.”
Project Harnesses the Efforts of Nine Hospitals To Reduce Blood Specimen Labeling Errors
Reducing Label Errors Was Goal of Project
FOR THE PATIENT SAFETY collaborative in Pennsylvania, nine hospitals participated in the effort to reduce the number of blood- specimen labeling errors. The hospitals were:
- Allied Services Rehabilitation Center (Viewmont Medical Laboratories), Scranton
- Berwick Hospital Center, Berwick
- Easton Hospital, Easton
- Geisinger Medical Center, Danville
- Geisinger Wyoming Valley Medical Center, Wilkes-Barre
- Lehigh Valley Health Network, Allentown
- Sacred Heart Hospital, Allentown
- Pocono Health Systems, Stroudsburg
- Wyoming Valley Health Systems, Wilkes-Barre