July 2009 was not a bad month for the point-of-care testing program at Nebraska Methodist Hospital in Omaha. First, the 430-bed hospital earned recognition from the Hospital Compare program sponsored by the Centers for Medicare and Medicaid Services, which placed Methodist among the elite two percent of institutions in the nation with the lowest mortality for heart attack patients. Capping that accolade, the American Association for Clinical Chemistry tapped the hospital’s point-of-care coordinator, Brenda Franks, MT(ASCP), for its Point-of-Care Coordinator of the Year Award.
Franks would be the first to say there’s no magic method for a POC testing program to arrive at such a plane, but she has a few words of advice for other POC testing programs. Among them: Get people to focus on the data. Pin down specific ways to reduce errors. Solve problems through coordinated teamwork. And make sure that POC solutions actually address the problems you are trying to fix.
Nebraska Methodist’s point-of-care program started out with a slight advantage over most. “When point-of-care glucose testing first became available 20-plus years ago, the laboratory here was privately owned, so we brought it on as a phlebotomy-based program,” Franks says. There was a benefit to that, because every POC glucose performed was put into the patient medical record via the hospital information system. “So even after our lab became part of the hospital, we always had our results in the computer—unlike what everyone else was doing, having nursing perform the test and writing it in charts. We never went through that phase.”
Today, Methodist, which has two separate CAP-accredited laboratories, has about 500 POC operators but still uses primarily phlebotomists, with nurses performing point of care in targeted places throughout the hospital. The laboratory pays a little extra to have two outputs for its LifeScan SureStep Flexx Meter for POC glucose results—one for the nursing interface and one for the laboratory.
“We went with the LifeScan SureStep Pro model shortly before I became POC testing coordinator 11 years ago,” Franks says. “They were leaders as far as data management; they had it on their meters before anyone else, and it was very important to us to be able to manage operators and review QC. We were ahead of the curve on that.”
The laboratory also employs Abbott i-Stats, ITC’s Avoximeter, and ITC Hemochron Signature Elite instruments interfaced through Telcor’s QML POC Data Manager for connectivity. “Our waived testing performed by nursing is also entered through the Cerner module into the medical record. We’re not entirely paperless—but we’re trying to be.”
As the hospital’s reputation in cardiac care suggests, implementation of tight glycemic control starting in 2004 has proved to be a major success story. The POC testing program used a performance improvement process to implement a tight glycemic control and insulin protocol.
“When we first started talking about insulin infusion at Methodist, we wanted to actually capture the data and let it drive our processes,” Franks says. “We had our perfusion team from cardiovascular surgery, we had our cardiovascular nurse practitioners, we had representatives from our diabetes center so we can screen and follow up patients after discharge from the hospital, and we had the laboratory and the public relations department there. Our strength here is our ability to work as a team, and that’s how we did that,” Franks says.
The results are evident in the CMS Hospital Compare charts showing the quality measures it uses to assess patient care. While the national average for maintaining blood sugar after surgery came in at 84, Nebraska Methodist’s score was 96.
Data collection also proved critical in addressing a related problem at Methodist: taming patient identification errors in the rushed environment of the emergency department. “In addition to my POC duties we have an electronic incident reporting system, and I’m the keeper of the pathology department incident reports, which I route to staff for review,” Franks explains. “I noticed in our nursing-based POC glucose program that we weren’t having patient ID errors in surgery or outpatient surgery. They had no trouble with positive patient identification, and of course our phlebotomy team doesn’t have errors because we live and breathe patient ID. But we were having serious issues in our emergency room.”
She found it frustrating. “On the incident report review, there weren’t only POC issues but also patient ID issues on specimens collected, whether urine, blood, or other fluids. So after over a year of sending reports to the nursing service here about performance, I was invited to the Unit-Based Council meeting—which includes their ER cores, who are like shift supervisors, and some key staff members and nursing management too.”
She presented to them all the ID errors and explained that Methodist’s ER was responsible for 50 percent of all the ID errors in the entire hospital. “They looked at me and said, ‘Brenda, there’s no way.’ But I said, ‘Actually, there is.’ It was all right there in the incident reports.”
“They took it very personally. They blacked out the names on the incident reports and stuck them on the refrigerator in the lounge. So finally, the people making the errors were acutely aware that errors were happening.”
What the laboratory and nursing realized was that they had a processing issue with the way patients were identified initially in the ED by the admitting interviewer who took down information after the patient was triaged. If a patient with a potential heart attack arrives, nurses start an IV and draw blood immediately, she explains. “At that point you don’t have labels or patient ID, but we changed that process. We got armbands on patients faster, we implemented scanning, and they decided if you start that IV and draw that blood, you don’t leave the room until the specimens are labeled. That actually was not happening previously.” The nurses also started policing each other, she adds. “If they saw someone coming out of the room, they’d tell them to get back in there until the patient was banded and specimens were labeled.”
“The ED really rallied to the data that was presented to them and the interaction with the lab,” says Thomas L. Williams, MD, the laboratory’s medical director, noting that the emergency department designed its own posters to remind people about proper technique. But two of the nurses managing the ED, Lori Groves, BSN, RN, and Terri Perry, BSN, RN, give generous credit to Franks for volunteering to attend the monthly Unit-Based Council meetings to share data and improve communication between the ED and the lab. “The participation and information provided by Brenda helped drop our error rate dramatically in a remarkably short time,” they said in recommending Franks for the AACC award.
In fact, that active interdisciplinary approach has helped improve laboratory testing across the board, Dr. Williams says. “We have been a lot more aggressive with that the last couple of years. I and one or two key people from the lab visit all areas of the hospital extemporaneously” as well as routinely. Four times a year, for example, “one of the pathology assistants and I go to a 6 AM meeting of the surgical nurses and talk about lab issues with them. We have really focused on specimen ID, completion of surgical requisitions, and team building.”
A separate quality improvement initiative spearheaded by the laboratory was directed at the Cellsaver and autologous growth factor program. What the perfusion department wanted was a way to figure out if it was removing the red cells and debris for the Cellsaver machines according to AABB guidelines, Franks says. “So I helped them design a system so that each one of the Cellsavers had a name in Cerner, just like a patient, and we preprinted labels and made a procedure so they could collect specimens and send them to the lab, and the lab would know what tests to order on those particular samples.”
“Brenda worked with perfusion, IT, and laboratory staff to develop processes for supplemental quality checks for hospital Cellsaver and autologous growth factor processing equipment,” Dr. Williams adds.
When accreditation inspectors arrived from the Joint Commission, they asked about performance of the Cellsaver device, Franks says. “And we said: ‘We’ll show you the performance data right here.’ We did that with our platelet-rich plasma as well—just using different parameters that we check to make sure we’re harvesting a high-quality product.”
Methodist’s chief perfusionist, Jeff Kangas, CCP, wrote the following in recommending Franks for the AACC award: “In the past few years, the Joint Commission required perfusion departments to be accountable for quality assurance for all their blood-related equipment and to document it accordingly. Brenda not only created a very user-friendly system to meet all of our POC QA and QC needs, but she also devised a very accurate inventory system that basically automated our quality control and reagent supplies and prompted us to keep our QC logs current.”
It was a somewhat more humble approach that the POC testing program used to address a quality control problem in ACT testing. When the laboratory was cited during a CAP accreditation round for not doing QC as scheduled, the solution involved, in a sense, recentralizing part of the POC testing program. “Originally our perfusion team ordered and stored their own ACT cuvettes,” Franks explains. “But we were having trouble monitoring QC every time they opened a new box.” The laboratory decided to redo the way it managed inventory, using what she calls a “crude but functional” system for cuvette tracking—starting by storing cuvettes in the lab instead.
“When a shipment comes in, I put a sticky label outside the box with a lot number, so they pull the label off when they need a box and put it in an inventory book. We didn’t have a way to positively guarantee that every single box of those cuvettes were QC’d before. But after that, when we got the inventory sheets we could match them up with our received inventory.” It was simply a redesign of the inventory system, she adds. “But it did make their lives a lot easier, and it made us a lot more compliant.”
Prothrombin time/INR standardization was another challenge addressed by the Methodist POC testing program. “When INRs really hit the news in August 2001, there was a facility in Pennsylvania that had an INR calculation error that led to overdoses for several patients and really bad outcomes,” Franks says. “So we decided to survey not only our hospital but also look at our sister hospital in Council Bluffs, Iowa, Children’s Hospital next door, and all the clinics. It really surprised us to find that we had so many types of instrumentation out there.”
As a result, the laboratory standardized everyone to one of three test platforms that used only sensitive ISI reagent, a move that turned out to be successful. Last year, the hospital acquired six Roche CoaguChek XS instruments because the device has an ISI of 1.0, requires a smaller sample volume than the previous POC device, and allows for room-temperature reagent storage. “That was really the kicker for us. It allowed us to move to POC coagulation testing in home health and hospice,” Franks says. Earlier, the laboratory hadn’t made that move because the former POC reagents still required refrigeration; “we felt we had the potential to compromise the integrity of that reagent if our home health department had to transport it to and from patients’ homes.”
There can be differences of opinion over the appropriate POC testing technology, Franks reports. Whether to transition from serum bilirubin to transcutaneous bilirubin screening for newborns, for example, involved some debate for the POC program at Methodist.
“We have a POC technology committee here, so it reviews all requests for new or expanded POC testing and accepts or rejects it. Then the request goes to the executive committee and gets a final blessing from the patient care committee after that. We did our evaluation and realized it would be of benefit to have the Respironics Bilichek device in our healthy newborn arena. We determined it was not a device we wanted for high-risk babies or home health, but it was a good screening test to identify babies at risk.”
Initially, nursing wanted to perform the testing. After discussion with nursing, however, “because the tests are kind of expensive and we did not have connectivity to enter results, we thought we should bring them into the lab. The nurses can order TcBs on our computer system just like CBCs or newborn screens, and that dispatches our phlebotomy team to the nursery to collect specimens. It’s still a POC test because it’s done at the patient bedside; it’s just that the operator is from the phlebotomy team.” The phlebotomists, she adds, love the test because they don’t have to stick the infant.
It’s worked out very well, she says. “In 2001, the Joint Commission sent out an alert that hospitals were sending home babies with high bilirubins because they were discharging them too early or missing the diagnosis and were having bad outcomes. And we had a couple of babies that had to be readmitted shortly before we implemented our protocol. But since we started using TcB we haven’t had any babies readmitted.”
The POC testing program at Methodist has made a particularly dramatic difference in the turnaround time for creatinines, tests that are crucial to making sure that the kidneys of patients receiving CT scans will be able to clear the contrast dye from their systems. With three CT scanners in house, Methodist has a high volume of scans both on the inpatient and outpatient side, says Tom Brennan, CNMT, service leader of the CT department at the hospital.
“CTs became more of a front-line diagnostic tool over the last decade because there’s so much more information that can be obtained from a CT versus a straight x-ray. So our department has grown and grown and grown. Sixty percent of all patients seen in the ER have a CT, and close to 90 percent of all CT scans require a contrast agent,” he points out.
“The way it used to happen,” Franks says, “is they would collect the specimen in the CT department and tube it to the lab. But because of new process changes in the core lab, we were unable to process those creatinines in a timely manner. So instead of a 20-minute turnaround time, we were having backups of an hour, while the patient waited in the radiology room and couldn’t drink anything, and the hospital couldn’t do anything with the patient.”
The core lab and radiology were ready for POC testing. “We know why they need those results in a hurry, and we were tired of getting those phone calls. So everyone was truly in agreement that that was where we needed to go.”
The laboratory trained a small group of operators in the CT department, so all RNs who start IVs pull off a sample for creatinine at the time they start an IV. “They have the i-Stat device right there and they perform the creatinine. It’s a two-minute assay, and by the time they are visiting with the patient about the procedure, they’ve already got the results and docked the meter.”
Having previously brought in i-Stats for cardiac surgery, the laboratory had an IT interface in place. “We were able to do creatinines, capture those results, and shuttle them through our interface to our Cerner Millennium HIS right away. On our core lab side, once the creatinine hits Cerner, then Cerner performs the eGFR calculation; the i-Stat works the same way, so each POC creatinine is reported with an eGFR.”
As a result, the patients no longer have to wait an hour, and there is an added benefit, she says. “Our physicians can call at 4 PM and send an add-on outpatient for a CT, and now we can accommodate them. So it has made everybody happier.” In fact, agrees Brennan, “It was probably the single biggest improvement we made in the 11 years I was involved in CT.”
In this instance, POC testing was the right answer, though Franks emphasizes that isn’t always the case. “There’s a misconception that if you’re having trouble getting results from the lab, throwing POC at it is going to fix it. But the decision to implement POC is dependent on the internal dynamics of each facility, so what works for one location is not always a perfect fit elsewhere.”
In 2004, for example, the hospital’s head and neck specialty surgeons requested that the laboratory institute point-of-care intraoperative PTH testing, Franks says. “When we finally looked at the data and shook it out, we decided we were going to leave it in the lab,” where it’s been very successful. Similarly, Methodist Hospital does not have POC cardiac markers in the ER. Nevertheless, she says, if someone comes to Methodist with a heart attack, the data would show that he or she will receive excellent care.
There are many reasons for that, since factors like connectivity, operator training, error prevention, and quality control are all central to a POC testing program’s success. But Franks and Dr. Williams agree that if there is a key strength to Methodist Hospital’s point-of-care program, it is a culture that encourages working as a team across hospital departments.
Anne Paxton is a writer in Seattle.