Fast times at the front end—speeding up ED tests

July 2007
Feature Story

Karen Lusky

It’s no joke, says pathologist William Neeley, MD: “How many patients were ever cured or treated in a hospital emergency department waiting room?”

The answer is none, of course. Yet waiting seems to be the rule, notes Dr. Neeley, medical director of the Detroit Medical Center University Laboratories.

Detroit Medical Center and other hospitals are, however, finding ways to make long wait times in their EDs the exception—an effort that can improve patient care and satisfaction and, in some cases, the bottom line.

Detroit Medical Center was “awash in red ink” in 2003 when it decided to put up billboards all over town announcing the hospital system’s ED physicians would see patients within 29 minutes, Dr. Neeley said at the 2007 Executive War College sponsored by The Dark Report.

The medical center’s 29-minute guarantee, along with cost-cutting measures in a “multi-prong attack on finances,” put the eight-hospital system back in the black, Dr. Neeley says. “The ED initiative was a very significant part of that success.”

Lab business consultant Earl Buck, vice president of operations management for Chi Solutions, Ann Arbor, Mich., says, “Anything a hospital does to improve turnaround of all services in the ED will ultimately affect its customer satisfaction and its volume of hospital admissions.” And satisfied customers spread the word, drumming up more business for the hospital. All of which can have a positive impact on an organization’s finances, Buck says.

Detroit Medical Center, or DMC, implemented the faster ED turnaround time in six of its eight hospitals, including Detroit Medical Center Receiving Hospital, three other hospitals on its main campus, and two off-campus hospitals. Each of the DMC’s six hospitals with the ED initiative in place has its own protocol for complying with the 29-minute timeline.

Patients who enter the emergency department at DMC’s Detroit Receiving Hospital, a level one trauma center, are seen immediately by a triage nurse who completes a mini-registration and assessment that takes about three minutes, says Monica Marshall, RN, BSN, MS, executive director of emergency services for the hospital.

Based on acuity level, the patients then proceed to the appropriate treatment area and never back to the waiting area, she says. A registration clerk goes to the patients to complete a bedside registration. The patients who don’t require more immediate attention see a physician within 20 minutes. Once the doctor sees the patient initially, the 29-minute clock stops, Marshall says.

Dr. Neeley views the team effort required to comply with the 29-minute timeline as an interdisciplinary wheel where every spoke, including the lab, emergency physicians, specialists, nurses, imaging, housekeeping, admissions staff, and others, has to play its part.

And the laboratory gets high marks in that regard. The published turnaround time for stat lab tests at DMC is 30 minutes, Dr. Neeley says, except hCG, which is 15 minutes. But Marshall says the ED finds the actual turnaround time is 15 minutes for most tests. The laboratory is so fast, she adds, that blood glucose is the only point-of-care testing done in the ED.

The main campus hospitals’ stat labs handle all the stat ED testing except for cardiac troponins and drug screens, which the stat labs transport via pneumatic tube to the core lab. (Two of the eight DMC hospitals located several miles from the core lab do all of their own laboratory testing.)

Autoverification, which is done through the lab information system, helps speed up the time to lab results by automatically releasing results within the acceptable ranges for an instrument. The lab automatically releases 93 percent of its hematology results and 83 percent of chemistry results.

Automated tracking tells the laboratory exactly when the samples are ordered and arrive in the lab and when they are placed on automated analyzers and the results reported. “So we have all these figures and data and monitor it daily,” looking for and analyzing the root cause of outliers, Dr. Neeley says.

Outliers generally stem from a combination of factors. For one, ED staff, who are responsible for drawing lab specimens, may not put a sample in the pneumatic tube immediately, or the stat lab staff may leave it sitting in the tube carrier for a while.

“If the lab doesn’t deliver the samples from processing to the instrument,” that’s another bottleneck, Dr. Neeley says. And the technologists and technicians have to focus on problem results outside the acceptable range for the instrument, which is another potential bottleneck.

Automated sample tracking can improve service, Dr. Neeley notes. “It’s the areas where you can’t automatically time stamp the sample that are problematic.” For example, the nurse may think she put the sample in the pneumatic tube carrier at 5 PM, but it might have been 5:10 PM. Or the staff person in the stat lab may think he or she pulled it out immediately, but it might have been there for 10 to 15 minutes, he says.

To tackle those current no-data zones, DMC plans in the future to use radiofrequency identification chips on the specimen labels. An RFID reader will automatically record the time the sample is put into and removed from the tube carrier.

In the ED, bottlenecks in patient flow still occur, Marshall says, especially when the inpatient beds are full. But they have managed to meet the 29-minute guarantee 98 percent of the time. When the ED misses the mark, it offers patients compensation in the form of tickets to sports events or museums or, perhaps in the future, a gas card.

Avera McKennan Hospital and University Health Center, a level one trauma center in Sioux Falls, SD, revamped its ED as a Lean project in 2005 to ensure patients see a physician within 20 minutes of arrival.

But it’s actually doing better than that. An April 2006 survey of McKennan ED patients found that patients were seen by a doctor, on average, 12 minutes after arrival, said Leo Serrano, FACHE, CLSup (NCA), director of clinical laboratories at Avera McKennan, in his Executive War College presentation, “Putting the Patient First—If Toyota Ran Your Hospital.”

More important, he added, 12 minutes was also the mode or most common time frame in which patients saw a physician.

The hospital’s ED initiative catapulted patient satisfaction with the ED from the 38th or 39th percentile into the 90th percentile, where it’s been for the last year, Serrano says.

Like the Detroit Medical Center’s 29-minute initiative, McKennan’s faster ED patient turnaround has given it a competitive edge in what Serrano says is one of the toughest markets he’s been in during his 40-year tenure as a lab manager. Though South Dakota is sparsely populated, Sioux Falls has two large medical centers—McKennan and another level one trauma center—within two miles of each other. And thanks to the ED initiative, McKennan has picked up business because people know they will be seen faster in the ED at McKennan.

Patients walking into the ED at McKennan go right to a triage nurse who does a quick assessment and mini-registration and draws their blood for lab testing, applying bar-coded labels to the samples.

The patient is then shown to an exam room, where the patient’s bar-coded blood samples are placed in a rack. Only when the patient has been seen by the doctor—and orders placed—does registration come to the room to ask for the person’s billing and insurance information.

If the doctor doesn’t see the patient within 20 minutes, the ED uses the “standard symptom-based” protocol for ordering lab tests based on the conditions the triage RN evaluated, Serrano says.

The laboratory and emergency department developed the automatic lab protocols by studying lab tests ordered in the ED for various symptoms. “Amazingly, the plethora of ED physicians ordered the same thing for the same conditions, so it was easy to get them” to buy into it, he says.

As soon as the laboratory tests are ordered, by the physician or based on the protocol, the ED staff puts the patient’s blood samples in a high-speed pneumatic tube connected to the core lab, located almost a half-mile from the hospital. The tube system is so fast that the samples usually arrive in the lab within three minutes after the order is placed—or seven minutes if it’s a busy time with a lot of samples traveling through the tubes.

The laboratory staff who receive the samples scan the bar codes and then start to process them.

The lab can automatically track when the staff takes the sample out of the tube but not when the ED staff puts it in, which is a weakness in the system, Serrano says. And like Detroit Medical Center, McKennan is studying solutions to that problem, including two-dimensional bar coding and RFID.

The laboratory guarantees a 12-minute or less TAT for CBCs, 18 minutes or less for coagulation testing, and 24 minutes or less for chemistries and cardiac markers. The countdown for the TAT starts when the sample arrives in the tube station and stops when the results are available for the physicians in the ED. Everything is autoverified except for the CBCs, because the instrumentation doesn’t allow for the latter, Serrano says.

To prevent the core technologist assigned to production from becoming sidetracked, a “float tech,” who floats in and out of the core laboratory as needed, handles instrument downtime and other issues.

The ED does a limited amount of point-of-care testing, including urine dipsticks and pregnancy tests. The physician or nurse does the pregnancy test and interprets it, and then sends it by tube to the lab with the identifying labels for validation that it was interpreted accurately.

ED staff also uses an Abbott i-Stat to do POC testing for cardiac patients when the physician wants to get them back to the cardiac cath lab, and for trauma patients, Serrano says. But the lab’s TAT is so good, he adds, that it doesn’t make sense to do more POC testing given its high cost compared with tests done in the core lab.

While the patient’s lab work is being done, the patient has x-rays or other imaging studies. “Instead of doing things in a linear fashion,” the ED performs them concomitantly, which is a Lean concept, Serrano explains.

At one point, the ED had a bottleneck where patients couldn’t be admitted from the ED to the hospital quickly enough because rooms weren’t ready.

“We thought the issue was a housekeeping one,” Serrano says, but a Lean project revealed that the nurses weren’t notifying housekeeping quickly enough after patients left. Nurses now notify housekeeping within 15 minutes of a patient leaving. “And usually within 75 minutes of a patient leaving a room, the room is ready,” he says. “That has an immense impact on the time patients are in the ED.”

Future plans call for a redesign that will allow nurses to hit a switch in the room to notify housekeeping automatically when a room is ready for cleaning.

McKennan Hospital monitors turnaround times and outcomes, and hasn’t seen an increase in the number of discharged patients who are rushed back to the hospital or have a negative outcome, Serrano says.

The lab has also found that the lab protocols, employed only if the physician doesn’t see the patient within 20 minutes, are used five percent of the time.

In Indiana, Memorial Hospital of South Bend has used a few standardized lab preorders for certain conditions as part of its triage diagnostic protocols for years. But two years ago, the hospital’s ED implemented non-protocol preorders for laboratory tests and imaging, a change that expedites care and has improved staff and patient satisfaction, says David Halperin, MD, an emergency physician at Memorial Hospital.

The physician working in the minor treatment area next to the nursing triage area usually does the non-protocol orders, “tailoring them to the patient’s complaint, age, and risk factors,” Dr. Halperin says. Sometimes the physician will also order IV fluids and analgesics.

In some cases, once the patient is in an exam room, the ED “back team physicians” will also institute preorders if they expect the person to face a longer-than-usual waiting time to be examined by a physician, adds Dr. Halperin.

The preordered lab results are often back by the time the physician sees the patient. The physicians treating the ED patient can access a Cerner database for the laboratory results, which has archives of previous lab results because the ED treats a large population of repeat patients, Dr. Halperin says.

The physician who examines and treats the patient can add testing as needed. But having the lab results in hand often gives the physician a head start on making a diagnosis. Say an elderly patient comes in complaining of light—headedness, and his sodium is 115 mEq/L, Dr. Halperin says. “The doctor would still assess the patient but knows the cause of his symptom is likely the low sodium.” And while the physician has to investigate the cause of the low sodium, the physician can “jump right to it.”

Dr. Halperin estimates that physicians treating patients tend to order more tests about a fourth of the time. “Sometimes you run more tests to get more information,” he says, “but you can make enough of a decision so that the other labs ordered aren’t a rate-limiting step.” A simple example is when a physician does a non-protocol preorder for a patient with a fever that shows the person has an elevated white blood cell count, and the physician wants to add blood cultures before giving IV antibiotics.

Do the lab preorders drive up lab utilization unnecessarily? It might do so if the ED had an essentially well population of people frequenting it for colds and minor traumas, Dr. Halperin says. “But our patient population is pretty sick. The more ill the ED population, the more efficient the utilization.”

Dr. Halperin doesn’t have data on how the preorder model affects the patient throughput. But “right now everyone likes preorders” not only because the approach is more efficient, he says, but also because it “changes patients’ perceptions of wait times. The patient perceives he’s waiting to get lab [tests] run.”

The non-protocol preordering also helps identify ill patients sooner based on their critical lab results, he adds.

The other hospital in South Bend is planning to move to an area outside of the city. And the “non-protocol preordering appears to be a useful tool as we ramp up to be the only hospital left in South Bend,” Dr. Halperin says.

Memorial Hospital of South Bend is in the analysis phase of a Six Sigma project to reduce cycle time and throughput in the ED, says Janet Sipp, a nursing educator who is a member of the Six Sigma team. So far the team has found that lab TAT and the first time the doctor gets back to see the patient affect cycle time, as does the time when the nurse gets back to the patient room after the doctor leaves to do discharge teaching and discharge the patient.

The team has yet to determine the current lab turnaround time in the ED and may put reducing TAT on hold until the hospital transitions from using an independent lab to its own lab this summer.

Whether lab service turnaround times affect patient throughput in an ED is “site specific,” says Chi Solutions’ Buck. For example, a Six Sigma study in which he was involved created a stat lab in the ED to assess the impact on patient throughput, but it improved lab test TAT by only two minutes, he says. The study found that the lab TAT was not a critical factor in shortening the ED length of stay, Buck says. More critical in that setting was the availability of specialty physicians to come to the ED to see patients.

But a Six Sigma project last year at Alegent Health Bergan Mercy Medical Center, Omaha, Neb., significantly improved the lab’s TAT in the ED for cardiac troponin and the chemistry metabolic profile, says Judy VonSeggern, Alegent’s manager of lab quality assurance, compliance, and education.

The TAT was measured from the time the test was ordered in the hospital information system to when it was reported in the lab information system.

The Six Sigma team didn’t measure the impact of the improved lab TAT on ED length of stay, but VonSeggern says it’s “reasonable to assume” that better test TAT could be helpful in that regard.

The impetus for the Six Sigma study was physician dissatisfaction with the lab TAT for the identified tests, which was “quite variable with some being quite long,” says VonSeggern. The doctors wanted the results in 45 minutes, and at the outset, the lab was meeting that expectation only 56.6 percent of the time with an average TAT of 44 minutes. When the team had completed the Six Sigma project, the lab was hitting the 45-minute mark (order to result reported) 87 percent of the time with an average TAT of 34 minutes.

To achieve the improved TAT, the Six Sigma team made a number of changes aimed at reducing identified “variances” in the process that caused the long TATs:

l Assigned a phlebotomist to do only ED draws during the peak times in the ED from 10:00 to 22:00. The phlebotomist, who performs other duties during a slow time in the ED, carries a direct communication phone so the ED staff can reach the person. If the phlebotomist is busy and doesn’t answer, the call rolls over to the lab, which immediately dispatches someone to the ED to do the lab draw.

l Implemented color-coded pneumatic tubes that are used strictly for sending lab specimens from the ED. This provides a visual cue to let the lab staff know that a specimen is from the ED.

l Added a “stat” spinner in the lab for ED specimens.

l Reinforced the Lean processes already in place in the lab.

For future improvement, the lab extended training for phlebotomists to include five to 10 additional shifts in the ED.

“The laboratory and the ED have been very happy with the outcome” of the Six Sigma project, she says. And ED patient satisfaction with laboratory services improved also—“a nice byproduct.”

There’s another positive byproduct to such initiatives, says Chi Solutions’ Buck: “Collaborative efforts between the lab and ED show our health care colleagues the value the lab can bring to the delivery of health care.”