November 2011
Feature Story

Anne Paxton

Outside a Petri dish, a “culture” can be a pretty intangible thing; it often refers to the attitudes, beliefs, and customs collectively shared and followed by a group, consciously or not. As former IBM CEO Lou Gerstner famously said, “Culture is what people do when no one is watching.” But when it comes to protecting patients in the hospital, the presence or absence of a culture of safety can make a very tangible difference, one that can be measured in concrete units related to potential patient harm: error rates, length of stay, morbidity, and mortality.

In promoting that culture of safety, pathologists should lead, say Michael B. Cohen, MD, and Emily E. Volk, MD, both members of the CAP Public Health Policy Committee. Addressing an audience at CAP ’11, they argued that pathologists should play a role beyond the laboratory as patient safety advocates who engage in process improvement for patient care.

“Pathologists know they are in a critical place to really influence patient safety,” says Dr. Cohen, professor and head of pathology at the University of Iowa, in an interview with CAP TODAY. “And you really need to move your thinking from the local milieu, the clinical laboratory, to a higher plane, the entire hospital. The whole patient care experience needs to be considered as you go about making process improvements.”

A teacher of biomedical ethics and an examiner for the national Malcolm Baldrige Award, which is given to institutions for excellence, Dr. Cohen says adverse events in the hospital can be compared to the “perfect storm” portrayed in the 1997 book by Sebastian Junger. If there is a bad outcome, it’s usually because multiple mistakes have been made. “You could have done everything right in one area, but typically what happens is that there are a series of errors that occur, just as in a piece of Swiss cheese sometimes all the holes line up.”

In a case study he and Dr. Volk described in their presentation, transfusions were mistakenly given to an elderly patient because of three or more failures: a wrong assumption was made about the transfusion trigger, automated systems were down, resulting in a false-negative antigen test, and there was a failure to recognize a transfusion reaction. But pathologists, Dr. Cohen says, may often be prone to thinking about correcting one thing, for example, the false-negative antigen test. “If pathologists want to truly focus on patient safety, they have to keep the whole process in mind.”

Perhaps surprisingly, there is still considerable room for improving the most basic of safeguards. Despite dramatically successful initiatives at some institutions since the publication of the influential Institute of Medicine report, “To Err Is Human,” in 1999, “We’ve barely made a dent in errors in patient identification,” says Dr. Volk, who practices with Clinical Pathology Associates and is medical director of the Department of Pathology and Laboratory Medicine for Baptist Health System, San Antonio.

Some solutions, such as requiring two patient identifications in every encounter, have helped. “But,” Dr. Cohen says, “if you live in a city where you have a particularly common name, say, ‘Cohen’ in New York City, or ‘Fernandez’ in Texas, or if the patient doesn’t speak English or comes in incapable of communicating, then it can be very challenging.” At Dr. Volk’s five-hospital system, with an average daily patient census of 750, “We often have patients with the same name on the same floor, and we’ve seen situations here where two patients with the same name have the same birthday,” she says.

Implementing bar coding, Dr. Cohen says, is still a work in progress. “It’s being implemented in multiple areas just in the pathology laboratory, but there are plenty of additional opportunities, and I think it’s a particular challenge for smaller hospitals and smaller group practices to be able to do it, because it’s not cheap.”

But pathologists are in a great position to advocate for investing in the technology to prevent patient misidentification, Dr. Volk believes. In her experience, bar coding and near-patient label identification can reduce errors to near zero. “This is the technology you use at Hertz Rent-a-Car, where they scan your car and the bill prints up right in front of you. We’re in the process of bringing this technology onboard, and as far as I’m concerned, it can’t happen fast enough. We should offer this to our patients in any health care setting where it’s feasible.”

Another simple technique in anatomic pathology is the use of different colored inks, as long as they do not interfere with fluorescent studies. While it hasn’t been mandated, “I think it’s becoming more and more common to use different colored inks when handling core biopsy specimens that grossly look very similar and could be easily mixed up,” Dr. Volk says. “That way, if in histology they pick up the wrong block and place the tissue in a slide that’s labeled for another patient, we have the opportunity to correct it.”

Most hospitals, faced with a bad outcome or what is often referred to as a “near miss,” will typically launch a root-cause analysis, Dr. Cohen says. “They will get a few people together with varied expertise to work on a relatively short time frame to really understand what happened and recommend changes to avoid another such episode.” Pathologists are not routinely called in on these analyses, but he argues they should be. “For example, in a wrong-side surgery, it’s highly unlikely that the pathologist was involved, but if the pathologist is advocating for patient safety, he or she could well be asked to participate in the root-cause analyses.”

Reports of bad outcomes or near misses are often not well received within the hospital, Dr. Cohen notes. “But you can learn a great deal from a near miss. Obviously you want a culture where these incidents are brought to the fore without any concern about retribution.” Institutions known for their culture of safety, such as Seattle’s Virginia Mason Medical Center, have seen a significant increase in the number of patient safety network reports, including reports from physicians, which have gone a long way toward correcting potential safety problems.

Hospitals can borrow from the airline industry, which has made flying much safer than being a patient, Dr. Cohen says. “Atul Gawande, in his book The Checklist Manifesto, makes a very compelling case that health care can learn much from the elaborate checklist pilots go through before taking off.” When putting in a central line, for example, “you need to follow a sequence of steps in a prescribed order, including multiple components of sterile technique that need to be done first. You want to make sure you do all the steps, and using a list instead of doing it from memory is a good way of ensuring this.”

In fact, Dr. Volk says, lessons can be drawn from the airlines at even earlier stages. “One of the first things they evaluate in the flight safety literature is the decision to fly.” The transfusion case study she and Dr. Cohen described illustrates how this concept could apply in the hospital. “Retrospectively, it is easy to criticize somebody else’s decision made in the heat of the moment, but this was a really unfortunate example of where the decision to transfuse was probably the first mistake.”

Historically, clinicians were chastised for ordering single units, but there has been a cultural shift in that thinking and blood transfusions are now more commonly treated like a liquid transplant, she says. “Now, when I talk to my colleagues who order transfusion, I recommend that doctors consider each unit they’re planning to transfuse as an individual and separate decision that has its own set of risks, its own set of complications, and its own ability to modulate the patient’s immune system.”

Transfusion safety is the only part of the laboratory where pathologists have an immediate direct therapeutic impact, Dr. Volk says, “but any laboratory test we perform has the potential to provide good or bad information on a patient, and if it’s the wrong patient, there is potential for great damage to patients.” With about 70 percent of medical decisions being made on the basis of information from the laboratory, pathologists and other laboratory professionals are front and center in caring for patients. “But it also translates into a tremendous amount of responsibility to make sure that impact is safe and accurate.”

Baptist Health System has partnered with HPI Consulting Group, a company that helps hospitals embrace a culture of safety. “They have had everyone in the hospital, from housekeeping to dietary to nursing, go through four hours of safety training, and every physician on staff was required to get an hour of safety training,” Dr. Volk says. “The concepts stressed were that it is a team effort to take care of patients, and we need to move away from the older paradigm of the paternalistic ‘doctor-in-charge’ kind of atmosphere.”

Dialogue is necessary, and narrative is a powerful tool, she believes. “Gawande talks about the elimination of nonchalance and the avoidance of silent disengagement in The Checklist Manifesto. I think pathologists can also play a role in eliminating nonchalance by participating in daily safety huddles, which we’ve started doing in our laboratories and throughout the hospitals. We share in our area any safety issues that may have come up, then the representatives from the different clinical areas—nursing, the laboratory, surgery, ER, etc.—have huddles with administration, so any standout safety events that occur in those areas are shared across departments.”

In line with Dale Carnegie’s central premise in How to Win Friends and Influence People, just calling people by name is important, because it eliminates anonymity, Dr. Volk says. “In laboratories, we run the risk of having our medical technologists, phlebotomists, and laboratory supervisors working in a sort of disconnected way. I truly believe people are more likely to do quality work and to pay attention when they know they’re not working in an anonymous fashion.” In the safety huddles, for example, “everyone has the opportunity to speak up. If staff has seen something that isn’t quite right, then they’re far more likely to raise their hand and speak up if they know the medical director knows their name.”

Eliminating nonchalance is an essential component of the culture of safety, she adds. “Every time we draw somebody’s blood, we’re essentially putting a bullet in the chamber of a gun. It’s a momentous event, and a potentially dangerous act, so we need to eliminate any nonchalance about it.” Pathologists can also work to discourage “shotgun” laboratory test orders, such as “Factors I–XIII, stat,” when clinicians are faced with a patient with a potential bleeding disorder. “Pathologists can work to develop algorithmic approaches that match the clinical presentation and the patient’s history,” Dr. Volk says. “Because every time you order a test, you run the risk of getting potentially misleading information, and going down a rabbit hole chasing something with no meaning to that patient, leading to interventions they didn’t need.”

But in ordering laboratory tests as well as transfusions, “some clinicians are just not up to speed with the more conservative approaches that are being encouraged.” As a JAMA article noted (Bennett-Guerrero E. 2010; 304:1568–1575), transfusion-ordering practices in just one clinical setting, cardiac surgery, varied from less than 10 percent of patients to 90 percent-plus over the course of a year.

Anticipatory policies and procedures are means of making a preemptive strike on potential errors. Dr. Volk refers to the concept of “high-reliability organizations,” which have very, very narrow windows for error. An aircraft carrier would be an example. “Medicine should be considered a high-reliability organization. But the reality is that our window for error is much wider than an aircraft carrier’s. We need to think ahead to situations that may occur that may place somebody in danger, then build our procedures around those situations to anticipate these dangers.” Or, as Dr. Cohen says, “We need to design processes that do not allow for error.”

Dr. Volk’s hospital followed the advice to anticipate and build procedures accordingly in setting its policy for platelet transfusion in someone who has TTP (thrombotic thrombocytopenic purpura), a rare condition that occurs when small blood clots suddenly form throughout the body, using up large numbers of platelets. “When there’s a combination of abnormal red cells called schistocytes on a smear with a low platelet count, we have the laboratory call in the pathologist to look at it right away, to avoid a potentially deleterious transfusion. Our policy was based on experience and a near miss, and we wanted to take advantage of that lesson.”

As experts in laboratory testing, pathologists are in a wonderful position to help build evidence-based order sets, and this is particularly powerful in the computerized physician order entry arena, she notes. Through careful, regular review of the laboratory test menu, use of a formulary approach, and attention to established test panels, pathologists can have a positive impact on laboratory use.

At Dr. Volk’s hospital, pathologists were not consulted when the original order sets were built. “So we had ‘free Dilantin’ put on an order set for seizure workups, rather than regular Dilantin.” But free Dilantin is for patients with abnormal serum proteins and is not performed in-house; it takes 48 hours to get a result from reference labs. “So clinicians unintentionally increased length of stay for patients who had seizure disorders who were on Dilantin.”

Happily, after a clinician complained about the slow turnaround time, “I was able to discover the root of the problem, and we got ourselves on the working group for computerized physician order entry. So even though I never enter orders, it turns out that my pathologist colleagues and I had a lot to offer in that process.”

Heading off interruptions is another important safety technique, Dr. Cohen notes. “An everyday example is when you’re driving and you get a call on your cellphone and you miss your turn. If I’m looking at slides under the microscope and get interrupted by a phone, that’s a distraction and I may miss something.” While it’s difficult to know how many errors in AP are attributable to distractions, says Dr. Volk,“it’s safe to say that a lot of folks working in AP are in distraction-filled environments.” Again borrowing from Gawande’s Checklist Manifesto, she says the sign-out room should perhaps be seen as an airline cockpit. “During takeoff and landing, they don’t talk about what their kids did last week; they talk only about getting that up in the air and landing it safely. We should do the same when performing frozen sections.”

Mandatory second review can add to the culture of safety, Dr. Cohen says. “None of us is infallible. So, medicolegal issues aside, we have a bylaw within the hospital that requires review of external pathologic material, especially in anatomic pathology, prior to definitive therapy at our institution.” For example, if a woman is referred to his hospital for chemotherapy for breast cancer or a man for prostate cancer surgery, “we’d like to see the original biopsies to confirm there is cancer there.”

There is evidence that mandatory second review is helpful, Dr. Volk agrees. “Often by the time the patient gets to the second hospital they have more clinical information and may be able to make more informed or refined diagnoses. How much diagnostic error is out there I think is debatable and probably an extraordinarily controversial topic.” To avoid becoming afterthoughts, secondary reviews should always be conducted with the same concentration and discipline as the primary review, she says.

But Dr. Volk thinks second reviews can be important checks on what her colleague, Daniel Mais, MD, has called “eminence-based pathology.” “If the initial biopsy was read at an institution considered academic or a highly regarded tertiary care center, then a set of pathologists who may be affiliated at a community hospital may have a tendency to be more reverent to the earlier diagnosis when they review the case. It’s a human thing, but we need to be careful that we try to be objective and do the second review in a blinded fashion so we’re not overly influenced by who did the diagnosis.”

Creating a culture of safety should be considered part of the College’s transformation project, Drs. Volk and Cohen say. “It folds in beautifully with the concepts the College is promoting,” Dr. Volk says. “I don’t think it’s established who in the hospital has the most control over safety procedures. Depending on the hospital, it’s a void waiting for a leader.” Work at the microscope is as important as ever, she adds, but pathologists should also make themselves known as safety advocates in the hospital beyond the laboratory.

Says Dr. Cohen: “Pathologists need to get out from behind the microscope and start seeing, at the institutional level, how they can help enhance the protection of patients.” For pathologists, improving the culture of safety, he says, is both opportunity and obligation.