CAP TODAY Pathology/Laboratory Medicine/Laboratory Management
Anne Ford
November 2015—Blood is thicker than water, the saying goes. And thanks to a recent Q‑Probes, the rates of mislabeled specimens submitted for ABO blood typing and of wrong-blood-in-tube errors are now as clear as water.
The mislabeling rate hasn’t changed much since a similar Q‑Probes study was performed in 2007. The 2015 Q‑Probes, “Blood Bank Safety Practices,” reviewed 41,333 specimens and found that 306, or 0.74 percent, were mislabeled. The previous study looked at 112,112 specimens, of which 1,258, or 1.12 percent, were mislabeled.
“That’s a little unfortunate,” says study coauthor Glenn E. Ramsey, MD, medical director of the blood banks at Northwestern Memorial Hospital and Lurie Children’s Hospital in Chicago and professor of pathology at Northwestern University’s Feinberg School of Medicine. “We’d like to see a greater improvement in the error rate.”
That frustrating discovery was just one of the latest study’s findings regarding safety practices in blood banking. Among those findings are the importance of correct first name, last name, unique identifier, and date of birth as labeling elements, and the role of dedicated phlebotomists in collecting the samples. Adverse factors include the prevalence of specimen relabeling, reuse of labels, and incorrect patient registration.
There is an opportunity to improve blood collection practices, Dr. Ramsey says. For example, in addition to a larger reduction in the rate of mislabeled specimens, he would like to have seen an improvement in the rate of wrong blood in tube. But the 2015 Q‑Probes WBIT rate of 0.043 percent and the 2007 study’s WBIT rate of 0.04 percent are virtually identical.
Paul F. Lindholm, MD, lead author of the latest study, says, “These seem like very small numbers, but they represent very undesirable outcomes. Wrong blood in tube has been a problem for many years, and it’s vexing to figure out how to get past this .04 percent number. How can we get to no wrong blood in tube? That’s what we’d like to achieve.” Dr. Lindholm is an associate professor of pathology at the Feinberg School of Medicine and director of coagulation and hemostasis and co-director of transfusion medicine at Northwestern Medicine.
One potentially promising finding pertains to date of birth. When it was used as an element of a specimen label, as it was with 87 percent of inpatient specimens and 90 percent of outpatient specimens in the 2015 Q‑Probes, it was correlated with a mislabeling rate of 0.152 percent. In contrast, not including date of birth was associated with a mislabeling rate of 1.26 percent.
“I’m not sure what it is about date of birth exactly,” says Dr. Ramsey. “In both studies, having the date of birth involved either on the requisition or on the labels seems to be correlated with lower mislabeling rates. This is not something that’s been pointed out in other publications I’m aware of, so perhaps it’s some kind of correlation with other measures that those laboratories are taking.”
Then, too, more laboratories appear to be embracing the use of barcode readers to identify patients by their armband during specimen collection. In 2007, the percent of labs doing so was 8.1 percent; that number has now increased to 37.9 percent. “That’s certainly a positive step,” Dr. Ramsey says.
He’s also pleased with the trend in the number of institutions that perform two blood typings on patients without a historical type before issuing nonemergency, nongroup O RBCs. Where the 2007 Q‑Probes saw just 60 percent of participants requiring this, 90 percent of participants in the latest Q‑Probes do so.
“In the previous study, there was a trend toward better outcomes or lower wrong-blood-in-tube rates in facilities that were doing two typings on first-time patients,” Dr. Ramsey explains. “It wasn’t statistically significant, but it was a trend. Now almost all the facilities are doing this. This is a positive step toward helping reduce the rate of mislabeled specimens—making sure that that first sample, which you’re going to compare subsequent samples to, is the correct type.”
For his part, Dr. Lindholm is encouraged to see an uptick in the number of institutions that prohibit the practice of producing and saving specimen labels for future use. “There is a practice out there where apparently people will generate labels, and they will put them in their pocket, and then they’ll label a tube with labels that they got out of their pocket. It’s bad, because it leads to specimen mislabeling,” he says. Fifty-three percent of participants in the 2015 Q‑Probes prohibit this practice, whereas less than 32 percent in the 2007 study did so. “But still,” he adds, “this is a practice that is not good, and yet it still seems to be done.”
Another less-than-ideal practice that continues is the collection and labeling of blood bank specimens by nonlaboratory personnel. In 2007, 93.5 percent of participants reported this practice; that number has now dipped to 86.7 percent. Says Dr. Lindholm: “The data show that if a specimen is collected by a nonlaboratory person, someone on a floor, the specimen is far more likely to be unsuitable, mislabeled, or both. This is another area of opportunity to improve.”
Dr. Ramsey speculates that the continued prevalence of nonlaboratory personnel collecting specimens may have to do with the increasing use of bedside electronic systems for patient and sample confirmation. “If these systems become more widely available, then perhaps that will be seen as a measure that would be able to improve the accuracy of the samples without having to have dedicated staff do it,” he says. “I don’t know that we’ve done the correlation in our data, but I suppose it’s possible. Hospitals have to decide how their resources are allocated, and perhaps in some cases it’s perceived as more expensive to have a phlebotomist staff member do it than to have the care teams do it.” Still, he adds, most hospitals that use dedicated phlebotomists see a much lower rate of specimen collection problems.