Medication errors: Stressed nurses fear they will make mistakes

Pharmacists have important role to play in prevention

Betsy Lehman’s death was as ironic as it was horrible. She was, after all, a health reporter for The Boston Globe. And she died in one of the many famous hospitals that stand in a solid wall of medical competence around Boston’s Longwood Avenue. How could Betsy Lehman have died from a medication error? Looking back, it was frightfully easy.

Lehman, 39, was being treated for breast cancer at the Dana-Farber Cancer Institute, where she was one of four patients on an experimental protocol that called for high doses of cyclophosphamide.

Unfortunately, Lehman and one other patient got doses of cyclophosphamide much higher than what was called for in the study — four times higher, in fact.

According to attorney Robert Pavlan, who represented one of the pharmacists in the case, at least a dozen medical professionals — pharmacists, nurses, and doctors — saw the protocol schema or summary sheet. All were sufficiently confused so that they did not catch the error, he maintains.

The schema was written in a fashion that seemed to support unusually high doses of cyclophosphamide, Pavlan says. It listed three other drugs prescribed as part of the experimental protocol in terms of daily dosages. The confusion occurred when it was assumed the prescription for the cyclophosphamide was a daily dose, when in fact it was a single amount prescribed over a period of days.

For all involved, it was a tragic miscommunication. And though the pharmacists were not alone in letting the poisonous doses go by, they have largely shouldered the blame in the court of public opinion and, Pavlan says, are the only professionals thus far to have suffered any official sanction.

The Massachusetts Board of Pharmacy recently reprimanded three of the pharmacists involved in the Dana-Farber poisoning, a decision Pavlan says was made for public consumption. "The Board wanted to avoid adverse press," he says. The trio of pharmacists kept their licenses, however, because the board
recognized the difficulty in evaluating cancer chemotherapy dosages in a hospital that’s big on using experimental protocols.

The Massachusetts Nursing Board is currently investigating 16 nurses in relation to the incident, and the state’s medical board is wrapping up its investigation of the doctors involved, as well.

While Pavlan argues that pharmacists are simply agents of the physician, the Board felt otherwise — essentially holding the pharmacists responsible for not reading the protocol before dispensing the cyclophosphamide. The correct dose of cyclophosphamide appeared about half-way through the thick protocol document. But Pavlan, who also is a pharmacist, terms it "ridiculous" to expect, in a practical sense, that one of the pharmacists would have stopped and read the full study protocol.

Lori Bassinger, executive director of the Board of Registration in Pharmacy in Massachusetts disagrees. "I think that is unfortunate, that [reading the protocol] is not seen as part of the pharmacist’s duty," says Bassinger, who adds that the board’s position is essentially that the pharmacist must do whatever is necessary, even reading a long protocol, to get the dosage right.

But adding to the confusion in the Dana-Farber case, Pavlan says, is the existence of other protocols that allow for high doses of cyclophosphamide for two and even three days at a time. "If the drug had been digoxin 0.25 mg q.i.d. x 5 days, then I would’ve said those pharmacists should have known better," he says.

Competence usually isn’t the issue

Few things sink as deep a pit in a pharmacist’s stomach as making a medication error, especially one that harms a patient. Competent pharmacists remember such incidents for years. "In pharmacy, we beat ourselves over the head with the errors we make. We’re almost neurotic," says Stacy Wiegman, PharmD, MS, a fellow at the Institute for Safe Medication Practices in Warminster, PA, a clearinghouse for medication error reporting.

"The worst errors usually occur with the most experienced personnel because you get into a habit," Wiegman says. For that reason, the Institute urges compassion for employees who err. "By focusing on the individual for errors, we’re punishing the best and brightest in the medical field," Wiegman says, which does nothing to correct a systems problem.

When it comes to medication errors, pharmacists must be vigilant, even outside their pharmacy walls. In fact, one looming "systems problem" is a surfeit of tired, overworked nurses, says Judith Schindul-Rothschild, PhD, RN, CS, associate professor of nursing at Boston College in Boston and author of the largest job satisfaction survey ever taken among RNs, published in the American Journal of Nursing (AJN) in November 1996.

"We did see veteran nurses writing to say they were making [medication] mistakes for the first time," Schindul-Rothschild says. "What shocked us was a marked increase in the number of nurses who say they are going to quit nursing."

Respondents said a big reason for wanting to give up a nursing career was the fear of making medication errors, Schindul-Rothschild says. "Nurses are working so hard; there’s so little time to physically take care of themselves that the setting itself is setting them up to make mistakes when they never have made them before in their career."

About a third of the nurses responding to a recent AJN survey reported an increase in medication errors, but the situation is much worse on subacute floors, where 47% of nurses claimed medication errors had increased.

"Managed care contractors have capitated rates to hospitals," Schindul-Rothschild explains. "So it’s common these days to see a hip replacement stay on the acute care side as briefly as possible, because on the subacute care side you can bill [the insurer] at a daily rate. That’s how hospitals make money."

Subacute care floors tend to have lower staffing levels than critical areas of the hospital, and to make use of more unlicensed personnel, Schindul-Rothschild says. "Patients fare much worse in the subacute setting," she says. "There’s a very high readmission rate."

While nurses can take measures to decrease medication errors, the bottom line is that every pill or IV injection has to be given by a walking, talking human being, Schindul-Rothschild says. "If those people are tired, they make mistakes."

How pharmacists can reduce the risk

Wiegman says there are numerous systems corrections hospital pharmacists can make that will minimize risk, both for themselves and overworked nurses. They include:

• Standardize and simplify.

"Any time you can knock a step out of a procedure, you can decrease medication
errors," Wiegman says. "We always advocate standardization."

• Keep a tight formulary.

The fewer drug choices you have, the less likely a mistake will occur.

• Responsibly purchase products.

Buy generic stock drugs, such as sodium chloride, potassium chloride (KCl), and sodium phosphate, from different vendors, Wiegman recommends. That way, the labels for each will have a distinctive look. And don’t play the penny-wise and pound-foolish game when it comes to stock injectables. After all, a liability lawsuit costs a lot more than a different brand of KCl.

• Staff education.

Staff education is not the strongest component in mistake prevention, but Wiegman says it’s essential when a new drug is to be introduced into the hospital. So always hold an inservice educational session when a new drug wins a spot on the formulary.

• Remove potassium chloride concentrate from nursing units.

Yes, the nursing staff will moan when you come collecting the KCl, but Wiegman says you can prevent a major cause of medication death by restricting access to the powerful drug. Why pick on KCl? Because nurses sometimes confuse Lasix and potassium, knowing that the latter counteracts side effects of the former. Instead of injecting Lasix, the nurse will inject potassium. If they really put up a fuss, you can deliver pre-mixed bags containing KCl.

• Get familiar with useful software.

Seems obvious, but Wiegman says there are hospitals that have drug interaction programs they haven’t learned to use. Also, strive to integrate as much patient information into your software as possible. Elements such as a patient’s height and weight may seem like trivial pieces of information, until you start dosing drugs based on body surface area.

• Set rules on sampling.

The pharmacy should act as the sole gateway for drugs, so if a sales rep is giving samples to physicians in the hospital, the pharmacy ought to know about it. Wiegman knows of several cases in which a new intravenous drug packaged in foil, Mevacron, was mistaken for IV metronidazole. "The packaging had been associated with metronidazole forever," she says. "In one hospital, three incidents happened in one night." How did the drug get into the hospital? A pharmacy sales rep dropped it off.

• Cooperate with nurses.

Nurses now not only hand out medications, but in some hospitals they act as the respiratory therapist or the physical therapist (or both). In other words, they’re busy — extremely busy. If overworked nurses have become a focal point of potential medication errors, as the AJN survey suggests, then it is important that hospital pharmacists be proactive in seeking ways to ameliorate potential problems. Make sure your drug instructions are simple, easy to read, and unambiguous. Standardize methods of distribution so as to avoid confusion. Consult with nurse managers periodically to identify any problems and to gain a better understanding of the work environment in which the drugs are dispensed.

• Be aware of potential problem situations.

Be vigilant, especially about product labeling. For example, several deaths have occurred because of misunderstanding of the packaging of Brevibloc, Wiegman says. In this case, there’s more medicine in an amp of Brevibloc than in a vial.

Habits can be deadly

"How many times has a doctor said, ‘Give me an amp of Brevibloc [during a code]’?" Wiegman notes. But an amp of Brevibloc contains a dose 25 times higher than that needed in a code. The true "code dose" of Brevibloc is packaged in a small vial containing 100 mg of the drug. The Institute for Safe Medication Practices’ data base lists five deaths associated with overdosing of Brevibloc.

Another recent labeling problem occurred with Camptosar, which listed both "20 mg per ml." and 5 ml on the label. Deaths occurred when patients were mistakenly given the entire vial in the belief it contained just 20 mg even though it contained 5 ml, Wiegman says. The manufacturer has since revised the label.

And there’s always the bugaboo of drug names that sound the same, the most famous recent example being Losec and Lasix. Merck eventually changed the name of Losec to Prilosec in the United States. Wiegman says a new plan by the Food and Drug Administration to short-circuit a sound-alike drug name before it hits the market will help manufacturers save millions of dollars.