More than one-third of the observed errors injured patients, including three life-threatening mistakes, according to the study. Two of those were caught by the operating room staff, and one was intercepted by researchers. Nobody died because of the mistakes.
The error rate is much higher than what has been previously reported but in line with rates found in inpatient wards and outpatient clinics, researchers said. There have been a few studies about medication errors in the operating room but they relied mostly on self-reported data, which typically underrepresent true rates.
The study was published in Anesthesiology and presented last weekend to the American Society of Anesthesiologists, where clinicians told the authors that similar problems are occurring at their hospitals.
"There was not a lot of surprise because everybody knew the self-reported error rates were too low," said lead author Karen C. Nanji, an anesthesiologist at Massachusetts General Hospital. "We just didn't know what the true rates were."
She added: "Medication errors are at least as high at many other hospitals. The types of errors are not unique to Massachusetts General Hospital in any way. Most of them are around common medications that are used across the board."
The medications most frequently associated with errors were propofol, a commonly used sedative in the operating room; fentanyl, a powerful pain medication; and phenylephrine, a medication given to increase blood pressure in patients with very low blood pressure.
Researchers say the environment of the operating room is very different from that of other hospital settings, where there is time for pharmacists and nurses to double- and triple-check medications before they're given to patients.
"In the OR, because everything is happening so rapidly, there's not time for all those checks," Nanji said. "If a patient needs a medication, we need to give it in the next minute."
She and the research team observed the randomly selected operations during an eight-month period and reviewed the anesthesia records. Researchers documented every drug that was given immediately before, during and after the surgery.
They also documented any mistake in the process of ordering or administering a drug and any drug-related harm, whether or not it was caused by an error. An allergic reaction in a patient not previously known to be allergic to a particular medication was still included, for example.
One of the most common errors involved the labeling of syringes. Many medications are clear liquids, and syringes need to be properly labeled so the correct drug can be administered. Massachusetts General has a bar-code syringe labeling system, but "in our case, most errors were occurring where the bar-code scanning system was not being used," Nanji said.
On average, about 10 medications were given during an operation. The study found that some kind of error was made in about 1 in 20 medications, which equates to every other operation.
"We've never had the data to focus on errors before, and now that we have this information, we are looking at targeted interventions," Nanji said. One idea under consideration is to electronically document medications before they are given to patients to allow the electronic system to check for proper doses and possible drug interactions or allergies.
Patient safety advocates say it's a welcome step when hospitals conduct these kinds of studies and make the data available to the public. But the findings still took some by surprise.
"Boy, we still have a lot of work to do," said Tejal Gandhi, president and chief executive of the National Patient Safety Foundation. "If it happens at MGH, it can happen anywhere."