Even at Massachusetts General Hospital in Boston, one of the best hospitals in the nation, medication errors occur in half of all surgeries, according to a study published Sunday in the journal Anesthesiology.
Previous estimates of medication errors in the operating room have relied doctors’ own estimates. When self-reporting, physicians reported errors in just 1 in every 133 operations.
“I think the rate itself a lot of people will find surprising,” said Dr. Karen Nanji, M.P.H., an anesthesiologist at Mass General who co-authored the research.
Nanji and her colleagues watched 277 surgeries from start to finish, witnessing medication being administered more than 3,500 times. They saw 193 errors, one in half of all operations, and 1 in every 20 times medication was given in the course of surgery.
Two-thirds of the errors were serious and 2 percent were life-threatening. In one error the researchers described as typical, nurses incorrectly set up an intravenous (IV) drip, so that a medication that should be dripped out slowly could have rushed out all at once into the patient’s vein, potentially resulting in death.
In another, doctors and nurses failed to act as a patient’s blood pressure dropped dangerously.
One patient with a known penicillin allergy was given a similar drug and developed a rash.
“Understanding a patient’s history of reactions to drugs and antibiotics is something that in 2015 we ought to be getting right,” said Dr. Allan Frankel, the co-chief medical officer of Safe and Reliable Healthcare and a former practicing anesthesiologist. Frankel was not involved with the study.
The most common errors that led to patient harm were wrong doses, omitted medications and delayed responses to changes in the patient's vital signs, such as low blood pressure.
But some of the errors were less serious, or caught in time. For example, a nurse might draw a medication into a syringe before realizing by its color that it isn’t the right drug.
“The reliability of the process should make it very difficult for a nurse to do that, but the fact that the nurse identified it is something to be heralded,” Frankel said.
He saw the researchers “drawing a line in the sand to say, ‘Can anesthesia can be made more reliable?’” Frankel said.
Dr. Dan Mayer, a retired emergency doctor who has worked on quality control, said that the researchers had set a high bar, but were probably right to do so.
“They were being picky,” he said. “About 100 of the errors that they found were either near-misses or with little potential for harm, but should we not count those? The whole culture of error reduction is to study near-misses and study what made them possible so they don’t become real errors.”
Earlier research has set the bar too low, he said. Even if the real number lies somewhere in between, it’s high enough that we should do something about it.
Systems Are Key
Fewer errors will require better systems.
In other parts of hospital care, there are already more checks and balances to prevent medication errors. But because surgery is so fast-moving, the operating room (OR) has remained out of reach for some of these processes.
To begin to reduce the number of errors that happen behind the closed doors of the OR, the researchers began the same type of work there that has brought errors down in other areas of Mass General and at other hospitals around the country.
“It’s like a 12-step program,” Mayer said. “The first step is you’ve got to admit there’s a problem.”
Some types of error can be eliminated altogether, such as giving medication to a patient with a known allergy. But mistakes are inevitable.
“It’s impossible to reduce all error,” Nanji said. With every new medication and technique that enters clinical practice, there’s additional opportunity for error.
Doctors and nurses are human, and humans forget to do something about 1 in 100 times and do something incorrectly about 3 in 1,000 times, Frankel said.
“The system has to be designed to adjust for our humanity,” Frankel said.
Good processes have to make it easy for medical staff to do their jobs efficiently while making it less convenient to skip over safety practices to improperly rig up an IV drip, for example.
Transparency, with Limits
Massachusetts General also showed a laudable commitment to improving care by being willing to air its dirty laundry in the study, Frankel said.
But Nanji said it wasn’t a tough sell to get Mass General to consent.
The hospital “has a long history of studying how to improve quality and safety so the medical community as a whole can improve.”
But what share of this transparency extends to patients?
Patients usually don’t learn of instances in which a dangerous mistake almost happened during their surgery. Nor do they learn about the mistakes that were made that could’ve led to a bad result but didn’t.
One patient who had a history of gastrointestinal bleeding was mistakenly given a nonsteroidal anti-inflammatory drug, which can cause bleeding. But because no bleeding took place, that patient likely wouldn’t be told about the error.
The patient who developed a rash from a penicillin-type drug would be told that doctors had mistakenly given the medication.
“The hospital has policies, all institutions have policies about reporting errors and adverse events to patients, and all these policies were followed by the hospital,” Nanji said.
Frankel didn’t see it as unfair to keep patients out of the loop.
“There was probably a handful [of errors] and not much more that would warrant a conversation with a patient,” he said.
To make informed decisions about where they’ll get the best care, patients should ask about their surgeon’s experience performing the procedure and working with the other personnel in the room. The hospital’s reputation for safety can convey to patients how reliable its practices are, Frankel said.