Errors spike as teaching hospitals get going

NEW YORK (Reuters Health) - When the new crop of interns and residents report for duty at the nation’s teaching hospitals each year -- patients beware. A new study found a spike in the number of medical errors occurring during the first few months of each new class’s training.

In the study, researchers examined five years of patient and administrative records for new anesthesiology trainees at Alfred Hospital, a university-affiliated hospital in Melbourne, Australia.

They looked at a total of 19,560 anesthetic procedures performed by 93 trainees in their first year of residency at the hospital.

The beginning of each academic year could be the worst time of year to be admitted to a teaching hospital, the international research team concludes in the British Medical Journal.

Medical errors jumped in the first month and gradually tapered off until the increase disappeared after the fourth month. Suspected for years, the phenomenon has a name -- the “July phenomenon” in the United States and the “August killing season” in the United Kingdom.

“Our data confirm the presence of a learning curve,” the investigators wrote.

They found that the rate of undesirable events was significantly higher in the early stages of a rotation, particularly for technical related skills, such as the placement of catheters, performing tracheotomies and getting oxygen to the patient’s brain during a procedure.

However, the error rate “also showed significant improvements throughout the year,” the investigators found.

The researchers focused on anesthesiology because the risks to patients from errors are inherently more dangerous than other specialties and because there exists a well-managed electronic system for reporting errors.

Based on their findings, Guy Haller, of Geneva University Hospital, Switzerland, and colleagues urge changes in hospital programs that train an estimated 132,000 doctors every year in the US and overseas.

They suggest staggering the dates when trainees begin their residencies and putting in place written standard working practices. More orientation training and time in simulations and closer one-on-one supervision even for the more experienced trainees might also help reduce errors.

In an accompanying editorial, Dr. Paul Barach of the Utrecht Medical Center in the Netherlands, notes that patients need to be aware of this problem, too.

In an interview with Reuters Heath, he said “as with all challenges around patient safety, patients must be aware of where the potential opportunities for harm are.”

“Ideally, we wouldn’t have any trainees in the system,” Barach said, but “all systems have to replenish and train the next generation.”

Barach cautioned that the findings from this one hospital in Australia may not apply elsewhere. In the United States, for instance, anesthesiology apprentices have one-on-one supervision during the first 4 weeks of their training.

The Melbourne study needs to be replicated in the U.S. and elsewhere, he said.

What the study does show, Barach noted, is that the problem is a “process or systems issue” that should be addressed across all disciplines, from nursing to pharmacology to surgery.

“The clinical systems are not designed to absorb new trainees therefore the challenge is how do we reshape the system to better accommodate these new trainees?” Barach said.

SOURCE: BMJ/British Medical Journal, online October 14, 2009 Online First Issue.