June 3, 2010— -- There is an old saying among some doctors -- do not let your friends and family schedule a surgery in July.
July is the month when graduates, fresh out of medical school, report to residencies in teaching hospitals. Anecdotally, at least, it's been a time when medical errors peak.
A new study decided to see if the so-called "July Effect" was real.
Researchers from the University of California at San Diego investigated more than 62 million U.S. death certificates between 1979 and 2006. Of those, 244,388 deaths were caused by a medication errors in a hospital.
Month to month, the statistics showed a relatively equal chance for a fatal medication error -- except at teaching hospitals in the month of July.
The study found that fatal medication errors spiked by 10 percent in July in counties with a high number of teaching hospitals, but stayed the same in areas without teaching hospitals.
The findings appear in the current issue of the Journal of General Internal Medicine.
David Phillips, the lead author of the study, said he was careful to rule out medication deaths caused by unexpected allergic reactions. He also didn't include medication deaths that happened outside of the hospital, which might be the fault of the patient, and which previous research has found is often related to mixing medication and alcohol at home.
"If you have people dying from medication deaths and nobody thinks there is an error, that doesn't spike in July. So there's something going on in teaching hospitals in July, and the most common thing people think of was residents starting," said Phillips, a professor of sociology at the University of California at San Diego.
Phillips said his study, which looked at computerized death certifications, had an advantage of looking at death certificates across the country, in a variety of hospital settings.
"With that we can safely generalize to the whole country," said Phillips, who researched the study with co-author Gwendolyn Barker. "The disadvantage is that you don't have very much detailed information per death certificate."
That means the study only suggests that new residents are to blame. Further studies looking at hospital records might find a direct link to residents' mistakes, or no link.