“An apple a day keeps the doctor away,” or so goes the old adage.
People who believe there’s some truth behind the saying are crunching on a great big Macintosh right now, or at least they will be once they hear about the results of a study from UC-San Diego which showed that every July, there’s a 10% spike in fatal medication errors in hospitals.
The scientists behind the study suggested that the phenomenon is partly attributable to the arrival of freshly-minted physicians who begin their medical careers in teaching hospitals around July 1st every year. The scientists dubbed the phenomenon the “July Effect.”
To reach these conclusions, David Phillips and Gwendolyn Barker reviewed death certificates from people who died in hospital settings between 1979 and 2006. They defined fatal medication errors as those for which the medication error was listed as the primary cause of death.
Death certificates don’t record whether a person died in a teaching hospital per se, but they do capture the county in which the death occurred. So Phillips and Barker calculated the percentage of hospital beds in each county that are affiliated with medical residency training programs and went from there.
According to the authors, “The greater the concentration of teaching hospitals in a region, the greater the July Effect for intra-institutional medication errors in that region.”
The link was supported by 2 striking observations they made during their review:
– The spike occurred every July, and only in July
– It was observed only in counties where teaching hospitals were present
The scientists considered other hypotheses to explain the July effect, but discounted them one-by-one:
Could the spike be associated with July 4 celebrations? No, the researchers concluded, because the holiday is celebrated nationally, yet the spike was observed only in counties where teaching hospitals were located.
Could the spike be caused by the widely recognized national increase in alcohol consumption during the summer, and an associated bump in fatal interactions between alcohol and certain medications? No, they said, because the spike was not seen in August.
Could the spike be caused by cyclical coding changes in hospital-created death certificates? Not likely, claimed the scientists, because their review of the literature on coding misclassifications failed to reveal any sort of spike in July or any other month, for that matter.
The authors did not address how or why the fatal medication errors occurred, although the reasons are likely to include overprescribing, incorrect dosing, failure to recognize drug interactions or early warning signs of drug allergies and side effects. Of note, e-prescribing systems that feature medication alert and reminder systems are designed to reduce such errors, and numerous studies have shown they are highly effective.
Phillips and Barker recommended follow-up studies to validate their findings, and one particularly interesting study would be to see whether the presence of e-prescribing systems had an impact on the July effect. Of course, the mere presence of an e-prescribing system doesn’t assure that medication errors will be avoided. The system needs to be used, and in many cases, at least some degree of training is required before the system can be mastered.
Meanwhile, the scientists also suggested that residency training directors and policy makers should reassess how new physicians are supervised and beef up resident educational programs when it comes to medication safety.
Glenn Laffel is Sr. VP, Clinical Affairs at Practice Fusion.
Submit a guest post and be heard.
