July is the worst month to check into a teaching hospital, UCSD researchers say

July is the worst month to check into a teaching hospital because of the influx of inexperienced residents, UC San Diego researchers reported Wednesday. The rate of fatal medical errors spikes in July, increasing by 10% compared with the average in other months, they reported in the Journal of General Internal Medicine. The increase occurred only in counties with teaching hospitals.

Many physicians have long suspected an increased rate of medical errors when new residents join hospitals, a phenomenon commonly referred to as the "July effect." But hard evidence supporting the possibility had been hard to come by, said social scientist David H. Phillips of UCSD, because most previous studies had examined small, non-geographically representative samples over a limited time period. Perhaps the best evidence to date was a five-year study of anesthesia trainees at an Australian hospital that showed an increase in errors in February, the first month of their academic year.

Phillips and graduate student Gwendolyn E. C. Barker studied all 62,338,584 U.S. death certificates for the period 1979 to 2006, ultimately focusing on 244,388 deaths linked to medication areas. They found an average increase of 10% in medication-linked deaths in July in counties with teaching hospitals but none in other counties. The proportion of such deaths was highest in those counties with the highest number of teaching hospitals. Studying deaths outside the hospital, they found no similar spike in deaths during the period, suggesting that it was not simply a summertime phenomenon. They found no spike in other causes of death in hospitals during July, however.

Their findings, they wrote, "provide fresh evidence for 1) re-evaluating responsibilities assigned to new residents; 2) increasing supervision of new residents; [and]  3) increasing education concerned with medication safety."

-- Thomas H. Maugh II

Comments

[Mitch Beales]

July may be the worst month to check into a teaching hospital but if you need to go to the hospital in July you're better off in a teaching hospital than a non-teaching hospital. Even with the July increase, death rates may be lower in teaching hospitals than in non-teaching hospitals. A 1989 study published in the prestigious New England Journal of Medicine (Volume 321:1720-1725) reports death rates of 108 and 116 per 1000 in private not-for-profit teaching and non-teaching hospitals respectively. The 7% higher rate of death from all causes in non-teaching hospitals reported in this study approaches the "July effect" of increased death rates due to medication errors documented in the Journal of General Internal Medicine study reported here.

Teaching hospitals are vital for medical education. Obviously they must address any issue that compromises patient care even in the very limited way reported here but it is irresponsible to imply that teaching hospitals are anything less than the finest medical institutions available, even in the month of July. Residents are often drawn from graduates of foreign medical schools in part because hospital training opportunities are so limited for students in American medical schools. This does not mean that foreign residents have more hospital training, only that American medical schools have higher standards that will not allow them to admit students unless opportunities for clinical training will be available for them. Increasingly, for-profit hospitals have dramatically reduced the clinical training opportunities for students. One way to address the "July effect" might be to improve the well supervised clinical training students receive before they become responsible for life and death decisions as residents. Your readers, for the sake of their own health, should seek out and support teaching hospitals whenever they can.