It is common knowledge to avoid major teaching hospitals in July. Such is when new residents, fresh from medical school, begin as physicians. In some studies the month has been associated with more errors, including notably fatal medication errors. However, the effect continues to be questioned and debated. Even a short review of the surgical literature finds that a preponderence of studies show no worse outcomes with surgical procedures in July as compared to other months. The most notable of these studies is likely this large retrospective review of all Medicare patients undergoing CABG, CEA, AAA repair, colectomy, pnacreatectomy, esophagectomy or hip ORIF between 2003 and 2006. They found no increased mortality or reported morbidity in those three Julys as compared to the other 33 months of the study. Other studies looking just at patients undergoing CABG or patients undergoing emergent appendectomy back up those results.
Add a recent study in the Journal of Neurosurgery to the pile of evidence that new resident physicians in July don’t endanger patients.
McDonald, Robert J., Harry J. Cloft, and David F. Kallmes. “Impact of Admission Month and Hospital Teaching Status on Outcomes in Subarrachnoid Hemorrhage: Evidence against the July Effect.” Journal of Neurosurgery 116 (2012): 157-63.
The study by a group out of the Mayo Clinic is a retrospective analysis of a huge proportion of all hospital admissions between 2001-2008 for non-traumatic subarachnoid hemorrhage. The study pulled all admissions with ICD codes associated with SAH from the National Inpatient Sample. This is an AHRQ national database contributed to by all hospitals in 44 states.
The admissions and their outcomes were studied with two linear regression models for both teaching and non-teaching hospitals. One looked solely at inpatient mortality. The other looked at “unfavorable” discharged; those patients with SAH being discharged to skill care.
There were 52,879 admissions for non-traumatic SAH in the NIS database between 2001-2008. 36,914 were admitted to teaching hospitals and 15,965 were admitted to non-teaching hospitals. There was no monthly variation, in either teaching or non-teaching hospitals, in either model. The authors failed to find any evidence of a “July effect.”
Of note however, there was a discrepancy in outcomes in terms of hospital teaching status. The probability of in-hospital mortality for patients presenting to a teaching hospital with non-traumatic SAH was 11% lower than that or patients presenting to a non-teaching hospital. The probability of “unfavorable” discharge, likewise, was 12% lower.
The results of this retrospective review of SAH hos -pital admissions within the 2001–2008 NIS failed to demonstrate significant month-to-month variation among outcomes including in-hospital deaths and/or discharges requiring skilled care. This pattern was observed in both teaching and nonteaching hospitals and suggests that a July effect is absent among SAH hospitalizations.
Also, for life threatening problems, such as subarachnoid hemorrhage, tertiary centers (more often than not teaching facilities) appear to be the place to go for care. At least in terms of outcome.