Rumor has it the worst time and place to be sick is in a teaching hospital in July, when new doctors-in-training enter the wards and others are promoted. A new study of heart attack patients shows this pattern of worse outcomes known as the "July effect" may indeed be true - but only for the sickest people.
"Patients who are already at high risk of inpatient mortality - because of their age and other (co-existing) diseases - are likely the most to be affected by physician inexperience in July," Dr. Anupam Jena said in an email. He led the research at Harvard Medical School in Boston.
But according to Dr. Gary Rosenthal, who wrote an editorial accompanying the study, getting hospitalized in July is "not a significant cause for concern."
"If this (July effect) phenomenon really exists, this study says it's only limited to the sickest patients," said Rosenthal, from the University of Iowa Hospitals and Clinics in Iowa City.
Jena and his team conducted the new study in light of conflicting reports about patient outcomes in July.
The researchers analyzed records of U.S. patients hospitalized for a heart attack in either May or July in 2002 to 2008.
Their study included information on 19,054 people who were at high risk of dying based on their age, gender and other health conditions and 57,163 low-risk patients.
People were admitted to one of 98 teaching-intensive hospitals - facilities that place an emphasis on training new doctors - or 1,353 non-teaching-intensive hospitals.
Prior reports had not included such comparisons, the researchers noted.
For the sickest patients, the chance of dying in the hospital was 30 percent higher if they were admitted to hospitals with a strong teaching focus in July versus in May, the study team reported in the journal Circulation.
At those hospitals, close to 23 percent of high-risk patients died in July compared to 19 percent in May.
In contrast, the proportion of patients who died in May and July was similar at hospitals with a less intensive teaching focus. Between 22 and 23 percent of the highest-risk patients died at non-teaching-intensive hospitals during both months.
Among lower-risk heart attack patients, the chance of dying was similar in May and July at both major teaching hospitals and non-teaching-intensive hospitals.
Rates of bleeding complications and stent placements did not explain the difference in deaths for the sickest patients treated at teaching hospitals, the researchers said. Neither of those differed between May and July at teaching or non-teaching hospitals.
"Our study is different because it recognizes that the July effect should not be present for all patients, but primarily those patients for whom small clinical errors or relative physician inexperience can substantively impact patient outcomes," Jena said.
Rosenthal agreed the new study makes a unique contribution by looking at a large number of teaching and non-teaching hospitals.
However, he said the people included in the study - patients with an acute heart attack - represent only about 2 percent of all admissions to U.S. hospitals. So the finding of a July effect may not hold true among people with other diseases.
Rosenthal also pointed out that when the researchers looked at all 12 months of the year, their findings showed patients in teaching-intensive hospitals generally fared better overall, with a lower risk of death than those in non-teaching-intensive hospitals.
"This study clearly shows that even in July, patients did just as well in teaching-intensive hospitals as they did in other hospitals," he said.