Admission to a hospital when most of the beds are already full can be deadly for patients, according to a new U-M Health System study showing high occupancy increases the risk of dying in the hospital by 5.6 percent.

For the study, published in the March issue of Medical Care, researchers evaluated a set of critical factors that can affect hospital deaths: hospital occupancy, nurse staffing levels, weekend admission and seasonal influenza.

Having more nurses made patients safer, decreasing risk by 6 percent. But weekend admission raised the risk by 7.5 percent and admission during widespread seasonal flu had the greatest impact by increasing the risk of death by 11.7 percent, according to the study.

Because of the size of the study, which included 166,920 adult patients admitted to 39 Michigan hospitals over three years, the findings can be generalized to hospitals nationwide, authors say.

“The study establishes that there is indeed a connection between hospital occupancy and death rates in U.S. hospitals,” says lead author Dr. Peter Schilling, a resident in orthopedic surgery at UMHS.

“It’s important to emphasize though that this study does not identify a specific occupancy level above which patient care suffers and deaths abruptly become more common. The key occupancy level may differ for each hospital,” Schilling says.

The findings are considered robust because each factor still had a significant impact even while evaluated in a model simultaneously. While this study is not the first to demonstrate that these factors are associated with in-hospital mortality, UMHS is the first to compare all four at once.

“The study further establishes each factor as a major predictor of hospital deaths but the good news is that each can be modified in some way,” says co-author Dr. Darrell Campbell Jr., chief of clinical affairs at UMHS.

Researchers calculated the occupancy of the hospitals every day for the years 2003-06. On average, patients in the study were admitted while hospital occupancy was 73 percent of full capacity. One-third of patients were admitted on high occupancy days, at average levels of 80 percent or more.

Study patients were admitted after being seen in the emergency department for a heart attack, congestive heart failure, stroke, pneumonia, hip fracture or gastrointestinal bleeding.

“Hospital occupancy changes from day to day, so patients shouldn’t try to choose a hospital based on its occupancy level,” says co-author Dr. Matthew Davis, co-director of the Robert Wood Johnson Foundation Clinical Scholars Program. “But these kinds of study findings should prompt hospitals to look at the flow of patients and processes of their care teams during high occupancy times. Those are more challenging moments when more things can go wrong.”

Dr. Michael Englesbe, assistant professor of surgery at the Medical School, also was an author on the study.