Each year nearly 300,000 U.S. patients get serious cases of pneumonia and sepsis — bloodstream infections — during their hospital stays. Almost 50,000 of them die.
That's the bottom line of a new study that its authors say is the first truly national look at the stubborn problem of hospital-acquired infections. It's based on 69 million hospital cases in 40 states.
These potentially avoidable infections result in longer hospital stays, costing more than $8 billion a year.
For all the attention devoted to the problem in recent years, study author Ramanan Laxminarayan says no study has ever been able make a credible estimate of deaths from infections that occur in the hospital. Earlier studies have been too small, he says, and could not distinguish between infections acquired in the hospital and those that patients had when they were admitted.
Authors of the new study, in the Archives of Internal Medicine, say they've gotten around this problem by looking, in part, at infections that occur among elective surgery patients. They're not likely to have been infected prior to admission, Laxminarayan says, or they wouldn't have had the surgery.
The researchers found the death toll from avoidable pneumonia and sepsis is bigger than from traffic fatalities. It's more than three times higher than that for AIDS, and roughly twice as much as annual deaths from firearms.
The authors consider the 48,000 death estimate an underestimate, since it counts only those from infections acquired and diagnosed during the same hospitalization. But many hospital-acquired infections are not diagnosed until after discharge.
The study is part of an ongoing project called Extending the Cure funded by the Robert Wood Johnson Foundation. (The foundation also funds NPR health programming.)
Dr. David Murphy and Dr. Peter Pronovost of Johns Hopkins say the new data are "disconcerting." "The knowledge that patients continue to experience harm from their interactions with the health system is unconscionable," they write in an accompanying editorial.
The new data focus attention, once again, on how to reduce avoidable hospital infections. Laxminarayan says Medicare's 2008 decision to stop paying for care associated with "never" events — infections that should "never" happen — won't do much. So far, he notes, Medicare's policy covers only three types of infections — those related to vascular catheters, urinary catheters or surgical sites following coronary bypass grafts.
In an analysis that's not in the published paper, the authors looked at how many deaths could be averted each year "if the Medicare rules were perfectly effective," Laxminarayan says. The answer: Fewer than 100.