Hospital Infection Rates Down Dramatically In U.S.

From the WebMD Archives

March 2, 2000 (Atlanta) -- The rate of infections that patients get in many hospitals' intensive care units (ICUs) in the U.S. has fallen significantly in the last 10 years, according to the CDC.

The information may be a breath of fresh air for the medical community, which is still recovering from an Institute of Medicine (IOM) report late last year that found 44,000 to 98,000 hospitalized Americans die each year due to medical errors.

During the 1990s, rates of bloodstream infections declined by 30% to 40% in the intensive-care units (ICUs) of hospitals taking part in the National Nosocomial Infection Surveillance (NNIS) system, the CDC says. "We saw similar decreases in other hospital-acquired infection rates in those hospitals," Bob Gaynes, chief of the hospital infections program at the CDC, tells WebMD.

NNIS is a voluntary system of larger U.S. hospitals that the CDC established in 1970 to monitor hospital-acquired, or nosocomial, infections and to guide these hospitals' prevention efforts. As of last year, 285 hospitals in 42 states were participating.

The most accurately measurable data was that on bloodstream infections, Gaynes says, but decreases in lung and urinary-tract infections were also reported.

"These were really dramatic changes at a time where you might even expect the trends to go in another direction," he tells WebMD. "Basically, because of changes in health-care delivery, patients that are hospitalized now compared to a decade or two ago have higher measures of severity of illness -- they're sicker."

Gaynes says the NNIS system is successful because of the "ongoing vigilance" inherent in the program as well as participants' use of the data to develop standards for preventing infections.

Though the information about the high-risk population found in ICUs is very good news, Gaynes says, there is still a need to develop infection-prevention standards in other areas of hospitals or even outside hospitals, such as in long-term care and ambulatory surgery facilities.

In all, though, Rick Wade, senior vice president of the American Hospital Association, tells WebMD, "it's good. This is an interesting example of what can be accomplished when you have the right set of circumstances, the right use of data, and the follow-through when you have the information to actually make changes at the hospital level."


Gaynes says the information not only helps those facilities that participate in the NNIS program but can benefit all hospitals. He says the information is on the Internet, it's in published form --, it's even on some commercial software. "I think it can serve as a model," he says.

Carmela Coyle, senior vice president of policy for the AHA, agrees. "I think that what this has shown is that you can change patient care. ... What [the CDC] did was feed all that information back to these organizations ... to give them and equip them with the information they need to actually make change, to actually improve patient care," she tells WebMD.

Confidentiality is a key aspect of the NNIS program, says Gaynes. That struck a chord with both Wade and Coyle. "While joining this, and being part of this, was voluntary," Wade tells WebMD, "there were protections so that once people volunteered to be part of this, they could feel free to report and share all of the information back with the project, so there was complete data with which to benchmark and make changes."

While this report brings back the memory of the IOM report on medical errors, Gaynes makes a distinction. He says the IOM defined a medical error as something preventable, while this report deals with "adverse health events," which are inadvertent consequences of medical treatment.

Coyle tells WebMD, "whether it's this system or an error reporting system, the bottom line objective to both those systems is learning, making changes, and improving patient care. While some of what the IOM's been talking about is around preventable error, I think what they also acknowledge is it's not all preventable; we're talking about human beings. There's no such thing as a perfect person. But how do you construct a system that helps us keep people from making mistakes?"


WebMD Health News
© 2000 WebMD, Inc. All rights reserved.