Staph superbug may be infecting 10 times more patients in U.S.

A dangerous, drug-resistant staph germ may be infecting as many asfive per cent of hospital and nursing home patients, according to a comprehensive study.

At least 30,000 U.S. hospital patients may have the superbug at any given time, according to a survey released Monday by the Association for Professionals in Infection Control and Epidemiology. The estimate is about 10 times the rate that some health officials had previously estimated.

'They acquired it in a previous stay in [a] health-care facility or out in the community.' —Study leader Dr. William Jarvis on when patients were infected

Some federal health officials said they had not seen the study, and could not comment on its methodology or prevalence. But they welcomed added attention to the problem.

"This is a welcome piece of information that emphasizes that this is a huge problem in health-care facilities, and more needs to done to prevent it," said Dr. John Jernigan, an epidemiologist with the U.S. Centers for Disease Control and Prevention.

At issue is a superbug known as Methicillin-resistant Staphylococcus aureus, which cannot be tamed by certain common antibiotics. It is associated with sometimes-horrific skin infections, but also causes blood infections, pneumonia and other illnesses.

The potentially fatal germ, which is spread by touch, typically thrives in health-care settings where people have open wounds.

But in recent years, "community-associated" outbreaks have occurred among prisoners, children and athletes, with the germ spreading through skin contact or shared items such as towels.

Sample much broader

Past studies have looked at how common the superbug is in specific patient groups, such as emergency-room patients with skin infections in 11 U.S. cities, dialysis patients or those admitted to intensive-care units in a sample of a few hundred teaching hospitals.

It's difficult to compare prevalence estimates from the different studies, experts said, but the new study suggests the superbug is eight to 11 times more common than some other studies have concluded.

The new study was different in that it sampled a larger and more diverse set of health-care facilities. It also was more recent than other studies, and counted cases in which the bacterium was merely present in a patient and not necessarily causing disease.

The infection-control professionals' association sent surveys to its more than 11,000 members and asked them to pick one day from Oct. 1 to Nov. 10, 2006, to count cases of the infection. They were to turn in the number of all the patients in their facilities who were identified through test results as infected or colonized with the superbug.

The final results represented 1,237 hospitals and nursing homes —or roughly 21 per cent of U.S. inpatient facilities, association officials said. The researchers concluded that at least 46 out of every 1,000 patients had the bug, with this breakdown:

• About 34 per 1,000 were infected with the superbug, meaning they had skin or blood infections or some other clinical symptom.
• Twelveper 1,000 were "colonized," meaning they had the bug but no illness.

Most of the patients were identified within 48 hours of hospital admission, which means, the researchers believe, that they didn't have time to become infected to the degree that a test would show it. For that reason, the researchers concluded that about 75 per cent of patients walked into the hospitals and nursing homes already carrying the bug.

"They acquired it in a previous stay in [a] health-care facility or out in the community," said Dr. William Jarvis, a consulting epidemiologist and former CDC official who led the study.

The infection can be treated with other antibiotics. Health-care workers can prevent spread of the bug through hand-washing and equipment decontamination, and by wearing gloves and gowns, and separating infected people from other patients.

The study is being presented this week at the association's annual meeting in San Jose, Calif., but has not been submitted for publication in a peer-reviewed medical journal.