JAMA Article Provides False Support for CDC's Do-Nothing Positi...( NEW YORK March 11 -- A new study...),Health

NEW YORK, March 11 /PRNewswire-USNewswire/ -- A new study in the Journal of the American Medical Association (JAMA) purports to show that screening for MRSA (methicillin-resistant Staphylococus aureus), a simple skin or nasal swab, is not effective in reducing MRSA hospital infections ("Universal Screening for Methicillin-Resistant Staphylococcus aureus at Hospital Admission and Nosocomial Infection in Surgical Patients," JAMA vol. 299, no. 10, March 12, 2008).

The findings of the authors will be seized upon by the Centers for Disease Control and Prevention (CDC) and advocates of the do-nothing status quo. But the study is seriously flawed -- rendering its findings meaningless.

Researchers used a 'rapid test,' but many patients were not tested until they had already been in the hospital for twelve hours. Furthermore, the results of the MRSA tests were not acted upon for another 221/2 hours on average. Most patients had completed more than half of their hospital stay before their results were known. Therefore, the precautions they needed -- isolation, proper antibiotics, chlorhexidine baths -- were taken late or not at all.

Unbelievably, almost a third of surgical patients (31%) who tested positive didn't get their test results until after their surgery. Therefore they too didn't receive any of the precautions they needed. Some people carry MRSA germs in their noses or on their skin without realizing it. The bacteria do not cause infection unless they get inside the body -- usually via a catheter, a ventilator, or an incision or other open wound.

No weekly MRSA testing was conducted, which is de rigour when conducting universal screening to prevent patients colonized with MRSA from passing it on to other patients in the hospital.

A previous study by the same lead author at the same location, The University of Geneva Hospital, found that universal screening on admission with preemptive contact precautions (the way it
'/>"/>