With the exception of perhaps some burn units, which treat patients particularly vulnerable to infections, "we've seen rates come down in virtually every type of ICU. So for the public to see hospitals with rates as high as 3 (and 4), that would really say that, well, it would raise a concern about the commitment to safety at that hospital."
About one third of hospitals have enough cases so far to justify a rate, but in late April, another three months of data will be posted, and in July, another three months until all hospitals have enough cases to compare.
For those hospitals that do have enough data, those with over 1.0 SIR are worse than national average, and some are much worse, including 37 with scores above 2.0.
The effort continues to reduce infection rates by 50% as soon as possible. It is funded with an $18 million contract from the federal Agency for Healthcare Research and Quality to the Health Research & Educational Trust, a non-profit research affiliate of the American Hospital Association.
Nancy Foster, the AHA's vice president for quality and safety and Pronovost say they're particularly pleased because of the type of data that is now posted.
Rather than using administrative discharge codes to tell whether a patient developed a central line infection, which are sorely inaccurate, CMS and CDC are using a much more specific clinical definition of CLABSI, as reported by each hospital, and collected by the National Healthcare Safety Network. The NHSN data, for example, includes central line patients who had a fever.