Mar 012017

In 2008 hospitals were informed they would no longer be paid for treating hospital acquired infections like ventilator associated pneumonia. Miraculously, the rates of VAP (self-reported by hospitals to the Centers for Disease Control and Prevention) fell dramatically by 60 to 70% between 2006 and 2012, to less than one VAP per 1,000 ventilator days in medical ICU patients. Wow!

Is this amazing decline in VAP rates real? Not according to a research letter by Metersky and colleagues in JAMA, who report the rate of VAP remained almost exactly unchanged between 2005 and 2013. About 10% of mechanically ventilated patients developed VAP in their analysis.

Rather than hospitals' self reported data, authors used the Medicare Patient Safety Monitoring System MPSMS, in which tens of thousands of records from patients are abstracted each year from a random selection of hundreds or thousands of hospitals. VAPs (and other clinical events) are identified according to a stable case definition. The Centers for Medicare and Medicaid Services (CMS) has a shop called the Clinical Data Abstraction Center (CDAC) for this.

Clinical and epidemiologic research on VAP is crippled by the lack of a gold standard diagnostic test or case definition, usually unblinded study designs, and disincentives by health care teams and centers to make the diagnosis. The result is that the large body of VAP research is hard to aggregate or interpret.

Acknowledging its own VAP data's fatal flaws, CDC has embarked on a quixotic quest to track a new invented entity with no demonstrable relevance to health outcomes, which CDC calls ventilator-associated events.

The MPSMS method seems to largely avoid these pitfalls and has the power of an enormous data set from a continually changing sample. I could not find the MPSMS case definition for VAP, so can't describe or critique the actual method.

The research letter should heighten skepticism for the numerous interventions proposed or advocated as effective prevention against ventilator-associated pneumonia. To date, the only measure that has been shown to prevent deaths from the development of VAP is prophylaxis with broad spectrum antibiotics upon arrival to the ICU. Almost no ICUs use this as a standard prevention technique.

Most clinical research using using endpoints other than mortality (such as prevention of VAP itself) should be viewed with skepticism due to the aforementioned study limitations.

The estimates of VAP incidence from the Centers for Medicare and Medicaid Services (CMS) are more credible than the CDC's. The rock-solid statistic of 10% suggests there may be a baseline rate of VAP that can't be reduced. (That is, unless forthcoming studies create a new critical care standard of systemic antibiotic prophylaxis for all ventilated patients, which does seem to reduce VAP and its mortality.)

If the MPSMS data reflects reality in U.S. ICUs, it's a win for CMS and the national coffers, but bad news for hospital quality officers and CFOs, who will be forced to pay the costs of treating patients for VAP on the false premise that all these infections are preventable.

Read more:

Trend in Ventilator-Associated Pneumonia Rates Between 2005 and 2013. JAMA. 2016 Dec 13;316(22):2427-2429.

Medicare Non-Payment of Hospital-Acquired Infections: Infection Rates Three Years Post Implementation. Medicare & Medicaid Research Review. 2013: Volume 3, Number 3.

Get our weekly email update, and explore our library of practice updates and review articles.

PulmCCM is an independent publication not affiliated with or endorsed by any organization, society or journal referenced on the website. (Terms of Use | Privacy Policy)


Are ventilator-associated pneumonia rates plummeting, or unchanged?