In Conversation with naomi o'grady

Dr. Naomi O'Grady is a Senior Staff Physician at NIH and first author of the CDC Guidelines. She recently wrote a thought piece for Annals of Internal Medicine. We sat down with her to have a discussion about her article, central line-associated bloodstream infection (CLABSI) and measurement of this infection. Here are some of our excerpts:

VC: CLABSI has become such a volatile HAI. Given all the penalties and public reporting, no wants to have one. And hospitals that do have one now have to report these data publicly - leading to all kinds of problems. What are your views on the policy for CLABSI reporting in the US? What impact has this had?

NG: ​The precise impact of public policy on CLABSI is unclear. In 2008, the Centers for Medicare and Medicaid Services (CMS) instituted financial penalties for hospitals with higher-than-expected CLABSI rates. At least two studies have reviewed the effects of such penalties on CLABSI rates: one suggested no effect, while the other reported a decline in infection. Whats important is to understand whether financial penalties have actually changed practice at the bedside. One fear is that such policies may dissuade clinicians from ordering blood cultures in patients with central venous catheters (CVCs). Less testing for CLABSI, in turn, could result in lower detection of CLABSI—improving hospital performance without changing patient outcomes. 

In addition to the non-reimbursement policies instituted in 2008, several states instituted public reporting of hospital-acquired infection rates for hospitals. In 2004, only 4 states had such a requirement, but in 2016, 38 states and the District of Columbia required public reporting. Again, it is difficult to ascertain how these policies impact care at the bedside. But, by design, they pressure hospitals to reduce CLABSI to zero. It is well known that economic incentives to influence practice can lead to unintended consequences. In the case of CLABSI , financial pressure could lead to changes in the way that CLABSI is defined or detected. 

VC: Lets talk a bit about measurement - there are challenges there, right?
NG: Well, determination of CLABSI rates is inherently dependent on the definition of CLABSI and methods used for case-finding. Changes in either of these influence the interpretation of CLABSI data over time. The National Healthcare Safety Network (NHSN) definition of CLABSI includes three components: 1) The presence of a laboratory confirmed bloodstream infection; 2) in a patient with a CVC; 3) with no alternative source of bacteremia identified. The last of these components is especially subjective. Policies tying public reporting to non-payment could influence hospitals to adopt an intense explanation for “alternate sources” of bacteremia. Of note, such relentless searches did not occur  when financial consequences were not present. Thus, while CLABSI rates over the past 2 decades may have been over-estimates of the actual rates, it is plausible that current accounting strategies may be underestimating CLABSI, especially if some are attributed to other sources of bacteremia. Even with rigorous application of the NHSN definition, inter-rater reliability can be less than optimal. 

VC: What about the new mucosal barrier injury (MBI) addition? Has that helped in getting closer to truth?
NG: The new mucosal barrier injury (MBI) category is intended to improve comparability of CLABSI rates by specifically classifying infections caused by certain pathogens as non-CLABSIs. The addition of MBI to CLABSI was a welcome change, especially for institutions caring for large numbers of cancer patients.  But, as expected, retrospective studies showed an overall decrease in CLABSI rates when MBI cases were considered separately. Thus, whether lower CLABSI rates after 2013 reflect actual improvement in patient safety or artifact related to changes in classification remains unclear.

VC: CLABSI rates have gone down over the past few years. Your comments make me wonder: is the decline real, or "fake news?"
NG: Many factors have contributed to changes in CLABSI rates over time. Some changes have made a real impact on patient care. Conversely, others may have altered the numbers without improving patient outcomes.

VC: So what should clinicians and hospitals do?
NG: Facing this uncertainty, I think our focus should continue to remain on ensuring CVCs are used for appropriate indications, removed when no longer necessary and bundled strategies for CLABSI prevention are rigorously implemented. In addition, hospitals should internally monitor CLABSI rates over time. In this way, internally consistent data rather than comparisons with other institutions employing alternative definitions or measurement strategies may be used to assess progress.

​VC: Any closing thoughts? 
NG: Perhaps we will never know how much of the recent improvement in CLABSI rates is real versus how much is “fake news.” Regardless, we must continue to refine our definitions and data. After all, we should be zeroing in on zero when it comes to CLABSI. Our patients count on it.