Montgomery County is now in the loop!

The next time a contagion sweeps through the NIH Clinical Center, Montgomery County officials will be on it, thanks to a new agreement between NIH, Maryland and Montgomery County. The back-story is that Montgomery County officials were unhappy that they weren’t informed promptly about the deadly KPC outbreak at NIH.

This raises the question of when a hospital should communicate with public health officials (and the public generally) about fairly common SNAFUs. At any given time, 5-10% of hospitals are dealing with clusters or outbreaks of multiple-drug resistant gram negative bacteria (KPCs, ESBLs, MDR-Acinetobacter, etc.), and even more are in the midst of MRSA, VRE, fungal or other outbreaks. The population at risk during these outbreaks is pretty clearly defined, and doesn’t include the general public. General notification can generate media frenzy, free-floating panic and anxiety, and waste precious time and resources for the personnel trying to contain the outbreak (responding to media, doing damage control of various types, etc.). Furthermore, most states don’t include common healthcare associated bacterial pathogens among their legally reportable diseases.

However, as state and local public health officials become increasingly involved in HAI issues, it would be wise to establish explicit criteria for when healthcare facilities should report clusters and outbreaks. Provided they have sufficient funding (which they currently do not!), public health departments should play a critical role in coordinating responses to HAI outbreaks, which often involve multiple healthcare facilities in a region (across the spectrum of acute, long-term, and long-term acute care).

So when do you think a hospital should notify their state and/or local public health department? Two cases of MRSA infection in the NICU? A single serious post-operative Group A strep infection? New introduction of a carbapenemase into the ICU?

Oh, and Happy Thanksgiving!

Riding the epidemic curve to glory, WGS edition

One reassuring lesson all healthcare epidemiologists learn is that every outbreak will, eventually, come to an end. The trick is to prevent outbreaks in the first place, or to recognize them early enough to intervene effectively. Many outbreaks, though, are recognized as they peak and are entering the “downhill” part of the epidemic curve. Any interventions prescribed by erstwhile epidemiologists are then, in retrospect, credited with helping to contain the outbreak (even if said interventions were completely idiotic). This phenomenon was described by Dr. Alexander Langmuir, the father of the CDC’s Epidemic Intelligence Service, as “riding to glory on the downhill slope of the epidemic curve”.

Now we can add the performance of whole-genome sequencing (WGS) to the list of activities that epidemiologists can “ride to glory” as the key to outbreak control. As WGS becomes faster and more affordable, reports have been published in NEJM, Science Translational Medicine and Lancet Infectious Diseases suggesting that WGS can be the key to real-time or “actionable” information to help contain outbreaks (due to MRSA, KPC-producing K. pneumoniae, and MRSA, respectively). As we’ve pointed out previously, though, it isn’t at all clear that WGS was important to real-time outbreak management, or that WGS is ready for prime-time and coming soon to a hospital near you.

Why? For any technology to see widespread adoption in clinical diagnostic laboratories, there must be sufficient automation (including of the analysis and interpretation of the massive amounts of WGS data), and it must provide a substantial advantage over existing testing approaches.

Which brings us to the more important question: given our crude approaches to outbreak control, how does WGS provide any immediate advantage over other same-day typing methods? Does the added discrimination really make a difference in whether we decide to isolate or cohort patient X, decolonize healthcare worker Y, or close unit Z to new admissions? These questions are particularly pertinent when we don’t yet understand the “within host” variation in genotype and how “within” versus “between” host variation can be applied to determine direction of transmission (or even the fact of transmission). Eli recently sent me a link to this interesting discussion of the Lancet ID report, which addresses some of these issues. 

The bottom line is that we have a relatively few crude tools for outbreak response: enhanced basic practices (e.g. hand hygiene, environmental disinfection), active surveillance, isolation, cohorting (of patients and/or HCWs), decolonization of carriers (both HCWs and patients), removal (temporarily or permanently) of HCWs implicated in transmission, closure of units, mitigation of any identified common sources, etc. The level of discrimination provided by WGS is a great research tool, and will undoubtedly help in retrospectively piecing together the most likely outbreak scenarios--but for now I’m not convinced it has much advantage over other typing methods for guiding real-time outbreak investigation and management.

Don't let authors control for factors in the causal pathway!

A downside of being an epidemiologist who thinks about antibiotic resistance 24-7 is that eventually it's hard to read the literature for fear of seeing another paper where authors make a critical mistake.  It's unfortunate, because the review process should catch these common errors and at least make the authors mention limitations in their discussion sections.  A case in point is a recent paper by Chen et al. in CID that sought to assess the impact of MRSA SCCmec type and vancomycin MIC on treatment failure. In this case, treatment failure was defined as all-cause 30-day mortality, persistent bacteremia, or recurrent bacteremia. They provide nice graphs showing higher mortality for hospital associated-MRSA strains (SCCmec type I, II, III) and for strains with vancomycin MICs > 2. So far, so good. But then, I looked at their multivariable model.

As expected, they controlled for septic shock or severe sepsis in their model. They should not have done this. If it were a comparative effectiveness study assessing various treatment outcomes, it would be appropriate to collect severity of illness including evidence of sepsis before treatment initiation and control for it in the model. However, Chen's study was only looking at outcomes based solely on strain differences. Thus, they should never have controlled for shock or sepsis. How else would a bacteremic patient die if not through sepsis? For a longer description of why controlling for factors in the causal pathway is suboptimal, see my post from 3/19/2012. I have pasted the key sections below. Note: We have been writing letters and review articles in CID pointing this issue out since the year 2000.

____
March 19, 2012: My first ever publication, and in some ways still my favorite, was a letter to the editor of CID that I wrote in 2000 pointing out a common flaw in outcome studies of infectious diseases. In the letter, I discussed a paper that looked at the outcomes (death) associated with methicillin-resistance in patients with S. aureus bacteremia. In the analysis, the authors controlled for septic shock in their regression model. I pointed out that shock is in the causal pathway between infection and death and, therefore, should not be controlled for in regression in models. This would be like controlling for car accidents when looking at the association between cell phone use and death. In infectious diseases, if you remove shock from the causal pathway, it is hard to see how you might otherwise die.

The error of controlling for intermediates is frequently repeated in ID outcome studies when, for example, authors control for illness severity using the APACHE score. If the APACHE is measured after the infection manifests, this variable would be in the causal pathway and should not be controlled for in the regression model. The APACHE should be measured before the infection manifests, as we did here. Jessina McGregor and JJ Furuno (both now at Oregon State) published a nice systematic review on optimal methods for ID outcome studies in CID back in 2007. Wouter Rottier (with Marc Bonten) just published a meta-analysis looking at the impact of confounders and intermediates (factors in the causal pathway) on ESBL-bacteremia outcomes. (JAC, March 5, 2012) I highly recommend that you read these studies prior to undertaking an ID outcome study.

Q: What are Iowans doing to prevent hospital-acquired infections?

A: Quite a bit, actually. Thanks for asking!

Just last week, University of Iowa researchers Tara Smith, Marin Schweizer, and Phil Polgreen all sat down with Iowa Public Radio's Ben Kieffer (picture to the left) to discuss the latest science in HAI prevention including the epidemiology of MRSA in animal populations, the importance of environmental control in hospital settings and the latest hand-hygiene surveillance technologies.

Click on over to have a listen!

Shocking: Another review of the evidence finds flu vaccine has no miracle powers

The Science section of the New York Times this week had an article on a new report regarding influenza vaccination. This report (free full text here) from the University of Minnesota is the kingdaddy of all analyses on influenza vaccination--a 3-year project funded by the Alfred P. Sloan Foundation that reviewed 12,000 papers back to 1936 and involved interviewing 88 influenza experts. The report is 123 pages without appendices and includes over 500 references. 

Here's my summary of the chapter on vaccine performance of the two major vaccines (trivalent inactivated influenza vaccine [TIV] and live attenuated influenza vaccine [LAIV]):

Population	TIV	LAIV
Children	Inconsistent evidence of protection	High level of protection (83%)
Healthy adults	Moderate protection (59%)	Lack of evidence of protection
Elderly	Paucity of evidence for protection	Inconsistent evidence of protection

What about use of the vaccine in healthcare workers? This can be found on pages 57-58 of the report and I have pasted below the important discussion of the HICPAC recommendation regarding offering influenza vaccine to healthcare workers):

The 2006 statement on influenza vaccination of healthcare personnel (HCP) from the Healthcare Infection Control Practices Advisory Committee (HICPAC) and ACIP illustrates potential concerns with using a grading scale.[63] This recommendation used the HICPAC grading scale, which is similar to the GRADE criteria in that it provides a structure for ranking the evidence. All recommendations were approved by the HICPAC and the ACIP. This document has been used widely as evidence to support HCP vaccination policies, including mandating vaccination. It offers six recommendations, and one was deemed to have the highest possible evidence, category IA. Category IA recommendations are “strongly supported by well-designed experimental, clinical, or epidemiological studies.”[63] The recommendation in the HICPAC document that received a category IA rating states:  “Offer influenza vaccine annually to all eligible HCP to protect staff, patients, and family members and to decrease HCP absenteeism. Use of either available vaccine (inactivated and live, attenuated influenza vaccine [LAIV]) is recommended for eligible persons. During periods when inactivated vaccine is in short supply, use of LAIV is especially encouraged when feasible for eligible HCP.”[63]

This recommendation is supported in part by this key summary statement in the HICPAC document: “Vaccination of HCP reduces transmission of influenza in healthcare settings, staff illness and absenteeism, and influenza-related morbidity and mortality among persons at increased risk for severe influenza illness.[64-67]” In the first study cited, the authors did
not find a statistically significant reduction in patient mortality associated with HCP vaccination, after adjusting for covariates.[64] In the second study, the authors concluded that “we do not have any direct evidence that the reductions in rates of patient mortality and influenza-like illness that were associated with HCW vaccination were due to prevention of influenza.”[65] In the third study, vaccination did not reduce the episodes of self-reported respiratory infection or the number of days ill with a respiratory infection, but it did reduce the time employees were 58 unable to work because of a respiratory infection.[66] In the fourth study, the authors reported reductions in absenteeism and illness among HCP that were not statistically significant.[67] The authors did, however, report serologically confirmed vaccine effectiveness of 88% for H3N2 and 89% for influenza B across three influenza seasons.[67] Since only two of the four studies cited provide some support for the HICPAC statement and the others no support, it is unclear how the quality of evidence in these studies received a category IA evidence grade. Another review conducted in the same time frame by the Cochrane Collaboration noted that the two RCTs cited in this recommendation were at “moderate risk of bias.”[68] They concluded that “both elderly people in institutions and the healthcare workers who care for them could be vaccinated for their own protection, but an incremental benefit of vaccinating healthcare workers for elderly people has yet to be proven in well-controlled clinical trials.”[68]

So this report questions the evidence base for even recommending influenza vaccination to healthcare workers. Yet, SHEA's position is so over-reaching that it calls for mandating vaccination and firing noncompliant healthcare workers. This is now the fourth independent analysis that does not support the SHEA position statement (read about the others here, here and here).

I continue to be fascinated by the post-modern disdain for evidence. A marvelous example from this week is the shock and utter disbelief suffered by Mitt Romney and his staff on learning that Barack Obama won the presidential election, despite nearly every poll indicating that Romney would lose. I guess I naively thought that somehow epidemiologists were immune to such bias but SHEA's flu vaccine position suggests otherwise.

One of the recommendations in the Minnesota report is that "scientifically sound estimates of influenza vaccines’ efficacy and effectiveness must become the cornerstone of policy recommendations." Amen. And it's time for SHEA to retract its policy!

MRSA in water treatment plants! Panic?!

There is a new study just published in Environmental Health Perspectives that is getting a lot of media attention (here and here and here). The (open-access) study by University of Maryland researchers identified MRSA (and MSSA) in water samples collected during 2009/2010 from two mid-atlantic and two mid-western water treatment plants. They found that 50% of samples contained MRSA and 55% contained MSSA. Among the MRSA strains, 83% were SCCmec type IV and 15% were SCCmec type II, while 68% were pvl gene positive. USA100, USA300 and USA700 PFGE strain types were identified.

Sounds pretty scary, and if you don't read carefully, you might conclude that you shouldn't drink municipal water. However, it appears that most of the MRSA/MSSA was detected in "influent" or pretreatment samples, which were 83% positive, while 17% of post-treatment "effluent" samples were positive and only 8% were MRSA positive. This suggests that water treatment is effective in eliminating S. aureus, so keep drinking the water.

Note:  The positive samples in the effluent were collected during periods when seasonal chlorination was not taking place. Thus, if your municipality uses chlorination, there is likely little S. aureus risk. Also, we should ask if we can generalize beyond this 4-facility sample.

Image Source: EPA via CDC

Not another damn map!

I'm pretty sick of maps. We have blue states, red states, periwinkle states, pink states and Ohio-colored states. Then we have light green states, green states and dark green states that show how many cases of Exserohilum meningitis have been identified in each. (Don't you wonder why there are no meningitis cases in Massachusetts? Seems like they might have known something!) If that weren't enough, it's time to start worrying about this map...

Come to think of it, I wish some of the other maps looked more like this map. Would make things a little less predictable, no? Texas, the same color as California - haven't seen that in awhile.