MSSA and MRSA: Both are dangerous!

Jessica Ericson and co-workers recently published a remarkable investigation of invasive Staphylococcus aureus infection in hospitalized infants. In it they describe a retrospective multicenter study of 348 NICUs in which a group of 3888 infants suffered invasive S. aureus infection between 1997 and 2012. They compare the demographics and mortality of infants with invasive MRSA and MSSA; determine the relative annual proportions of MSSA and MRSA; and calculate the risk of death after an invasive MSSA and MRSA infection. It's a fascinating study and I recommend reading it.

Among other results, they find that infant mortality following invasive MRSA and MSSA infection is essentially the same. Moreover, in their cohort of patients, MSSA was responsible for a larger burden of disease and death in infants than MRSA. Based on their findings, and consistent with previous studies, the authors recommend that

Measures to prevent S. aureus infection should include MSSA in addition to MRSA.

This is an important point. The goal of infection prevention is to protect patients; both MRSA and MSSA are deadly, so we should be mindful of each. Commonly, patients are screened for MRSA carriage only, and isolated and decolonized if found to be positive. Previous research suggests that it may be possible to reduce the incidence of both MRSA and MSSA infection by screening for S. aureus universally, and this latest study shows why this is critically important.

After drafting this post I realized that Mike Edmond, on the Controversies in Hospital Infection Prevention blog, had already written a great piece in connection with this paper. You should read the post. In it, he captures the issue powerfully in a single line: 

I often joke that I've never had a patient tell me that they don't want a MRSA infection, but they'll take an MSSA infection. 

Indeed, both pathogens deserve attention and respect, as has long been known. 

(image source: Wikipedia)

MERS as (another) messenger of prevention

It's hard for me to know how to interpret the MERS situation in South Korea. At a high level, a recently recognized viral respiratory pathogen has traveled halfway around the world and is causing morbidity and mortality in a small section of an immunologically naive population. It appears to be associated with hospitals. What do we take away from this? Lessons will be learned when the event subsides and people study what happened, but to me, MERS reminds us that outbreaks of pathogens for which there are no vaccines or drug therapies underscore the importance of prevention.

When possible, preventing pathogens from physically reaching or entering a host by respiratory, percutaneous, alimentary, blood et al pathways is preferable to relying on pharmaceutics. Drugs tend to be complex and costly to develop, can take a long time to enter the marketplace, and -- especially in the case of antibiotics and antivirals -- they can become obsolete over time. Moreover, drugs are often toxic to the patient. Prevention is applicable in situations when appropriate drugs don't exist (e.g., for newly emerged pathogens), when it isn't possible to administer drugs in a timely manner, or when patients cannot tolerate them. 

Consider two anecdotes related to the spread of MERS virus in South Korean hospitals. As described by Choe Sang-Hun, it appears that the index patient in the South Korean event had "coughed and wheezed his way through four hospitals before officials figured out, nine days later, that he had something far more serious and contagious." Furthermore, ED wait times in Korea can be extraordinarily long by US standards. Another patient, who waited two-and-a-half days in the emergency department before a hospital bed became available, infected 55 additional individuals during their wait. Apparently, 2.5 days isn't an unusually long waiting time in some Seoul hospitals. 

Applying effective prevention measures to patients suspected of infection is the only way of stopping the chain of transmission in such environments. Unfortunately, it is unclear how to achieve good infection control for MERS and a range of other pathogens. Eli Perencevich described the issue clearly, as usual, in the Controversies in Hospital Infection Prevention blog recently: 

. . . we don't actually know how to achieve good infection control for MERS and the other diseases he [Tom Frieden] mentioned [measles, DR-TB, SARS, Ebola]. If only we invested in studies to understand how to best implement PPE in these [hospital] settings. One could imagine improved PPE technology, refined PPE donning and doffing algorithms and enhanced environmental cleaning as potential targets for future studies examining optimal protection from MERS. Not coincidentally, many of these are the same targets that Mike, Dan and I mentioned in our Ebola+PPE editorial several months ago. If we invest in infection prevention technology and implementation research, our health care system will be safer regardless of the pathogen du jour.

And that's the point that MERS makes me think about. Yes we need antimicrobials and vaccines that work against specific pathogens, of course we do, but developing such drugs is a major effort. Biochemical pathways must be understood, pathogen life histories and survival strategies must be elucidated, and the host response must be characterized among many, many other things. Doesn't it make sense that research on pathogen-agnostic approaches to prevention, which don't require such specific and complex information, might be simpler and broadly applicable? 

Investing in research on infection prevention approaches, and how to implement them sustainably in realistic clinical environments, would pay benefits far beyond helping to thwart the spread of exotic and newly emerged pathogens. We may learn how to better control and prevent the usual suspects of hospital associated infection, which, afterall, are responsible for a tremendous burden of disease day in and day out.

(image source: Wikipedia)

HAI and aliens: The Drake equation in epidemiology

In a 1961, Frank Drake introduced the following equation for the number N of civilizations in our galaxy with which radio-communication might be possible,

N = R x F_p x N_e x F_l x F_i x F_c x L

As described by the SETI institute,

• R is the average rate of star formation in the galaxy, 
• F_p is the fraction of those stars that have planets, 
• N_e is the average number of planets that can potentially support life per star that has planets, 
• F_l is the fraction of planets that could support life that actually develop life at some point, 
• F_i is the fraction of planets with life that actually go on to develop intelligent life (civilizations), 
• F_c is the fraction of civilizations that develop a technology that releases detectable signs of their existence into space, and 
• L is the length of time for which such civilizations release detectable signals into space. 

Drake's purpose in writing this equation was to facilitate discussion at a meeting. Its importance is not the numerical prediction of communicative civilizations in the galaxy (note there are 7 factors in the equation and errors in each term will combine to make any calculation wildly uncertain) but rather in the framing of issues related to the search for alien life. That said, the equation tells a story. Assuming that these are the relevant factors, then if any the terms are zero, N is zero and we are likely to be alone. If none of them are zero, then even if they are exceedingly small, there is a chance that there is life somewhere in the galaxy. Moreover, it's unlikely that any of these terms are zero, given the huge size of the galaxy. In epidemiological terms, then, the equation helps to frame our thinking about the potential prevalence of life in the Milky Way galaxy.

Given that NASA opined recently that we're likely to have strong indications of life beyond Earth within a decade, it made me wonder about Drake-like equations in medicine and epidemiology. As a toy example, suppose that we write the number of patients contracting hospital-acquired infections (HAIs) yearly in the US as the product of several factors, say

N = N_{hospital visits} x P_contact x P_{develop disease} x P_{disease reported}

where

• N_{hospital visits} is the number of patients visiting hospitals annually,
• P_contact is the probability that a patient comes into contact with infectious material (e.g., via environmental contamination or an infectious patient or HCW)
• P_{develop disease} is the probability of developing disease if infected, and 
• P_{disease reported} is the probability that an infection is recognized and reported.

According to the CDC, there are 35.1M hospital discharges annually in the US, so N_{hosp visits}~35M. Now suppose that P_contact and P_{develop disease} are both low, say 1% , and that we have excellent surveillance so that P_{disease reported} ~1. If that could be true, then we would expect to see 3,500 HAI per year. We should be so lucky. Being more realistic, however, might yield a 10% change of coming into contact to infection, P_contact~0.1, and a higher probability of contracting disease if infected, say 50%, so that P_{develop disease} ~ 0.5. In that case we get N=1.75M HAI annually, which is close to the CDC estimate of 1.7M.

How could this be decreased? The number of hospital visits, N, is unlikely to decrease drastically, so that's not really a control variable. Perhaps we could develop interventions to decrease P_contact and P_{develop disease}. Obviously there is tremendous focus on reducing P_contact through handwashing, alcohol based had rubs, contact precautions, better environmental cleaning, etc already. If P_contact could be reduced by a factor of 10, from 0.1 to 0.01 -- seemingly a tall order -- N could be dropped to 175K. That may not be possible, but suppose we could achieve a factor of 2 improvement so that P_contact ~ 0.05. If we could combine that with a similar decrease in P_{develop disease} by, say, better use of antimicrobials, then N could in turn fall from 1.7M to 438K. Thus, combination strategies could have great impact.

This is simply a back of the envelop calculation: the equation above is but an approximation and the estimates are completely arbitrary. Moreover, parameters will vary from facility to facility and even between patient populations (imagine how P_{develop disease} is likely to vary between transplant versus general surgery patients). That said, this toy model illustrates a simple point: Breaking a problem down into smaller pieces can be helpful in thinking about it.

While this way of thinking is not alien (pun intended) to biostatistics and epidemiology, and clearly has limitations, I think it's helpful for framing issues in one's mind. In addition to clearly laying out assumptions in whatever is being contemplated (in this case, HAI), toy model approaches can suggest what may be needed in order to get a better answer.

(image source: Wikipedia)

The digital epidemiology of Staphylococcus aureus

Digital epidemiology encompasses an emerging set of analytic techniques and approaches to data collection. Data in these studies are almost always born digital -- they are not recorded or transcribed by hand -- and often the research involves online networks in one guise or another. While these methods are being utilized increasingly, studies combining both digital network data and microbiological data on the spread of hospital associated pathogens have, so far as I know, been missing. 

Obadia et al have published an exemplary study doing just this for the case of MRSA and MSSA in a long term care center. Many researchers have in the past adopted a very reasonable and plausible hypothesis regarding the spread of staph in hospitals: namely, that it depends to a large extent upon person to person contact. If that's true, then obviously the ways in which patients and healthcare workers (HCWs) interact with one another, i.e., the patient-patient and patient-HCW contact networks, must be important for understanding spread. To my knowledge, until this study, nobody has really documented this with clarity at the individual level.

Obadia et al have illustrated this relationship between staph infection and contact network structure quite clearly by utilizing wireless proximity sensing and spa typing. They demonstrate how to employ digital technology to measure who interacts with whom, how frequently, and for how long, over long periods of time, and how to combine that data with microbiological surveillance in order to observe how transmission depends on the web of contacts in a facility. The authors found that close proximity interaction (CPI) paths existed between those colonized with like staph strains, and that those path lengths were significantly shorter than paths between random pairs in the study population. This is in agreement with what is expected from the transmission hypothesis. Their study also highlighted the importance of HCWs as links in the chain of contacts between infected patients.

One important implication of this work is that it might be possible to prevent infections by managing and monitoring close contact paths between patients and patients and between HCWs and patients. The approach may also be useful for developing targeted surveillance strategies that can detect spread and break the contact pathways most likely to result in further spread. I recommend reading the paper, and also the excellent comments regarding it by Eli Perencevich at the Controversies in Hospital Infection Prevention blog.

Overall, I think this study is a great illustration of the power of digital epidemiology methods for gathering detailed data in order to understand how disease is spread in the real -- as opposed to the simplified, theoretical -- world. We need more like it to inform both our thinking about hospital associated infections and analytic models of such pathogens.

(image source: CDC)

Some broad threats to public health

A recent Twitter thread highlighted several current threats to public health and I thought the points were sufficiently important to immortalize in a blog -- not necessarily because any one point is of primal importance (although each one alone is stunningly important for public health), but rather because we often forget to think holistically about public health. The reality is, of course, that many areas must combine in order to make good public health possible.

The thread highlighted three elements of public health that are all compromised to some extent at present: the effectiveness of antimicrobial drugs, the coverage of vaccination against vaccine preventable infectious diseases, and the preservation of sanitation infrastructure.

A few words about each of these. The specter of pathogens resistant to current antimicrobial drugs is well known. This topic is widely covered in the news media, in the scientific and medical literature, and even in political discourse. There is also a rich conversation on social media. Much has been written about the coming -- or, if you're a patient infected with a resistant pathogen, the present -- post-antibiotic era. The threat to public health is so great that the issue is now commanding economic and political attention, which hopefully will result in action soon.

And yet, antimicrobial resistance is not the only important threat to public health. The incidence of many vaccine-preventable diseases is increasing, not because pathogens are evolving and becoming mismatched to vaccines, but because significant numbers of people are electing to forgo having children vaccinated. The reasons why are varied and complex, but often they originate in mistrust between people and those who make and provide vaccines. Part of that mistrust was eroded by deeply flawed published research that has since been discredited; meanwhile, the effects and attendant impacts on human health continue. Moreover, vaccines are getting more expensive, and have been for years, which probably doesn't help the goal of increasing coverage, either.

Lastly, the sanitation infrastructure in many US cities is old, undersized, and crumbling. (It's not only the sanitation infrastructure that is failing or threatening to fail; transportation and power distribution are similar stories.) As a result, human waste is frequently released into the environment. This is remarkable for many reasons, not the least of which is that sanitation is one the oldest and most recognized cornerstones of public health. The undesirability of having human excrement handled improperly is so obvious that there's no reason to belabor the point here.

It's tempting to refer to these issues as horsemen of the public health apocalypse, but that would be bombastic and incomplete. There are other important threats, including the safety of the food supply, the high incidence of healthcare associated infections (both susceptible and drug resistant), the growing prominence of chronic diseases of the aging and the attendant demands on healthcare resources, and the continued emergence of new pathogens from nature.

To close, it's good to resurface from the depths of one's own research periodically. It can result in context and perspective, which is badly needed in any field of research. Much has been written about the use of Twitter in healthcare and biomedical research. Maybe this is another: it can force you to come up for air.

(image source: Wikipedia)

A fictional story: Counterpoint and extension

Today we welcome our first guest blogger, Dave Bridgelend, a mathematical modeler who focuses on business modeling and simulation. Dave has created business models for government agencies, financial services companies, telecommunications companies, energy companies, and healthcare organizations. He is the co-author of the book "Business modeling: A practical guide to realizing business value".

David Hartley recently described an attempt to reduce hospital acquired infections at the (fictional) Generic General Hospital. The attempt seemed successful, until the hospital CEO considered the broader impact of the effort. The budget for routine HAI prevention activities was raided to fund the improvement, and that budget diversion would likely cause unintended consequences.

But there is a bit more to this story. After the GGH CEO lectured her staff on unintended consequences, the GGH patient advocate spoke up. She pointed out that GGH already scored well on the HCAHPS measures of hospital acquired infections, earning ratings of "Better than the US national benchmark" on CAUTI and MRSA, and ratings of "No different form the US national benchmark" on the other four HAI measures.

"And when patients choose a hospital, they look at HAI a bit, but most of their decision is driven by what they hear from their personal networks. Last year when Martha was in the hospital for a week, the hospital made a big stink about her daughter staying in the room with her.  When Joe broke his leg, the hospital was too loud at night for him to sleep. When Carol had her hip replaced, she said the bathroom was dirty."

She continued: "I understand we want to improve GGH. But let's focus our attention on what the patients care about. Further reducing HAI won't bring them here. Keeping the bathrooms clean will."

(image source: Wikipedia)

Vaccines: A tool for the post-antibiotic era?

In honor of World Immunization Week this week, I recently read two books by Paul Offit: Vaccinated: One Man's Quest to Defeat the World's Deadliest Diseases and The Cutter Incident: How America’s First Polio Vaccine Led to the Growing Vaccine Crisis. Both are excellent. Vaccinated is essentially a biography of Maurice Hilleman, but it also reviews how several of the important vaccines currently in use were developed and marketed. The Cutter Incident tells the story of incompletely inactivated lots of polio vaccine manufactured by Cutter Laboratories, which caused 40,000 cases of polio nationwide in 1955, including 200 cases of paralysis and 10 deaths. There are many pearls and much wisdom to be found in the pages of these two books; I recommend reading them.

Certainly the utility of vaccines is well demonstrated and their development and application is one of the major accomplishments of modern medicine. In the US alone the improvement of population health as vaccines have become available is remarkable. Globally, it has been estimated that vaccines prevent nearly 6 million deaths annually worldwide.

The books got me thinking about future potential vaccines. In one passage, Offit recounts the development of a pneumococcal vaccine and quotes Robert Austrian talking about the rationale for his work:

The only alternative then to protect those at high risk of early death is to prevent them from becoming ill.

This beautiful and simple idea -- a medical and public health truism if ever there was one ("an ounce of prevention is worth a pound of cure") -- strikes me as relevant to HAI and antibiotic resistant infections. Think what healthcare might be like if there were vaccines for many of the bacterial infections that are currently problematic and often resistant to antibiotics, like Staphylococcus aureus, Clostridium difficile, and Neisseria gonorrhoeae.

Several antibacterial vaccines are available, including ones for pertussis, tetanus, diphtheria, meningococcus, pneumococcus, Haemophilus influenzae type b (Hib) disease, cholera, typhoid, and anthrax. However, there are reasons that vaccines for S. aureus, C. diff, and N. gonorrhoeae (as well as others) don't yet exist. For one, the immunology can be complex, as Offit explains in the discussion of the pneumococcal vaccine. Proctor describes the situation for Staph aureus in a recent review, as do Fowler and Proctor in another review. Also, the cost of developing, testing, and licensing can be steep relative to the profits of a licensed, marketed vaccine. Yet another issue is the specter of adverse events, both real and perceived. On this point, Offit notes that

. . . a technology that would clearly save lives sits on the shelf. "We could make a group B strep vaccine tomorrow," said one senior pharmaceutical company scientist. "But it would have to be given to pregnant women and we couldn't handle the liability." 

Dempsey et al offers a recent, interesting, and partially validating study to this view of a potential group B strep (GBS) vaccine. Such issues are difficult.

That being said, perhaps vaccines should be emphasized more in the conversation regarding antibiotic resistance. I've wondered in the past about the effectiveness of developing new antibiotics when there seems to be little reason to believe, given the past track record, that they will be used responsibly. A new generation of antibiotic drugs could become useless within a few years if effective antibiotic stewardship isn't practiced globally. Vaccines, if they could be made, may offer protection against what may soon be untreatable infections. Or put differently, perhaps vaccines could be an important tool in a post-antibiotic era.

Of course, there are issues to be better understood and addressed. Recent work illustrates that Bordetella pertussis is evolving in response to the vaccine, raising the possibility that future vaccines may be associated with similar dynamics. Also, vaccines to human commensals like Staphylococcus aureus might promote overgrowth of other commensal organisms. Studies have investigated this for the case of Streptococcus vaccination and MRSA colonization and infection. Moreover, it's unclear whether people would really embrace more vaccinations given the current and recent climate surrounding vaccines.

Regardless, one seldom hears about vaccines in the conversation about antibiotic resistance. It seems like funding should address making new vaccines as well as development of new antibiotic drugs -- because Robert Austrian was right. 

(image source: CDC/PHIL)

A fictional hospital story: Is it really fictitious?

This vignette is an amalgam of conversations with several colleagues over the last couple years. It is completely fictitious.

Surgical site infections (SSIs) had increased in the last year at Generic General Hospital (GGH) and the president of the hospital wanted action. The infection control staff met and decided upon a set of interventions for SSIs that they then implemented in their surgical step down unit. The interventions included a campaign to raise awareness of SSI risks, an increased emphasis on hand hygiene, and the implementation of a new wound care protocol. During a 60 day trial period, intensified surveillance in the unit suggested that these interventions had reduced SSIs by 50 percent.

While it might appear to be a success story, the CEO of Generic Medical System (GMS), the owner of GGH and several sister facilities, including Generic Rehabilitation, Generic Skilled Nursing Care, and Generic Community Oncology, was curious. When briefed on the success, she wanted to know if the intervention was sustainable in the step down unit and whether it was generalizable to infection control in other GMS institutions, which also had moderate infection rates. She further asked if the resources taken from the small infection control budget at GGH for this pilot had resulted in increased incidence of HAI in other hospital wards and units. And of course she needed to know if the interventions would save money in the short and longer terms.

At a meeting called to address these questions, GGH staff decided that only continued surveillance would answer the issue of sustainability. They believed the intervention was generalizable to sister facilities (but when challenged later, they couldn't explain why, other than offering “expert opinion”). The staff didn’t have data on potential changes in the incidence of infection in other parts of GGH because the money and resources dedicated to the step down unit pilot program had resulted in decreased surveillance in other wards. One of the three infection control nurses covering the entire hospital was dedicated to the pilot, leaving a 33% reduction in person-power needed to collect and chart the statistics in the rest of GGH. Nobody had any idea of cost savings.

The CEO wasn't pleased. She explained to her leadership team the importance of sustainability, generalizability, avoidance of unintended consequences, and improving the bottom line. She noted that hospitals are zero sum enterprises: they have finite budgets, and in the absence of grants or donations to support trial patient safety interventions, what is focused on one activity must be taken away from other activities. In the example of the SSI interventions, she noted that the intensified surveillance dedicated to assessing that program diverted surveillance resources and left “blind spots” elsewhere in the institution.

She illustrated the point in the following way. Of the $10K GGH annual budget for HAI prevention, $2K was spent on posters and buttons promoting hand hygiene in the step down unit. Normally, that $10K was evenly distributed amongst wards across the hospital, but with this special project only $8K was left for routine prevention activities in other units. It wasn’t unreasonable to expect that there might be increases in HAIs in those wards, assuming the yearly $10K investment was having an effect in the first place.

To illustrate, she asked them to suppose that had been an increase in HAI in the transplant unit following the decreased prevention investment there. During the study SSI evaluation period -- the period of intensified surveillance in the step down unit -- it probably would have gone unnoticed, as part of the infection control staff were assessing the interventions in the surgical step down unit, instead of collecting routine HAI surveillance data in the rest of the hospital. In that case, the negative impact in the transplant ward would have been an unobserved and unintended consequence of the otherwise successful interventions in the surgical step down unit.

The CEO lectured her staff on the phenomenon of unintended consequences, cautioning them that the degree to which changes in standard operating procedures (e.g., hospital infection control or other patient safety interventions) result in unintentional consequences is unknown and requires additional research. She lamented the lack of a research group to address these and related issues. Perhaps one day, she thought, someone would do the research and develop evidence-based guidelines on the relevant issues. Until then, the leaders of GMS had limited options for reducing HAI throughout their network of facilities.

There are several key points illustrated in the vignette. First, unintended consequences of seemingly good ideas can occur. Second, existing programs can be impacted when new initiatives are launched if additional resources aren't provided. Third, because generalizable approaches are desirable, it is critical that researchers anticipate questions such as those the CEO asked her staff.

(image source: Wikipedia)

Infection prevention knowledge: Permutations on "known" and "unknown"

Pathogens are everywhere, and we are increasingly aware of how widely they can be disseminated in healthcare environments. Researchers have found, for example, that

• Well-child visits are a risk factor for subsequent influenza-like illness visits. Infections are thought to spread in waiting and exam rooms.

• Hospital water taps can be contaminated with bacteria including Legionella spp., Acinetobacter spp. and other Gram-negatives. 

• Hospital surfaces and equipment are sometimes contaminated with nosocomial pathogens including Clostridium difficile, VRE, and MRSA.

• Doctors' ties and white coats are anything but sterile.

• Toilet seats can harbor MRSA and probably other pathogens. 

• Hospital privacy curtains can be contaminated with pathogenic bacteria including MRSA and VRE.

• Stethoscopes can carry MSSA and MRSA, among others. 

The list could go on and on. We know that many pathogens can persist on surfaces for a considerable time, and while it's not clear the extent to which contaminated surfaces play a role in HAI in general, there is reason to believe they are important in many infections. The unfortunate reality is that patients suffer nosocomial infection; contaminated surfaces can only add to the risk of infection.

I never thought I'd invoke Donald Rumsfeld in a discussion of infection control, but he once described a useful construct for thinking about infection prevention (among other things). He's quoted as saying

. . . there are things we know that we know. There are known unknowns; that is to say, there are things that we now know we don't know. But there are also unknown unknowns – there are things we do not know we don't know.

If we add another, obvious, category -- things we don't know that we know -- then a 2x2 table can be written for types of knowledge. Done for infection prevention and control, it might look like the table below.

It would be interesting to organize what we know and don't know into such a table. That's a big thing to do; it requires assessing what specific practices are truly evidence-based ("known knowns", like have been described for central line infections and ventilator-associated pneumonia), identifying best practices that aren't necessarily well studied ("unknown knowns"), and enumerating gaps in our knowledge ("known unknowns", such as the role of contaminated surfaces in HAI). Of course, we can never identify the things we don't realize that we don't know (the "unknown unknowns"), but the hope would be to ultimately understand the other three quadrants well enough so that we are sure that the unknown unknowns aren't important. Obviously, that's hard to do.  

It seems to me that seriously trying to fill in the quadrants is an important step towards a complete theoretical picture of infection. Probably the "known knowns" quadrant is smaller than we would hope, and the "known unknowns" quadrant is significant. I wonder how large the "unknown knows" category -- the things we don't realize we know -- is?

(image source: the E. coli micrograph, Wikipedia; the 2x2 table, David Hartley)