Bad law = bad medicine: PA plan to test health workers may spur drug resistance

Provision aimed at patient safety may actually undermine it

In a move that runs counter to national public health guidelines and may contribute to the rise of drug-resistant pathogens, Pennsylvania has passed a state law that could lead to routinely culturing a wide variety of health care workers for methicillin-resistant Staphylococcus aureus (MRSA) and other multidrug-resistant organisms (MDROs).

Moreover, the Pennsylvania provision for screening health care workers in the absence of an outbreak or ongoing transmission runs directly counter to current guidelines by the Centers for Disease Control and Prevention, potentially putting every hospital in the state out of line with the prevailing standard of care. As a result, the Hospital & Healthsystem Association of Pennsylvania (HAP) is essentially telling member facilities to report they have no way to comply with such a provision because there are no established protocols to do so. Not so fast, says the state health department, which is charged with drawing up some kind of regulations to comply with the law and is in the process of doing just that. The whole thing arose from an ill-fated attempt at political compromise among lawmakers and the governor's office, which originally wanted all health care workers screened.

"These kind of [universal screening] strategies are promoted only by people who are unfamiliar with the biology of MRSA, the antimicrobials available, and the development of resistance," warns William Schaffner, MD, chairman of the department of preventive medicine at Vanderbilt University Medical Center in Nashville. "The less trenchant your knowledge of the subject is, the simpler it appears and the more Draconian the solutions that you find proposed. Then we in infection control find ourselves playing both defense and catch-up. We simply have to be better connected [with the state legislatures]."

In addition to a requirement for active surveillance cultures (ASC) for high-risk patients such as those admitted from long-term care facilities, the Pennsylvania law calls for hospitals to include in their infection control plans "procedures and protocols for staff who may have had potential exposure to a patient or resident known to be colonized or infected with MRSA or MDRO, including cultures and screenings, prophylaxis and follow-up care." The health care worker screening provision has caused an uproar since it was included in the July 12, 2007 law (Senate Bill No. 968).

"We are struggling with that." says Sharon Krystofiak, MS, MT(ASCP), CIC, infection control manager at Mercy Hospital in Pittsburgh. "We have no idea where they pulled this verbiage out of thin air. It says something like according to established protocols. There aren't any. It never has been [a recommended practice]. Unfortunately, Pennsylvania has taken the tact of passing laws and then figuring out what they mean."

The infection control community in Pennsylvania is rallying to try to get some reasonable interpretation of the provision as enforcement and compliance mechanisms are honed in the weeks and months ahead. "The aspect of culturing staff is particularly broad — any staff that are exposed to MRSA," says Patrick J. Brennan, MD, chief of healthcare quality and patient safety for the University of Pennsylvania Health System in Philadelphia. "You can interpret that to be anybody that takes care of a patient with MRSA. I'm not sure how they intend to implement that or what the implications are going to be. I think a lot of this is going to be subject to department of health interpretation. If this is strictly and broadly interpreted it could be very onerous."

Could contribute to rise in drug resistance

Casting such a wide net for health care workers who may be only transiently colonized with MRSA will open up a host of unintended consequences, clinicians warn. Those include a possible rise in drug resistance due to the overuse of mupirocin in an attempt to eradicate MRSA or "decolonize" workers. Such practices have long been known to select out resistant pathogens, leading to the old infection control axiom of "use it and lose it" when it comes to antibiotics. "Absolutely, it's not even a concern — It's a guarantee," says Steve Weber, MD, a health care epidemiologist at the University of Chicago. "All the recommendations for using mupirocin say avoid widespread or empirical use because it will be associated with mupirocin resistance."

As a result, the drug will be less effective when truly needed and a legal provision legislative framers said was designed to bolster patient safety could actually undermine it. "You start using mupirocin widely and that is an absolutely predictable formula for developing mupirocin resistance," says Schaffner. "The literature is very clear on this. Some of these [health care workers] I'm sure will get prolonged or repeated courses. If you want to create mupirocin resistance, that's the way to do it."

With resistance already appearing in staph mainstay drug vancomycin, any erosion of mupirocin efficacy would be a serious matter. "We're really strapped — you can use chlorhexidine showers [as a decolonization option] but it's not as effective as mupirocin," he says. "And mupirocin is not a magic wand. You don't automatically decolonize everyone who gets a five-day course, for example. Then another course will be given. I can see that it won't take very long before we start getting substantial portions of MRSA that are resistant to mupirocin."

A cautionary tale

Indeed, the situation in Pennsylvania is being viewed as a cautionary tale for other states facing increasing involvement of legislatures in the clinical practice of infection control. In particular, more states are considering requiring ASC to identify patients with MRSA before they can spread the bacteria to others. However, it appears Pennsylvania is the first state to include screening health care workers, which is generally done by nasal swab and culture.

"There are exceptions, but the literature on MRSA largely suggests that health care workers' nasal carriage is not the source of spread of MRSA to patients within health care facilities," Schaffner says. "This [Pennsylvania] legislation appears misguided and is unlikely to have a noteworthy impact on MRSA transmission within health care facilities. It will raise all kinds of issues having to do with how frequently institutions should do [screening], who indeed is considered a health care worker in the context of this legislation, and what institutions ought to do after they have identified such an individual."

ASC for high-risk patient groups is controversial in its own right, but is generally seen as a potentially effective way to identify and isolate MRSA colonized patients so they can't spread infection to others. The health care worker screening provision — which sounds reasonable enough in the abstract — is another matter entirely.

Epidemiologists and ICPs have long emphasized that it is generally futile and counterproductive to routinely search for health care workers colonized with staph in the absence of an outbreak. Culturing the nares of health care workers does not address the primary threat to patients, which is the transient colonization of worker's unwashed hands as they go from patient to patient. Thus, hand hygiene between patients is the cardinal rule for health care workers to follow.

"We reserve screening for individuals who have an epidemiological association with some transmission event," explains Michael Bell, MD, medical epidemiologist in the CDC division of health care quality promotion. "If there is some reason to consider screening that would be fine. But if you screen at random there is the challenge of interpreting the results. If somebody comes back with one or two organisms, do you continue to reculture them for some period of time? Do you try to decolonize them? If you can't decolonize, do you fire them? I don't think you would be allowed to actually. There are all sorts of ramifications of a random culture that makes that a routine practice that we do not normally advise."

Rarely, persistently colonized workers will cause infections in patients, a legitimate issue that does warrant screening to protect patients. The CDC's latest guideline on MDROs includes a section on the issue of colonized workers that states: "Occasionally, HCP [health care personnel] can become persistently colonized with an MDRO, but these HCP have a limited role in transmission, unless other factors are present. Additional factors that can facilitate transmission include chronic sinusitis, upper respiratory infection, and dermatitis."1 In a report published just this year, a neonatal specialist persistently colonized with a MRSA strain that eventually became mupirocin-resistant was implicated as a recurrent source of transmission in a newborn nursery.2

In general, the CDC recommends that hospitals should obtain cultures of health care personnel when there is epidemiologic evidence implicating the worker as a source of ongoing transmission. When decolonization for MRSA is used, perform susceptibility testing for the decolonizing agent, the CDC recommends. Limit decolonization efforts to culture-positive workers who have been epidemiologically linked as a likely source of ongoing transmission to patients. Consider reassignment of workers if decolonization is not successful and transmission to patients persists, the CDC emphasizes. "With MRSA, what we see is that people have skin eruptions, broken skin, eczema, who become colonized on that part of their skin and can be transmitters of infection," Bell says.

Such targeted interventions are much preferred by epidemiologists rather than the routine screening that could result from trying to follow up — as it is worded in the Pennsylvania provision — "potential contact" with patients with MRSA or other MDROs. "The majority of human beings carry Staphylococcus aureus of one form or another for some period of time," Bell says. "It is part of the skin and nasal floral in many cases. That is actually not abnormal. It is only when the organism goes some place that it shouldn't — a surgical wound or an IV catheter — that's a problem. There are plenty of organisms that we all carry, and if you tried to eradicate them you would actually cause disease because we need some of these organisms to ensure our health."

For example, pathogens in fecal flora are routinely transmitted by unclean hands, but simply practicing hand hygiene interrupts the chain of transmission. "The answer is not to check who is carrying fecal flora — because we all are," he says. "Nor is the answer to try and eradicate it. The key is to keep it from going where it needs to go [to cause infection]. Hand colonization, transient or otherwise, is really the major culprit."

Another factor to consider, however, is the rise of community-associated MRSA (CA-MRSA), which means health care workers may be just as likely to become colonized in the community as at work. "With the epidemic of CA-MRSA, you can speculate that you will have more and more health care workers who happen to be colonized or infected unrelated to their hospital work," Weber says. "So it may well be that we could have a higher prevalence of health care workers who are incidentally colonized."

The changing epidemiology of staph strains could reopen some basic questions about whether colonized workers could indeed be responsible for a low level of ongoing transmission to patients. Though no fan of legislating clinical practice — or widespread screening of workers for that matter — Weber warns that the traditional dogma about MRSA may be changing.

"I understand that even acknowledging that there is a remote or finite possibility of transmission [from colonized workers to patients], I stand at odds with some of my colleagues in the field," Weber says, "but with this bug and the way the epidemiology is changing in the last five or six years, I don't think any of us are in a position to exclude anything. There is a lot we don't know. But we also don't know enough to enact legislation or even accept this [worker screening] as a standard practice. I would reserve screening health care workers for situations where you have a cluster or a sharp increase in transmission on a specific unit or care area where there is epidemiologic evidence to suggest a point source."

Studies have found colonization rates in health care workers from 0.8% up to 9.6%, he notes. Again the rise of CA-MRSA may change that equation, particularly the USA300 strain that has been detected in some 40 states. In some individuals, carriage can be brief or go on for months and even years.

"In general, people fall into categories of non-carriers, persistent carriers or so called intermittent carriers — people who from month to month and assay to assay don't reliably show themselves to be positive," Weber says. "You want to have a good handle on those kinds of wrinkles before you enact legislation. Without that kind of information it is going to be hard to interpret or develop fair and equitable employment policies that will actually reflect the biological status of the individuals."

Indeed, in addition to decolonization efforts there are issues of workers' compensation and reassigning or furloughing staff in an era of a nursing shortage. "It's a big can of worms," Krystofiak says. "There could be serious financial and personnel consequences based on something that is totally unsupported by anything that has ever come out as a recommendation."

No one on the clinical side is minimizing the importance of preventing MRSA infections. Indeed Krystofiak emphasizes patient isolation and hand hygiene while maintaining an active data system that can red-flag up to 4,000 patients previously admitted with an MDRO. "If they are readmitted we put them into isolation on admission," she says. "Guilty until proven innocent. We have been very proactive on this, but all of our staff members are reading this and [wondering] do you think we are positive? I wouldn't think so [because they] have been putting on the appropriate gowns and gloves. If you sanitize your hands before and after patient care, it shouldn't be an issue."


  1. Siegel JD, Rhinehart E, Jackson M, et al. Centers for Disease Control and Prevention. Healthcare Infection Control Practices Advisory Committee. Guideline for Isolation Precautions: Preventing Transmission of Infectious Agents in Healthcare Settings 2007. On the web at
  2. Mean M, Mallaret MR, Andrini P, et al. A neonatal specialist with recurrent methicillin-resistant Staphylococcus aureus (MRSA) carriage implicated in the transmission of MRSA to newborns. Infect Control Hosp Epidemiol 2007; 28:625-628.