Reply to Evans et al

TO THE EDITOR—We appreciate the letter from Drs Evans, Roselle, and Kralovic who lead the National Veterans Affairs (VA) Infectious Disease Service [1]. As VA investigators and hospital epidemiologists, we have worked together collegially for years.

First, we would like to clarify our work [2]. We evaluated the effectiveness of individual facility level policy for contact precautions on the outcome of methicillin-resistant Staphylococcus aureus (MRSA) acquisition based on the gold standard of surveillance cultures and MRSA healthcare-associated infection (HAI) rates. We reviewed MRSA prevention policies from 74 facilities from 2011 to 2015. Most facilities used standard precautions for MRSA before 2013 and then adopted contact precautions, as depicted in Figure 3 in our paper. We looked at individual patient data across 75 414 admissions. Not surprisingly, MRSA acquisition was associated with MRSA colonization pressure, pressure ulcers, and need for healthcare worker assistance with activities of daily living. Surprisingly, on a patient level, we found no effect of contact precautions versus standard precautions on MRSA acquisition or MRSA HAI rates. Of note we did find an overall decrease in MRSA HAI rates regardless of policy.

Dr Evans and colleagues comment that “infection control is working” in VA nursing homes, and we agree. As noted above, we also documented declining MRSA HAI rates similar to work by Dr Evans [3]. We were interested in whether one part of the VA MRSA initiative bundle, contact precautions, led to the MRSA HAI rate decrease. Based on our work, we do not believe the policy of contact precautions led to the decrease in MRSA HAI rates because it did not decrease MRSA acquisition. Contact precautions are used to prevent transmission and acquisition, which would then hypothetically decrease HAIs in the causal pathway. If contact precautions do not decrease transmission and acquisition, they cannot logically decrease HAIs.

We respectfully disagree with Dr Evans and colleagues and assert that it is valid to evaluate contact precautions as a policy in an effectiveness study. Contact precautions are a policy intervention and have to be evaluated as such. That they are not adhered to 100% of the time is a limitation of real-life use. We disagree that facilities with low compliance should be “more appropriately assigned by the authors to the standard precautions group for statistical analysis.”

What should we do? We should consider more targeted approaches to the use of gowns and gloves as is currently recommended by Centers for Disease Control and Prevention guidance [4]. We should focus their use on residents at high risk of acquiring and becoming infected with antibiotic resistant bacteria. In long-term care, most staff-patient contact occur outside the resident’s room [5], More time could be spent on resident hand hygiene, which is part of the current VA MRSA Initiative.

We owe it to our veterans to use the best data we have to develop the most effective and least onerous policies for preventing infection.

Note

Potential conflicts of interest. D. M. reports a merit award to evaluate impact of isolation use in long-term care from VA Health Services Research and Development; grants from the Centers for Disease Control and Prevention (CDC), the National Institutes of Health, and the Agency for Healthcare Research and Quality; travel expenses from the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America; and honoraria from Springer Nature Inc. M. C. R. reports grants from the CDC and National Institute of Allergy and Infectious Diseases. L. P. has no potential funding or conflicts to disclose. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References