Healthcare personnel (HCP) frequently become contaminated with antibiotic-resistant bacteria on their gloves and gown when caring for patients who are colonized or infected with these organisms. Reported rates of contamination with methicillin-resistant Staphylococcus aureus (MRSA) range from 17% to 20%. Reference Okamoto, Rhee and Schoeny1,Reference O’Hara, Calfee and Miller2 Almost all of these studies were conducted in the intensive care unit (ICU) setting, the nursing home, and/or long-term care settings. Reference O’Hara, Calfee and Miller2,Reference Roghmann, Johnson and Sorkin3 To our knowledge, no studies of transmission to HCP gloves and gown have been conducted in non-ICU acute-care hospital wards.
The aim of this study was to determine the rate of MRSA transmission to HCP gloves and gowns in hospital wards and to assess whether the bioburden on the patient was associated with transmission.
Methods
In this prospective cohort study, MRSA patients were randomly selected among all eligible non-ICUs from the electronic health record at a 750-bed, academic, urban, tertiary-care center in the United States between August 2018 and March 2019. All patients were on contact precautions and were admitted in non-ICU settings.
We observed and cultured the gloves and gowns of 10 sequential HCP interactions per patient following patient care but prior to doffing gloves and gown, as described previously. Reference Snyder, Thom and Furuno4–Reference Jackson, Harris and Magder6 Each HCP was observed, and care activities were recorded on a standardized data collection form.
From MRSA patients, we collected samples from the anterior nares, chest, antecubital fossa, and perianal area using Eswabs (Copan Diagnostics, Murrieta, CA). Methodology and laboratory procedures are described elsewhere. Reference O’Hara, Calfee and Miller2
Descriptive statistics were performed to summarize patient demographic and clinical characteristics. The Pearson χ2 test, t test, Spearman correlation, and logistic regression were performed, as appropriate, to assess relationships between risk factors and glove and gown contamination. All analyses were conducted using SAS version 9.4 software (SAS Institute, Cary, NC). The Institutional Review Board of the University of Maryland approved the study and waived patient consent.
Results
Our cohort consisted of 55 unique patients with a history of MRSA colonization or infection. Demographic and clinical characteristics are presented in Table 1. Of these 55 patients, 22 (40%) were on contact precautions for a positive clinical culture, 17 (31%) were on contact precautions for a positive nasal surveillance culture, and 16 (29%) were on contact precautions for a prior history of MRSA with no record at the current institution of a positive clinical or surveillance culture. Of the 39 patients with a record of positive clinical or surveillance culture, nearly half (n = 18, 45%) had their last positive report >1 year prior, and the median number of days since the last positive MRSA culture or surveillance culture was 225 (range, 5–5,857). Only 5 patients (12.8%) had had a positive clinical or surveillance culture for MRSA in the prior 7 days. We observed 517 HCP–patient interactions, of which 16 (3.1%) led to contamination of an HCP glove, 18 (3.5%) led to contamination of an HCP gown, and 28 (5.4%) led to contamination of either a gown or a glove.
Description of Patients on Contact Precautions for MRSA in Non-ICUs From August 2018 to March 2019 (N = 55)
Note. MRSA, methicillin resistant Staphylococcus aureus; ICU, intensive care unit; HIV, human immunodeficiency virus; SD, standard deviation.
a Psychiatry, mother baby unit, labor and delivery, telemetry, and pediatric non-ICU wards were not included in the study.
b At the time of enrollment in the study, 46 patients (83.6%) had at least 1 patient characteristic.
HCP type
The HCP were aggregated into 2 groups: “direct patient care” and “no direct patient care.” The direct-care group included nurses (n = 264, 51%), physicians (n = 63, 12%), patient care technicians (n = 54, 10%), occupational and physical therapists (n = 17, 3%), respiratory technicians (n = 4, 0.7%), and radiology technicians (n = 2, 0.3%). The no-direct-care group included environmental service employees (n = 40, 8%), food service employees (n = 45, 8%), case managers (n = 3, 0.5%), social workers (n = 4, 0.7%), patient transporters (n = 3, 0.5%), clinical nutritionists (n = 2, 0.3%), and phlebotomists (n = 3, 0.5%).
As shown in Table 2, in the direct-care group, 26 of 404 HCP (6.4%) showed transmission of MRSA to glove or gown, compared to 2 of 113 providers (1.8%) in the no-direct-care group (odds ratio [OR], 3.93; 95% confidence interval [CI], 0.89–17.44).
Association Between HCP Type, Patient Bacterial Bioburden and Transmission of MRSA to Glove or Gown in Patients in Non-ICU Units Colonized With MRSA
Note. HCP, healthcare personnel; MRSA, methicillin-resistant Staphylococcus aureus; ICU, intensive care unit; HCP, healthcare personnel; OR, odds ratio; CI, confidence interval; CFU, colony-forming units.
a OR adjusted for healthcare personnel contact time.
b Direct patient care: An aggregated group of HCP with direct patient contact that includes nurses, physicians, patient care technicians, respiratory technicians, radiology technicians, and occupational and physical therapists.
c No direct patient care: An aggregated group of HCP with no direct patient contact that includes environmental service employees, food service employees, case managers, social workers, patient transporters, clinical nutritionists, and phlebotomists.
d MRSA bioburden (CFU/mL or CFU/mL2) detected at any 1 or more of the sampled sites. Of the 55 patients from whom we obtained samples, 54 (98.1%) had nasal swabs, 54 (98.1%) had chest swabs, 54 (98.1%) had arm swabs, and 41(74.5%) had perianal swabs.
e MRSA bioburden (CFU/mL or CFU/mL2) not detected at any sampled sites.
Time spent by the HCP in patient rooms
Overall, HCP spent a median of 5 minutes (interquartile range [IQR], 3–10 minutes) in a patient’s room. We detected a significant difference in the mean time spent in the room by HCP with MRSA transmission to glove or gown (mean, 11.1 minutes; SD, 10.0) compared to HCP without MRSA transmission (mean, 7.2; SD, 6.3; P = .002).
MRSA bacterial burden
The median MRSA bioburden in the nares was 0 log10 CFU/mL (range, 0–3.6): 0 log10 CFU/mL (range, 0–3.5) in the perianal region, 0 log10 CFU/cm2 (range, 0–0.3) on the arm skin, and 0 log10 CFU/cm2 (range, 0–6.2) on the chest skin. Detectable bioburden on patients was negatively correlated with the time since MRSA was first detected (rs = −0.06; P < .001). In addition, 9 of 97 observations (9.3%) with any detectable bacterial bioburden at any site resulted in transmission of MRSA to HCP, compared to 11 of 310 observations (3.6%) in which no detectable bioburden occurred at all sites (OR, 2.84; 95% CI, 1.02–7.87).
Discussion
Among non-ICU patients on contact precautions for MRSA, HCP gloves and gowns become contaminated 5.4% of the time, and direct patient contact was associated with higher rates of transmission to HCP. Patients who had any detectable MRSA on their skin or in their nose or perianal region were more likely to transmit MRSA to HCP. Finally, the longer the time spent by HCP in the patient room, the greater the likelihood of MRSA transmission, as has been reported in previous studies. Reference O’Hara, Calfee and Miller2,Reference Jackson, Harris and Magder6
The MRSA transmission rate to HCP gloves or gown of 5.4% in the non-ICU setting is much lower than the 16% rate reported by a large, multistate cohort study conducted in the ICU setting using identical methods. Reference O’Hara, Calfee and Miller2 However, the study conducted in the ICU required patients to have positive clinical or surveillance cultures for MRSA in the 7 days prior to enrollment. In this study, only 5 patients (12.5%) met this criteria. Most patients were on contact precautions for a positive clinical or surveillance culture that occurred more than a year prior, yet the use of a clinical indicator to initiate contact precautions, often based on previous cultures, is a standard practice. The major limitation of this study is that the sample size is much smaller than the ICU-based studies conducted by our group and by others. Reference O’Hara, Calfee and Miller2,Reference Roghmann, Johnson and Sorkin3,Reference Pineles, Morgan and Lydecker5 Our smaller study population limited our ability to stratify by last positive culture, which may account for the wide confidence intervals. Also, some of the patients included in the study may have no longer been colonized with MRSA. Finally, genetic relatedness using molecular epidemiological techniques was not studied.
The use of contact precautions for patients with MRSA is recommended by the Centers for Disease Prevention and Control. 7 However, many hospitals in the United States are moving away from the use of contact precautions for MRSA. Reference Marra, Edmond and Schweizer8,Reference McKinnell, Eells and Clark9 The Society for Healthcare Epidemiology of America issued a guidance document on the duration of contact precautions for acute-care settings. Reference Banach, Bearman and Barnden10 More evidence is needed to help guide the optimal use of contact precautions: the use of contact precautions for the right patient, in the right setting, for the right type of encounter. Our findings suggest the need for a larger, geographically diverse, cohort study of ward patients with MRSA. If our results can be replicated, it may be appropriate for some hospitals to consider safely removing contact precautions from ward patients while retaining contact precautions for higher risk groups such as ICU patients.
Acknowledgments
We thank Stephanie Hitchcock from the Pathology Department of the University of Maryland for specimen processing.
Financial support
This work was supported by the NIH National Institute of Allergy and Infectious Diseases (grant no. R01 AI121146-01 to A.D.H.).
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
