Nursing Home Roommates and MDRO Carriage: MRSA, ESBL, and VRE Study Findings

Nursing home residents who shared a room had higher odds of carrying the same multidrug-resistant organism (MDRO) than residents who did not share a room, according to a poster presented at IDWeek in Washington, DC, held from October 21 to 24. Universal decolonization was associated with lower matching carriage across all 3 organism categories examined, although the researchers could not establish causality or confirm transmission through genomic testing.

The study, “Impact of Roommate Status on Multidrug-Resistant Organism Carriage in Nursing Homes Along with Mitigating and Exacerbating Factors for MRSA, ESBL, and VRE,” included 4,851 residents from 44 nursing homes. Investigators evaluated matching carriage of methicillin-resistant Staphylococcus aureus (MRSA), extended-spectrum beta-lactamase producers (ESBL), and vancomycin-resistant enterococci (VRE).

“Without genomic data, it is unclear whether matching carriage represents transmission or common background colonization,” the researchers noted. This distinction is important because the study assessed whether both residents in a pair tested positive for the same MDRO category, rather than establishing that one resident transmitted an organism to the other.

Examining Roommate Status Across 44 Nursing Homes
The poster identifies nursing homes as important reservoirs for MDROs. Unlike hospitals, where single rooms are increasingly common, nursing homes predominantly have multibed rooms and encourage communal activities. Shared rooms may create opportunities for transmission through close contact, shared surfaces, and interactions with health care workers.

Researchers conducted a secondary analysis of residents randomly sampled during MDRO point prevalence sweeps. Sampling occurred at baseline in 2016–2017 and during universal decolonization in 2018–2019. Participating facilities included 28 nursing homes in the Protect Trial and 16 in the SHIELD Orange County regional collaborative.

Each sweep included bilateral nares swabs for MRSA and axilla/groin swabs for MRSA, VRE, and ESBL. Samples were drawn from a random selection of 50 occupied beds, or from all occupied beds, per nursing home.

Investigators generated all potential resident pairs within each nursing home sweep and classified them as roommates or non-roommates. Roommate pairs cohorted because of their MDRO status were excluded. The study reported 158,540 non-roommate pairs and 2,034 roommate pairs.

Higher Odds of Matching Carriage Among Roommates
After adjustment, roommates had 40% higher odds of concordant MRSA carriage than non-roommates. The adjusted odds ratio was 1.4, with a 95% confidence interval of 1.1–1.8 and a P value of less than 0.001. Mean predicted probabilities of matching MRSA carriage were 17.7% for roommates and 15.3% for nonroommates.

For ESBL, roommates had 50% higher odds of matching carriage. The adjusted odds ratio was 1.5, with a 95% confidence interval of 1.0–2.4 and a P value of 0.05. Predicted probabilities were 6.6% and 5.2%, respectively.
The strongest roommate association was observed for VRE, with 120% higher odds of matching carriage. The adjusted odds ratio was 2.2, with a 95% confidence interval of 1.1–4.4 and a P value of 0.04. Predicted probabilities were 3.8% for roommates and 2.8% for non-roommates.

The models accounted for sex, wounds, and indwelling devices. They also adjusted for a stay exceeding 30 days and decolonization status.

Resident Factors and Decolonization Were Associated With Carriage

Model-predicted probabilities identified additional factors associated with matching carriage. Wounds and devices were associated with increased matching carriage, while stays exceeding 30 days were associated with increased matching MRSA and ESBL carriage. Male pairs had higher concordant MRSA and VRE carriage, whereas female pairs had higher concordant ESBL carriage.
Universal decolonization, compared with routine care, was associated with lower concordant carriage across all 3 categories. Matching MRSA carriage was 36% lower, at 10.5% vs 16.1%. Matching ESBL carriage was 66% lower, at 1.2% vs 3.5%, and matching VRE carriage was 93% lower, at 0.09% vs 1.2%.
The researchers described this association as “suggesting a potential protective effect through reduction of skin bioburden.” Their findings also suggested that higher-prevalence organisms, such as MRSA and ESBL, may spread through broader facility-wide exposures.

Findings Have Important Limitations
Concordance was based on culture positivity rather than genomic clonality, and the study could not assess causality. Only 2 body sites were evaluated. The researchers noted that the lack of perirectal sampling likely affected ESBL and VRE estimates.

The findings associate shared rooms with matching MDRO carriage and identify resident characteristics that may influence this relationship. They also associate universal decolonization with lower concordant carriage, supporting further consideration of how room sharing and infection prevention interventions interact in nursing homes.

Poster authors: Gabrielle M. Gussin, PhD; Ken Kleinman, ScD; Raveena D. Singh, MA; Thomas Tjoa, MPH, MS; Gabriel Gadia, BS; Justina Bui, BS; and Susan S. Huang, MD, MPH.

Affiliations: Division of Infectious Diseases, University of California Irvine School of Medicine, Irvine, California; and University of Massachusetts Amherst, School of Public Health, Amherst, Massachusetts.
Source: IDWeek abstract 2417814.