ľֱ

Decolonization in Nursing Homes Cuts the Risk of Dangerous Infections

— Protocol borrowed from the ICU continues to show success

MedpageToday
A photo of a female nurse washing a seniors leg in a nursing home.

Universal decolonization in nursing homes -- by way of a chlorhexidine wash during routine bathing paired with a nasal antiseptic -- reduced the risk of infection-related hospitalizations, a cluster-randomized trial showed.

Across 28 nursing homes, the proportion of hospital transfers due to infection in facilities randomized to the protocol dropped from 62.9% during a baseline period to 52.2% over the course of the 18-month intervention (risk ratio [RR] 0.83, 95% CI 0.79-0.88).

Meanwhile, no change was observed for facilities that continued their routine care (62.2% to 62.6%, respectively; RR 1.00; 95% CI 0.96-1.04), resulting in a significant between-group difference, reported Loren Miller, MD, MPH, of the David Geffen School of Medicine at University of California Los Angeles, and colleagues.

Furthermore, secondary outcomes of the study, which was published in the , suggested that the intervention led to drops in hospital transfers for any reason along with a reduction in multidrug-resistant organism (MDRO) carriage among the nursing home residents.

"Given the growing concern of harm from antimicrobial-resistant germs in nursing homes, the impact of this decolonization bathing program is quite important and provides a concrete action that facilities can use to protect nursing home residents," co-author Nimalie Stone, MD, of CDC's Division of Healthcare Quality Promotion in Atlanta, said in a .

While universal decolonization had already been proven in randomized trials involving patients in and in , its routine use in nursing homes had not been tested.

The cleansing protocol in this study involved an antiseptic chlorhexidine wash during routine bathing and showering (2% or 4% formulation) along with nasal iodophor (povidone-iodine) administered twice daily by a nurse on Monday through Friday every other week.

"Compared to other healthcare strategies, this is a relatively simple win for nursing homes," co-author Susan Huang, MD, MPH, of the University of California Irvine School of Medicine, said in the release. "We hope nursing homes will want to adopt it."

Nursing home residents are at particularly high risk for healthcare-associated infections, which each year in the U.S. lead to an estimated 150,000 admissions and up to 380,000 deaths in this population.

Given their advanced age, underlying conditions, use of medical devices, and risk for wounds, residents are vulnerable to MDROs, noted Miller and colleagues, and the prevalence of MDRO carriage is far higher in nursing home residents compared with hospital patients (65% vs 10-15%).

For the so-called Protect trial, the researchers randomized 28 nursing homes providing skilled nursing care 1:1 to either the decolonization protocol or to continued use of the facility's usual routine care. Facilities were excluded if they already performed decolonization or were specialized in psychiatric, dementia, or pediatric care.

Data included a total of 28,956 nursing home residents, with 15,004 during the 18-month baseline period (September 2015 to February 2017) and 13,952 during the 18-month intervention period (July 2017 to December 2018), which followed a 4-month phase-in period for staff training.

Residents had a mean age of about 76 years, and a majority were women. Residents' racial and ethnic makeup across the study periods was 40-51% white, 11-16% Black, 14-20% Asian, and 19-21% Hispanic. More than 40% had diabetes, about a fourth had chronic lung disease, and roughly 20% had renal failure.

Adherence to the decolonization protocol was considered "reasonably high," according to the researchers: ranging from 87.4% with chlorhexidine during routine bathing to 95.6% at initial admission; and from 60.3% with nasal iodophor at admission to 67.4% for routine administration.

When looking at hospital transfers for any cause as a proportion of total discharges, researchers found a decrease in risk in the nursing homes assigned to the decolonization protocol, dropping from 35.5% to 32.4% (RR 0.92, 95% CI 0.88-0.96). The usual routine-care group meanwhile experience an uptick from 36.6% to 39.2% (RR 1.08, 95% CI 1.04-1.12).

Miller and colleagues determined that the number needed to treat with the universal decolonization was 9.7 to prevent a single hospitalization due to infection and 8.9 to prevent one for any cause. To put that into context, "a 100-bed nursing home could prevent 1.9 infection-related hospitalizations per month," they wrote.

Among the nursing homes that included sample collection during both periods, prevalence of MDRO carriage among residents in the routine-care group dipped only slightly, from 48.3% during baseline to 47.2% toward the end of the intervention period. In the decolonization arm, prevalence dropped from 48.9% to 32.0%, respectively.

Adverse events associated with the protocol were rare (35 in total) and mostly involved mild rashes possibly related to the enhanced bathing routine (chlorhexidine was discontinued in most of those cases).

Study limitations included that three of the 14 nursing homes assigned to the intervention never implemented the decolonization strategy, and that some sites had low adherence due to the requirement that nurses administer the nasal iodophor.

  • author['full_name']

    Ian Ingram is Managing Editor at ľֱ and helps cover oncology for the site.

Disclosures

The study was supported by a grant from the Agency for Healthcare Research and Quality.

Miller and Huang disclosed relationships with Medline Industries, Stryker Corporation, and Xttrium Laboratories. Stone had no disclosures. Other co-authors disclosed relationships with Abbvie, Medline Industries, Stryker Corporation, Thermo Fisher Scientific, and Xttrium Laboratories.

Primary Source

New England Journal of Medicine

Miller LG, et al "Decolonization in nursing homes to prevent infection and hospitalization" N Engl J Med 2023; DOI: 10.1056/NEJMoa2215254.