Preventing SARS-CoV-2 Transmission in Health Care Settings in the Context of the Omicron Variant

The world is in the midst of an unprecedented surge in SARS-CoV-2 cases attributable to the Omicron variant. This new variant is 2 to 3 times more contagious than the Delta variant. According to a report from the UK, 1109 of 14 606 non–household contacts of individuals with the Omicron variant developed infection vs 2922 of 102 997 non–household contacts of individuals with the Delta variant (7.6% vs 2.8%, respectively; adjusted odds ratio [OR], 2.63 [95% CI, 2.43-2.84]).¹ The increased contagiousness of the Omicron variant is not only leading to a surge in community infections but it is also leading to more transmissions in hospitals. During the last week of December 2021 in England, for example, 2525 of 12 424 (20.3%) patients hospitalized with SARS-CoV-2 were first diagnosed more than 7 days after hospital admission. By comparison, the analogous figures for the first week of November 2021 were 434 of 5208 (8.3%) patients.

The increase in hospital-onset infections associated with the Omicron variant belies the fact that nosocomial transmission of SARS-CoV-2 has been part of the COVID-19 pandemic from the beginning. The frequency of hospital-based transmission, however, is likely unappreciated. This is because few hospitals systematically test patients throughout and following their hospital stays. Most hospitals only test patients for SARS-CoV-2 at the time of admission and therefore may miss some infections acquired after admission, especially because approximately 40% of SARS-CoV-2 infections are mild or asymptomatic and thus do not trigger repeat testing. Furthermore, hospital stays for many non–COVID-19–related conditions are short, so some infections will only develop after discharge and will be missed or misattributed to posthospital exposures.

Some jurisdictions have gone to great lengths to discern the true frequency of hospital-acquired SARS-CoV-2. This is best done using whole-genome sequencing of the virus to demonstrate relatedness to the infections of other patients and staff as well as to eliminate spurious temporal and spatial associations. In a study from the University of Oxford that evaluated 803 inpatients and 329 staff members from 4 hospitals who were diagnosed with SARS-CoV-2 during a 7-week period in late 2020, an estimated 188 of 803 (23.4%) patient infections were deemed to potentially be nosocomial.² Whole-genome sequencing helped identify multiple cases that were potentially acquired at the hospital despite receiving care on different wards (suggesting unidentified staff intermediaries) or being diagnosed at admission (suggesting exposure during a prior admission).

Most hospitals have already implemented multifaceted infection control programs to prevent nosocomial SARS-CoV-2 transmission. These typically include universal mask wearing, strong encouragement or mandates for staff vaccination, requiring symptomatic staff members to stay home, contact tracing, and testing of all inpatients at admission. These measures have markedly reduced hospital-based transmissions, but the increase in nosocomial infections associated with the Omicron variant raises the urgent question of what more can be done to protect patients and staff.

Hospitals could invoke 3 additional measures to further reduce the risk of nosocomial SARS-CoV-2 transmission as the Omicron variant continues to surge.

Studies conducted before Omicron became the dominant variant estimated that vaccinated people are two-thirds less likely to carry SARS-CoV-2 than unvaccinated people and half as likely to transmit SARS-CoV-2 to others. However, the Omicron variant is partially refractory to current vaccines and so booster doses are needed to increase protection. A case-control study conducted in the UK of 760 647 symptomatic individuals infected with the Omicron variant vs symptomatic individuals with negative tests estimated that after 6 months, 2 doses of messenger RNA vaccine only lowered the odds of symptomatic disease by 6% (OR, 0.94; 95% CI, 0.92-0.95).³ Booster doses, however, increased protection to 68% against symptomatic disease (OR, 0.32; 95% CI, 0.31-0.33). Booster effectiveness does decrease over time, declining to approximately 50% after 10 weeks from boosting, but this is still substantially more than the protection afforded by 2 shots alone.

As with the Delta variant, booster doses will likely decrease transmission of Omicron in addition to preventing infections. In an analysis of 2225 people infected with the Omicron variant in Denmark, household members who had received a booster were less likely to become infected compared with vaccinated household members who had not received the booster after adjusting for age, sex, and the vaccination status of the source (25% vs 32%, respectively; adjusted OR, 0.54 [95% CI, 0.40-0.71]).⁴ In an adjusted analysis that included both the Omicron and Delta variants and also took into account viral loads, people who had received the booster were also less likely to transmit the virus to household members compared with vaccinated household members who had not received the booster (adjusted OR, 0.58; 95% CI, 0.41-0.83).⁴

These data support the idea that mandating booster doses in health care workers is likely to reduce inadvertent transmissions to patients. The US Supreme Court recently affirmed the US government’s right to require health care workers to get vaccinated if they work in facilities that receive federal funding.

The US Centers for Disease Control and Prevention only stipulates a soft recommendation in favor of admission testing for SARS-CoV-2 at the time of hospital admission. However, with the Omicron variant surging in the community, the pretest probability of newly admitted asymptomatic patients being silent SARS-CoV-2 carriers is higher than ever. In one large hospital system in Massachusetts, 282 of 2960 asymptomatic patients (9.5%) admitted during the last week of 2021 were positive for SARS-CoV-2. Failure to detect and isolate potentially infectious SARS-CoV-2 carriers presents a constant threat of within-facility transmission. Some hospitals not only test patients at admission but continue testing patients every few days thereafter to detect cases that were incubating prior to admission (and thus missed by the initial admission test) as well as cases acquired in the hospital. The Omicron variant’s short incubation period (median of 3 days) and high contagiousness makes this more critical than ever. Detecting newly positive patients is essential because patients are most contagious within the first few days of infection. Rapidly detecting newly infected patients allows facilities to quickly isolate these patients to prevent transmission to staff and other patients.

This is especially important for patients in shared hospital rooms. The estimated risk of infection for a patient admitted to a shared room with an occult positive SARS-CoV-2 carrier is 30% to 40%. Serial testing of patients in shared rooms could help decrease this risk. Placing portable high-efficiency particulate air filters between patients in shared rooms may also decrease transmission risk. There may also be a role for more frequent testing of asymptomatic staff; however, the discovery of additional positive staff members may further exacerbate hospitals’ current staffing crises.

Transmission risk is a function of infectious dose.⁴ The greater the viral exposure, the greater the risk of transmission. Conversely, measures that reduce viral exposure reduce transmission risk. Surgical and procedural masks reduce viral exposure by an estimated 40% to 60% depending on mask fit.^5,6 Exposure reduction is multiplied if both parties in an interaction are wearing masks. Mask wearing is associated with a clear decrease in transmission risk but it does not eliminate it. Even before the Omicron variant, transmissions from masked clinicians to patients, unmasked patients to clinicians, and mutually masked patients and clinicians following sustained interactions were well-documented.⁷ Outbreaks have also been documented in many hospitals despite universal mask wearing policies.⁸

The greater contagiousness of the Omicron variant magnifies the risk of mask failure. It is not yet clear whether this is because the Omicron variant binds more efficiently to respiratory epithelial cells or if it reproduces more aggressively in the upper airways compared with prior variants. But it does mean that smaller amounts of exposure are likely able to lead to infections. The solution is more effective respiratory protection. N95 respirators decrease aerosol exposures by 95% or greater, far exceeding the protection provided even with mutual mask wearing by patients and clinicians.⁵ N95 respirators have the further advantage of providing more effective source control compared with surgical masks.⁹ This means N95 respirators can also protect patients from occult positive clinicians and other hospital personnel and therein further reduce nosocomial transmission.¹⁰ Some object that universal use of N95 respirators is not practical because they are too uncomfortable to wear for long periods. This likely reflects many hospitals’ preferential use of older, hard-shell N95 models. Newer soft-shell models are considerably more comfortable and breathable.

These measures all work in conjunction in the hierarchy of infection control. No one measure is perfect. But by improving administrative controls (vaccinations and testing), engineering controls (ventilation and filtration), and personal protective equipment (better respiratory protection), transmission of the Omicron variant can be substantially reduced. This is of particular urgency because Omicron outbreaks in hospitals further exacerbate critical staff shortages and threaten hospitals’ capacity to accommodate the unprecedented surge in inpatient admissions.

Published Online: January 24, 2022. doi:10.1001/jama.2022.0262

Correction: This article was corrected January 31, 2022, to add the word “and” after “2020-2021” in Dr Karan’s conflict of interest disclosure.

Corresponding Author: Michael Klompas, MD, MPH, Harvard Medical School and Harvard Pilgrim Health Care Institute, 401 Park Dr, Boston, MA 02215 (mklompas@bwh.harvard.edu).

Conflict of Interest Disclosures: Dr Klompas reported receiving grants from the US Centers for Disease Control and Prevention, the Agency for Healthcare Research and Quality, and the Massachusetts Department of Public Health and receiving personal fees from UpToDate. Dr Karan reported serving as a paid researcher for the Independent Panel on Pandemic Preparedness and Response in 2020-2021 and as an expert witness on face mask use.