Brigham and Women's Provides Blueprint for Hospital COVID-19 Protection

Chanu Rhee, MD, MPH (IDWeek)

Investigators from Brigham and Women’s Hospital reported new observational data showing coronavirus 2019 (COVID-19) infections to originate in the hospital were especially rare during the first wave of the pandemic in the US.

Over 12 weeks from March-May 2020, the investigators traced just 2 COVID-19 cases to nosocomial origin.

The findings, presented at IDWeek 2020 by Chanu Rhee, MD, MPH, Assistant Professor at Harvard Medical School, provide clinical evidence for a series of rigorous infection control practices applied in the Boston hospital. They may also inform institutions in COVID-19 hotspots or those at continued risk of institution outbreaks.

In an interview with Contagion® during IDWeek, Rhee provided context into his team’s findings, what practices Brigham and Women’s put in place at the beginning of the pandemic, and how such measures can inform both hospitals and other community settings attempting to mitigate COVID-19 spread.

Contagion: Prior to research, what was our working theory or understanding of nosocomial COVID-19 infection among hospitals in the early days of outbreak?

Rhee: In the early days of the pandemic, there were reports of widespread outbreaks of SARS-CoV-2 within skilled nursing facilities and other congregate settings. However, little was known about nosocomial SARS-CoV-2 infection among patients in acute care hospitals. As such, there was a lot of anxiety about whether the infection control measures recommended by the CDC were adequate to prevent spread within the hospital and many patients have foregone essential medical care due to fear of contracting COVID-19.

Contagion: What observed measures were put into place to manage COVID-19 hospital spread?

Rhee: At our hospital, the core components of our infection control program included the following:

• Screening of all patients for COVID-19 symptoms on admission and daily thereafter
• Requiring all hospital staff to attest to the presence or absence of symptoms prior to starting their shift; staff with symptoms are referred for COVID-19 testing and not allowed to work if symptoms are present
• Liberal use of PCR testing initially for all symptomatic patients on admission or with any new symptoms during hospitalization. In late April we implemented universal testing for all patients at the time of admission, including asymptomatic patients.
• Dedicated COVID-19 units with airborne infection isolation rooms.
• Personal protective equipment (PPE) in accordance with CDC recommendations.
• PPE donning and doffing monitors
• Universal masking of staff and subsequently patients and visitors
• Restriction of visitors except in extenuating circumstances

We also developed hospital-wide protocols on when clinicians were permitted to discontinue isolation precautions in patients with suspected COVID-19, which generally required 2 negative RT-PCR tests at least 12 hours apart, and used protocols based on CDC-guidance to determine when patients with confirmed COVID-19 could be released from isolation.

Contagion: How did this research better inform infection control measures and resourcing?

Rhee: COVID-19 presents many infection control challenges since many infected individuals are asymptomatic yet contagious; current diagnostic tests are imperfect, especially if patients are tested too early in the incubation period; and transmission can occur through multiple mechanisms including droplets, fomites, and aerosols. Our analysis, however, demonstrates that a multifaceted infection control program based on US CDC guidance can minimize the risk of nosocomial transmission of SARS-CoV-2.

Over the first 12 weeks of the pandemic, our hospital cared for over 9000 patients including nearly 700 with COVID-19. Despite the high burden of COVID-19 in our hospital, we only identified 2 patients who likely acquired their infection in the hospital, including one who was most likely infected by his spouse prior to visitor restrictions and universal masking.

Contagion: Adversely—given the rarity of hospital spread when such measures are applied—what can other community settings at risk of significant spread take away from these findings?

Rhee: While our study cannot determine which interventions are most important to prevent spread of COVID-19, my take-away is that a comprehensive program is needed to minimize the risk. In other words, there is no one silver bullet.

One important observation that emerged from our case reviews was that several patients were only tested for the first time three or more days after hospitalization, sometimes because of atypical symptoms that were initially attributed to non-COVID conditions. These cases highlighted the importance of implementing universal testing on admission, and we observed fewer late-onset cases after this intervention. However, universal testing is not infallible as several patients initially tested negative while asymptomatic and then tested positive after symptoms began several days later. This underscores the fact that PCR can miss some infections early in the incubation period.

Another important theme that emerged from our case reviews was the need to conduct serial testing of patients with clinical syndromes highly suspicious for COVID-19. At least three cases with concerning syndromes initially tested negative but were positive on repeat testing. Based on our early experience, we instituted a protocol requiring at least 2 negative RT-PCR tests in symptomatic patients before discontinuing isolation.

Contagion: Are there any plans for continued or follow-up research?

Rhee: In addition to the analysis presented in the study, which encompassed the period from March 7 – May 30, we have continued to closely monitor our hospital for any new potential cases of nosocomial SARS-CoV-2 infection. We found no additional cases of nosocomial infection over the ensuing months. However, in late September we did have a cluster of SARS-CoV-2 infections that occurred on two units in our hospital that affected both patients and staff. More information can be found on our website here.

Our investigation determined that the cluster likely began with a highly contagious patient who initially tested negative twice on admission. With a combination of pre-emptive precautions, intense contact tracing, and serial testing of all patients and staff associated with the affected units, and numerous other measures, we have been able to fully contain this cluster. We are currently evaluating the lessons learned from the cluster to determine how best to further improve our infection control policies moving forward.

The study, “Incidence of Nosocomial COVID-19 in Patients Hospitalized at a Large U.S. Academic Medical Center,” was presented at IDWeek 2020.