Before the coronavirus disease 2019 (COVID-19) pandemic, the UNC Medical Center (UNCMC) Infection Prevention team was actively working with the emergency management department and various other departments within the facility and at the state level to plan and prepare for providing medical care to patients with high-consequence pathogens. The Johns Hopkins Hospital biocontainment unit and UNCMC physician leadership research and experience with high-consequence pathogens guided our work to plan for patient placement. Reference Garibaldi, Kelen and Brower1–Reference Narra, Sobel and Piper4 When the Centers for Disease Control and Prevention (CDC) announced that COVID-19 entered pandemic status, infection preventionists attended federal training on barrier precautions and controls for highly infectious diseases. 5,6 Transmission of severe acute respiratory coronavirus virus 2 (SARS-CoV-2) to healthcare personnel (HCP) was well documented outside the United States, so swift action was needed to prepare a dedicated space for the care of COVID-19 patients. Reference Xiao, Fang, Chen and He7
Unlike a typical biocontainment unit, we did not have dedicated clinical space designed for unidirectional movement, ample room for donning and doffing personal protective equipment (PPE), or dedicated air-handling systems designed for negative-pressure airflow or filtration of an entire patient unit. We therefore developed a plan to incorporate these principles into existing infrastructure, which included some airborne infection isolation rooms (AIIRs). We then scaled these design elements up to care for large numbers of patients on multiple units.
Our goals were to prevent occupationally acquired severe acute respiratory coronavirus virus 2 (SARS-CoV-2) infections for all HCP providing patient care, to minimize PPE usage consistent with protecting HCP without compromising patient care, to improve care efficiencies, and to provide a comfortable and safe working environment.
Methods and approach
UNCMC is a large academic medical center with >950 beds and employing >11,500 HCP. Our approach was to create areas of containment designated as high-risk for COVID-19 along with areas designated as low risk. A low-risk zone (LRZ) was an area where staff adhered to universal pandemic precautions (UPP), using PPE consisting of an extended-use surgical mask and eye protection. The high-risk zone (HRZ) was an area where staff were required to wear full PPE for special airborne and contact isolation precautions at all times. Full PPE was defined as a reusable surgical gown, gloves, a controlled air-purifying respirator (CAPR) with surgical mask or an N95 respirator with eye protection. A medicine intensive care unit (ICU) was the first unit selected for COVID-19 containment based on previous plans for other high-consequence pathogens. However, realizing that a surge of patients was likely, we simultaneously prepared plans for an acute care unit (ACU). Each unit had a similar layout with some structural differences that made it impossible to obtain unidirectional personnel flow. As a result, we realized that we would need an innovative and standardized way to create safe, effective workspaces while minimizing disturbances and optimizing patient care. In addition, we needed to formulate a method for containment that could be replicated in other areas of the hospital, regardless of the layout or number of beds.
To measure HCP satisfaction with this process, an electronic survey approved by our institutional review board was sent to all staff to assess their perceptions of the medical center’s efforts to protect them from acquiring COVID-19 in the fall of 2020. The survey collected occupational and primary work location data, and respondents could provide additional comments if desired.
Standard workflow and visual management
Standardizing the guidelines and workflows among the HRZs meant the functionality of the units was similar, leading to less confusion among ancillary departments, consulting providers, and floating staff. We intentionally retained nurses’ stations, clean utility and equipment rooms, and automated dispensing cabinets for medications in the LRZs. Teams temporarily relocated offices, conference rooms, and call rooms to create the HRZs. Each containment unit and its corresponding nurses’ station had multiple entrances (Fig. 1), including stairwells. Signage and stanchions prevented accidental HRZ entrance. If no close access to low-risk areas existed, an LRZ path 1 m wide was demarcated on the floor with tape to create a visual cue between the high- and low-risk sides of the hallway (Fig. 2). This path allowed entrance to areas that would otherwise require relocation.
Layout of acute care unit with high-risk and low-risk zones denoted. (Adapted from image courtesy of Katherine Rowe, MSN, RN, PCCN, UNC Medical Center.)
Demarcated low-risk footpath within a high-risk zone.
Before implementing the first HRZ, we created a packet of visual management aids for each area based on input from the clinical units. These aids ensured standard and consistent visual cues throughout the hospital (Fig. 3). The units received ample signage, visual cues, and relevant containment unit workflows for implementation. Laminated signage, colored duct tape, and equipment tags with zip ties allowed the units to expand the HRZs quickly.
Containment unit entrance signage.
Staff training and preparedness
Before COVID-19, HCP had not received education on containment of high-consequence pathogens, including donning and doffing appropriate PPE for a novel respiratory virus. The infection prevention team provided comprehensive online learning modules, educational videos, and PPE guidance. To provide support for just-in-time education and to ensure high PPE compliance, we created separate donning and doffing stations for each HRZ on a unit, which were, ideally, staffed with a PPE monitor. Reference Summerlin-Long, Selimos and Brewer8 A PPE monitor is a trained, designated person who observes PPE donning and doffing. PPE monitors received education about the importance of proper PPE use and provided effective feedback. Redeployed staff or those facing closure or reduced staffing in their primary department filled PPE monitor roles. The PPE monitors promoted compliance with all steps for donning and doffing in the correct order. Doffing PPE was considered the time of highest risk for contamination, and having an observer assisting and maintaining high compliance with these steps added an increased measure of safety. Reference Casanova, Rutala, Weber and Sobsey9–Reference Okamoto, Rhee and Schoeny11
PPE and supplies
Adjustments to nurse staffing models (ie, nurse-to-patient ratio) determined that 6 patients would be the threshold to convert an area to an HRZ with donning and doffing stations. Without an HRZ, each room entry required a new set of PPE. By caring for 6 patients inside the HRZ, HCP could care for all patients together without doffing between patients. Wearing a reusable surgical gown and a set of inner gloves allowed HCP to change only outer gloves and to use alcohol-based hand rub between patients. If a patient inside the HRZ had a concurrent multidrug-resistant infection, such as methicillin-resistant Staphylococcus aureus, HCP would also don a disposable isolation gown and then doff only the isolation gown and outer gloves when leaving the room.
The infection prevention team developed guidelines and standard work procedures for the movement of people, equipment, and supplies in and out of the HRZ to ensure infection control and to promote safety through consistency. Staff disinfected equipment twice upon leaving the HRZ according to the manufacturer’s instructions, once inside the zone and again after exiting. Trash and soiled linen removal required new workflows in partnership with environmental services staff. If a decreasing census required conversion of portions of the HRZ back to non–COVID-19 patient beds, environmental services staff performed terminal cleaning in all rooms and enhanced cleaning in shared spaces, followed by ultraviolet room disinfection.
Predetermined patient transportation routes from the emergency department to each of the COVID-19 units were indicated by directional markers in hallways. Routes between units, procedural areas, and discharge allowed coordination between unit staff, critical care staff, and patient transport staff.
Air handling
Initial concerns about the long-distance viral transmission via aerosols led us to consider the degree of negative pressure gradient needed in rooms and hallways not designed with such capacity. AAIRs were insufficient to house all patients on the containment units. We considered changing the physical structure, adding air outlets into external walls of patient rooms and hallways, and/or erecting sealed barriers to contain airflow in these spaces. Ultimately, weighing the evidence supporting that SARS-CoV-2 was spread primarily over short distances via aerosols, with our aggressive timeline for preparation and considering practical limitations, we chose not to alter the physical design of our unit spaces. 12 As an alternative measure, portable high efficiency particulate air (HEPA) filter units were available for placement in patient rooms when aerosol-generating procedures (AGPs) occurred.
Clinical staffing and workflow
While in the HRZ, staff were unable to eat, drink, or use the restroom. Nursing staff on each unit rotated between the HRZ and LRZ based on HCP availability and patient acuity. Scheduled exits from the HRZ allowed for breaks at a frequency acceptable to meet these personal needs and to provide respite from wearing full PPE.
Once donned and in the HRZ, HCP did not have access to low-risk spaces, including automated dispensing cabinets, clean supplies, and nutrition areas. Therefore, at all times, it was necessary to have staff in low-risk areas facilitate obtaining and passing medications, supplies, and laboratory specimens. Retaining staff continuously within the HRZ provided immediate patient access, a benefit given the instability of many COVID-19 patients. Nurses used partners to work together between the zones, covering a group of patients and taking turns within and outside the zones.
Flexibility for fluctuations
Patient placement occurred via standardized workflow (Fig. 4). The inpatient containment units were located in the same tower of UNCMC to decrease the footprint of COVID-19 patients throughout the hospital. Due to census fluctuations, surge plans were in place to expand the high-risk portions of the units when necessary. This flexibility allowed units to maintain beds for non–COVID-19 patients simultaneously but outside the HRZ. To expand containment-unit capacity, 2 surge units, one intermediate and the other acute, were added.
Patient placement workflow at UNCMC. Note. ACU, acute care unit; ICU, intensive care unit; ID, infectious disease; OSH, outside hospital. *Contraindications for shared suite placement include concurrent infection requiring additional isolation precautions (eg, C. difficile), COVID-19 exposure; required aerosol-generating procedure, unable to comply with masking, other non–infection-related clinical reasons as indicated by diagnosis and/or behavior or status (eg, forensic patients, psychiatric patients, etc).
Pediatric and asymptomatic obstetric patients with COVID-19 were admitted to specific units within their specialized hospitals of the Medical Center, depending on the level of care required. Fortunately, the COVID-19 census on these units never became high enough to support the resources required to maintain a containment area, but conversion plans were available. In addition, the staffing office deployed PPE monitors to the units when available.
To meet the specialized needs of psychiatric patients, we created an HRZ in a pre-existing portion of an inpatient psychiatric unit and an associated offsite inpatient psychiatric facility. Reference Schultz, Miller and Stancill13 Patients could enter a small milieu at scheduled times. The centralized nurses’ station provided an ideal space to allow staff to don PPE. After entering the contained area, staff could perform safety checks according to the protocol and doff safely at the hallway doffing station before re-entering the nurses’ station. This setup also allowed staff within the nurses’ station of the LRZ to see their colleagues in the HRZ.
The COVID-19 census varied over time, and the containment units needed to reduce the number of beds in the HRZ to create beds for non–COVID-19 patients. Nursing and physician leadership decided when to convert the number of COVID-19 beds with bed management, transferring patients to consolidate the HRZ if necessary. Currently, supplies from the patient rooms and hallways are used in another HRZ or discarded. Staff perform thorough disinfection of all equipment and carts. Environmental services staff terminally clean the HRZ, including the hallway and patient rooms. All rooms receive a curtain change and ultraviolet light disinfection out of an abundance of caution.
Results
Patient-to-HCP transmission findings
From April 1, 2020, to May 31, 2021, UNCMC admitted 1,512 unique COVID-19 patients within the primary COVID-19 containment units (ICU and ACU). All HCP were required to complete a daily wellness screening and to report any positive symptoms. HCP with signs or symptoms of COVID-19 were furloughed and tested for SARS-CoV-2. Occupational health staff interviewed HCP who tested positive to determine the likely source of exposure.
During this period, 17 (9%) of 189 nursing staff working on the ICU and ACU tested positive for SARS-CoV-2. Of these 17 cases, 16 were determined to be due to community or family/household exposure. The other case had an unclear source of exposure with no known exposures outside work or PPE breaches at work. Sources of exposure for 2 additional non-nursing HCP working on these units were unclear. No instances of possible patient-to-HCP transmission were noted on the surge units while functioning as containment units with HRZs.
Comparatively, 39 instances of patient-to-HCP transmission occurred in 11 units not designated as COVID-19 containment units. Causes of exposure included HCP failure to maintain adequate UPP, HCP failure to retest patients developing symptoms after admission, exposure during AGPs from patients with unknown COVID-19 status, and HCP caring for the roommate of a patient with an unknown COVID-19 status.
Patient-to-patient transmission findings
Nurses sometimes were required to care for COVID-19 and non–COVID-19 patients during rotations on the same shift, depending on staffing and unit census. The COVID-19 containment units also housed non–COVID-19 patients outside the HRZs most of the time. Importantly, there were no instances of patient-to-patient transmission between patients in high-risk and low-risk areas on the same unit.
PPE usage
Based on a block of the 6 patients needed to maintain an HRZ, we calculated the minimum number of times various HCP would need to enter a patient’s room during a 12-hour shift, and we evaluated the number of doffing opportunities. The number of PPE doffing opportunities decreased by 65% when using the HRZ model (Fig. 5).
Comparison of estimated minimum number of person protective equipment doffing opportunities for every 6 COVID-19 intensive care unit patients per 12-hour shift.
Improved HCP comfort and satisfaction
Overall, staff working exclusively on COVID-19 units were the most satisfied with UNCMC efforts to keep them safe and were similarly satisfied with PPE availability as others working in direct patient care, based on our survey. Approximately half of respondents working exclusively on COVID-19 units agreed that PPE monitors played an essential role in keeping staff safe. More respondents working equally in COVID-19 and non–COVID-19 units agreed with the importance of the PPE monitor (Fig. 6).
Percentage of healthcare providers who agreed or strongly agreed with infection prevention strategies for COVID-19.
Discussion
Converting pre-existing patient care areas into containment units with HRZs created a safe and efficient working environment. With the HRZ method, we were able to accommodate patient surges and prepare additional units in advance. By the end of 2020, UNCMC had designated 1 intensive care unit and 1 intermediate care unit, 2 acute care units, 2 inpatient psychiatric areas, and 1 behavioral health holding area in the emergency department as containment units with HRZs at various times. Our experiences with the HRZs were positive, and staff felt safer knowing the locations of COVID-19 patients.
This research was initiated to prevent workplace transmission to HCP. Prolonged, protected exposures in the HRZs did not adversely affect the staff, who had only 3 possible cases of workplace-acquired COVID-19. Full protection is conferred with all HRZ staff wearing full PPE and adjacent staff in the LRZ wearing UPP PPE. Although there is no strong evidence of viable SARS-CoV-2 RNA in patient corridors, staff were still protected. Reference Birgand, Peiffer-Smadja, Fournier, Kerneis, Lescure and Lucet14,Reference Zhang, Kelly and David15 Over >11,000 patient days and countless HRZ entrances, AGPs, and doffing events, the use of HRZs proved to be a very efficient means for providing patient care and protecting our HCP. Although 3 SARS-CoV-2–positive HCP had an unknown exposure source, there was no way to definitively determine whether transmission occurred in the workplace from a patient source. The low rate of transmission to HCP in the HRZs suggests that it was unlikely. The lack of patient-to-patient transmission between patients in the COVID-19 containment areas and non–COVID-19 patients adjacent to the containment area also demonstrated this to be a successful model. Reducing PPE usage had the added benefit of minimizing doffing events, and therefore minimizing contamination possibilities. The HRZ method of containment unit reduced PPE usage without compromising safety and was satisfying for HCP to work in, as evidenced by survey responses. Breaks and rotating out of the HRZ provided HCP time to recharge and refocus while supporting their partner. This method was sustainable at the ideal state for more than a year, with some limitations for staffing PPE monitors for another year. When PPE monitors returned to their home departments, UNCMC created temporary positions to staff this critical role.
Survey responses from respondents working exclusively on COVID-19 units indicate high satisfaction with PPE availability. However, this satisfaction is lower than that of all responses combined. Comments from COVID-19–unit staff indicated challenges with availability or maintaining consistent supplies of smaller N95 respirators earlier in the pandemic. Staff on these units were also consistently using the PPE daily, and staff outside these units had limited opportunities for needing PPE. N95 respirators were prioritized for the containment units because our supply chain strategized to provide the safest products. Our organization uses tiered safety huddles for escalating supply safety concerns, and this was our mechanism for resolution.
COVID-19–unit staff also provided the lowest response among all groups concerning the importance of PPE monitors. Again, the timing of the survey may have influenced the responses to this question; staff on the COVID-19 units may not have felt that the PPE monitors were as crucial as they once were since the containment units had operated for at least 6 months at the time of the survey. On the other hand, the consistent coaching from PPE monitors likely built confidence and familiarity with the donning and doffing steps over time. Comparatively, respondents working equally on COVID-19 and non–COVID-19 units reported high agreement with the importance of PPE monitors. This higher proportion may be due to the respondents’ comparative experiences working on non–COVID-19 units, indicative of the PPE monitors’ support and overall safety.
During the Ebola outbreak, the US government invested substantial amounts of money in a small number of treatment facilities dedicated to care for high-consequence pathogen patients. Reference Herstein, Biddinger and Gibbs16 We show here that multidisciplinary collaboration at a facility not designated to care for patients with high-consequence pathogens could provide effective care while maintaining HCP safety. Keys to success were unit design, a modified PPE regimen with PPE monitors, and HCP buy-in. Patients with COVID-19 received priority placement in AIIRs within the HRZs. Because the existing infrastructure on each unit was maintained without any alterations to isolate airflow, it appeared that airborne transmission was not an important mechanism in the transmission of SARS-CoV-2. To protect HCP from patients with this novel respiratory virus, using a containment unit with an HRZ effectively provides source control and minimizes high-risk exposures in a population at heightened risk for acquiring COVID-19. In conclusion, the use of HRZs and standardized methods for caring for COVID-19 patients are critical strategies to keep HCP and patients safe from transmission.
Acknowledgments
We thank all healthcare personnel caring for COVID-19 patients at UNC Medical Center and all of the departments that helped support this massive undertaking.
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
