Exposure Profile of Health Care Personnel Infected With COVID-19 in a Tertiary Care Hospital. A Brief Report

To the Editor:

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spreads through respiratory droplets, aerosols, and occasionally fomites.^1,2 Adopting environmental controls such as droplet and airborne precautions (ie, using masks, gloves, gowns, and eye wear), social distancing, and hand and environmental hygiene can substantially reduce transmission.³ An important goal of infection control programs is to prevent the transmission of COVID-19 (coronavirus disease 2019) to health care personnel (HCPs).⁴

Several factors increase the risk of acquiring infection in HCPs including exposure to infected patients, working overtime with more responsibilities, inadequate personal protective equipment (PPE), insufficient diagnostic testing, and so on.^4,5 On top of the aforementioned risks, HCPs are also at risk of acquiring COVID-19 infection from infected coworkers.⁴ In a study conducted in Greece, almost half of the high-risk exposures in HCPs were to their infected coworkers.⁶

We evaluate exposure types in COVID-19–positive HCPs of our tertiary care hospital. Our primary aim is to evaluate work-related versus community COVID-19 exposures in HCPs in relation to their job types. Our secondary aim compares patient- versus employee-related COVID-19 exposures in HCPs based on job type.

Data were obtained from the University of New Mexico's Occupational Health from March 2020 through July 2021. The study was approved by the University of New Mexico's institutional review board. There were a total of 1087 COVID-19–positive HCPs of an employee population of 12,500.

Positive results were primarily obtained by nucleic acid amplification test polymerase chain reaction (PCR), but also included rapid molecular PCR and rapid antigen in the symptomatic. Confirmatory PCRs were obtained in those with symptoms who tested negative on antigen test.

Job types were classified into direct patient care, indirect patient care, and no patient care. Direct patient care was defined as those who work directly with patients or body fluids, such as physicians, advanced practice providers, nurses, technicians, rehabilitations services, language interpreters, housekeeping, and so on. Indirect patient care included those working in a clinical area but not directly with patients such as pharmacists, administration, maintenance, food and nutrition services, and so on. Those working in no patient care were neither working in a clinical building nor interacting with patients.

Each positive employee was triaged by a nurse. Sources of exposure were categorized as occurring in the community or workplace. If in the workplace, exposures were determined to be from patients or employees through contact tracing.

Normality of the data was checked graphically. Categorical variables were presented as number of observations and percentages and were compared using χ² or Fisher exact test. A log linear model was used for hypothesis testing of categorical variables. Continuous variables were presented as means and SDs and were compared using t test and analysis of variance. Variables included age, sex, job type of HCPs, and the source of exposure to COVID-19 in HCPs.

The overall rate of infection in HCPs was 8.7% during the time period studied. Participant characteristics by job type are shown in Table 1. A total of 1087 participants were included. The age was similar across the job types, with overall mean of 38 years that ranged from 18 to 75 years. Overall, there were more females (ie, 76%), and the distribution between men and women across job types was homogenous.

TABLE 1 - Characteristics of Health Care Personnel Who Were Infected With SARS-CoV-2, by Job Type

Characteristics	Total Positives	Job Type
	Mean (SD)	Direct	Indirect	None
Age, y	38 (11.8)	37.2 (11.4)	40 (12.2)	40 (12.2)
	% (n)
Sex
Male	23.9 (260)	23.2 (164)	29.2 (46)	23.5 (50)
Female	76.1 (827)	76.9 (545)	71.8 (117)	76.5 (163)
Exposure type
Community	89.1 (968)	88.3 (626)	93.9 (153)	87.8 (187)
Work-related	11.0 (119)	11.7 (83)	6.1 (10)	12.2 (26)
Work exposure type
Employee-employee	82.4 (98)	75.9 (63)	90.0 (9)	100.0 (26)
Patient-employee	17.7 (21)	24.1 (20)	10.0 (1)	0 (0)

Community exposures were the source of COVID-19 exposure in 89% of COVID-19–positive HCPs, whereas workplace exposures accounted for 11%. Within the workplace, employee-to-employee exposures were the source of COVID-19 transmissions in 82% of HCPs compared with patient-to-employee exposures, which were 18%.

Job type had an influence on age, F_2,1079 = 9.03, P = 0.001. There was no significant difference noted between age and community- versus work-related exposures (Table 2). Similarly, there was no difference between age and patient- versus employee-related work exposures.

TABLE 2 - Association of Age With Exposure Types (Two-Sample t Tests)

	Obs	Age, Mean (SD), y	95% Confidence Interval	P
Exposure type
Community	965	38.2 (11.7)	37.5–39.0	0.523
Work-related	119	38.9 (12.2)	36.7–41.2
Work exposure type
Employee-employee	98	38.9 (12.3)	36.4–41.3	0.728
Patient-employee	21	37.7 (11.6)	32.6–43.1

There was neither significant relationship found between gender and community- versus work-related exposures, χ²₂ (n = 1085) = 0.111, P = 0.739, nor between gender and employee versus patient work-related exposures, χ²₁ (n = 119) = 1.970, P = 0.160.

There was no significant relationship between job type and community- versus work-related exposures, χ²₂ (n = 1085) = 4.6298, P = 0.099 (Table 3). A significant relationship was found between job type and work-related exposures (Fisher exact test, n = 119, P = 0.015), with those exposed from patients more likely to work in direct patient care, whereas those exposed from employees more likely to work in indirect patient care or no patient care areas. The effect size was small (Cramer V = 0.2649).

TABLE 3 - Association Between Exposure and Job Type

Characteristics	% Infection, Direct Patient Care	% Infection, Indirect Patient Care	% Infection, No Patient Care	χ 2 (P)	Cramer V
Community	64.8	15.8	19.4	4.6298	0.0653
Work-related	69.8	8.4	21.9	(0.099)
n = 1,087
Employee to employee	64.3	9.18	26.5	0.007	0.2649
Patient to employee	69.8	4.8	0.0	(0.015)
n = 119

Numbers in bold font indicate significance.

P = 0.05.

Health care personnel are more likely to contract SARS-CoV-2 in the community than in the work environment. When at work, employee-to-employee exposures pose a greater risk than patient-to-employee exposures for all job types. Those who are exposed from patients are more likely to work in direct patient care.

Our findings that employee-to-employee exposures are more common than patient-to-employee exposures are consistent with other studies. Garzaro et al⁷ assessed transmission of COVID-19 among 2411 HCPs in a tertiary care hospital in Italy and found that the HCPs in management positions with no direct patient care were a larger source of infection to other HCPs than patients. Similarly, Maltezou et al⁶ assessed 3398 HCPs who were exposed to COVID-19 in a tertiary care hospital in Greece. Among 3398 HCPs, 1031 were high-risk exposures, of which 700 (68%) were exposed to SARS-CoV-2 from other HCPs, whereas only 331 (32%) were exposed from patients. In our work-related exposures, 82% were between employees, whereas the remaining 18% were due to infected patients. Furthermore, we found a significant relationship between HCPs with indirect and no patient care job duties and an increased risk of contracting COVID-19 from their coworkers compared with patients.

We found that community-based exposures accounted for 89% of the overall COVID-19 exposure in HCPs. The remaining 11% were known workplace exposures. With such high rates of community exposures, employees themselves, particularly those who are asymptomatic, are a source of COVID-19 transmission in HCPs, whereas patients, who are tested on admission, account for a small proportion of exposures.⁸

Overall, females in our study accounted for two-thirds of HCPs who were infected with COVID-19 and had higher rates of community exposures (89%). This is consistent with literature and is likely due to females constituting a bigger proportion of the workforce in health care.⁹ In addition, they have assumed more caregiving responsibilities during the pandemic, including for sick children and family members,^10,11 increasing their likelihood of community exposures.

In our employee population, work-related HCP exposures were largely due to a breech in PPE, particularly in employee-to-employee exposures. Contact tracing and risk assessments found break-room exposures, where employees eat and drink together, to be a common theme. Exposures from patients were also largely due to inappropriate PPE and often in the setting of either not yet knowing a patient's COVID status or in a patient who had converted positive after admission. Being in the same room as aerosol-generating procedures without appropriate PPE, particularly the use of a respirator, was also a contributing factor.

We believe perceived risk impacts work-related exposures. Despite the implementation of extensive source control measures, including environmental controls, administrative controls, and PPE, employees tend to be less adherent in following guidance with each other than when around patients. It is unclear if there is a perception among HCPs that patients pose a greater risk when compared with coworkers, and thus, PPE guidance is more closely followed when working directly with patients.

Our study has some strengths. The sample size for primary aim was 1087, which was adequate. Most of our participants were young; hence, age was not a confounder. Contact tracing was done for each work-related exposure, ensuring an identifiable source. Limitations include a smaller sample size in aim 2, and it can be difficult to determine causation of respiratory illness when rates of community spread are high.

In conclusion, our study found that employee-to-employee–related exposures are a significant source of COVID-19 infections in HCPs in the work environment, especially for those working in jobs that have indirect and no patient care. Although those with patient-related exposures are more likely to work in direct patient care, the number of exposures from other employees in this group was also higher. This is likely due to a laxity or nonadherence to administrative and PPE hazard controls.

We recommend that hospitals educate HCPs about the risk of employee-to-employee exposures and encourage adherence to hazard controls, such as appropriate PPE and social distancing both inside and outside the patient care areas. Such adherence can be encouraged by educating HCPs of these risks, posting flyers in and outside patient care areas, rewarding HCPs for adherence, limiting the number of employees in break rooms or cafeterias, hiring more employees to lessen the work burden and fatigue of each employee, and so on.

Sher Ali Khan, MD, MPH
L. Olivia Hopkins, MD, MSCR
Preventive Medicine Residency Program
University of New Mexico Health
Sciences Center
Albuquerque, NM