Literature searches of publications (peer reviewed and not yet peer reviewed) were conducted every two weeks from June 22, 2020, to October 8, 2020, using a COVID-19 literature library maintained by the Centers for Disease Control and Prevention (CDC), at which point the CDC migrated its library to the World Health Organization’s (WHO) COVID-19 Global Research Database (WHO 2020). Literature searches were conducted in the WHO library once every two weeks until September 23, 2021, using the search strings listed in the Supplemental Information. Additional publications were identified through other literature reviews and the Department of Homeland Security’s Master Question List for COVID-19 (caused by SARS-CoV-2)(DHS 2022).

All literature on SARS-CoV-2, published in English between March 19, 2020 (earliest upload date in the CDC literature library) and September 23, 2021, were included, including preprints and other non-peer-reviewed publications irrespective of where the study occurred. Original articles were prioritized, but research letters and reviews were also included. Studies performed in all types of workplaces were included (e.g., hospitals, dental settings, restaurants, manufacturing plants, etc.). Studies that conducted sampling exclusively in homes of patients were also included to capture exposure risks to home healthcare workers. Any literature not relevant to workplace exposure and safety was excluded.

Results were screened first by title, and those that remained were screened by abstract. For each publication screened, a researcher created a summary noting the key take-away findings of the study, study limitations, and additional relevant information included in the publication. For publications with novel data or important findings, a shortened version of the summary was written and added to a condensed report. As preprints were subsequently published, the summaries and condensed reports were updated, and the citations revised.

Since biological fluids can serve as potential sources of exposure in the workplace, several studies have been conducted both in the field and the laboratory to determine the presence and persistence of infectivity of SARS-CoV-2 in these fluids.

Respiratory tract secretions such as mucus, sputum and saliva are biological fluids that can contain viral material and make up respiratory droplets that are released into the air. Viable SARS-CoV-2 was isolated from viral cultures of upper respiratory tract specimens from infected individuals in two studies (Bullard et al. 2020; Wolfel et al. 2020). SARS-CoV-2 viral RNA was not found in sweat samples collected from hospitalized COVID-19 patients (Arslan et al. 2021). One study did not detect SARS-CoV-2 viral RNA in urine, semen, and expressed prostatic secretion samples from recovered COVID-19 male patients (Ruan et al. 2021), but two additional studies evaluated urine for viable SARS-CoV-2 (infectious) virus, and one found viable virus (Kashi et al. 2020). SARS-CoV-2 RNA was not found in vaginal fluid or cervical exfoliated cell and was only found in 1 anal swab sample from 27 women with positive SARS-CoV-2 RNA respiratory samples (Cui et al. 2020). SARS-CoV-2 viral RNA was found in fecal samples of 6 of 8 COVID-19 hospital patients and in the urine of 1 of 8 patients. Vero E6 cells were inoculated with fecal samples and showed no cytopathic effects indicating a lack of infectivity (Albert et al. 2021).

Studies looking at tears and ocular secretions have had mixed results. Several studies conducted conjunctival swab testing of COVID-19 patients using SARS-CoV-2 reverse transcription-polymerase chain reaction (RT-PCR) analysis (Mohaqiq et al. 2021). Out of nine studies, investigators in three studies did not detect SARS-CoV-2 RNA. In the other six studies, investigators found that a combined 5/131 samples from 3/52 known COVID-19 patients were positive for SARS-CoV-2 RNA (Mohaqiq et al. 2021). A case report showed that an ocular sample from a patient with conjunctival hyperemia detected SARS-CoV-2 RNA and was tested for viable virus with Vero E6 cells in which a cytopathic effect was observed 5-days post-inoculum (Colavita et al. 2020).

There have been several laboratory-based studies looking at the persistence of infectivity of the virus in various bodily fluids. A study tested SARS-CoV-2 persistence of infectivity in suspensions of human feces and urine. Each sample was inoculated with 0.3 milliliter of virus stock with an infectious titer of 10⁶ 50% tissue culture infectious dose (TCID50). This study showed viable SARS-CoV-2 virus was recoverable for up to 48 hours in feces and 3-4 days in urine (Liu et al. 2021b). A study using 10⁶ surrogate viral particles found, after contaminating a toilet bowl rim, toilet seat top, and toilet seat underside prior to flushing, <10³ particles could be found on the seat, handle, floor, and walls via dispersal and deposition following flushing (Sassi et al. 2018). SARS-CoV-2 has been shown to remain viable in both highly acidic and highly basic conditions (Sun et al. 2020).

Investigators added SARS CoV-2 to commercially available human nasal mucus and sputum and assessed persistence of infectivity under three environmental conditions - 4°C/40% relative humidity (RH), 21°C/40% RH, and 27°C/85% RH (Matson et al. 2020). Fifty μL of prepared solution containing 1x10⁵ TCID50/mL was added to either sealed tubes or onto polypropylene disks and sampled at specific time points. SARS-CoV-2 persistence of infectivity was greater in liquid droplets than when dried on surfaces using human nasal mucus and sputum. Lower temperature and lower relative humidity were associated with longer half-lives, but no virus survived more than 48 hours. They found viral RNA for up to a week in all samples despite the absence of viable virus (Matson et al. 2020). SARS-CoV-2 stability was tested in nasal mucus, sputum, saliva, tears, feces, urine, breast milk, blood, and semen. Viral stability was assessed by inoculating Vero E6 cells and recovering infectious virus at various time points. Samples were tested in liquid (sealed tubes) and on stainless steel surfaces (except human urine, which was only tested in liquid). The virus was stable for up to 21 days in nasal mucus, sputum, saliva, tear, urine, blood, and semen; it remained infectious significantly longer under winter (5°C and 75% RH) and spring/fall conditions (13°C and 66% RH) than under summer conditions (25°C and 70% RH). The virus was stable up to 24 hours in feces and breast milk (Kwon, Gaudreault and Richt 2021).

One of the additional occupational exposure concerns from bodily fluids was the potential for exposure to their presence in wastewater. RT-PCR testing for SARS-CoV-2 RNA in wastewater was commonly used in multiple countries but did not look at infectivity for the virus (Anand et al. 2022; Baldovin et al. 2021; Miyani et al. 2020; Tomasino et al. 2021). A review of published studies looked at the risk of occupational exposure to SARS-CoV-2 in wastewater residuals and biosolids. Although SARS-CoV-2 RNA can be detected in wastewater, it was noted that few studies have been able to isolate infectious virus in fresh feces (Jeong et al. 2020; Liu et al. 2021b; Xiao et al. 2020). As wastewater goes through the various treatment processes, the amount of infectious virus is expected to decrease similarly to other viral pathogens. The recommendations to reduce exposures to other pathogens should also be effective to reduce the risks of SARS-COV-2 infections (Brisolara et al. 2021).

Downstream processing of air and surface samples to test for SARS-CoV-2 RNA or viability are similar and considerations for these methods are largely the same. As a part of a laboratory study, Tastanova et al. (2021) evaluated 6 different in-house and commercial RT-PCR detection methods and determined that each method demonstrated similar limits of detection (1 or 2 viral copies/μL), sensitivities (93.6% to 97.8%), and specificities (98.7% to 100%). The RT-PCR results were given in terms of a cycle threshold (Ct) or quantification cycle (Cq) (Bustin et al. 2009), which is the number of amplification cycles needed for detection of the nucleic acid target during the amplification process, and therefore has an association with the amount of viral genetic material in the sample. Tastanova et al. (2021) also used an external quality assessment quantitative test sample and linear regression modeling to determine a Ct value of 40 as the maximum value a sample is considered positive, and a value of 40 has been utilized by many studies as a method limit of detection (Jin et al. 2021; Lei et al. 2020; Zhou et al. 2020). A study by Zhou et al. (2020) found that Ct values of >30, corresponding to an envelope gene copy number of <10⁵/mL, are unlikely to be culturable, however, SARS-CoV-2 has been isolated above this threshold (Lednicky et al. 2020; Lednicky et al. 2021).

Most studies have utilized culture media such as Dulbecco’s Modified Eagle’s Medium (DMEM) or Eagle’s Minimum Essential Medium (EMEM) plus bovine serum albumin (BSA), fetal bovine serum (FBS), or a combination of both to mimic the protein content present in clinically relevant media such as saliva, mucus, or sputum. A cross-comparison of the SARS-CoV-2 decay rate across studies that utilized similar experimental conditions (≈ temperature and RH), but different suspension media, reported a significant reduction in the decay rate when SARS-CoV-2 was suspended in nasal mucus or sputum rather than culture medium (Matson et al. 2020; van Doremalen et al. 2020). A separate study that compared the decay rate of SARS-CoV-2 with and without BSA found that the decay rate increased significantly in the presence of BSA (Pastorino et al. 2020). These findings would appear to be contradictory as the inclusion of BSA, used to mimic the protein content of respiratory fluids such as saliva, functioned as a preservative in the Pastorino study (Pastorino et al. 2020); however, respiratory fluids also contain enzymes and mucins that may impede the persistence of infectivity of SARS-CoV-2. Vero E6 (African Green monkey kidney) cells were used to culture virus from air and environmental samples in many viability studies and was the dominant method during the time of this review (Chia et al. 2020; Dumont-Leblond et al. 2021; Dumont-Leblond et al. 2020; Lane et al. 2021; Lednicky et al. 2020; Lednicky et al. 2021; Ratnesar-Shumate et al. 2021; Zhou et al. 2020).

Studies have established that SARS-CoV-2 is viable in a variety of bodily fluids including feces, urine and potentially tears. The virus can remain stable in fluids (mucus, sputum, tears, blood, semen) for weeks with infectiousness lasting significantly longer under winter and spring/fall conditions than under summer conditions. The virus can remain stable in feces for up to 1 day. Knowing the stability of the virus in fluids is essential in directing the efforts for cleaning and disinfection protocols necessary to eliminate infectious material. Similarly, wastewater may be a useful tool for tracking trends for SARS-CoV-2 RNA, but only a few studies were able to isolate infectious virus in fresh feces.

The duration, proximity, and frequency of contact with someone infected with SARS-CoV-2 plays a role in dictating the risk of infection. The vast majority of documented SARS-CoV-2 transmission is associated with close contact in the home or in indoor spaces for a prolonged period of time. SARS-CoV-2 RNA was commonly detected in hospital bathrooms of COVID-19 patients’ surfaces such as paper towel dispensers, toilet bowls, toilet flushers, and door handles. Bus handrails and support bars have also detected SARS-CoV-2 RNA, but other public settings such as schools or supermarkets had infrequent detection. Indirect contact transmission is possible but is less likely as it requires an intermediate step for transmission to occur. Simply contacting a contaminated surface before touching the mouth, eyes, or nose does not guarantee transmission, as the viral load of SARS-CoV-2 on a surface at the point of contact and the transfer efficiency of the virus from hand to mucous membrane will ultimately determine the risk of transmission. Therefore, if the viral load is low, then the likelihood of infection will also be low. The results from persistence of infectivity studies of SARS-CoV-2 on surfaces were similar to other studies of other respiratory viruses, such as SARS-CoV-1. Several studies suggested a biphasic decay rate with rapid decay during the initial hour and a slower decay rate thereafter. Additionally, reduced survival rates were seen in more porous materials. These findings suggest that porous surfaces have a reduced viral infectivity particularly after the initial hour of contamination. As the scientific community begins to collectively agree on the dominant transmission pathways of this virus, approaches to disrupt these pathways should help to reduce the spread.

Aerosols have become a significant concern in the transmission of SARS-CoV-2 as studies have found exhaled particles when breathing, speaking and singing will transmit respiratory droplets that can stay suspended in the air. In addition, the number of exhaled particles of an individual post-infection increased day 3 thru 7, which provides a timeframe when transmission could increase. However, other studies have found that the transmission potential was greatest in the first 2 days before and 3 days after onset of symptoms in the index patient (Ge et al. 2021). No difference was typically seen in survival between aerosol SARS-CoV-1 and SARS-CoV-2 or between different aerosol SARS-CoV-2 isolates, but one study found higher aerosol efficiency (i.e., ratio of nebulizer concentration to aerosol concentration) with SARS-CoV-2 which suggested better infectivity and virion integrity compared to those of SARS-CoV-1 or MERS. Some studies have shown that small aerosols can contain significant amounts of viral material and can stay aloft and viable for minutes to hours. As of May 2021, the CDC updated the modes of SARS-CoV-2 transmission to be categorized as inhalation of virus, deposition of virus on exposed mucous membranes, and touching mucous membranes with soiled hands contaminated with virus (CDC 2021). Epidemiological studies evaluating superspreader events, such as the Guangzhou restaurant outbreak, provided evidence that airborne transmission was a likely pathway (Hamner 2020; Li et al. 2021; Lu et al. 2020; Miller et al. 2021; Park et al. 2020; Zhang et al. 2021). In addition, the study at the meat packaging plants also supported airborne transmission, as well as corroborated laboratory studies that found environmental conditions such as low temperature and high humidity could lead to increased transmission (Dombrovsky et al. 2020; Günther et al. 2020). Environmental conditions impact surface and aerosol SARS-CoV-2 viability with warmer temperatures and sunlight increasing droplet evaporation and viral decay rates. Strategies to mitigate aerosol and surface transmission need to take into account the environmental conditions to yield the most successful results.

Environmental sampling studies such as the ones performed in healthcare facilities provided evidence that higher viral loads were found in aerosols and surfaces when sampling was typically performed closer to the patient and early in the infection. In a review of healthcare studies, SARS-CoV-2 RNA was mostly found in less than 27% of the surface samples and less than 50% of the air samples, with only a few studies being able to successfully culture surface or air samples (Figure 1). While these studies provided insight into the areas that are the most heavily contaminated with SARS-CoV-2, their major limitation was the inability to specifically quantify the amount of viral load. The lack of studies finding viable virus, especially in the healthcare settings, where one would expect viable virus, could reflect inadequacy of current sampling methods. A negative RNA sample for SARS CoV-2 could indicate an absence of the virus or it also could indicate that the virus was present but was not found, as current methods are insufficiently sensitive. Similarly, a positive SARS CoV-2 RNA sample, indicates the presence of viral RNA fragments, but unless the sample is also cultured successfully, does not determine viability which is critical when discussing transmission. The limited number of studies in which there was successful culture indicates finding viable SARS-CoV-2 is likely more of a methodological problem than lack of viable virus. The lack of finding viable virus prevents occupational-based health decisions (like the types of PPE required) to be determined solely from sampling strategies.

Collection of viable SARS-CoV-2 in the air remains challenging in field studies. The inconsistent results from sampling across multiple healthcare studies are likely due to the variety of methods utilized (type, flow rate of aerosol sampling pump, and duration of collection), ventilation, percentage of fresh air, and air exchange rates, all of which illustrate the strengths and weaknesses within each study (Figure 2)(Dumont-Leblond et al. 2021; Dumont-Leblond et al. 2020). One of the many reasons for varied outcomes from aerosol sampling could be due to the potentially high air exchange rate in the hospital rooms, which varied from minimal air movement to 16 air changes per hour or was not recorded at all (Dumont-Leblond et al. 2020; Jin et al. 2021; Li et al. 2020; Tan et al. 2020). Additionally, sampling for SARS-CoV-2 was conducted in a variety of locations in hospitals including intensive care hospital patient rooms, general ward patient rooms, patient bathrooms, patient mobile toilet facilities, in hospital hallways, and during hospital PPE doffing (Chia et al. 2020; Guo et al. 2020; Jin et al. 2021; Kenarkoohi et al. 2020; Lei et al. 2020; Liu et al. 2021b; Pochtovyy et al. 2021; Razzini et al. 2020; Santarpia et al. 2020; Stern et al. 2021; Tan et al. 2020). The lack of culturable virus from air samples could be the result of potential damage done to the virus during high volume sampling or extended duration of aerosol sampling, as the virus can be lost at multiple stages of the collection process (Brown et al. 2015; Dumont-Leblond et al. 2020; Jin et al. 2021; Pochtovyy et al. 2021). Only one study has evaluated performance of aerosol samplers, finding lower recovery of SARS-CoV-2 with midget impingers (Ratnesar-Shumate et al. 2021). The sampling collection method can have a large impact on maintaining viability of virus, and therefore yield inconclusive estimates of any infectious aerosol concentrations (Brown et al. 2015; Mainelis 2020).

As discussed previously, the proximity to an infectious person, their actions, and time relative to onset of illness, will impact the number of infectious particles that are being aerosolized or spread via bodily fluids, and therefore the number of viral particles that could be detected by sampling. Additional parameters affecting surface sampling include what type of surface was sampled, decontamination procedures in place where the surface was sampled, when the samples were collected relative to when disinfection measures were implemented, and when the surface was contaminated. A range of experimental conditions should be considered when evaluating the decay rate of SARS-CoV-2. The effects of 1) the selected viral suspension medium, including the similarity to respiratory secretions of an infected individual and whether preservative agents such as bovine serum albumin, fetal bovine serum, or both were integrated into the medium, 2) the droplet size of the suspension medium 3) the selected drying period duration prior to the onset of the experiment, and 4) the efficiency of virus recovery can all significantly influence the reported decay rate of SARS-COV-2. In addition, the differences in terminology for results between studies can make it difficult for comparisons, such as terms like decay rate and K-infectivity. When validation is impracticable, researchers should be aware of the limitations of the methods they select and provide detailed explanations of the methods so that the research can be appropriately assessed and compared with other studies.

Environmental parameters during air or surface sampling, such as temperature, relative humidity, and the presence or absence of UV radiation impacts the viability of the virus and, therefore, will also significantly impact the study results and limit the ability to compare results across studies. Not all elements of the methods and conditions are fully described within each study which makes cross-comparison of methods challenging and the experiments difficult to repeat.

Strategies for after-sampling methods for quantifying SARS-CoV-2 in air and on surfaces have varied across studies (Figure 3). While most sampling studies have utilized various air collection samplers or swabs on surfaces, methods for preserving samples during transport and storage, extracting SARS-CoV-2 RNA from sample medium, and quantitating viral nucleotides have differed from study to study. It is unclear whether these differences have significantly impacted the accuracy and precision of the results, but this lack of clarity has made it harder to compare results across studies. The impact of how samples are transported (what medium), the time period between collection and analysis (potential loss), and differences in the RNA extraction method between studies on method performance from an end-to-end (collection through analysis) perspective is poorly understood. Moreover, there is a high degree of variation in the number and combination of genes that have been targeted across studies, which may impact the sensitivity (more target genes lead to higher likelihood of detection) and comparability (due to differences in target genes combinations) of study results. Determining if a sample has viable virus is performed with viral assays which are considerably more complex and difficult than bacterial or fungal assays (Cox et al. 2020). Lastly, many of these methods do not have a set limit of detection of the entire sampling procedure and the collection, including the initial capture method, or extraction efficiencies. Studies run controls with the RT-PCR analysis in order to determine the Ct value to establish sample positivity, but based on a review of hospital studies, the methods for RNA analysis and positivity Ct values varied from study to study (Birgand et al. 2020).

Without validating the method utilized including all the steps during the sampling and analysis processes, limited statements can be made regarding viral loads and infectivity. Public health decisions can only be made with reliable and verified methodologies, and therefore, until sampling can be performed with consistent and successful methodologies for SARS-CoV-2, sampling for viable virus cannot be a determining factor for workplace safety. This rapid review has demonstrated that there is a tremendous amount of work still to be done in regard to sampling for the quantification of SARS-CoV-2 in the environment.

From June 2020 to September 2021, a comprehensive assessment of existing knowledge pertaining to SARS-CoV-2 transmission pathways was conducted biweekly using a rapid review framework. This rapid review quickly assessed very large amounts of information to get specific papers to subject matter experts. This approach allowed for real-time recommendations in the workplace to evolve as transmission pathways were better understood during the pandemic. This process enabled a small group of exposure assessment researchers to review most papers in their area of expertise as they became available. This group of exposure assessment researchers was also utilized to answer questions from workers, workplaces, and other government agencies using the most recent research. During the first summer of the pandemic, a comprehensive research review was used to write the NIOSH research agenda (NIOSH 2021) and also to fund many projects within NIOSH that addressed the research gaps.

The “knowns” of transmission changed rapidly during this period. Initially, contact and droplet transmission were deemed by the WHO as a critical exposure pathway, and as more scientific data became available, aerosol transmission was determined to be the primary pathway. As the pandemic progressed, researchers began to realize that infection and the onset of illness was important, as this varied from SARS-CoV-1, and articles began pointing out the timing of sampling in relation to onset of illness of SARS-CoV-2 patients. Similarly, sampling and environmental conditions such as ventilation, air exchange rates, and proximity of samplers to patients also became reported more frequently. Additionally, the understanding of pre-symptomatic and asymptomatic patients’ contribution to the role of transmission became evident (Wei et al. 2020a; Wei et al. 2020b). As stated earlier, keywords were utilized to narrow the focus of the SARS-CoV-2 research, the articles were screened, and if deemed significant, a summary was generated. Other critical topic areas, such as Engineering Controls or Personal Protective Equipment, had different search parameters and captured alternative articles that were not typically placed in more than one topic area, even though there could have been overlap in content. Therefore, subject matter experts captured additional studies that were deemed highly significant outside of the framework of the rapid review search strategy. This approach yielded a large number of articles for qualitative review by Occupational Environment and Exposure Assessment subject matter experts. Perhaps a more concise transition of information, from keyword to selected summarized key articles, or a ranking system of articles based on expertise determined by number of the previously published peer-reviewed manuscripts in the field could have provided a more efficient real-time assessment.

Future SARS-CoV-2 sampling efforts would greatly benefit from the development of validated sampling methods in a laboratory prior to field sampling. The validation of sampling methods early in the pandemic would have allowed more accurate comparisons of results and appropriate interpretation of sampling findings. Without validated sampling and analysis methods, study findings can be difficult to interpret when addressing key questions, assessing exposure and transmission risk, and developing appropriate mitigation efforts.

While sampling methods to detect SARS-CoV-2 RNA and to isolate infectious SARS-CoV-2 are still developing, results are not currently suitable for the definitive determination of worker safety and transmission risk. Therefore, workers and managers must rely on best practices in order to minimize exposure. Workplace best practices, while not the focus of this literature review, are strategies that can be implemented to potentially reduce exposure including encouraging employees who are sick or have confirmed or suspected SARS-CoV-2 to stay home when possible. While this will reduce some transmission, asymptomatic and pre-symptomatic individuals can still spread the virus (Wei et al. 2020a; Wei et al. 2020b). Findings from the epidemiological, sampling, and experimental literature on SARS-CoV-2 have largely confirmed that many of the catch-all guidance measures for limiting airborne transmission of COVID-19, such as practicing social distancing and wearing masks, were and continue to be effective. Models that have estimated the impact of universal mask wearing have reported considerable reduction in transmission when mask-wearing compliance is over 50% (Tian et al. 2021). Other models have shown a clear reduction in transmission when maximum gathering and building capacity ordinances were enforced (Tian et al. 2021). Additional workplace strategies include supporting personal hygiene practices, improving workplace ventilation and filtration, utilizing outdoor areas as much as possible, reduced touching of common surfaces, and increased surface disinfection, among others.

Based on this literature review, our overarching findings include:

While there are a vast number of articles included in this review, we have purposefully narrowed the research scope, and therefore have not included all SARS-CoV-2 research. Topic areas including Epidemiology, Personal Protective Equipment, and Engineering Controls were not included in this review unless determined by a subject matter expert to be critical. Therefore, by attempting to focus the research scope, all potentially relevant articles may not have been captured. Another limitation was the intention of a quick and efficient review over comprehensiveness, which could also contribute to relevant articles not being captured. In an effort to collect information with a consistent approach, subject matter experts reviewed articles bi-weekly to identify the studies that were most impactful, and qualitatively evaluated the caliber of the studies.