Health risks of pest control and disinfection workers after the COVID-19 outbreak in South Korea

Abstract

The exposure patterns of pest control and disinfection workers have changed after the coronavirus disease 2019 (COVID-19) outbreak, but the health risks of chemical exposure have not been assessed. We identified these workers' chemical exposure patterns and risks before and after the COVID-19 outbreak. We used data conducted between 2018 (pre-pandemic) and 2021 (post-pandemic) from three-year cross-sectional surveys on pest control and disinfection workers. Inhalation and dermal exposure concentrations were estimated using equations based on a biocidal product risk assessment model from the Korean National Institute of Environmental Research. The non-carcinogenic and carcinogenic risks of chemicals were calculated using the United States Environmental Protection Agency risk assessment model. We found that the annual work frequency (50^th percentile) of foggers using disinfectants increased the most among all the work types, from 140 uses/year to 176 uses/year after the COVID-19 outbreak. Moreover, all chemicals' non-carcinogenic and carcinogenic risks increased regardless of exposure routes. In the worst scenario (95^th percentile), the margin of exposure for citric acid, benzethonium chloride, benzyl-C12-16-alkyldimethyl chlorides, and sodium chlorite of inhalation exposure, and isopropyl alcohol and benzyl-C12-16-alkyldimethyl chlorides of dermal exposure were acceptable (>100) before the COVID-19 outbreak but became unacceptable (<100) after the COVID-19 outbreak. Carcinogenic risks of dichlorvos from inhalation and dermal exposure were above acceptable levels (>10⁻⁶) before and after the COVID-19 outbreak but comparatively high after the COVID-19 outbreak. Additionally, significantly more workers experienced health symptoms after the COVID-19 outbreak (p<0.05), with the most common being muscle lethargy (31%), skin/face stinging (28.7%), and breath shortness/neck pain (24.1%).

Graphical abstract

Introduction

The pest control and disinfection companies in Korea have continuously expanded their business since the 1960s. Their primary services mostly include pest control followed by prevention of infectious diseases (KCDC, 2017). However, due to the coronavirus disease 2019 (COVID-19) outbreak in 2020, many places were disinfected simultaneously, which led to a nationwide shortage of human resources (MJJNEWS, 2022). Consequently, all leading facilities changed their primary services from pest control to virus disinfection (Dailyculture, 2022; MedicalTimes, 2020). Such a shift changed workers’ chemical exposure characteristics including exposure chemicals, amounts, and pathways.

The hazards of disinfectants are more harmful to humans compared to those of insecticides (Casey et al., 2017; Purohit et al., 2000). Disinfectants containing quaternary ammonium compounds, glutaraldehyde, and ethanolamine had sensitizing and irritating properties (Heederik, 2014). Therefore, their prolonged exposure was known to cause asthma, dermatitis, and eye irritation (Casey et al., 2017; Gonzalez et al., 2014). Also, unlike insecticides, mainly used in summer, disinfectants are used almost daily in hospitals and indoor environments, resulting in constant exposure. Since the COVID-19 outbreak, the pest control services that were earlier functional between May and September, worked throughout the year, which ultimately increased the chemical exposure of their workers. Moreover, workers have been using ultra-low volume (ULV) sprayers to cover the heavy workload, which increases harmful chemical exposure and damage to the respiratory organs (Choi et al., 2021; Swain et al., 2021). Therefore, these changes in exposure patterns have increased the health risks of workers since the outbreak of COVID-19 (Rim, 2021).

To the best of our knowledge, no studies have conducted a health risk assessment of pest control and disinfection workers considering the changes in exposure patterns after the COVID-19 outbreak. Over the past years few studies conducted on pest control and disinfection workers have reported that they experienced skin and eye irritation, respiratory disorders, and cancer due to biocide chemical exposure (Ambroise et al., 2005; Lindström et al., 2018; MacFarlane et al., 2009; Suojalehto et al., 2021). However, these studies only considered the intermittent exposure to insecticides before the COVID-19 outbreak. Since pest control and disinfection workers have been exposed to harmful chemicals at higher levels than the general population after the COVID-19 outbreak, those workers are expected to experience health risks and clinical symptoms after the COVID-19 outbreak.

Therefore, in this study, we investigated the exposure of pest control and disinfection workers to the biocide product and chemicals used before and after the COVID-19 outbreak and the changes in exposure patterns. Finally, we also assessed changes in health risks of pest control and disinfection workers before and after the COVID-19 outbreak.

1. Materials and methods

1.1. Study participants

This study utilized data on pest control and disinfection workers between 2018 (before the COVID-19 outbreak) and 2021 (after the COVID-19 outbreak) from three-year cross-sectional surveys to compare the health risks of workers before and after the pandemic. Generally, pest control and disinfection companies expand their business over time, leading their workers to increase the type and use of chemicals. Therefore, if we compare the health risks with the same study population, we cannot determine whether the increased health risk is due to their business expansion or the COVID-19 outbreak. For this reason, we chose different study populations belonging to an institution that can represent the pest control and disinfection workers during each study period. Then, we compared each population's work experience and demographic characteristics to see if there were any differences according to potential confounding factors. To select study participants, we were provided a list of 309 randomly selected companies among more than 1000 companies from the Korea Disease Control Association (KDCA), a non-profit corporation under the Korea Disease Control and Prevention Agency. Most of the pest control and disinfection companies actively conducting business in South Korea belong to the KDCA to share disinfection techniques and receive training from the state. Therefore, the KDCA is an institution that can best represent the characteristics of general pest control and disinfection workers in South Korea. Among the workers from these companies, we targeted current field workers and those who took safety education organized by KDCA. As a result, 477 and 533 participants were recruited before and after the COVID-19 outbreak. After excluding those who did not consent to the survey and complete the questionnaire, 156 (response rate: 32.7%) and 174 (response rate: 32.6%) participants were included before and after the COVID-19 outbreak, respectively. Through a preliminary survey, we confirmed that the participants’ responses from one company were similar. Nevertheless, to prevent bias that we recruited only workers from certain companies, we asked other companies’ workers to participate if more than 5% of all respondents participated from one company. The final study subjects engaged in 117 and 136 pest control and disinfection companies before and after the COVID-19 outbreak, respectively, and 1 to 4 workers participated per company (Appendix A Fig. S1).

1.2. Questionnaire survey

We surveyed the pest control and disinfection workers to obtain exposure factors (such as amount of product used and work duration) for the health risk assessment. The questionnaires were developed based on the survey designed to get biocidal exposure factors for the risk assessment provided by the National Institute of Environment Research (NIER), and the survey intended to conduct a risk assessment on farmers and pesticide applicators (NIER, 2018; Yun et al., 2012). The surveys comprised four sections—pest control and disinfection type, biocide products used, exposure factor, and self-reported health symptoms. Regarding self-reporting symptoms, we asked, “Have you ever experienced the health symptoms before the pest control and disinfection work?” and “Have you ever experienced the health symptoms after you started working on pest control and disinfection work?”. Therefore, we could ascertain the newly developed symptoms after the work by comparing the response to the questions. Furthermore, by asking for any disease diagnosed by a physician and the year of diagnosis, we excluded the participants with the underlying disease in the groups with newly developed symptoms. The surveys were conducted by sending an online survey link via email or using contact numbers of the participants. All surveys were reviewed and approved by the Institutional Review Board of Korea University (before the COVID-19 outbreak: 1040548-KU-IRB-18-138-A-2; after the COVID-19 outbreak: KUIRB-2021-0370-01), and informed consent was obtained from all individual participants included in the study.

1.3. Selection of target chemicals for the risk assessment

In a preliminary survey, we get lists of insecticide and disinfectant products and ingredients approved by the Pharmaceutical Affairs Act, regulated by the Korean Ministry of Food and Drug Safety before the COVID-19 outbreak, and approved by the Consumer Chemical Products and Biocides Safety Control Act, regulated by Korean Ministry of Environment (KMOE) after the COVID-19 outbreak. Afterwards, we compared with chemicals that our study subjects reported. As a result, workers all used chemicals except natural substances, phytoncide, allowed by the law. They used 37 of 46 chemicals permitted by the law before the COVID-19 outbreak and 56 of 63 chemicals permitted by the law after the COVID-19 outbreak. Since the KDCA recommends some legally licensed products and highly disinfected and safe chemicals, we can assume that there is no significant difference in the use of chemicals, even in companies that we have yet to investigate. Therefore, we targeted chemicals that study subjects used for the risk assessment, given that their use represented the chemicals that general pest control and disinfection workers used in the field. Then, to compare the risks of chemical exposure before and after the COVID-19 outbreak, we selected chemicals commonly used before and after the COVID-19 outbreak. We identified the hazards of biocidal chemicals and screened target chemicals for the risk assessment using the Product ingredient safety index (PISI) developed for safety and health risk evaluation of formulated product ingredients (Raslan et al., 2020). According to the degree of hazards, carcinogenic, mutagenic and reprotoxic (CMR) chemicals were scored on a scale of 0–4, and non-CMR chemicals on a scale of 0–3 (Appendix A Table S1). Considering the exposure pathways and potential health effects of pest control and disinfection workers, we collected the hazard classification and toxicological information on inhalation toxicity, dermal toxicity, and carcinogenicity from the international toxicology database (ACGIH, 2018; ATSDR, 2020; BAuA, 2020; DFG, 2020; ECHA, 2012; GHS, 2013; HSE, 2020; IARC, 2020; IFA, 2020, 2021; INRS, 2020; MAK Commission, 2010; NIOSH, 2020b; NTP, 2020; OSHA, 2020; SER, 2021; US EPA, 2022). We then selected chemicals representing moderate risk (2 points on a scale) and high risk (3 and 4 points on a scale) in inhalation toxicity, dermal toxicity, and carcinogenicity as target chemicals for the risk assessment.

1.4. Toxicity endpoints

Toxicity endpoints were collected from national toxicological reports (ECHA, 2012; NIOSH, 2022; OECD, 2022; SCCNFP, 2001; US EPA, 2022). To identify the long-term risks (for >1 year), we initially collected chronic toxicity endpoints according to exposure route and subchronic and subacute toxicity values (in the absence of chronic toxicity). We collected no observed adverse effect concentration (NOAEC) as inhalation toxicity reference value and no observed adverse effect level (NOAEL) as dermal toxicity reference value (Kim et al., 2018). Then the toxicity reference values were converted to chronic toxicity reference values for workers. Chronic toxicity reference values (e.g., chronic NOAEC and NOAEL) for each chemical were adjusted to extrapolate the experimental data to an actual human exposure, intra-species and inter-species differences, differences in the duration of exposure, and differences in exposure route following the risk assessment guidelines provided by NIER (Appendix A Table S2) (NIER, 2017). For example, the toxicity reference value for deltamethrin was calculated as 1 × (1/1.4) × 70 kg × (1/10 m³) × (75%/75%) = 5 mg/m³, which is an allometric scaling factor from dog tests (1.4), inhalation rate for workers (10 m³), oral absorption rate (75%), inhalation absorption rate (75%) (Appendix A Table S3).

1.5. Exposure assessment

Considering that pest control and disinfection workers are primarily exposed to chemicals through inhalation and via the skin, we estimated risks from inhalation and dermal exposure. Exposure scenarios were categorized according to the purpose of usage and the application types of biocides. For example, the chemicals used in gel-type products or for wiping were considered under dermal exposure, and those used in hand spraying or fogger were considered under both inhalation and dermal exposures. Exposure concentrations to each chemical were calculated using equations based on a biocidal product risk assessment model developed by the NIER. The exposure factors (such as the amount of product used per application, annual working frequency, and working duration per application) were gathered using the survey questionnaire, and other experimental values (such as airborne fraction, product discharge rate, skin contact area, possible skin contact area) were obtained from exposure factors for biocide applicators presented by NIER (2021). We assessed the health risks based on the 50^th percentile and 95^th percentile (the worst-case) of exposure concentrations (Edokpolo et al., 2015).

1.5.1. Estimation of inhalation exposure concentration

We estimated inhalation exposure concentrations according to product type and method of use. For chemicals used in hand spraying, an instant release (floor or object spraying) scenario was applied (Eq. (1)), and for chemicals used in a fogger, a constant spraying scenario was applied (Eq. (2)). Additionally, we applied the assigned protection factor to prevent overestimated exposure concentrations.

$$\begin{array}{ccc}& & C_{\mathrm{a}}=\left[(R\times T\times W_{\mathrm{f}}\times F_{\text{air}})/V\right]\times e^{-qt}\hfill \\ & & C_{\text{inh}}=C_{\mathrm{a}}\times \text{APF}\times t_{\mathrm{n}}\times n/24\hfill \end{array}$$ (1)

$$\begin{array}{ccc}& & C_{\mathrm{a}}=(R\times W_{\mathrm{f}}\times F_{\text{air}})/(V\times (q+v/h)\times (1-e^{-(q+v/h)\times T})\hfill \\ & & \times {\mathrm{e}}^{-(q+v/h)\times (t-T)(}\left(t>T\right)\hfill \\ & & C_{\text{inh}}=C_{\mathrm{a}}\times \text{APF}\times t_n\times n/24\hfill \end{array}$$ (2)

Here, C _a (mg/m³) represents concentration of biocide chemicals in the air, R (mg/hr) is the rate at which the product is released, T (hr) is spraying time, W _f is weight fraction in product (maximum concentration based on the worst case assumption (SCCS, 2021)), F _air is airborne fraction, V (m³) is the average volume of working space, q (1/hr) is ventilation rate, and t (hr) is the exposure time. C _inh (mg/m³) is the inhalation exposure concentration, APF is the assigned protection factor (default value 10% (OSHA, 2018), t_n (hr/use) is exposure time per application, n (time/day) is frequency of use, v (m/hr) is the Stokes law for particle settling velocity, and h (2.5 m) is the assumed average height of working space. F _air was obtained from exposure factors presented by NIER (Appendix A Table S4) and other factors from the survey. The inhalation lifetime average exposure concentration (LAEC) (µg/m³) was calculated by averaging inhalation exposure amount during the average life span (70 years) (US EPA, 2012).

1.5.2. Estimation of dermal exposure dose

We estimated dermal exposure concentrations according to the product type and method of use. For products used in wiping or gel-type, a semisolid product application scenario was applied (Eq. (3)), and for chemicals using in hand spraying or fogger, during (Eq. (4)) and after spraying scenarios (Eq. (5)) were calculated and summed (Eq. (6)).

$$\begin{array}{ccc}& & L_{\mathrm{d}}=A_{\mathrm{c}}\times W_{\mathrm{f}}\times (1-F)\times A_{\mathrm{s}}\hfill \\ & & D_{\text{der}}=L_{\mathrm{d}}\times \text{APF}\times \text{ABS}\times n/\text{BW}\hfill \end{array}$$ (3)

$$L_{\mathrm{d}1}=R_{\text{der}}\times t_1\times W_{\mathrm{f}}$$ (4)

$$\begin{array}{ccc}& & A_{\mathrm{c}}=A_0/(S_{\max }\times {10}^4)\hfill \\ & & L_{\mathrm{d}2}=A_{\mathrm{c}}\times W_{\mathrm{f}}\times A_{\mathrm{s}}\hfill \end{array}$$ (5)

$$D_{\text{der}}=(L_{\mathrm{d}1}+L_{\mathrm{d}2})\times \text{APF}\times \text{ABS}\times n/\text{BW}$$ (6)

Here, L _d (mg) represents dermal uptake amounts, A _c (mg/cm²) is the amount of skin absorption, F is the airborne fraction, and A _s (cm²) is the skin contact area. D _der (mg/(kg·day)) represents the dermal exposure dose, APF is the assigned protection factor (default value 10% (OSHA, 2018), ABS is the dermal absorption fraction, and BW (kg) is body weight. L _d1 (mg) is the dermal uptake amount during hand spraying or fogging, R _der (mg/min) is the rate at which product is applied to the skin, t ₁ (min/use) is spraying time per application. A ₀ (mg) is the amount of product use, S _max (m²) is the maximum skin contact area. L _d2 (mg) is the dermal uptake amount after spraying. BW, t ₁, and A ₀ were obtained from the survey, and others from the exposure factors presented by NIER (Appendix A Table S4). The dermal lifetime average daily dose (LADD) (mg/(kg·day)) was calculated by averaging dermal exposure amount during the average life span (70 years) (US EPA, 2012).

1.6. Risk characterization

The health risks of pest control and disinfection workers were evaluated for both carcinogenic and non-carcinogenic chemicals. For non-carcinogenic chemicals, the margin of exposure (MOE) was used to assess non-cancer risk (Kim et al., 2018). The MOE is defined as the ratio of NOAEL for the estimated dose or concentration (ECHA, 2016). The MOE value of < 100 represents a non-carcinogenic effect of concern (ECHA, 2016).

The cancer risks were estimated only on chemicals with cancer slope factors. The value of lifetime cancer risk (LCR) was estimated by the following equation (US EPA, 2012).

$$\text{LCR}=D\times \text{CS}{\mathrm{F}}_{\mathrm{r}}$$ (7)

Here, the D (µg/m³ or mg/(kg·day)) represents the LAEC or LADD according to the exposure route, CSF_r represents the cancer slope factor by the route. A LCR value greater than 1.0E-06 represents a carcinogenic effect of concern (ECHA, 2016).

1.7. Statistical analysis

First, we performed the chi-square tests to compare the demographic characteristics of study participants before and after the COVID-19 outbreak. Then, after testing the normality by employing the Kolmogorov-Smirnov test, we conducted Mann-Whitney U tests and Fisher's exact tests to compare the exposure factors and self-reported health symptoms among the study participants included before and after the COVID-19 outbreak. All statistical analyses were performed using the R version 4.0.2, and the statistical significance level was set as a two-sided p-value of <0.05.

2. Results

2.1. Characteristics of pest control and disinfection workers

Table 1 shows the demographic and working characteristics of pest control and disinfection workers before and after the COVID-19 outbreak. Of the workers, more than half of workers were male and were over the age of 50 before and after the COVID-19 outbreak. In addition, about half of the workers had more than ten years of working experience before and after the COVID-19 outbreak. After the COVID-19 outbreak, there was a significant increase in the number of workers using ULV foggers (p < 0.001), while the number of workers using hand sprayers decreased (p < 0.001) after the COVID-19 outbreak. Insecticides were less frequently used after the COVID-19 outbreak compared to that before the COVID-19 outbreak (p < 0.001), while the use of disinfectants increased after the COVID-19 outbreak compared to those before the COVID-19 outbreak (p = 0.025).

Table 1.

Demographic and working characteristics of pest control and disinfection workers before and after the coronavirus disease 2019 (COVID-19) outbreak.

Characteristics	Before the COVID-19 outbreak (n=156)	After the COVID-19 outbreak (n=174)	p-valuea
	N (%)	N (%)
Sex
Male	132 (84.6)	122 (70.1)	0.078
Female	24 (15.4)	52 (29.9)
Age (years)
<40	23 (14.7)	23 (13.2)	0.472
40–49	49 (31.4)	39 (22.4)
50–59	54 (34.6)	63 (36.2)
≥60	30 (19.2)	49 (28.2)
Weight (kg)
<60	16 (10.3)	27 (15.5)	0.490
60–69	43 (27.6)	54 (31.0)
70–79	58 (37.2)	49 (28.2)
≥80	39 (25.0)	44 (25.3)
Smoking
Never	53 (34.0)	69 (39.7)	0.211
Past smoker	48 (30.8)	41 (23.6)
Current smoker	55 (35.3)	64 (36.8)
Drinking
Never	11 (7.1)	30 (17.2)	0.754
Past drinker	30 (19.2)	31 (17.8)
Current drinker	115 (73.7)	113 (64.9)
Working years (years)
<5	27 (17.3)	32 (18.4)	0.477
5–9	53 (34.0)	42 (24.1)
10–19	57 (36.5)	77 (44.3)
≥20	19 (12.2)	23 (13.2)
Work type (multiple choice)
Use hand sprayer	145 (92.9)	103 (59.2)	<0.001
Use ULV fogger	91 (58.3)	138 (79.3)	<0.001
Wipe surface	110 (70.5)	125 (71.8)	0.791
Insecticide use
Yes	153 (98.1)	155 (89.1)	<0.001
No	3 (1.9)	19 (10.9)
Disinfectant use
Yes	77 (49.4)	151 (86.8)	0.025
No	79 (50.6)	23 (13.2)
Protective actions (multiple choice)
General gown	135 (86.5)	134 (77.0)	0.026
Professional gown	34 (21.8)	77 (44.3)	0.001
Goggle	30 (19.2)	70 (40.2)	<0.001
Mask	109 (69.9)	145 (83.3)	0.004
Gloves	95 (60.9)	131 (75.3)	0.005
Safety helmet	36 (23.1)	47 (27.0)	0.411
Ear cap	10 (6.4)	7 (4.0)	0.327
Ventilation rate (work−1)b
<1	50 (32.1)	42 (24.1)	0.201
1–2	77 (49.4)	91 (52.3)
≥2	29 (18.6)	41 (23.6)

N: number of respondents; COVID-19: coronavirus disease 2019; ULV: ultra-low volume.

^aChi-square test;

^bVentilation rate indicates the frequency of the air volume change at the workplace.

2.2. Biocidal active ingredients used in pest control and disinfection

Table 2 shows the number of biocidal products and active ingredients used by the pest control and disinfection workers before and after the COVID-19 outbreak. This number was significantly increased after the COVID-19 outbreak compared to that before the COVID-19 outbreak. The number of disinfectants increased by 368.8% from 16 to 75, and the number of active ingredients increased by 133.3% from 12 to 28.

Table 2.

Biocidal active ingredients used in pest control and disinfection before and after the COVID-19 outbreak.

Product	Before the COVID-19 outbreak (n=156)			After the COVID-19 outbreak (n=174)
	Nproduct	Nchemical	Most used ingredient	Nproduct	Nchemical	Most used ingredient
Insecticide (% change)	78	26		104 (↑33.3)	28 (↑3.7)
Pyrethroid	51	15	Deltamethrin	76	17	Deltamethrin
Organophosphorus	11	5	Temephos	11	4	Temephos
Pyrazole	2	1	Fipronil	4	1	Fipronil
Carbamate	0	0	-a	1	1	Propoxur
Neonicotinoid	4	2	Imidacloprid	4	2	Imidacloprid
Trifluoromethyl aminohydrazone	7	1	Hydramethylnon	6	1	Hydramethylnon
Biological insecticide	2	1	Bacillus thuringiensis	1	1	Bacillus thuringiensis
Other	1	1	Demethicone	1	1	D-limonene
Disinfectant (% change)	16	12		75 (↑368.8)	28 (↑133.3)
Quaternary ammonium compounds	7	3	Benzethonium chloride	30	6	n-Alkyldimethylethylbenzyl ammonium chloride
Chlorine series	2	1	Hypochlorous acid	16	3	Hypochlorous acid
Alcohol series	2	2	Isopropyl alcohol	10	5	Ethyl alcohol
Peroxy acid compounds	0	0	-	7	4	Hydrogen peroxide
Acid anionic series	3	4	Citric acid	9	9	Citric acid
Other	2	2	Phytoncide	3	1	Dichloro-m-xylenol

COVID-19: coronavirus disease 2019; N_product: number of products; N_chemical: number of chemicals.

^a“-” denotes a null value.

2.3. Target chemicals for the risk assessment

The hazard scores of chemicals based on PISI are presented in Table 3 . Among twenty-four chemicals commonly used before and after the COVID-19 outbreak, eighteen chemicals with medium (score of 2) and high (score of 3) scoring for inhalation and dermal exposures were selected as target chemicals for the non-carcinogenic risk assessment. Among these chemicals, dichlorvos, benzalkonium chloride, and benzyl-C12-16-alkyldimethyl chlorides had the highest scores (3 points) for inhalation and dermal exposures. Among carcinogens, dichlorvos and fipronil had three points for carcinogenicity. However, as fipronil did not have cancer slope factors for any exposure route, only dichlorvos was evaluated for the carcinogenic risk assessment.

Table 3.

Product ingredient safety index (PISI) scores on inhalation (0–3 points), dermal (0–3 points), carcinogenicity (0–4 points), and target chemicals for the risk assessment.

Chemical name	Inhalation	Dermal	Carcinogenicity	Target chemical
Insecticide
Deltamethrin	2	0	0	√b
Etofenprox	2	0	0	√
Bifenthrin	2	2	0	√
λ-cyhalothrin	2	2	0	√
Cyfluthrin	3	0	0	√
Permethrin	0	3	0	√
D-phenothrin	1	1	0
Cypermethrin	0	3	0	√
Octachlorodipropylether	0	0	0
Temephos	3	2	0	√
Chlorpyrifos	3	1	0	√
Dichlorvos	3	3	3a	√
Imidacloprid	0	0	0
Hydramethylnon	0	0	0
Bacillus thuringiensis	0	0	0
Fipronil	3	2	3	√
Disinfectant
Benzethonium chloride	0	2	0	√
Benzalkonium chloride	3	3	0	√
Benzyl-C12-16-alkyldimethyl chlorides	3	3	0	√
Hypochlorous acid	0	0	0
Isopropyl alcohol	3	1	0	√
Ethanol	3	1	0	√
Citric acid	2	0	0	√
Sodium chlorite	3	1	0	√

^aThe inhalation unit risk of dichlorvos was 0.083 (µg/m³)⁻¹ (US EPA, 1989), and the dermal cancer slope factor was 3.4483 ((kg·day)/mg) (Chen et al., 2021).

^b“√” denotes a target chemical for risk assessment as they scored medium (score of 2) and high (score 3) for inhalation and dermal exposures and carcinogenicity.

2.4. Toxicological endpoints of target chemicals for the risk assessment

Table 4 summarizes calculated chronic toxicological endpoints according to the exposure routes for the target chemicals. We computed the inhalation and dermal toxicity reference values of eleven chemicals in insecticides and seven in disinfectants. In the inhalation exposure scenario, etofenprox showed the lowest toxicity reference value with a corrected NOAEC of 0.0134 (mg/m³) among the target chemicals. In the dermal exposure scenario, dichlorvos showed the lowest toxicity reference value with a corrected NOAEL of 0.01 (mg/(kg·day)) among the target chemicals.

Table 4.

Summary of calculated toxicological endpoints for target chemicals.

Target chemicals	NOAECcorr (mg/m3)	NOAELcorr (mg/(kg·day))
Insecticide
Deltamethrin	5	166.67
Etofenprox	0.0134	166.67
Bifenthrin	3.75	8.33
λ-cyhalothrin	33.5	1.67
Cyfluthrin	22.61	62.67
Permethrin	0.0737	500
Cypermethrin	4.38	3.26
Temephos	0.13	0.1974
Chlorpyrifos	0.1079	0.83
Dichlorvos	0.24	0.01
Fipronil	-a	0.83
Disinfectant
Benzethonium chloride	5.83	312
Benzalkonium chloride	0.089	10
Benzyl-C12-16-alkyldimethyl chlorides	7.96	6
Isopropyl alcohol	449.57	250
Ethanol	8.71	10,800
Citric acid	1050	2400
Sodium chlorite	0.58	416.67

NOAEC_corr: corrected no observed adverse effect concentration; NOAEL_corr: corrected no observed adverse effect level.

^a“-” denotes a null value.

2.5. Exposure factor and estimated exposure dose

Fig. 1 illustrates exposure factors by product type before and after the COVID-19 outbreak. Annual work frequency for fogging using insecticide and hand spraying, fogging, and wiping using disinfectants increased significantly after the COVID-19 outbreak (all p < 0.05). In particular, the annual work frequency (50^th percentile) of hand spraying, fogging, and wiping using disinfectants increased from 180 use/year, 140 use/year, and 120 use/year, respectively, before the COVID-19 outbreak to 260 use/year, 176 use/year, and 162 use/year, respectively, after the COVID-19 outbreak (Fig. 1a). However, the amounts used per work and the duration per work did not change significantly before and after the COVID-19 outbreak (Fig. 1b and c).

Fig. 1.

Exposure factors by product type before and after the coronavirus disease 2019 (COVID-19) outbreak: (a) work frequency, (b) amount, and (c) work duration. The black hexagram indicates a significant difference before and after the COVID-19 outbreak (p<0.05). COVID-19: coronavirus disease 2019.

Appendix A Tables S5 and S6 show estimated exposure concentration during the workers' average working period or lifetime according to the inhalation and dermal exposure of the chemicals. The inhalation and dermal exposure concentrations of chemicals increased after the COVID-19 outbreak compared to that before the COVID-19 outbreak. In particular, while the etofenprox, permethrin, and dichlorvos largely increased their 95^th percentile of exposure concentration among the insecticides after the COVID-19 outbreak, all disinfectant chemicals significantly increased their 95^th percentile of exposure concentration after the COVID-19 outbreak. The inhalation amounts of insecticide and disinfectants used by ULV foggers largely increased compared to those used in hand spraying (Appendix A Table S5). Also, the dermal exposure amounts of disinfectants with foggers significantly increased compared to that with hand spraying after the COVID-19 outbreak (Appendix A Table S6).

2.6. Risk characterization

2.6.1. Non-cancer risks

Fig. 2 shows the non-cancer risks with inhalation exposure according to the product type before and after the COVID-19 outbreak. When exposed to hand spraying or fogging, the MOE for all chemicals decreased after the pandemic outbreak compared to that before the pandemic outbreak. In the worst scenario (95^th percentile), the MOE for citric acid exposure by fogging was acceptable before the pandemic outbreak (MOE = 159) and decreased to an unacceptable level after the pandemic outbreak (MOE = 11). In addition, the MOEs for benzethonium chloride, benzyl-C12-16-alkyldimethyl chlorides, and sodium chlorite exposure by hand spraying were acceptable before the pandemic outbreak (MOE = 371, 282, and 442, respectively) and reached an unacceptable level after the pandemic outbreak (MOE = 76, 11, and 17, respectively).

Fig. 2.

The margin of exposure (MOE) for selected chemicals by two different inhalation exposure routes before and after the COVID-19 outbreak. MOE: margin of exposure.

Fig. 3 demonstrates the non-cancer risks with dermal exposure according to the product type before and after the COVID-19 outbreak. When exposed to hand spraying, fogging, or wiping or gel, the MOE for all chemicals except deltamethrin decreased after the pandemic outbreak compared to that before the pandemic outbreak. In the worst scenario (95^th percentile), the MOE for benzyl-C12-16-alkyldimethyl chlorides exposure by wiping was acceptable before the pandemic outbreak (MOE = 444) and decreased to an unacceptable level after the pandemic outbreak (MOE = 15). The MOEs for isopropyl alcohol and benzyl-C12-16-alkyldimethyl chlorides exposure by fogging were acceptable before the pandemic outbreak (MOE = 1531 and 472) and decreased to an unacceptable level after the pandemic outbreak (MOE = 54 and 16).

Fig. 3.

The MOE for selected chemicals by three different dermal routes before and after the COVID-19 outbreak. *The results of fipronil while using gel.

2.6.2. Cancer risks

Fig. 4 demonstrates lifetime cancer risks due to dichlorvos exposure according to the exposure route and working type. After the COVID-19 outbreak, the median inhalation cancer risk of hand spraying and fogging increased from 4.29E-07 to 6.38E-07 and from 3.02E-07 to 5.27E-07, respectively, but did not reach the safety threshold. In the worst scenario (95^th percentile), the inhalation cancer risk of hand spraying and fogging increased from 1.04E-06 to 1.30E-06 and from 1.07E-05 to 2.10E-05 after the COVID-19 outbreak (Fig. 4a). After the COVID-19 outbreak, the median dermal cancer risk of hand spraying and fogging increased from 9.70E-06 to 9.40E-06 and from 2.75E-05 to 3.66E-05, respectively. In the worst scenario (95^th percentile), the dermal cancer risk of hand spraying and fogging increased from 8.76E-05 to 1.30E-04 and from 2.98E-04 to 3.90E-04, respectively, after the COVID-19 outbreak (Fig. 4b).

Fig. 4.

Cancer risks of dichlorvos by exposure route before and after the COVID-19 outbreak: (a) inhalation and (b) dermal.

2.7. Change in health symptoms

Fig. 5 represents the percentage of health symptoms reported by pest control and disinfection workers according to working 10 years before and after the COVID-19 outbreak. Participants who worked ≥10 years experienced significantly more shortness of breath, sore throat, cough, runny nose, muscle weakness, skin/face stinging, or itching, headaches, and eye-related problems than those who worked <10 years, both before and after the pandemic. In addition, 23.7% and 28.8% of those working <10 years and ≥10 years respectively, experienced at least one health symptom before the pandemic, and the rate increased to 26.4% and 37.9%, respectively, after the pandemic.

Fig. 5.

Percentage of health symptoms reported by pest control and disinfection workers. Data are presented as a function of their working years (<10 years or ≥10 years). The red hexagram indicates a significant difference in health symptoms according to working years (p<0.05). The blue hexagram indicates a significant difference in health symptoms before and after the COVID-19 outbreak (p<0.05).

3. Discussion

This study evaluated the changes in the health risks of pest control and disinfection workers according to the changes in chemical exposure patterns after the COVID-19 outbreak. The pandemic has raised the non-carcinogenic and carcinogenic risks of all biocide chemicals used by pest control and disinfection workers due to increased inhalation and dermal exposures. In particular, the non-carcinogenic risks of four chemicals (citric acid, benzethonium chloride, benzyl-C12-16-alkyldimethyl chlorides, and sodium chlorite) regarding inhalation exposure, and two chemicals (benzyl-C12-16-alkyldimethyl chlorides, isopropyl alcohol) concerning dermal exposure, changed from an acceptable risk level to unacceptable after the pandemic. The number of workers experiencing health symptoms also increased; the most common symptoms in the order of muscle lethargy, stinging of the skin and face, shortness of breath, and neck pain.

3.1. ULV sprayer use and changes in chemical exposure patterns

Due to the small scale of most pest control and disinfection companies in Korea, there is little awareness of the hazardous effects of chemicals. Since workers lack the knowledge of the harmful effects of chemicals, they use highly toxic chemicals including carcinogens, to activate insecticidal effects in a short period (Al-Sayah, 2020). Such chemical misuse continued even after the pandemic due to increased use of ULV sprayers (Ghafoor et al., 2021). World Health Organization (WHO) and United States Centers for Disease Control and Prevention (US CDC) do not recommend using ULV sprayer for disinfection as it may cause adverse health effects by dispersing disinfectant aerosols (US CDC, 2021; WHO, 2020). However, ULV sprayers have the advantage of disinfecting a large area in a short time and are effective in inactivating viruses or microorganisms (Choi et al., 2021). ULV sprayers take less than 10 minutes to inactivate pests or microorganisms while wiping takes >10 min in the same space and shows over 80% of disinfection efficiency (Choi et al., 2021; Farajollahi et al., 2012; Suman et al., 2012). For this reason, the frequency of use of foggers with disinfectants increased from 140 times a year before the pandemic to 274 times a year after the pandemic (Fig. 1). In our previous study, the disinfectant aerosols dispersed from ULV sprayers resided in the air for approximately one hour (Choi et al., 2021), and some chemicals used in foggers showed inhalation risks in this study. Therefore, management guidelines for ULV foggers are needed for the safety of pest control and disinfection workers.

3.2. Risk of biocidal chemicals by exposure route and application method

This study showed that even if workers use the same chemicals, the risks are different according to the exposure route and application method. Previous studies also confirmed that different exposure routes increased the health risks (Choi et al., 2013; Martínez Vidal et al., 2002; Protano et al., 2009). Baharuddin et al. (2011) reported that inhalation exposure was more detrimental to the human body than dermal exposure, as the insecticide was approximately 10% absorbed when it was exposed to the skin, while it was 100% absorbed in the body when exposed to inhalation, which was consistent with our results in this study. In the worst-case exposure (95^th percentile), the risks of all chemicals except benzyl-C12-16-alkyldimethyl chlorides, temephos, and dichlorvos showed acceptable risk levels during dermal exposure, but the risks of inhalation exposure were unacceptable. Additionally, even if the exposure route of chemicals was the same as for inhalation, exposures might differ according to the product application type (Cao et al., 2018). In this study, the exposure amounts of all chemicals used in a ULV fogger were higher than that used in a hand sprayer after the pandemic. This is because knapsack (or fogger) spray chemicals affect the respiratory tract, while motorized sprayers (or hand sprayers) spray chemicals on floors (Baharuddin et al., 2011; Taheri et al., 2021). Therefore, it is crucial to standardize measurements for reducing chemical exposures of pest control and disinfection workers. Moreover, workers should be made aware of the hazardous effects of chemical they use and trained for proper use of protective equipment to avoid exposure.

3.3. Risk comparison with other occupational groups

We compared the observed health risks of pest control and disinfection workers in this study to other workers using pesticides. Only 6/18 chemicals investigated in this study have been studied previously, and mostly these studies focused on dermal exposure (Appendix A Table S7) (An et al., 2015; Atabila et al., 2018; Caffarelli et al., 2004; Choi et al., 2006; Hughes et al., 2008; Kongtip et al., 2013; Massalatchi Illyassou et al., 2017; Samiee et al., 2021). In most of the studies, the participants were mainly farmers working with manual sprayers, and dermal exposure was considered as the most critical exposure route for pesticides based on the measurements of oral, inhalation, and dermal exposure amounts (Brand et al., 2007; Grandjean, 1990). Previous studies have shown that the dermal exposure of farmers to deltamethrin, bifenthrin, λ-cyhalothrin, cypermethrin, chlorpyrifos, and dichlorvos had adverse health risks. However, in this study, only dichlorvos was reported as producing unacceptable health risks during dermal exposure. It may be because we considered that workers were wearing personal protective equipment; however, pest control and disinfection workers had different exposure patterns from those in the prior study due to the high frequency of the use of ULV sprayers. In this study, five chemicals (deltamethrin, bifenthrin, λ-cyhalothrin, cypermethrin, chlorpyrifos), except dichlorvos, did not show adverse health risks in dermal exposure, but showed these risks during inhalation exposure. In addition, we also identified the health risks of these five chemicals using MOE, considering both dermal and inhalation exposures. Therefore, compared to other occupational groups (such as farmers) that treated biocidal products, pest control and disinfection workers should comprehensively consider the risks of both dermal and inhalation exposure routes.

Meanwhile, the symptoms of shortness of breath, sore throat, cough, runny nose, muscle weakness, skin/face stinging, or itching were commonly observed in pest control and disinfection workers in this study. The biocidal chemicals that these workers use can cause respiratory disorders and skin irritation in the human body. Hydrogen peroxide and sodium chlorite, which are commonly used as disinfectants, are classified as specific target organ toxicants repeated exposure (STOTRE) 3 (respiratory sensitization) and acute inhalation toxicity 2 under the globally harmonized system (GHS) toxicity classification (GHS, 2013), and cause symptoms such as shortness of breath and bronchitis in humans when inhaled (Choi et al., 2020; Rim, 2021; US EPA, 2020). Moreover, these health effects are also observed in studies conducted on farmers. Baharuddin et al. (2011) reported that farmers who sprayed insecticides had experienced nausea, reddish face, skin itchiness, numbness, and excessive sweating. However, the previous studies including ours, are all cross-sectional studies, therefore, we cannot interpret the observed associations as a causal relationship. Moreover, studies on the health effects of exposure to biocides targeting pest control and disinfection workers are also limited. Therefore, further longitudinal studies are required to validate the adverse health effects of exposure to biocides on pest control and disinfection workers.

3.4. Strengths and limitations

This study has the advantage of quantitatively evaluating the biocidal exposure of pest control and disinfection workers. The Korean government has made an effort in the biocide management by implementing the Biocidal Product Act after the humidifier disinfectant accidents in 2011. Moreover, the KMOE prepared exposure factors for biocidal product risk assessment in 2021 and made it mandatory to conduct a health risk assessment for biocidal product manufacture (NIER, 2021). Among other industries, the pest control and disinfection industries mostly use biocide, therefore, we should prioritize managing biocidal chemical exposure of these workers based on this risk assessment. However, little research has been conducted due to the difficulty of recruiting pest control and disinfection workers interested in the health changes they experience and those are willing to participate in the studies as their businesses are small and show high turnover rate. In such circumstances, our study might have an impact because, to our knowledge, it is the first to observe the chemical exposure of pest control and disinfection workers and changes in exposure that have occurred after the pandemic.

The limitations of this study should also be addressed. First, as a cross-sectional study, we could not warrant the causality. However, we could identify the newly developed symptoms after the disinfection work by questioning whether the workers had symptoms before work and whether they experienced the symptoms after work. It makes us possible to explain the association between pest control and disinfection work and health symptoms. Second, the study is based on self-reports, and the study participants might have intentionally reported less use of biocides as they were concerned that it would impact the use of biocidal products and possibly damage their business due to strict regulations. Therefore, workers might be more exposed to chemicals than they reported. Third, since we calculated the risks assuming that all workers wore personal protective equipment, the actual risks are likely to be greater than those reported in the results of this study. In future research, we suggest longitudinal studies to confirm the health effects of exposure to biocidal chemicals among pest control and disinfection workers. Additionally, personal exposure measurements among those workers are needed to reduce uncertainties in risk estimation.

4. Conclusions

This study presented changes in health risks of pest control and disinfection workers before and after the COVID-19 outbreak. We observed that the frequency of use of the ULV sprayer and disinfectant was significantly increased after the COVID-19 outbreak, and the non-carcinogenic and carcinogenic health risks of workers have increased regardless of exposure routes. In the worst exposure scenario (95^th percentile), the risks of some disinfectants were acceptable before the COVID-19 outbreak but became unacceptable after the COVID-19 outbreak owing to inhalation and dermal exposures. Moreover, the number of workers who experienced muscle lethargy, stinging of the skin and face, shortness of breath, and neck pain increased after the COVID-19 outbreak compared to those before the COVID-19 outbreak, and were more prominent in those who worked ≥10 years. This study suggests that the increased chemical exposure due to the COVID-19 pandemic has worsened the health risks of pest control and disinfection workers.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.