Genomic and serological assessment of asymptomatic severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in child labor

ABSTRACT

Since working children have limited access to testing and monitoring for COVID-19, we decided to measure SARS-CoV-2 prevalence among them and compare it to non-working children. Our objective is to compare the frequency of SARS-CoV-2 genome and anti-SARS-CoV-2 antibody among working and non-working children. Volunteer child labor studying at Defense of Child Labor and Street Children and randomly selected 5–18-year-old (same range as child labor group) unemployed children participated in this study. The groups, respectively, had 65 and 137 members. This is an analytical cross-sectional study that surveys molecular prevalence of SARS-CoV-2 infection by RT-PCR, and seroprevalence of SARS-CoV-2 antibody by ELISA in working and non-working children. The IBM SPSS statistics software version 25 was used for data analysis. The χ2 or Fisher’s exact test was used to analyze categorical dependent variables, for calculating odds ratios and 95% confidence intervals. Among the children enrolled in this study, molecular prevalence of SARS-CoV-2 turned out to be 18.5% in working children while it was 5.8% in unemployed children [aOR: 3.00 (CI95%: 1.00–7.00); P value: 0.003] and seroprevalence turned out to be 20% in working children vs 13.9% in non-working children [aOR: 1.000 (CI95%: 0.00–2.00); > P 0.001]. Equal SARS-CoV-2 viral load as adults and no symptoms or mild ones in children, coupled with working children’s strong presence in crowded areas and their higher rate of COVID-19 prevalence, make them a probable source for spread of the virus.

Introduction

Virology of SARS-CoV-2

In December 2019, a cluster of pneumonia cases with unknown etiology were reported in Wuhan, China. In early January 2020, a novel coronavirus, named severe acute respiratory distress syndrome coronavirus 2 (SARS-CoV-2), was identified as the causative agent [1]. From Wuhan, the virus quickly spread across the globe and urged the World Health Organization (WHO) to announce a pandemic on 11 March 2020 [2]. SARS-CoV-2 is a member of the Coronaviridae family and was initially isolated from the patient’s respiratory tract. An infection with this virus can cause a wide-spectrum of respiratory and systemic symptoms, collectively referred to as coronavirus disease 2019 (COVID-19) [3]. Similar to previous coronaviruses SARS and MERS, the newly emerged virus is classified as a member of the Orthocoronavirinae sub-family of betacoronavirus genus [4]. Diagnosis of SARS-CoV-2 infection is based on either serological immune assays (IA) or molecular assays, such as real-time reverse transcription polymerase chain reaction (RT-PCR). Currently, RT-PCR is the gold standard for diagnosing SARS-CoV-2 infection [5]. Antibody testing for SARS-CoV-2 enables the identification of past exposure to the virus, even after the virus has been cleared by the immune system. Combining both serological methods and RT-PCR allows the identification of resolved and current infection, respectively, which are therefore valuable techniques in epidemiological studies [5–8].

SARS-CoV-2 in children

So far, few studies have been done to evaluate SARS-CoV-2 infections and their complications in pediatrics. COVID-19 remains mostly asymptomatic or induces mild symptoms in children, while severe or critical cases are rare cases and might be accompanied by nonspecific symptoms [9]. Although children seem less susceptible to severe COVID-19, infected children can still serve as potential sources of transmission to more vulnerable people [10,11]. Therefore, studying transmission dynamics in children has significant importance for developing prevention policies to contain COVID-19 spread.

Economic aspects of SARS-CoV-2 and child labor

The SARS-CoV-2 pandemic affected economies and caused businesses to close-down. During previous Ebola virus outbreaks, the detrimental impact on local economies exacerbated child labor [12]. A similar scenario is unfolding in the current COVID-19 pandemic, with millions of children at risk of child labor. Both the act of closing schools to prevent COVID-19 spread and the common misconception that children are less susceptible to COVID-19 can push children into labor [13,14].

It is estimated that there are currently 152 million children in child labor globally, of which half of them are between 5 and 11 years of age [15]. Child labor is a socio-economic problem that is closely associated with poverty, illiteracy, (illegal) immigration, and human trafficking. Furthermore, these children have an increased risk of various physical and psychological health issues [14–16]. For instance, studies have reported that working children had a higher incidence of infectious diseases, including respiratory infections [17]. Hamdan-Mansour and colleagues reported that influenza was more common in child labor compared to non-working children [18]. Furthermore, Tiwari et al. also reported that tuberculosis and hilar gland enlargement/calcification were more common in working children in the gem polishing industries compared to non-employed children [19].

A study indicated that children from poor socioeconomic backgrounds are more prone to COVID-19 [20]. A lot of these children suffer from chronic malnutrition, which makes them more vulnerable to a wide range of diseases [21]. Furthermore, long working hours in hazardous environments increase the chance of exposure to all sorts of pathogenic agents, including SARS-CoV-2 [14].

SARS-CoV-2 in child labor still remains unexplored, and to our knowledge, this is the first study that evaluates SARS-CoV-2 exposure in children in child labor. In this study, the age- and employment-related prevalence of SARS-CoV-2 in two different groups of children was investigated.

Material and methods

Study design

This study is an analytical cross-sectional study with a control group. Children aged between 5 and 18 years old were recruited by a non-randomized convenience sampling method and divided into two separate groups: 1) unemployed children and 2) employed children. Employed children were recruited through a non-governmental organization school, named Defense of Child Labor and Street Children (South of Tehran, Iran). This organization educates labor children and holds classes on reading/writing and life skills. On the sampling day, employed children with their parents or their legal guardians were invited to the school and researchers explained the goals and method of the study. An informed written consent was obtained from all parents or child’s legal guardians. Unemployed children were recruited from different hospitals across Tehran, Iran. These children were admitted to the hospitals for various medical treatments and their parents signed the written informed consent forms for participation of their children in our study.

This study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Tarbiat Modares University of Tehran (code number: IR.TMU.REC.1399.046). The participants were asked and informed about being included in the study and were able to withdraw from it at any moment.

Data collection

The children’s demographic data was gathered by filling out a questionnaire (including age, sex, education, employment status, and having a permanent residence). Dacron nasopharyngeal and oropharyngeal swabs were collected and transferred to the Pasteur Institute of Iran for the evaluation of SARS-CoV-2 RNA. Because most of our participants were asymptomatic or had mild symptoms, to increase the chance of viral RNA detection, nasopharyngeal and oropharyngeal swabs were simultaneously collected for each participant. Peripheral venous blood samples (2 ml) were collected and transferred to the virology laboratory of Tarbiat Modares University of Tehran.

Molecular assessment

The viral RNA from nasopharyngeal and oropharyngeal specimens was extracted using the QIAamp Viral RNA Mini Kit according to the protocol that is fully automated on QIAcube. The Sansure Biotech’s Novel Coronavirus (2019-nCoV) Nucleic Acid Diagnostic Kit (PCR-Fluorescence Probing) was applied for SARS-CoV-2 genome detection with real-time RT-PCR method. The kit is designed to detect open reading frame 1ab (ORF1ab) and N gene with FAM and ROX as the reporter dye. The conditions were set in accordance with the manufacturer’s protocol. The RNase P gene is used as an internal control for monitoring the quality of sampling and avoiding false-negative results, with CY5 as a reporter dye. We also used ‘no template’ control to detect false positives caused by contamination and a positive template control for the assessment of RT-PCR performance. The following RT-PCR temperature cycles were used: 50°C for 30 min, 1 cycle; 95°C for 1 min, 1 cycle; 95°C for 15 sec, 60°C for 31 sec, 45 cycles; 25°C for 10 sec, 1 cycle according to the manufacturer protocol.

Serological assessment

Blood samples were centrifuged to separate the serum. The collected sera were stored in −20°C prior to detecting SARS-CoV-2 antibody using an enzyme-linked immunosorbent assay (ELISA). We applied the Pishtaz Teb’s SARS-CoV-2 IgG ELISA Kit, which is designed to detect IgG antibodies directed against the SARS-C oV-2 nucleocapsid protein. The kit was used in accordance with the manufacturer’s protocol. We used the kit’s positive control serum, which includes buffering solution and deactivated human serum containing anti-SARS-CoV-2 IgG antibody. Phosphate buffer and human serum devoid of anti-SARS-CoV-2 IgG antibody included in the kit were used as a negative control.

The presence of SARS-CoV-2 IgM was assessed in samples that were PCR positive and IgG negative. Here, the Pishtaz Teb’s SARS-CoV-2 IgM ELISA kit was used in accordance with the manufacturer’s protocol.

Statistical analysis

The collected data was analyzed using IBM SPSS Statistics software version 25. The χ2 or Fisher’s exact test was used to compare categorical data and to calculate odds ratios and 95% confidence intervals.

Results

Demographic information

In total, 65 children in child labor and 137 unemployed children were included in the study. Of the children in the child labor group, 56.9% were boys and 43.1% were girls with an average age of 11.63 years old and none of these children had the Persian nationality. The non-working group consisted of 137 children, 59.1% boys and 40.9% girls, with an average age of 11.26 years old (Table 1). Of the 65 children in child labor who participated in this study, 41 (63.1%) worked as street peddlers and 24 (36.9%) were service workers; on average, they worked 52.88 hours per week (min. 7 hours, max. 91 hours). The mean age of onset was 8.66 with a minimum of 5 years and a maximum of 15 years (Table 1).

Table 1.

Baseline characteristics of participants.

Qualitative variables		Child labor group				Comparison group				Total
		Frequency		Percentage		Frequency		Percentage		Frequency		Percentage
Gender	Female	28		43.1		57		41.6		85		42.1
	Male	37		56.9		80		58.4		117		57.9
	Total	65		100		137		100		202		100
Education	Preschool	1		1.5		36		26.3		37		18.3
	Primary school	62		95.4		47		34.3		109		54.0
	Secondary school	2		3.1		54		39.4		56		27.7
	Total	65		100		137		100		202		100
Job	Service works	24		36.9		0		0		24		11.9
	Street peddler	41		63.1		0		0		41		20.3
	Unemployed	0		0		137		100		137		67.8
	Total	65		100		137		100		202		100
Age	5–12 years	43		66.1		79		57.7		122		60.4
	13–19 years	22		33.9		58		42.3		80		39.6
	Total	65		100		137		100		202		100
Nationality	Iranian	0		0		137		100		137		67.8
	Non-Iranian	65		100		0		0		65		32.2
	Total	65		100		137		100		202		100
	Average (SD)					Minimum				Maximum
Quantitative variables with normal distribution	Child labor group		Comparison group		Total	Child labor group	Comparison group		Total	Child labor group	Comparison group		Total
Age	11.63(2.8)		11.26(4.85)		11.38(4.29)	5	5		5	18	18		18
Age of onset of work	8.66(1.98)		0			5	0			15	0
Working hour per week	52.88(18.24)		0			7	0			91	0

Molecular results

The prevalence of molecular detection of SARS-CoV-2 in working children was significantly higher than in unemployed children (18.5% and 5.8%, respectively).

In the child labor group, 52 cases (80%) were tested negative by RT-PCR, 12 cases (18.5%) were tested positive, and 1 (1.5%) was missing data. Among the 12 positive cases, 6 (50%) were asymptomatic. Three (25%) children had only one symptom (two cases suffered from a cough and one child reported headache symptoms), and the other three had two or more symptoms (headache with gastrointestinal symptoms, headache with respiratory symptoms, fever, weakness, and cough). Among symptomatic patients, cough and headache were reported in two cases (66.6%), and weakness, nausea, and fever were reported in one case (33.3%). While in our comparison group, out of 137 individuals, 129 (94.2%) were tested negative and 8 (5.8%) were tested positive (logistic regression test, adjusted OR: 3.000 (CI95%: 1.00–7.00); P value 0.003). Among the positive cases, four (50%) were asymptomatic, two (25%) had just one symptom that was fever, and the remaining two had more than two symptoms including fever, gastrointestinal symptoms, and respiratory symptoms (the complaints reported by two groups are shown in Table 2).

Table 2.

The complaints reported by two groups (number and duration).

		Child labor group		Comparison group		Total
		Frequency	Percentage	Frequency	Percentage	Frequency	Percentage
Symptoms	No symptoms	37	56.9%	132	96.4%	169	83.7%
	One symptom	18	27.7%	3	2.2%	21	10.4%
	Headache	6	33.3%	0	-	6	28.5%
	Cough	6	33.3%	0	-	6	28.5%
	Sore throat	2	11.1	0	-	2	9.5%
	Fever	2	11.1	3	100%	5	23.8%
	Muscular pain	2	11.1	0		2	9.5%
	2 or more symptoms	10	15.4%	2	1.5%	12	5.9%
	Headache	8	80%	0	-	8	66.6%
	Cough	6	60%	2	100%	8	66.6%
	Fever	5	50%	2	100%	7	58.3%
	Sore throat	3	30%	0	-	3	25%
	Weakness	3	30%	0	-	3	25%
	Shortness of breath	2	20%	0	-	2	16.6%
	Nausea	1	10%	0	-	1	8.3%
	Chill	1	10%	0		1	8.3%
	Diarrhea	0	-	2	100%	2	16.6%
	All symptoms	28	43%	5	3.6%	33	16.33%
	Headache	14	50%	0	-	14	42.4%
	Cough	12	42.9%	2	40%	14	42.4%
	Fever	7	28.6%	5	100%	12	36.3%
	Sore throat	5	17.9%	0	-	5	15.1%
	Weakness	3	14.3%	0	-	3	9.1%
	Muscular pain	2	7.1%	0	-	2	6.0%
	Shortness of breath	2	7.1%	0	-	2	6.0%
	Nausea	1	3.6%	0	-	1	3.0%
	Chill	1	3.6%	0	-	1	3.0%
	Diarrhea	0	-	2	40%	2	6.0%
	Total	65	100%	137	100%	202	100%
Onset of symptoms	No symptoms	37	56.9%	132	96.4%	169	83.7%
	<1 week	9	13.9%	4	2.9%	13	6.4%
	1–4 weeks	8	12.3%	1	0.7%	9	4.5%
	>4 weeks	11	16.9%	0	0	11	5.4%
	Total	65	100%	137	100%	202	100%

We surveyed the relationship between SARS-CoV-2 incidence and age, gender, education, job, and household contact. Only household contact (aOR: 1; P value < 0.001) and education (aOR: 1; P value < 0.001) had a direct impact on SARS-CoV-2 incidence (Table 3).

Table 3.

RT-PCR results of SARS-CoV-2 genome existence in respiratory specimens.

RT-PCR	Child labor group		Comparison group		Total
	Frequency	Percentage	Frequency	Percentage	Frequency	Percentage
Negative	52	80%	129	94.2%	181	89.6%
Positive	12	18.5%	8	5.8%	20	9.9%
Missing data	1	1.5%	0	-	1	0.5%
Total	65	100%	137	100%	202	100%
Adjusted OR	aOR: 3.000 (CI95%: 1.00–7.00) P value: 0.003

Serological results

Of the 65 working children, 13 (20%) were IgG positive, 51 (78.5%) were negative, and 1 (1.5%) was missing data. Of the 137 children in the non-working group, 19 (13.9%) were IgG positive and 118 (86.1%) were seronegative (aOR: 1; CI95%: 0.00–2.00; P value 0.000). Three out of 65 working children and 4 out of 137 in the control group tested positive for both ELISA and PCR tests, while only PCR or ELISA was positive in the other subjects. Three out of 65 working children and 4 out of 137 in the control group were positive for both ELISA and PCR tests. The rest of the population had only one positive test (Table 4).

Table 4.

ELISA results of SARS-CoV-2 antibody existence in blood specimens.

ELISA	Child labor group		Comparison group		Total
	Frequency	Percentage	Frequency	Percentage	Frequency	Percentage
Negative	51	78.5%	118	86.1%	169	83.7%
Positive	13	20%	19	13.9%	32	15.8%
Missing data	1	1.5%	0	-	1	0.5%
Total	65	100%	137	100%	202	100%
Adjusted OR	aOR: 1.000 (CI95%: 0.00–2.00) P value: 0.000

We also assessed IgM presence in samples that had positive PCR results while being IgG negative. There were 12 samples with such conditions in both groups, all of them were tested negative for IgM (8 in working children and 4 in non-working children). We have tested the IgM to trace someone who is PCR positive and its IgG in the negative. All children were asymptomatic and were IgM negative.

Discussion

Findings and the existing literature

Our study included 65 working children as a target group and 137 unemployed children as a comparison group. Both of our serological and molecular results showed that children in child labor are at higher risks of COVID-19 compared to non-working children: 18.5% vs. 5.8% in molecular and 20% vs. 13.9% in serological assay. This signifies the importance of monitoring working children for COVID-19 and identifying the infected ones for public health policies.

Utilizing RT-PCR method may introduce the risk of false-negative results caused by multiple possible factors including PCR reagents from distinct sources, overall quality of specimen collection, virus load fluctuations in different SARS-CoV-2 infection phases, and the multiple steps of RNA preparation. Furthermore, RT-PCR is unable to detect past infections of SARS-CoV-2 [7]; thus, we evaluated antibody presence in children’s serological specimens.

A retrospective study in China demonstrated that among 366 pediatric patients, 6 (1.6%) were infected with SARS-CoV-2. The average age of patients was 3 years old, and all of them were infected by their parents. Fever, dry cough, and vomiting were the common symptoms among them [22].

In another research, 319 pediatric patients ranging from 14 days to 18 years old were tested for SARS-CoV-2 by PCR, out of which 15 (4.7%) cases were positive (8 boys and 7 girls). The average age of COVID-19 patients was 10.5 years old. In this study, all of the patients had household contact. Among them, 33.3% of the children had no symptoms and were tested solely because of their history of household contact. In other cases, clinical symptoms ranged from mild to moderate including fever, cough, diarrhea, vomiting, runny nose, and difficulty of breathing [23]. In our study, 0% of the children in child labor and 50% of the non-working children were asymptomatic, which suggests the important role of children in the spread of SARS-CoV-2. Our study also revealed that two (16.7%) of the children in child labor who were tested positive for SARS-CoV-2 had a history of household contact.

A case report in Iran, which included nine children aged 2–10 years old who were hospitalized due to COVID-19, reported fever, cough, chills, myalgia, weakness, retraction, tachypnea, and crackle as common symptoms. None of these patients had underlying diseases, and all of them had household contact prior to infection [24].

The recent pandemic caused by SARS-CoV-2 virus is a threat to the livelihood of vulnerable and impoverished demographics. The shutdown of businesses and national economies has caused a major financial crisis for the lower-income families. Children in particular, rather than contamination, are at higher risks of maltreatment and neglect. The extra financial pressure will force parents to send their children to work, so they add extra income to the family and help them meet their basic needs. Becker suggested that COVID-19 effects may cause a general drop in family income through parental unemployment, sickness, or death [25,26].

One study demonstrated that once potential endogeneity is allowed in the bivariate probability model framework, there is statistically a positive association between the children in child labor in Bangladesh and the probability of reporting any sort of exhaustion, physical injury, or health issues. Results in rural children were even stronger. When extended to the correlation of total working hours and the probability of reporting any injury or illness, their method suggested a non-linear relationship between children’s health status and the amount of working hours [27]. Unhealthy conditions in workplace may also increase the children’s vulnerability to both mental and physical health deterioration. However, our results showed no direct relationship between the type of work, number of working hours, and SARS-CoV-2 infection, although this may be due to our limited number of participants.

In another study, of the 127 evaluated children, 80.3% of the injuries occurred in urban areas. Most injured patients were working to supplement their family income or pay their family’s debts. The most common place of injury was industrial workrooms, and the most commonly reported injuries were cuts. None of the reported patients were using preventable devices when they were injured [28].

Many studies conducted over different parts of the world showed a negative correlation between working children and their general health conditions. For example, a study conducted in India reported that the children in child labor suffer from gastrointestinal tract infections, respiratory tract infections, anemia, skin diseases, vitamin deficiencies, and a high prevalence of malnutrition [29]. Likewise, our study demonstrated that based on their BMI, 75.8% of working children suffered from malnutrition.

Another study showed that the prevalence of influenza in child labor is higher than in unemployed children [18], similar to our conclusion that SARS-CoV-2 as a respiratory infection has a higher prevalence among working children. According to our findings, this is the first original research about SARS-CoV-2 prevalence in working children.

Limitations

In this study, we only had access to children in child labor who were supported by an NGO organization and we had a limited timescale (September until October).

Future directions

Further studies, on a larger scale of both time and population, are required for a better in-depth analysis of working children who are employed in various fields of work and different living environments, for example, public shelters, the total number of their working hours, and the prevalence of COVID-19.

Conclusion

The analytical results of this study showed that the prevalence of COVID-19 in employed children was almost three times higher than in unemployed children. The results from various studies demonstrated the occurrence of asymptomatic or mild SARS-CoV-2 infection in children while having equal viral loads as adults. Therefore, monitoring employed children for SARS-CoV-2, providing them with personal protective equipment for COVID-19, and placing them in priority for vaccination might be helpful against the fast spread of the virus and contribute to further success and effectiveness of social health policies against COVID-19.

Funding Statement

This study was funded by Tarbiat Modares University, Faculty of Medical Sciences (grant number: Med-86143). We thank them for their financial support and for providing assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).