Vulnerabilities and risk perceptions of contracting COVID-19 among Nepali migrant workers

Social, economic, and demographic characteristics influence public disaster risk perception, including the risk of COVID-19. Migrant workers are one the most vulnerable groups to disasters. More than four million Nepali migrant workers are employed abroad, and millions are working in cities and towns in the country. This study analyzes how the social, economic, and demographic conditions of returning Nepali migrant workers determine their risk perceptions of COVID-19. An online national survey was administered from May 10 to July 30, 2020, targeting returning Nepali migrant workers nationwide. A total of 782 responses from migrant workers were recorded, covering 67 of 74 districts. Using descriptive statistics and binary logistic regression models, the results show that migrant workers in blue-collar jobs, female, older than 29 years, with pre-existing health conditions, from low-income families, and from larger families are more likely to perceive higher risks of COVID-19. The migrant workers who believe in non-pharmaceutical COVID-19 control measures, including awareness campaigns and stay-at-home orders, have higher risk perceptions of the virus than other groups. The research contributes to identifying the program and policy priority areas to address the needs and COVID-19 vulnerabilities of returning Nepali migrant workers during and after the pandemic.


Introduction
Risk perception is a significant component of the entire emergency management field as it heavily influences individual preparedness and responses to hazards. This concept refers to the public's interpretations of the world and how they evaluate risks (Cori, Bianchi, Cadum, & Anthonj, 2020). A myriad of studies has been conducted to better understand how risks are perceived for specific types of hazards. For example, Ho et al. (Ho, Shaw, Lin, & Chiu, 2008) examined if risk perception is influenced by disaster types and victim characteristics. They found that victims' sense of controllability and perceived impacts were high for landslide events but not for floods (Ho et al., 2008). The risk perception literature has also demonstrated a strong emphasis on the technological hazard of nuclear power. Several studies have illustrated how risk perception leads to public opposition in siting nuclear power plants near an individual's residence (He, Chen, Kong, & Liu, 2020;Kilinic et al., 1007;Lindell & Earle, 1983). On the other hand, additional studies have concluded that residents living within close proximity of a nuclear power plant overall perceive the safety risks to be low (Crowley, 2021;Pidgeon, Lorenzoni, & Poortinga, 2008;Venables, Pidgeon, Parkhill, Henwood, & Simmons, 2012).

Analytical frameworkpandemic risk perception and vulnerabilities
The Coronavirus Disease of 2019  has presented the world with a diverse array of challenges regarding mitigation, preparedness, response, and recovery. Mitigation techniques have emerged as some of the most critically important strategies for slowing the spread of the virus (Parodi & Liu, 2020). However, past studies have concluded that the extent to which individuals adopt COVID-19 mitigation techniques is heavily influenced by their risk perceptions of contracting the virus as well as the consequences if they become infected (Figueiras, Dias Neto, Maroco, & Kern de Castro, 2022;Heydari et al., 2021;Lu, Kong, & Shelley, 2021;Siegrist, Luchsinger, & Bearth, 2021;Yuanchao, Linxiu, & Sun, 2021). This strong and positive association between adopting mitigation techniques to reduce the spread of COVID-19 and individual risk perceptions has led to a considerable number of studies examining specific factors that influence one's risk perceptions of the virus. These factors include trust in the media, pre-existing health conditions, employment status, employment type, family size, income, age, and gender.
One of these factors pertains to the media. Ye and Lyu (Ye & Lyu, 2020) found that public trust in the media helps to reduce the infection rates of diseases. More specifically, Zanin et al. (Zanin, Gentile, Parisi, & Spasiano, 2020) describe the significant influence of the mass media on both respondents' knowledge of and feelings towards COVID-19. They argue that better awareness of the virus likely leads people to be more uncertain and subsequently accept stricter mitigation measures. At the same time, they caution that the distribution of fake news has the potential to result in incorrect perceptions of the risk of contracting COVID-19 (Zanin et al., 2020).
The literature demonstrates mixed findings pertaining to the impacts of pre-existing health conditions on individual risk perceptions of the pandemic. Studies conducted by He et al. (He et al., 2020), Laires et al. (Laires et al., 2021), and Lu, Kong and Shelley (Lu et al., 2021) found there to be positive associations between the presence of pre-existing health conditions and higher risk perceptions of COVID-19. Furthermore, Siegrist, Luchsinger, and Bearth (Siegrist et al., 2021) identified health risk perception to be the strongest predictor of an individual's acceptance of COVID-19 mitigation practices. This was because participants with higher health risk perceptions demonstrated more hygienic behavior and interacted with fewer people compared to participants with lower health risk perceptions (Siegrist et al., 2021). However, other studies contradict these findings. For example, Greathouse et al. (Greathouse, Nagia, Rayyan, & Bailey, 2021) observed that the presence of a pre-existing health condition did not appear to impact behaviors or concerns related to the virus. They attribute this to a lack of patient education regarding how their pre-existing conditions increase their risk of complications from COVID-19 (Greathouse et al., 2021). Moreover, Yadav et al. (Yadav et al., 2021) concluded that pre-existing health conditions were inversely related to fear.
Employment status was also viewed in the literature as having a significant impact on risk perceptions, but like the presence of preexisting health conditions, the findings were not consistent. A study conducted by Asnakew et al. (Asnakew, Asrese, & Andualem, 2020) suggests that employed individuals are more likely to perceive themselves as being susceptible to COVID-19 than those who are unemployed. But other studies argue that individuals' risk perceptions of the virus vary based on their employment type. An and Sun  examined the impact of farmers' risk perceptions of the virus on working outside their hometowns in China. They found that farmers generally believe that the risk of contracting the virus is higher in urban areas than in rural areas and that farmers' risk perceptions heavily influence their decisions to work outside of their hometowns . On another note, Han et al. (Han et al., 2021a) studied the relationship between turnover among airline workers and perceived risks of contracting the virus and experiencing job instability during the pandemic. They concluded that airline employees' perceived risks of virus infections and job instability significantly intensified the relationships among attitude, commitment, and employee career turnover intentions (Han et al., 2021a).
In contrast to the presence of pre-existing health conditions and employment status, the literature appears to have reached a consensus regarding the impacts of family size on risk perceptions of contracting COVID-19. Several studies assert that an increase in family size is positively associated with an increase in risk perceptions pertaining to the virus [ 20,23,24]. Income was another demographic variable that displayed an overall consistency for the findings on risk perceptions, as several studies suggest that individuals with higher incomes are likely to have higher risk perceptions of contracting COVID-19 (Asnakew et al., 2020;He, Mol, Zhang, & Lu, 2013;Irigoyen-Camacho et al., 2020). However, Ruengorn et al. (Ruengorn, Awiphan, Wongpakaran, Wongpakaran, & Nochaiwong, 2021) observed that individuals with a monthly income loss of 50 percent during the COVID-19 pandemic faced a higher risk of anxiety.
Studies on the roles of age and gender in risk perceptions of the demographic variables presented mixed findings. While a considerable amount of studies found that risk perceptions of contracting COVID-19 increased with age (Asefa, Qanche, Hailemariam, Dhungana, & Nigussie, 2020;Asnakew et al., 2020;ernandez-Castillo et al., 2021;He et al., 2020;Laires et al., 2021;Lu et al., 2021;Yadav et al., 2021), other studies concluded that an increase in age was associated with a lower risk perception of contracting the virus, but a higher risk perception of experiencing complications (de Bruin, 2020;Rosi et al., 2021). Similar findings were displayed for the relationship between gender and risk perception, as the majority of studies found that females perceived the risk of contracting COVID-19 to be higher than males (Asnakew et al., 2020;ernandez-Castillo et al., 2021;He et al., 2020;Lu et al., 2021;Siegrist et al., 2021), but Karasneh et al. (Karasneh et al., 2021) found males to have a higher risk perception than females. Furthermore, Sledge and Thomas (Sledge & Thomas, 2021) assert that males are at a higher risk of dying from COVID-19 than females, but females view both state and non-state actors as more important for responding to the public health crisis than males do.

Nepali migrant workers during COVID-19
A significant number of the Nepali working population migrate abroad in search of alternative livelihood and employment. Due to political instability and stalled economic growth, Nepali labor migration has increased steadily in recent years. In fiscal years 2019/20 and 2021/ 22, more than 1.1 million labor approvals were issued by the government, allowing Nepali citizens to leave the country for employment (MoLSS (Ministry of Labor and Employment and Social Security), 2022). Remittance flow into the country increased in 2021 compared to 2020, reaching USD 8.2 billion in the fiscal year 2021 (World Bank, 2022). It was USD 7.5 billion in the fiscal year 2020 (MoLSS (Ministry of Labor and Employment and Social Security), 2022). It is estimated that remittances account for about 25 percent of Nepal's Gross Domestic Product (Paoletti, Taylor-Nicholson, Sijapati, & Farbenblum, 2014).
These migrant workers faced disproportionate impacts of the virus due to poor working and living conditions and the inaccessibility of medical and health services in Nepal and the destination countries. The implementation of a government-issued lockdown in 2020 resulted in hundreds of Nepali migrant workers being stranded at the Nepal-India border and kept in quarantine centers (Aacharya & Shah, 2020). Moreover, the quarantine centers built by the Nepali government were quickly overwhelmed by the large number of migrant workers returning to Nepal during multiple waves of COVID-19 outbreaks and thus presented sanitation concerns due to overcrowding . In addition, the conditions of these quarantine centers brought psychological distress, as BC et al. (BC et al., 2021) conclude that the centers in Nepal's Karnali province impacted the mental health of the occupants, resulting in high rates of depression and anxiety.
Despite growing remittance and outgoing numbers, the COVID-19 pandemic has resulted in financial consequences for Nepali migrant workers. The personal economic impacts differed between international and internal Nepali migrant workers (Ghimire, Nepal, Crowley, Ghimire, & Guragain, 2022). Ghimire et al. (Ghimire et al., 2022) found that internal Nepali migrant workers were almost 7 times more likely to plan to return to their current jobs following the lockdown and were 63% less likely to report that someone in their family lost a job due to the pandemic compared to international Nepali migrant workers.

Justification for research
Studies have recently been conducted to better understand the risk perceptions of contracting COVID-19 among migrant workers. One of these was put forth by Khanal et al. (Khanal, Teijlingen, Sharma, Acharya, & Sharma, 2021), who examined the relationship between personal characteristics and perceived threats pertaining to COVID-19 among Nepali migrant workers. Their findings suggest that there is no relationship between participants' demographic variables and their perceived susceptibility, but age, caste, and educational status affect participants' perceived levels of severity. However, their research focused exclusively on Nepali migrant workers who returned from India (Khanal et al., 2021). Given the importance of the role of risk perceptions in influencing COVID-19 mitigation actions and the vulnerability of migrant workers to acquiring and spreading the virus, there is a need to better understand these relationships through a larger scope. The main objective of this study is to address the research gap by examining COVID-19 risk perceptions among returned Nepali migrant workers from all over the world. The variables that were examined consist of the factors that were discussed in the analytical framework as frequently impacting one's risk perceptions of COVID-19.

Data collection
A national survey was designed and administered to better understand the COVID-19 risk perception of returning Nepali migrant workers from abroad during the first outbreak in the country. The survey focused on the social, economic, health, and psychological impacts of COVID-19 among migrant workers. It had five sections: awareness and perception of the virus, health impacts, perceptions of non-pharmaceutical mitigation measures and the virus, livelihood and employment impacts, and demographic information of respondents. The study was reviewed and approved by the Office of Research Compliance, Human Studies Program of the University of Hawaii with approval number 2020-00409, and pre-tested among experts in Nepal for clarity and comprehensiveness of the questionnaire.
The survey was distributed from May 10 to July 30, 2020. The Government of Nepal imposed a nationwide lockdown and stay-at-home order from April to July 2020. During this lockdown, Nepali migrant workers from India and other countries returned to their hometowns at the highest rates. The internal migrant workers also returned to their villages from cities and other geographic locations due to the shutting down of factories, offices, schools, transportation services, and businesses. Using purposive and convenient sampling approaches, the survey was distributed through the authors' personal and professional networks nationwide, focusing on the checkpoints and provincial and local quarantine centers. For the returning migrant workers from India, national, provincial, and local governments established quarantine centers nearby the entry points. For the migrant workers returning from third countries, the government has arranged quarantine centers in Kathmandu and transportation services to deliver them to their villages for isolation. Key local contacts were established to reach out to the returning migrant workers from India through land border checkpoints. They were in Provinces 1, 5, and Far-Western. Local contacts mobilized enumerators with strict COVID-19 social distancing and personal protection guidelines to reach out to the returning migrant workers in the quarantine centers and transit locations. They also conducted the survey over the phone with returned workers.
A total of 1107 responses were recorded. Among them, 782 responses were from migrant workers included in this analysis. The remaining 325 responses were from people either working locally or unemployed due to various reasons. Fig. 1 shows survey responses by the provinces of Nepal. Response ranges vary from below 36 to more than 165. The map also shows the number of international migrant workers by province. Gandaki Province had almost the same number of transnational migrant workers as Bagmati Province in 2019. The responses to the survey were below 36 for Gandaki. This is because of the challenges of reaching migrant workers during the survey administration. The geographic remoteness and the enforcement of stay-at-home orders during the first COVID-19 lockdown added difficulties in reaching them. The highest responses were received from Bagmati and No. 5 provinces. Only one international airport in the country is in Bagmati Province, and the authors focused on returning migrant workers by international flights during the initial global outbreak of COVID-19. The No. 5 Province had multiple checkpoints to screen returning migrant workers from India. Both provinces were targeted to increase responses from returning international Nepali migrant workers. There were 56-165 responses from Far-Western and No. 1 provinces and 37-55 from the Madhesh Province.

Variable selection
The survey had questions about the COVID-19 risk perception, nonpharmaceutical mitigation measures, socioeconomic impacts of the pandemic mitigation, and demographic characteristics. One of the questions in the survey was, 'Do you believe that there is a possibility of getting yourself or your family member infected from the  showing the respondents' COVID-19 risk perception. Previous studies have shown that risk perception of disasters is determined by socioeconomic characteristics and a person's exposure to the information about the impacts and lived experiences. Table 1 shows the variables that explain the socioeconomic characteristics, demographic data, and the respondents' awareness of COVID-19 mitigation measures. For the economic impacts, the survey included questions about respondents' occupations before the COVID-19 lockdown, their plans to return to the same employment, and household income. The survey included respondents' gender, age, household size, and family members with preexisting medical conditions. For COVID-19 non-pharmaceutical mitigation measures, the survey included questions focusing on respondents' beliefs on the effectiveness of COVID-19 awareness campaigns and stay-at-home orders.

Data analysis
A binary logistic regression model was used to analyze the survey data (Allison, 2012) because the likelihood of COVID-19 infection to an individual was measured in a binary manner (Yes/No question). The regression model to test the risk perception was built as shown below (Allison, 2012;Kim, Yamashita, & Ghimire, 2021): Where, Y is the risk perception variable with two possible values denoted by 1 and 0, and x is the vector of the explanatory variables, including the socioeconomic characteristics of respondents. A β 0 is the intercept parameter, and β is the vector of slope parameters. Explanatory vari-ables are types of jobs, having chronic medical conditions, faith in nonpharmaceutical measures to curve the spread, family size, age, poverty status, and gender.

Results
The survey was focused on socioeconomic characteristics, pandemic experiences, and beliefs on non-pharmaceutical mitigation measures. Table 1 shows the frequencies of responses to different questions related to these domains. The majority of migrant workers (79.81 percent) who responded to the survey were involved in occupations other than bluecollar. Most of them were engaged in temporary occupations. Out of 782 respondents to the survey, about 79 percent said they plan to return to the same occupation after the end of the stay-at-home order.
About a quarter, i.e., 28.77 percent of the migrant workers, mentioned that they have at least one family member with chronic medical conditions, making them more sensitive to COVID-19-related infections. Most survey participants (51.41 percent) believed that they or their family members were more likely to get infected by the virus. In contrast, only 41.94 percent of 782 respondents believe that COVID-19 awareness campaigns can effectively control the virus. But their belief in the government's lockdown and mobility control is high. A total of 54.87 percent of the migrant workers who responded to the survey mentioned that lockdown and stay-at-home orders reduce the COVID-19 spread.
Demographic characteristics in the survey were household size, household income, respondent age, and gender. The majority of migrant workers who participated in this study have large family sizes. Out of 782 respondents, only 44.88 percent have 4 or fewer members in their households. More than 8 percent of the respondents have more than 10 in their family, and around 20 percent of total respondents have 6-10 members in their family. More than 78 percent of the respondents are male, matching the higher number of male migrant workers. The majority of respondent migrant workers were relatively poor. The highest number of respondents (26.84 percent) were making 100,000 or fewer Nepali rupees per year, which is USD 837.52 (Exchange rate: USD 1 = Rs.119.4, 08/01/2020). Out of 782 total, 15.23 percent, 14.60 percent, 14.76 percent, and 8.48 percent of the respondents said that their families make 100000 to 200000, 200000 to 300000, 300000 to 400000, and 400000 to 500000 Nepali rupees, respectively. About 20 percent of the respondents mentioned that their family income is more than 500000 Nepali rupees. The high fluctuations in household incomes are due to the large income gap between domestic and international migrant workers.
There is a strong association between socioeconomic characteristics and risk perception among returning migrant workers. Details are presented in Table 2. The risk perception was measured by the perception of the likelihood of catching COVID-19. The regression model shows that the migrant workers working as daily wage laborers perceive that they are 1.81 times more likely to get infected with COVID-19 than other groups (OR = 1.82, 95 percent, CI = 1.15, 2.87, p < .01). Similarly, returning migrant workers are 44 percent less likely to return to their pre-pandemic occupations than other groups (OR = 0.562, 95 percent, CI = 1.37, 0.86, p < .0076), and they are 1.72 times more likely to have a family member of self with a chronic medical condition (OR = 1.72, 95 percent, CI = 1.23, 2.40, p < .02). But, these returning migrant workers are 1.50 times more likely to believe that public awareness campaigns against COVID-19 are successful in stopping the virus (OR = 1.50, 95 percent, CI = 1.1.09, 2.06, p < .012), and they also 1.48 times more likely to believe on the success of the stay-at-home order to control the virus (OR = 1.48, 95 percent, CI = 1.072, 2.05, p < .016). Returning migrant workers are 1.39 times more likely to have a family size 5 or higher (OR = 1.39, 95 percent, CI = 1.02, 1.91, p < .04) and 1.75 times more likely to be female (OR = 1.7, 95 percent, CI = 1.18, 2.61, p < .01). They are 29 percent less likely to be aged 29 years or younger (OR = 0.71, 95 percent, CI = 0.52, 0.97, p < .03), and they are 1.12 times more likely to have a family income below the poverty line (OR = 1.12, 95 percent, CI = 1.03, 1.22, p < .01).

Findings and discussion
Disaster risk perception varies based on socioeconomic characteristics (Agrawal, 2018). Job types, poverty status, age, and gender influenced COVID-19 risk perception among returning migrant workers. An important finding from this study is that these returning migrant workers believe that non-pharmaceutical measures, such as stay-at-home orders and awareness campaigns against the virus, control the spread of COVID-19.
COVID-19 has exposed the pre-existing vulnerabilities of migrant workers. They always live in a sub-optimal housing environment in their job locations due to exclusionary regulations and policies (Hajariya, 2016;Zhu, Zhao, Wang, & Yammahi, 2017). Studies show that they live in crowded settings with poor health and hygiene services (Derose, Escarce, & Lurie, 2007;ILO (International Labor Organization), 2020). In such an environment, communicable diseases and pandemics are more likely to spread. In the case of international migrant workers, hiring companies provide housing, which is crowded, and multiple individuals share living spaces. Due to this, they are more vulnerable to disasters, including pandemics. Their higher COVID-19 risks and vulnerabilities are shown by multiple outbreaks of infection in labor camps in different employment-providing nations for Nepali migrant workers, including UAE, Saudi Arabia, Qatar, Singapore, Malaysia, and others (ILO (International Labor Organization), 2020). Within the country, migrant workers live in crowded settings in Kathmandu and other cities. Non-specialized working laborers are less likely to afford housing in the urban setting and are not trusted by the landlords to rent apartments. They come to urban areas to earn cash and support their families in their hometowns. Their employment is seasonal, and they are wage laborers. To save money, they are forced to live in crowded housing (ILO (International Labor Organization), 2020). Higher COVID-19 risk perception among them shows underlying conditions and vulnerabilities of migrant workers to COVID-19 and communicable diseases in their work environments.
Due to the uncertainty of the economy, large employers, such as construction companies, shopping malls, and house help providers, are not likely to return to normalcy soon (Times, 2020). This situation has implications for the employment of migrant workers. Most of these migrant workers are employed in non-specialized occupations (ILO (International Labor Organization), 2020) and lost their employment at the beginning of stay-at-home orders and city lockdowns. They are uncertain when and how they will return to employment. In contrast, these migrant workers do not have enough savings to support their families during prolonged stay-at-home orders and lockdowns.
Nepali migrant workers are sensitive to disruptions and changes in the economy. Due to limited economic resources, they have to continue working to sustain their families regardless of physical and mental risks in jobs (Agrawal, 2018). Their economic vulnerability to COVID-19 is higher compared to other groups. The survey results showed that they are more likely to have a family member with chronic medical conditions. The COVID-19 trend has shown that individuals with pre-existing medical conditions are more likely to have severe infections from the virus (Treskova-Schwarzbach et al., 2021).
In Nepal, more than 66 percent of COVID-related deaths are among seniors and individuals with non-communicable diseases (MHP (Ministry of Health and Population), 2019). They are mainly pre-existing chronic conditions. The survey results have shown that Nepali migrant workers are more likely to have family members with chronic medical conditions. The vulnerability of migrant workers to the virus is more severe than other groups due to their poor economic status and higher possibility of getting infected by COVID-19.
Due to the nature of employment, living conditions, and economic situations, these migrant groups are one of the most vulnerable  Despite their vulnerability and higher COVID-19 risk perceptions, Nepali migrant workers have higher trust in government campaigns to curb the spread. The survey results showed that they are more likely to believe in awareness campaigns and stay-at-home orders to control COVID-19. This finding contradicts many trends of low trust in governments to contain the virus (Han et al., 2011(Han et al., , 2021b. Being one of the most vulnerable groups to COVID-19 and other disasters, having higher trust in government interventions could be a niche opportunity to reduce travel-related spread.
On demographic characteristics, migrant workers are more likely to have larger families and have a higher risk perception of COVID-19. These workers are older and have family members with pre-existing medical conditions. Due to their mobility, migrant workers are more likely to transmit the disease from one place to another through travel. If caught up with the virus, they are more likely to infect other people in their families.
Gender is another important social determinant of higher COVID-19 risk perceptions. Female Nepali migrant workers are more likely to perceive higher infection risks than males. Nepalese females are responsible for taking care of families (Thieme et al., 2005). Females are also more likely to have a higher risk perception of disasters than males (Agrawal, 2018).
Migrant workers are more likely to be older-aged compared to other employment groups. These migrant workers are continuously working during the pandemic. They are ready to take jobs with higher exposure to the virus because they are from relatively low-income families. They are more likely to borrow money from financial institutions and individual lenders to cover their travel expenses to their employment locations.
COVID-19 risk perceptions among Nepali migrant workers are influenced by the nature of the job, health condition, family size, age, poverty status, and gender. This finding is not unique from other studies on disaster risk perception. But this study unveils the relationship between the higher COVID-19 risk perception of migrant workers and the incubating vulnerabilities caused by the work environment, housing conditions, service delivery, systemic ignorance, and continued low social and economic statuses (Hajariya, 2016;Kim et al., 2021). Targeting this group for non-pharmaceutical interventions and vaccination will be an effective strategy to control the spread and infections. During the pandemic, the general public perceives government initiatives as less effective. Still, Nepali migrant workers believe that government non-pharmaceutical measures, such as stay-at-home orders and awareness campaigns, are more effective in addressing the virus's risk.

Conclusion
Migrant workers are one of the most vulnerable groups to disasters. The COVID-19 pandemic unveils their vulnerabilities. Nature of employment, health condition, family characteristics, age, poverty status, and gender influenced the COVID-19 risk perception among Nepali migrant workers. Blue-collar workers, who are older, female, with larger family sizes, and families living below the poverty line, are more likely to perceive a higher risk of COVID-19. This finding matches the conclusions of the previous research that high vulnerability correlates with higher risk perceptions of disasters (Agrawal, 2018). In the case of Nepali migrant workers, the combination of social and economic characteristics explored in this study makes them highly vulnerable to COVID-19.
Migrant workers are mobile. They travel from their hometown to the cities and abroad in the course of employment. They also return to their hometown to meet their families and relatives regularly. As a result, their situation exposed them to COVID-19, and they are likely to be vectors of bringing the disease to their families, friends, and hometowns. One critical finding from this study is that these Nepali migrant workers believe in government interventions to curve the COVID-19 spread.
Public health planners and policymakers should target migrant workers to strengthen non-pharmaceutical control measures to contain the spread. Measures should be initiated as soon as these migrant workers register at Nepal's southern border checkpoints and the international airports in the country.
Most Nepali migrant workers are less likely to return to prepandemic employment, and the majority of them are blue-collar workers who perceive higher risks of COVID-19. Studies have shown they are interested in staying in their villages and towns to continue agriculture and other self-employment opportunities. This is a great opportunity to revive agriculture with added human resources, leading the country to food self-sufficiency. Central, Provincial and Local Governments should design programs to address the needs of these workers to keep them in their villages and towns.
Targeted vaccination programs should be introduced for migrant workers and their families to reduce COVID-19 vulnerabilities among them. This research finds that Nepali migrant workers are vulnerable to COVID-19 due to their health conditions. Workers with higher vulnerabilities perceive higher risks from COVID-19. In addition, they can be vectors to bring the disease from one place to another through their travels. Increasing vaccination accessibility among them can reduce the virus spread and infections.
This research was conducted during the first stay-at-home order and lockdown in Nepal when there was very little understanding of the virus. Social stigma against migrant workers was paramount. There was no medical option to cure the virus. As a result, the data gathered at that time might not offer a complete picture of the COVID-19 risk perception among migrant workers. Similarly, this study does not consider other demographic and contextual variables, such as the country's caste system, which could have a crucial role in determining the COVID-19 risk perception.
This study did not account for the destination countries and the length of stay for these migrant workers to analyze the COVID-19 risk perception. Nepali migrant workers staying in the Middle East may have different risk perceptions of the pandemic compared to those working in India because their work environments and social settings are different. Further research should analyze the pandemic risk perception of Nepali migrant workers based on their destination geographies and length of stay abroad.
Future research should also focus on how the COVID-19 risk perception changed among Nepali migrant workers after introducing the vaccine and other medical remedies. Additional research is necessary to explore policy alternatives to reduce disaster vulnerabilities among Nepali migrant workers in their work locations and hometowns.

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 paper.