Precarious employment and migrant workers’ mental health: a systematic review of quantitative and qualitative studies

Objectives Evidence suggests that precarious employment can have detrimental effects on workers’ health, including mental health. Migrant workers are discussed to be especially vulnerable to such effects. Thus, we systematically reviewed existing research on the association between precarious employment and migrant workers’ mental health. Methods Three electronic databases (Web of Science, PsycINFO and PubMed/Medline) were searched for original articles on quantitative and qualitative studies published from January 1970 to February 2022 in English, German, Turkish and Spanish. Multiple dimensions of precarious employment were considered as exposure, with mental health problems as outcomes. Narrative synthesis and thematic analyses were performed to summarize the findings of the included studies along with risk of bias and quality assessment. Results The literature search resulted in 1557 original articles, 66 of which met the inclusion criteria – 43 were of high quality and 22 were of moderate quality. The most common exposure dimensions analyzed in the studies included temporariness, vulnerability, poor interpersonal relationships, disempowerment, lacking workers’ rights and low income. The outcome measures included stress, depression, anxiety and poor general mental health. The prevalence of these outcomes varied between 10–75% among the included quantitative studies. All qualitative studies reported one or more dimensions of precarious employment as an underlying factor of the development of mental health problems among migrants. Of 33 quantitative studies, 23 reported evidence for an association between dimensions of precarious employment and mental health. Conclusion The results of this review support the hypothesis that precarious employment is associated with migrant workers’ mental health.

Globally, the number of migrants is increasing steadily due to climatic disasters, globalization, socio-political pressures and conflicts (1,2). According to the International Labor Organization (ILO), there are approximately 258 million international migrants in the world, 164 million of whom are workers. About two-thirds of these workers live in high-income countries where they are an important labor resource (3). However, migrant workers tend to contribute to the labor force in their host countries through precarious employment, which is characterized by unfavorable work conditions. These conditions include job insecurity, low income, lack of worker rights and protection, lack of power to exercise rights, work with temporary or no contracts, participation in multiple part-time jobs, lack of employment compensation, unpredictable working schedules, and exposure to unfair and authoritarian treatment (4)(5)(6)(7)(8)(9)(10).
Precarious employment has become more prominent recently, especially in high-income countries (4,9). Moreover, the impact of the COVID-19 pandemic has the potential to further increase the prevalence of precarious working conditions (11,12). Apart from that, there is evidence that precarious employment can have negative effects on workers' quality of life and well-being (6), ultimately leading to health problems (13)(14)(15)(16). For example, some studies indicated an asso-ciation between precarious employment and physical health problems such as musculoskeletal issues (17), cardiovascular diseases (18), occupational accidents and injury (19,20). Results of other studies also show that precarious work conditions have an influence on mental health problems such as depression (21), anxiety and stress (22), suicidal ideation (23,24), sleep issues and burnout (24)(25)(26). Accordingly, there are a number of systematic reviews that have synthesized the evidence for detrimental mental health effects of different dimensions of precarious employment (5,9,(27)(28)(29)(30)(31).
However, none of the reviews so far specifically focused on migrant workers. Migrants are especially vulnerable and likely to be exposed to precarious employment because of language barriers, employer prejudice and discrimination, lack of professional networks, lack of sufficient knowledge related to health and the labor system (32)(33)(34) and poor social support (35,36). In addition, migrant workers have been reported to be exposed to various forms of harassment at the workplace: prejudices by employers and workmates (33), unfair treatment and discrimination (37)(38)(39) or being forced to work or perform tasks that are incompatible with their contracts at the threat of deportation (37,(40)(41)(42). This situation is often worse in the case of undocumented migrants who are even more vulnerable and disadvantaged due to a lack of work or residency permits. As a consequence, they are at risk from being exploited by their employers (38,40,41,43,44). Moreover, there is considerable evidence that migrants differ from non-migrants in characteristics such as general health status, access to healthcare or health-related risk factors (45,46).
Because of those peculiarities, studies examining precarious employment and its putative health effects in the general population cannot be automatically and directly transferred to migrant populations. It is thus crucial to comprehensively understand and systematically evaluate migrant workers' experiences under precarious work conditions and such conditions' effect on migrant workers' mental health. Apart from a better scientific understanding, such findings may also help to develop migrant worker-friendly occupational health policies in the future. Thus, the main aim of this review of qualitative and quantitative studies was to analyze and summarize existing quantitative and qualitative research on the association between precarious employment and migrant workers' mental health. To achieve this aim, the scientific questions were addressed in quantitative studies: (i) What is the prevalence of precarious employment among migrants? (ii) What is the association between precarious employment and mental health among migrants, including its direction and effect size? and in qualitative studies: (i) What dimensions of precarious employment are migrant workers exposed to?
(ii) What are the mechanisms underlying the relationship between precarious employment and mental health among migrant workers?

Protocol
We described our methods by means of a detailed protocol (47) that was developed in accordance with the Preferred Reporting Items for Systematic Review and Meta-analysis Protocols (PRISMA-P) 2015 guidelines (48), and was registered within the International Prospective Register of Systematic Reviews (PROSPERO) (registration number: CRD42019132560).

Eligibility criteria
The Population, Exposure, Comparator and Outcomes (PECO) framework was used as a guide for the eligibility criteria of the included publications (49). The included papers were original quantitative observational studies and all types of qualitative studies that involved international migrant workers of working age (≥15 years), and published in English, German, Spanish or Turkish in peer-reviewed journals from 1 January 1970 to 14 February 2022. All original studies with 'precarious employment' as an exposure and mental health problems as an outcome were included.

Search strategy and information sources
Keywords were determined on the basis of multidimensional precarious employment definitions, eg, the dimensions of precarious employment identified with the Employment Precariousness Scale (EPRES) (8,50,51) and an earlier reviewer with a similar scope (5) and related to three main headings, namely, 'migrant' (population), 'precarious employment' (exposure) and 'mental health' (outcome). Search strategies were developed with these keywords using the Medical Subject Headings (MeSH) thesaurus and freetext based on the eligibility criteria (supplementary material, www.sjweh. fi/article/4019, table S1). With these keywords, information sources were searched in the following order. First, three electronic databases (Web of Science, PsycINFO and PubMed/Medline) were searched for original observational studies published from January 1970 to February 2022. Then, we hand-searched the reference lists of previously published related systematic reviews. We also hand-searched for relevant studies published in the American Journal of Industrial Medicine, BMC Public Health, Ethnicity & Health, and Gaceta Sanitaria in the last year Koseoglu Ornek et al from which we obtained the largest number of eligible studies in the database search. We included all 2019 issues that were published online in the hand-search. Lastly, the reference lists of all included studies were screened based on the eligibility criteria.

Data collection, selection process and extraction
First, two authors independently evaluated the titles, and abstracts of the identified articles. Secondly, two authors independently assessed the fulltext of all candidate articles. Any disagreements were reconciled by discussion with consultation of a third researcher to build consensus if necessary. A flowchart showing details about the selection process is illustrated in figure  1. Two reviewers independently used a standardized form for data extraction from each included study in the data collection stage. A third reviewer was consulted for resolution in case of differences or disagreement between the reviewers' evaluation. We contacted 12 corresponding authors to obtain certain data from their respective studies such as participants' age or gender distribution if the information was not included in the published manuscript.

Risk of bias and quality assessment
Two reviewers independently assessed the quality of the included qualitative and quantitative studies. The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of the quantitative studies (31). The Critical Appraisal Skills Programme (CASP) was used to assess the qualitative studies. The assessed quality of the included studies was graded according to the total score obtained from NOS and CASP. Discrepancies between the reviewers were discussed until a consensus was achieved.

Data analysis and synthesis
The results of the included studies are reported as a systematic review narrative synthesis. No meta-analysis was conducted due to variations in the included studies' precarious employment dimensions, mental outcomes and measurement methods. Additionally, for studies that did not report confidence intervals (CI), we calculated the 95% CI using RevMan (version 5.4) (52). Thematic analysis was performed to synthesize the qualitative studies (53) using MAXQDA (54). One researcher developed an initial coding frame that a second researcher then checked. Themes and subthemes emerged on the basis of the initial coding. The results of the qualitative and quantitative studies were decided to be combined according to their common features and contents after discussion and analysis among the researchers. According to this decision, a form was developed based on themes and categories. The results of the review presented following the guidelines for systematic reviews and meta-analyses of observational studies (MOOSE) (55).
In total, 65 studies were assessed for quality. About two-thirds of them (26 qualitative, 17 quantitative) were of high quality, while about one-third (4 qualitative, 15 quantitative) were of moderate quality (supplementary tables S2 and S3).

Study characteristics
Three studies (two quantitative, one qualitative) were published in Spanish, the others in English. None of the studies were published in German or Turkish. The vast majority of studies were conducted in North America (43%) or Europe (38%). The youngest participant age in all the studies was 16, and the majority of participants were female workers.
Nineteen of the quantitative studies consisted of a study population that included migrant and non-migrant workers. Two studies involved the same participants, but provided data on different variables (40,79). Most of them were cross-sectional studies. They were published between 1998 and 2022, with more than 70% being published after 2009. All but one quantitative studies collected their data after the 2000s. The General Health Questionnaire-12 (GHQ-12) was commonly used to evaluate mental health. Aside from the Employment Precariousness Scale, which was used in three studies (77,78,80), there was no specific tool for assessing precarious work conditions (table 1).
The qualitative studies were published in 2003 to 2022, but the underlying data were collected from 1998 to 2019, and most of these papers (73%) were published after 2011 (table 2). Except for one paper, the data from all the qualitative studies were collected in 2006 and afterwards. Semi-structured questionnaire forms were generally used for data collection in focus group discussions or individual interviews. Purposive sampling was the preferred data collection method.

Prevalence of precarious employment and mental health problems
The majority of the quantitative studies reported more than one precarious work condition; 26 reported the prevalence of precarious employment and 28 provided an estimate for the prevalence of mental health problems among their participants. In some quantitative studies, the majority of participants were exposed to one of the dimensions of precarious work: non-permanent contracts (84%) (82), low-income (73%) (37), lack of health insurance (58%) (82), unfair treatment (54%) and job insecurity (63%) (57), precariousness (67%) (80).
Seven themes emerged from the codes of the qualitative studies: mental health and six other themes representing dimensions of precarious employment. All precarious work conditions and mental health problems reported in the quantitative studies likewise fit into the investigated Koseoglu Ornek et al

Precarious employment and its association with mental health
Some quantitative studies provided just the frequency of exposure, while others presented estimates for the association between exposure and mental health. The main dimensions of precarious employment and their associations with mental health problems are summarized in table 1. Twenty-three quantitative studies reported effect estimates that provided evidence for an association between the respective dimension of precariousness and poor mental health (figure 2). In total, 341 codes were identified from the qualitative studies. Seven themes and 44 categories emerged. Since these identified themes also included precarious work conditions defined in the quantitative studies, the findings regarding precarious employment and its association with mental health problems were given below based on the themes (figure 3). More detailed information about the themes, categories and quotes that best represented these themes and categories are provided in supplementary table S4.

Continues
their contracts, being paid less than are non-migrants and discrimination and racism at workplace. The most evident category within this theme was "fearing termination for insubordination", which was present in more than one third of the included qualitative studies. It consisted of four subcategories: feeling obliged to work during sickness, receiving insufficient overtime pay despite being requested to render overtime work, being requested to perform additional work or tasks without added payment, working on assigned tasks or jobs without prior consultation notice to the work (24, 32, 34, 37-39, 41-44, 85-87, 89, 90, 92, 93, 95, 97, 102). Being undocumented further worsened these working conditions (24,37,85,102). The studied workers felt too vulnerable to use their rights when needed due to their fear of termination. Thus, many of them went to work despite experiencing work accidents or being in strong pain. Some of them hid their health problems or any work accident/injury to protect their work positions, as some employers prefer not to work with workers who are sick or were in accidents (34,41). Some workers were exposed to retaliation and threats when they complained about their work conditions (95). Many of the participants expressed that they had accepted their precarious work conditions to survive (86). Furthermore, the inability to ask for better work conditions was common among the undocumented workers (24,43,85,102). Additionally, many participants were aware of their contractual rights but felt too vulnerable to struggle for them (41,44,85,86,89,95,101). Overall, the undocumented migrant workers were more vulnerable compared with the registered migrant workers. Thirteen quantitative studies reported discrimination at the workplace (21, 22, 57, 60, 62-65, 68, 76, 79, 81, 100), and ten of them showed a statistically significant association between discrimination/unfair treatment and mental health problems Theme 3: Temporariness. This theme emerged in 88% of the qualitative papers. It covered four categories: unstable jobs, temporary/short-term jobs, lack of job security and certainty and lack of income security or stability (32, 34, 38, 41, 42, 44, 83-86, 88-98, 101, 102). Temporary jobs and instability prevent workers from obtaining unemployment benefits (43,90). Some of them found work through informal resources (44), temporary placement agencies (95) or meeting points in cities where employers recruit workers daily (44). Many of them worked on short-term contracts or without any contract, leading to job inconsistency (44, 83-86, 90, 95, 101), economic instability and lack of employee rights (101). Job insecurity was very common, regardless of the type of precarious contract (32,38,89,90,93,98). Nine quantitative studies (25, 33, 57, 66-68, 74, 75, 80) reported job insecurity, and five of them provided the prevalence of job insecurity (25,57,67,74,75). Seven studies reported an association between job insecurity and mental health problems (33, 57, 66-68, 75, 80). Many of the studied migrants in the quantitative studies worked on fixed-term, temporary or verbal contracts or no contracts at all (40,57,58,72,73,76,82). Six of the quantitative studies reported an association between precarious contract types and poor mental health (58,60,72,73,76,82) and three of them provided gender-stratified analyses. In one of them, migrant women workers reported mental health problems more often compared to non-migrant women workers, even though this difference was not statistically significant (82). Undocumented male migrant workers who worked with temporary contracts reported a higher risk of mental health problems compared to the non-migrant men who worked with a permanent contract (73). In another study, the association between precarious employment and mental health was stronger among women workers than among men (77).
Theme 4: Lacking workers' rights. This theme emerged in more than two thirds of the qualitative studies (24, 32, 34, 35, 37-39, 41, 42, 44, 85, 87, 89, 94-99, 101, 102). Workers who worked on an hourly basis did not enjoy the benefits of labor rights, such as health insurance, paid vacation or paid sick leave; this applied to the undocumented workers as well (34,37,38,44,89,94,96,98). Some of the workers were not able to take breaks for basic needs; some were not allowed to go to the lavatory without permission, and some received pay deductions for spending 'more' time in the lavatory (94). Seven quantitative studies reported workers' rights (21, 23,40,59,64,69) with three of them yielding evidence for an association between lack workers' rights and mental health outcomes (21, 59, 64).
Nine of the included quantitative studies reported low income among the participants, with six of them providing its prevalence (21, 23, 26, 56, 58-60, 66, 70). Four of the studies observed evidence for an association between low income and mental health problems (23,56,58,60).
Theme 7: Disempowerment. This theme was apparent in 63% of the qualitative studies covering the following categories: lack of unions, uncertain pay amount or schedule, unusually small salary amount or lack of payment during official holidays, working time arrangements, and other issues (eg, language barriers and insufficient or non-existent knowledge about labor rights) (24, 32, 34, 38, 41-44, 83, 85, 89, 91-96, 99, 101). The majority of the workers had no unions (43,89) and/ or had insufficient information about access to labor unions or any work organization that supports workers. With no official documentation and language problems, the participating workers felt helpless relative to their employers (24,32,43,101). Thus, they were required to work overtime or on public holidays without any payment (38), paying low salaries (24,32), giving no payment at all (92) or paying wages at uncertain times (34,93). Some undocumented workers who fought for payment for days worked were threatened by their employers with calls to police, which might lead to deportation (24,32).

Discussion
The main purpose of this review of qualitative and quantitative studies was to summarize and analyze existing research on the association between precarious employment and mental health of migrant workers. The review showed that the included studies examined different dimensions of precarious employment and a variety of mental health problems. Some of them reported the prevalence of precarious employment and mental health problems and/or an association between precarious employment and mental health problems. A considerable number of those studies did observe evidence for such an association.
In total, the review included 65 studies from four continents, with almost 90% of the studies being performed in high-income countries, which have had the largest numbers of migrant workers in the last century (3). Most of the included studies were of high quality. The data of almost all the studies were collected and published after the 2000s, and the number of published studies has increased especially in recent years. This development is consistent with the rising prevalence of precarious work conditions in the last decades due to globalization and decline of social policies (5,9). Mounting awareness of precarious employment and its influence on health and life quality of workers among the general public may also have contributed to the rise in the number of studies in this field.
The results of the present review indicate that a high prevalence of precarious employment among migrant workers with migrants being frequently exposed to various dimensions of precarious employment. On top of this, the results from those studies that also included non-migrant workers indicate that the prevalence of precarious employment tends to be higher among migrant compared to non-migrant workers while at the same time the effects of precarious work conditions on mental health also seem to be more pronounced among migrants (77,81,82). Those findings support the assumption of a special vulnerability to precarious employment among migrants outlined at the beginning.
Temporariness, (temporary, fixed-term or verbal contracts or no contracts at all), was the most commonly reported dimension of precarious employment among the included studies. This is in line with the results of related reviews that also concluded that temporariness is the most commonly observed dimension of precarious employment among workers including migrants (5,19,27,29). Furthermore, migrant workers in the studies included in our review commonly expressed a lack of rights or perceived powerlessness to exercise workers' rights, income insufficiency for basic needs, income insecurity or perceived powerlessness to negotiate work schedules or salaries. For example, two qualitative studies reported that women workers could not even use pregnancy leave and were forced to work until the last days of pregnancy (44,89). Some studies reported that migrant workers were exposed to highly unacceptable behaviors, such as racism, drinking of soap and scolding with physical violence (34,42,87,93).
In terms of mental health problems, the most commonly investigated ones were stress, depression, anxiety, poor general mental health and sleep problems. The prevalence of these outcomes varied between 10-75% among the quantitative studies included in the present review. Some of the included studies provided data about a putative association between precarious employment and mental health problems, yielding prevailing evidence for such an association. For example, temporariness was a risk factor for mental health problems in most of the studies examining this dimension of precariousness. The influence of this dimension on mental health (2,5,29,30) and occupational accidents and injuries of workers (19,20) has been widely examined by other systematic reviews, and their results are similar to those of this review. Moreover, the migrant workers found exposure to discrimination and authoritarian behaviors to be hurtful experiences and are thus important factors of mental health. Reviews by Sterud (31) and Jurado (103) also supported this finding as they found a relationship between perceived discrimination and poor mental health among migrant workers. Furthermore, the results of qualitative studies show that workers' interpretation of precarious work conditions also seems to play an important role in developing mental health problems. Cultural background, education level, professional work experience in original country, perspective of life and workers' rights, awareness of the occupational system, and rights in the host country may affect their interpretation. For example, workers' interpretation and acceptance of precarious work conditions as a permanent endeavor or a temporary period may affect the direct relationship between precarious employment and mental health. Some workers resort to precarious employment because of urgent economic needs or a need to obtain a positive reference from an employer, which is required in finding new work or securing a work contract that will extend their residency in the host country. Thus, they may feel the need to endure precarious work conditions, such as discrimination, long working hours without compensation, unpredictable work schedules, lack of workers' rights and authoritarian or disrespectful behaviors at the workplace. These experiences and feelings of workers may have added impact on their development of mental health problems. Having low social support and family concerns, limited access to and information about health care systems and traumatic life experiences prior to migration may also increase workers' levels of mental health problems (103).
The findings of this review also provide scientific data on precarious employment and its influence on mental health based on differences in gender and type of migrant status. Being undocumented in the host country seemed to be the worst condition, and it exposed workers to combined dimensions of precarious employment (24,41,43). In addition, female migrant workers developed higher rates of mental problems due to exposure to precarious employment compared with male migrant and non-migrant workers (73,82).

Strengths and limitations
Search strategy and publication bias. This review is the first to provide scientifically comprehensive data about the association between precarious employment and migrant workers' mental health. Wide-ranging definitions of and approaches to precarious employment, migration, and mental health problems were adopted to identify all relevant studies in the field. In addition to three digital databases, related informal sources and the references of the initially included studies were manually searched to decrease the possibility of overlooking relevant studies. This search strategy covered all relevant observational studies that were published in four languages (English, Spanish, German and Turkish) in the last 50 years. This search produced a reasonable number of qualitative and quantitative studies published in English and Spanish.
A limitation of our review is that no meta-analysis was conducted due to variations between the precarious employment dimensions, mental outcomes and measurement methods in the included studies. For this reason, it was also not feasible to construct funnel plots for assessment. However, the findings from another systematic review suggest that publication bias does not seem to play a major role in the research area of precarious work (5).
Moreover, some sources of bias in the individual studies need to be taken into account. First, most of the quantitative works were cross-sectional studies using different measurement tools to evaluate multiple types of precarious employment and various mental health outcomes. Only three studies assessed precarious employment with a validated and reliable tool (EPRES); the rest used different forms of questionnaires hampering the comparability between those studies. Furthermore, also mental health was measured with different scales or questionnaires. For example, the 12-item general health questionnaire (GHQ-12) was widely used to measure mental health; however, some authors interpreted the level of mental health over the total score (linear outcome variable), whereas some used a cut-off to define a dichotomous outcome variable. Furthermore, there was a lack of data regarding the duration of exposure to precarious employment, whether the workers had any mental health problems before beginning to work and work experiences prior to migration. Finally, limited data on the causes of migration was provided in the included studies, which might have an influence on the development of mental health problems.

Implications
The results of this review underline the importance of preventing mental health problems by reducing or mitigating precarious work conditions among migrant workers. Additionally, our results may increase social awareness about precarious employment and its influence on mental health, thereby aiding the establishment of human-based, worker-friendly policies at workplaces. This review might also help protect workers from exploitation by enabling experts to control certain elements in workplaces, which seem likely to have precarious work conditions, such as factors related to workers' rights. Furthermore, our data may play a role in the ongoing discussion about precarious employment being potentially useful for the planning of long-term preventive programmes. We recommend that future qualitative and quantitative studies holistically examine multiple dimensions of precarious employment and their influence on the mental health and well-being of workers using a multidimensional definition of precariousness. When doing so, studies may especially aim to disentangle the underlying mechanisms by being based on sound theoretical frameworks such as occupational stress models (9). For instance, one may argue that some factors, eg, job insecurity, are not working conditions per se but rather mediators on the pathway between precarious employment conditions and mental health. By applying and operationalizing a priori defined theoretical frameworks, a better understanding of such mechanisms may be obtained. Such analyses may also elucidate to what extent the association between precarious employment and mental health depends on migrants' individual characteristics and competencies or factors related to the migration process (eg, time since migration, reason for migration) (104). Moreover, future research may identify work-related or individual factors that act as job resources mitigating the effects of precarious employment in terms of the Job Demands-Resources model (105). Concerning study designs, most studies included in our review were cross-sectional. Longitudinal studies that can establish a temporal relationship between exposure and outcome would thus be useful in the future.

Concluding remarks
The results of the present review indicate that migrant workers are exposed to various dimensions of precarious employment and frequently experience mental health problems. An association between different dimensions of precarious employment, most prominently temporariness, and mental problems was reported by a considerable proportion of the included studies. However, it was observed that the theme of disempowerment and lacking workers' rights were less frequently examined compared to other themes, especially in quantitative studies. Altogether, those findings support the hypothesis that precarious employment is associated with migrant workers' mental health. We recommend that future research should better disentangle the underlying mechanisms by being based on sound theoretical frameworks as provided by occupational stress models. Ultimately, the results of the present review may be used as evidence for developing a new policy to resolve precarious employment. In addition, this review may also be used as a guideline for developing a better migrant-friendly policy in the future.