Broadening the scope of live-in migrant care research: How care networks shape the experience of precarious work.

Live-in migrant care work is increasing across welfare states. In the context of ageing populations and changing healthcare systems, live-in care enables ageing in place without overburdening family caregivers. However, research has shown that live-in care often puts migrant care workers and their recipients in precarious positions. While the outcome of precarious work has gained attention, it is not self-evident. Based on interviews with migrant care workers in the Netherlands, as well as formal and informal caregivers, and the staff of intermediary organisations, this study shows that besides client conditions, the interactions of actors in the care network shape the circumstances and experience of migrant live-in care work. The findings suggest that managing the care networks to which migrant care workers belong makes it possible to mitigate the associated precariousness.

What is known about this topic

Due to changes in national healthcare systems, live‐in migrant care work is increasing in many welfare states.

The position of caregivers and care recipients is often precarious in this service.

What this paper adds

Live‐in migrant care workers belong to care networks, alongside intermediary organisations, informal caregivers and traditional formal care services.

Precariousness is not self‐evident, but one of many potential outcomes of the interplay between various actors in the broader care network.

INTRODUCTION

Many welfare states today are dealing with ageing populations and changing ideas about good care and dignified ageing (Buffel et al., 2014; Gallagher, Li, Wainwright, Jones, & Lee, 2008; Lewis & West, 2014). That healthcare systems should adjust to the new circumstances is a widely shared idea (Schwiter, Berndt, & Truong, 2018). Live‐in migrant care work (LIMC work) is one of many potential forms of long‐term care provision. While LIMC is well‐established in countries with a relatively limited welfare system (Italy, Spain, Austria), it is a remarkable new development in countries with ‘generous’ welfare systems (Sweden and the Netherlands; Da Roit & van Bochove, 2017; Hellgren, 2015).

Dutch LIMC work is a niche market that has been studied only recently (Da Roit & van Bochove, 2017; Davies & Mans, 2015; van Grafhorst, 2014). Exploratory research shows that some 10 Dutch organisations offer this type of 24‐hr care service (Da Roit & van Bochove, 2017), employing a few hundred LIMC workers. So far, there are no signs of LIMC workers working outside this scope. LIMC workers in the Netherlands are usually women in their forties, trained nurses or nursing assistants coming from EU member states in Central and Eastern Europe. Care recipients are usually older people with Alzheimer's or Parkinson's disease. Some are younger adults, suffering from multiple sclerosis, for instance.

Dutch media and research reports highlight the risks of LIMC work, drawing parallels with the circumstances of housemaids in the early 20th century, or with countries such as Italy, where LIMC workers are often subject to abuse and exploitation (Davies & Mans, 2015; Tonkens, 2011). The focus on adverse implications reflects the centrality of ‘precariousness’ in the international literature. Most studies address the uncertainties and risks that caregivers face, but more recent studies demonstrate the potential negative outcomes for care recipients (Ayalon, 2009; Salami, Duggleby, & Rajani, 2017). Studies often treat precariousness as self‐evident rather than focusing on factors that influence or mediate the outcome. However, local circumstances do seem to matter. While Hellgren (2015: 237) concludes that different welfare state regimes produce a similar ‘migrant precariat’, Salami et al. (2017: 1677) argue that ‘countries have different policies that affect the circumstances of live‐in caregivers or domestic workers’.

In this paper, we unravel precariousness by identifying factors that shape the outcomes of LIMC work in the Netherlands. Following recent work by Broese van Groenou, Jacobs, Zwart‐Olde, and Deeg (2016) and Jacobs, Broese van Groenou, Aartsen, and Deeg (2016), we look at the composition of care networks, including both formal and informal caregivers. Our central research question is: Who is involved in the care of clients receiving LIMC and how do interactions in the care network shape LIMC workers’ experience of precariousness? More specifically, we focus on the involvement of, and interactions between, intermediary organisations, LIMC workers, care recipients’ family members, and ‘traditional’ formal caregivers. This paper aims to provide a better understanding of the precariousness of LIMC work and to incorporate LIMC work in the study of care networks.

BACKGROUND

The precariousness of live‐in migrant care work

Migrant care is dominantly described as precarious work. In a broader context, Kalleberg (2009: 2) defines precarious work as ‘employment that is uncertain, unpredictable, and risky from the point of view of the worker’. LIMC work is precarious on the labour market level and the household level. First, the labour market position is poor because of low salaries and limited social rights and protection (Lutz, 2008). Comparing the position of LIMC workers in Sweden and Spain, Hellgren (2015: 223) concludes that, ‘[d]espite their very different policy trajectories (often undocumented), female migrant workers similarly occupy precarious positions in the respective labour markets’. However, other authors argued that differences exist in governmental and organisational policies, which limit cross‐country generalisation (cf. Da Roit & Weicht, 2013; Leiber & Rossow, 2019; Salami et al., 2017).

Second, on the household level, boundaries between work and leisure and between professional and personal relationships are often unclear, which can lead to tough emotional conditions and the risk of exploitation. Close informal relations between LIMC workers and care recipients often occur with an unequal distribution of power. Dealing with this demands a high level of ‘emotional labour’ (Bauer & Österle, 2013). Salami et al. (2017: 1676) argue that treating the migrant as part of the family might offer support, but it can also be a strategy family members use ‘to more easily manage or exploit’ migrant caregivers. Ayalon (2010) showed that living with a migrant caregiver could also lead to precariousness among care recipients, as there is no control of the quality of care, and physical or financial abuse of care recipients can remain unnoticed. Salami et al. (2017: 1676) conclude that future research ‘should examine the underlying conditions’ of precariousness on the household level.

This paper focuses on LIMC work in a relatively highly regulated welfare state (more information on the Dutch case is provided at the end of this section). Following Salami et al.’s recommendation, our study looks at underlying factors that influence the experience of precariousness in households. Our care network approach helps explain why some LIMC workers within the Dutch context experience precariousness, whereas others do not.

A care network perspective

A care network perspective focuses on a collection of individuals who provide care for someone on a regular basis (Jacobs et al., 2016; Tonkens, 2012), not just the dyadic relationship between LIMC workers and recipients. According to Broese van Groenou et al. (2016), about a quarter of Dutch community‐dwelling, older care recipients are looked after by a mix of formal and informal caregivers. The authors expect that mixed care networks will gain importance in the near future due to cutbacks in professional home care and policies that promote informal care provision. Mixed care networks differ in composition and function (Broese van Groenou et al., 2016). The various types of caregivers involved may complement each other, but disagreement about important decisions may create tension (Carpentier & Ducharme, 2003).

While there is little knowledge about mixed care networks in general (Jacobs et al., 2016), in the case of LIMC work, there is altogether no care network perspective. Some attention is given to care chains, but this literature focuses on transnational networks (Isaksen, 2012; Yeates, 2012). When studying actual care delivery in the receiving country, the focus is often only on the dyad relationship between care recipient and migrant caregiver, or on a triad including the care recipients’ children, mainly in their role as employer (Salami et al., 2017). We argue that the composition and functioning of mixed networks for LIMC recipients is crucial in shaping the precariousness for LIMC workers. Although our findings include insights into the situation of care recipients, our empirical data focuses on caregivers.

The Dutch case

Our study of how care networks shape LIMC workers’ experience of precariousness is situated in the Netherlands. The emergence of a Dutch LIMC market is remarkable, as Dutch public long‐term care (LTC) policies are traditionally among the most generous, inclusive and expensive in the world (Da Roit & van Bochove, 2017). In recent years, Dutch LTC funding and provision has changed. Access to residential care is now restricted and household help is no longer an individual right (Da Roit & van Bochove, 2017; van den Broek, Dykstra, & van der Veen, 2017). The Dutch government expects people needing care and support to first mobilise informal networks, before turning to publicly funded services (van Bochove, Tonkens, Verplanke, & Roggeveen, 2018). Government policies encourage ageing in place (van Dijk, Cramm, & Nieboer, 2013). However, informal or professional home care is not always enough for people needing almost constant assistance (Davies & Mans, 2015).

Dutch LIMC organisations established from 2005 onwards promise an alternative to residential care or intensive family care. Compared to ‘traditional’ home‐care organisations, these private organisations offer relatively cheap services which clients can generally pay for with publicly funded cash‐for‐care benefits. Research into these intermediary organisations is still in its early stages. The exploratory study by Da Roit and van Bochove (2017) characterised Dutch LIMC work as agency‐based, professional and formal. LIMC organisations engage in recruitment, matchmaking, training and supervision and previously offered such services as temporary employment, au pairs and private home care. LIMC workers are EU citizens with Dutch work permits. The organisations claim that they pay the minimum wage. Despite its relatively professionalised and formalised nature, in practice, tensions arise, for instance because the workers’ responsibilities have unclear boundaries (Da Roit & van Bochove, 2017; Davies & Mans, 2015; van Grafhorst, 2014). Our study looks at how the various actors involved deal with the uncertainties and vulnerabilities associated with LIMC work.

METHODS

Research design

To obtain a closer understanding of the role of care networks in shaping LIMC workers’ experience of precariousness, we conducted a qualitative study. Interviews with open‐ended questions allowed us to gain insight into the respondents’ experiences, beliefs and actions. By including multiple perspectives, we tried to develop a holistic view (Creswell, 2009). Following an adaptive theory approach, the aim of this study was not generating a new theory from scratch, but building on existing theories (Layder, 1998). During the data collection and analysis, we used sensitising concepts, but also remained open to unexpected findings. Below, we further explain different aspects of our research design.

Sample

Dutch LIMC organisations are private organisations that promote themselves online as agencies that offer ‘(foreign) live‐in caregivers’ (in Dutch: [buitenlandse] inwonende zorgverleners) or ‘care au pairs’ (zorg au pairs). They serve clients across the Netherlands, and sometimes Belgium. In the exploratory research by Da Roit and van Bochove (2017), eight LIMC organisations were identified, five of which participated in their study. The five organisations were contacted again for this follow‐up study, as well as an additional one that was identified by other organisations as an important player in the field. We invited the managers to be interviewed on the current state of affairs and asked them to provide access to other actors, particularly migrant caregivers.

Four managers were willing to be interviewed: three managers of the organisations that also participated in the earlier study, and one manager of the additional organisation that was contacted. Two managers (organisation A and B) allowed us to interview LIMC workers. Organisation A selected six LIMC workers representing different age categories and levels of experience. Through snowball sampling, we selected a seventh worker. Organisation B circulated our invitation to participate, which led to two more respondents. The remaining managers said that participation of LIMC workers was too hard to organise and they needed to protect their workers from public attention. No managers permitted interviews with care recipients, usually because the recipients' medical condition made talking difficult and the managers did not want to bother them. Because of privacy restrictions, conducting observations of clients was not allowed either. The Discussion section describes the limitations of the selection procedure in more detail.

In total, 20 respondents were interviewed: four managers and three care coordinators of LIMC organisations; nine LIMC workers; three relatives of care recipients; and one district nurse. The respondent codes are based on the type of respondent (Manager = M, Care coordinator = CC, etc.) and the organisation through which they were recruited (Organisation A, B, C, D; Table 1).

Table 1

Characteristics of the sample

Organisation	Respondents (n = 20)
LIMC organisation A	Manager (M‐A) (n = 1) LIMC workers (LIMC‐A) (n = 7) Relatives of care recipients (RCR‐A) (n = 3)
LIMC organisation B	Manager (M‐B) (n = 1) LIMC workers (LIMC‐B) (n = 2) Care coordinators (CC‐B) (n = 2)
LIMC organisation C	Manager (M‐C) (n = 1) Care coordinator (CC‐C) (n = 1)
LIMC organisation D	Manager (M‐D) (n = 1)
District care	District nurse (DN) (n = 1)

All migrant caregivers were women aged between 26 and 51 who came from Hungary or Romania. The country of origin was representative for organisations A and B, but not for C and D, which mainly work with Slovakians. Of the LIMC workers interviewed, some are paired in teams, rotating every two weeks (one staying with the recipient, the other returning to the home country). Others work alone and stayed in the Netherlands for longer periods.

Interviews

Five LIMC workers were interviewed in person in the client's home; the other four were interviewed via Skype. Six LIMC worker interviews were conducted in Dutch; two in English. In one case, a Hungarian‐Dutch interpreter translated the questions and answers. The remaining interviews were done in Dutch, either in person or on the telephone. The interviews were semi‐structured, using a topic list with broad topic areas related to precariousness and care networks (including the relationship with the client and their relatives, collaboration with other caregivers, working conditions, well‐being and social support). We allowed respondents to touch upon other themes relevant to them so that new insights could arise (cf. van Dijk et al., 2013). Interviews took between 45 min and 2 hr. All interviews were tape recorded and transcribed verbatim. To avoid losing nuance, we did not translate the transcriptions until the reporting stage. The first author translated the quotes. The second author checked the translations, and later these were reviewed by a professional language editor and NL‐EN translator.

Analysis and reporting

Data were analysed using a preliminary conceptual framework, combining deductive and inductive reasoning (Wilson & Chaddha, 2009). Initial coding was guided by the sensitising concepts of precariousness and care networks, focusing on different expressions of precariousness, strategies to deal with precariousness, the composition of care networks, and collaboration and tensions between actors in care networks. These codes were then categorised in secondary themes emerging from the data, such as clients’ conditions, intermediary organisations’ risk management strategies, and LIMC workers’ soft skills. In presenting our results, we organised the findings in terms of actors involved and the influence they have on shaping LIMC workers’ experience of precariousness.

We emailed a report of the analysis (van Bochove, zur Kleinsmiede, & Ashu, 2017) to the organisations. Some managers thanked us for sharing the report, but none responded to the content. We also discussed the findings with an inspector of the Dutch Health & Youth Inspectorate, which yielded insights on how regulators perceive this type of service. We briefly refer to these insights in the Findings section. This article is based on the Dutch report, but since the report was mainly descriptive, for the purposes of academic publication we revisited the data and coding and used additional literature to interpret our findings.

Ethical considerations

Before the interview started, the aim of the study was explained and verbal informed consent was obtained to collaborate in the study and to start recording. We promised anonymity to all our respondents; therefore, we use codes and pseudonyms to protect their identity. Particularly when contacting LIMC workers, we emphasised that their participation was voluntary and we assured them that their responses would remain confidential. This study did not require ethical approval, as the Dutch Medical Research Involving Human Subjects Act only requires consent of an ethics committee in the case of medical intervention (cf. Rutz, van de Bovenkamp, Buitendijk, Robben, & de Bont, 2018).

RESULTS

The circumstances of LIMC provision vary widely according to the recipient's care network. After discussing the relevant client conditions, we focus on the various types of care providers—intermediary organisations, LIMC workers, care recipients’ family members, and ‘traditional’ formal caregivers—and the role they play in the care network. Although we draw analytic distinctions between the actors, our findings show that their interactions are vital in shaping the precariousness that LIMC workers experience.

Care recipients: variation in physical, housing and social conditions

Often one of the first things LIMC workers did in an interview was describe their care recipient's situation because this influences how hard or easy (physically and emotionally) their job is. First, the recipient's medical condition makes a difference. Some clients need constant attention in daytime and also wake up several times at night, whereas others can be left unattended a while, giving the LIMC worker time off to watch TV, read or sit in the garden.

Housing conditions are the second important condition. LIMC workers spend most of their time in the care recipient's home and value having their own space. This can vary greatly. One LIMC worker remembered a negative experience:

I had my own room, but it was very small. Only a bed and that was it. No window, always dark and it stank. It wasn’t nice, really not nice. (LIMC1‐A)

Another respondent was pleased with the space she now has:

They [care recipient and his wife] don’t use upstairs anymore. I have my own TV room there, and a bathroom. I also have the big room, that’s the art room. It’s like the office, but with paint and pencils and everything, so I can go there and paint whenever I like. (LIMC1‐B)

In these two cases, health and housing conditions coincide: a client with high functional needs in a small house in the first, and a relatively healthy client in a large house in the second. The other respondents reported situations somewhere in between these extremes.

A third important client characteristic is the intensity of social contact, which can moderate the medical and housing conditions, as regular visits by clients’ relatives, friends and neighbours offer LIMC workers distraction from work and reduce feelings of social isolation. When such contact is absent and client and LIMC worker spend almost all their time together, it can become emotionally burdensome for the worker (cf. Bauer & Österle, 2013). In such situations, the precarious positions of care recipients and caregivers reinforce one another, which can lead to a ‘chain of precariousness’ (Hellgren, 2015).

Intermediary organisations: preventing and monitoring risks

As earlier research indicates, intermediary organisations do more than match supply and demand (Da Roit & van Bochove, 2017; Elrick & Lewandowska, 2008). We found that organisations try to avoid or soften the risky aspects of LIMC work in three ways: selecting low‐risk clients and caregivers, strategic matchmaking and monitoring care.

Managers of intermediary organisations described the profiles of risky care recipients and caregivers. Risky recipients have no social safety net and the ‘wrong’ motives. Potential clients should be able to guarantee that the LIMC worker can take time off. Having relatives, friends or neighbours in the care network makes this more likely. One organisation explicitly rejects clients without an informal support network.

This man, paralysed in a car accident, had no children or anyone. He was the client and at the same time our main contact. We wouldn’t serve this kind of client. (M‐B)

As a consequence, LIMC services, at least on paper, are not available for isolated older people with substantial care needs, which limits the claim that these services provide an alternative to residential care (Da Roit & van Bochove, 2017).

Several managers said they pay attention to the reasons why potential clients hire an LIMC worker. Clients who ‘find the looks more important than the experience’ (M‐D) or who have no real care need but say that they are ‘feeling so lonely’ (M‐B) are suspicious and usually rejected. Of course, this does not guarantee that the motives of accepted clients are always apt, and even if they are, their medical conditions can also make them cross certain behavioural boundaries, as several managers indicated.

According to the managers, risky LIMC workers have no prior experience of living abroad and have young children. In these cases, the ‘emotional labour’ LIMC workers would have to perform to deal with homesickness is deemed too demanding (cf. Bauer & Österle, 2013). In contrast, LIMC workers who have lived in a ‘western’ country before, with grown‐up children and a background in healthcare or long‐term care, are suitable. Even though LIMC workers are not allowed to perform nursing tasks in the Netherlands (they are not registered nurses), managers find diplomas and prior experience important. Experienced LIMC workers are more likely to detect changes in the health of the recipient and alert other caregivers.

Matchmaking is an important next stage in the process. Organisations try to find an LIMC worker who matches the client's needs, and arrange Skype meetings between the LIMC worker and the client and their family members. The goal is to see if ‘they click with one another’ (M‐A; M‐B). Particularly in the first months after matchmaking, the organisation evaluates whether the involved parties are satisfied. This is not always easy to predict. Various LIMC managers and coordinators said that ultimately it has not to do with the clients’ medical situation or the skills of the LIMC worker, but whether they match on a personal level. For instance an LIMC worker with a gentle/warm character could suit one care recipient, but another might prefer someone with a more rational/distant personality.

A final important task of intermediary organisations is monitoring the care process. Some organisations have fixed evaluation moments, whereas others evaluate when the LIMC worker or other actors in the network indicate that conditions have changed or the workload has become a burden. One manager (M‐C) said: ‘Each time, we check whether the client is still getting enough care’. If not, other solutions—such as a nursing home—are discussed.

Live‐in migrant care workers: the importance of soft skills and social support

Almost all LIMC workers point to personality traits that are important in coping with their work. Some describe caring for others as a calling rather than a job (LIMC2, 5‐A; LIMC1‐B). Whether or not they find their work stressful is related to the client characteristics discussed above, but LIMC workers also said it matters how they, as caregivers, regard and deal with potentially difficult aspects of the job. For instance they mention being ‘not easily bored’ (LIMC1‐B) and having ‘lots of patience’ (LIMC2‐B): they do not mind spending their free time in and around the house and deal patiently with clients who cannot communicate clearly or keep forgetting things. Several respondents mentioned having the necessary social skills and the ability to communicate well with the client's relatives as important qualities. One informal caregiver (RCR1‐A) praised the LIMC worker because ‘she gets on well with people’. One client's brother said of his sister's LIMC worker:

She’s very friendly and helpful, very sociable. If we get invited out [by friends], they always ask her to come too. And it’s not like ‘we have to invite you’, but because they really like chatting with her. (RCR2‐A)

Several respondents recalled examples of less sociable LIMC workers. For instance the client's brother cited above said that a previous worker ‘was more introvert’ and did not appreciate the company of the client's visiting relatives who just wanted to offer support: ‘She didn't like having people around. At one point, she wouldn't let anyone in anymore’ (RCR2‐A).

Besides personality traits, a support network also mitigates the potential risks of LIMC work. Several LIMC workers mentioned the people they can turn to. Sometimes these people are in the client's care network, but they may be relatives or friends of the LIMC worker, which suggests that the boundaries between mixed care and additional support networks are blurred.

Some LIMC workers have a family‐like relationship with the client and their relatives. One LIMC worker, for instance said that the care recipient's wife ‘is like a grandmother to me’ (LIMC1‐B). While Salami et al. (2017: 1676) mention the risks of being treated ‘like one of the family’—it makes it easier to exploit LIMC workers—the workers we interviewed usually found familial‐like bonds soothing rather than burdensome.

Regarding the additional social support LIMC workers receive, we saw a difference between the two types of workers mentioned earlier: ones staying in the Netherlands for longer periods, looking after a client on their own, and those working in pairs who replace each other every two weeks, travelling back and forth between the Netherlands and their home country. LIMC workers who stay in the Netherlands for longer periods often build up a non‐client‐related social network and, for instance go out with friends. One (LIMC2‐B) met her Dutch partner during her stay in the Netherlands. LIMC workers working in pairs see their own family and friends every two weeks and while in the Netherlands, often Skype or phone them. Previous research shows that managers encourage LIMC workers to travel back and forth because they can get rest and keep in touch with their families at ‘home’ (Da Roit & van Bochove, 2017). Some interviewed LIMC workers also mentioned the advantages (LIMC2, 6‐A), whereas others emphasised that it can be stressful to live in two countries (LIMC3, 4, 7‐A). Their stay in the Netherlands is entirely focused on caregiving and they are unable to build a local social network.

Several respondents gave examples of what can go wrong when a client's heavy care needs or social isolation coincide with a lack of social support for the LIMC worker. One (LIMC4‐A) said she has to work ‘nonstop’ in the client's house and gets headaches. Three managers gave examples of LIMC workers (ab)using alcohol to cope with such problems. One manager (M‐D) remembered a client calling the office to say that his ‘help had fainted’. In hospital, it turned out that the ‘help’ had overdosed on alcohol. The manager said that this LIMC worker had not only put herself in danger, but also the client, as she had violated the key principle of ‘staying with the client, no matter what’.

Care recipients’ family members: solving and causing problems

Informal caregivers often play an important role in care networks and have enormous impact on how LIMC workers experience their work. The bonds LIMC workers have with care recipients’ non‐resident family members vary from very close and warm, to rather distant, and, in some cases, troublesome.

Starting with the positive end of the spectrum: some LIMC workers looking after a client who needs a lot of attention and cannot easily communicate feel supported by the informal caregivers, usually the client's children. Several examples suggest that client relatives do not always put the client's or their own needs first. For instance an informal caregiver said that he and his relatives urged an LIMC worker to go back to her home country when her father had a cerebral haemorrhage. Although the LIMC worker felt bad about it, they told her to ‘make the right decision for your own family, not ours’ (RCR2‐A).

In most of the cases we encountered, clients received help from informal caregivers. One LIMC worker said her client's son ‘is like a brother’ to her and he takes over caring for his father 1.5 days a week (LIMC2‐B). In some cases, however, relatives are not present that often. Some respondents gave examples of children who promised to help, but ultimately did not (LIMC1, 3, 4‐A). According to a district nurse, this is because relatives think ‘the girl is there anyway’ (DN).

The experience of 39‐year‐old ‘Barbara’ (pseudonym for respondent LIMC1‐A) shows the variation in relatives’ involvement. Barbara currently lives with a 94‐year‐old woman who has six children. Two children regularly do administrative tasks but the others rarely visit their mother. The client does not need constant attention, but Barbara cannot leave the house. Her previous experiences demonstrate the extremes. On the positive pole is the helpful daughter of a previous client who joined her mother and Barbara in outdoor activities and was ‘always asking how she could help’. After this client passed away, Barbara remained friends with the daughter. On the negative pole are relatives of another client who caused more work rather than alleviating it. The client's parents were in need of care themselves and Barbara often had to look after that client's young child as well.

The dad wanted a beer every day at 4 pm. I always had to check if his glass was empty and if so, refill it. If I didn’t [notice in time], he immediately corrected me: ‘You didn’t see it!’ And I always had to bring the dad to the doctor or hospital. (LIMC1‐A)

This case resembles the kind of LIMC worker exploitation that is frequently reported in the literature (Salami et al., 2017). Barbara eventually quit working for this client, because she ‘only got one salary’ for taking care of the whole family. The LIMC organisation arranged another client for her.

The findings presented here show that care recipients’ relatives have a big impact on LIMC workers’ experience, but the impact is not self‐evidently positive or negative.

Traditional formal caregivers: complementing migrant and family care

The final group of actors present in many (but not all) client care networks are ‘traditional’ formal caregivers, such as home‐care nurses, district nurses, GPs and physiotherapists. Formal caregivers play various roles. Like informal caregivers, they can temporarily substitute for the LIMC worker, so she can have some time off. In this case, an informal caregiver will usually arrange for a district nurse or domestic help to keep watch at fixed times on fixed days (LIMC1‐B; RCR3‐A).

In various cases, home care or district nurses visit the client to do nursing tasks that LIMC workers are not allowed to carry out, such as applying morphine patches, taking blood samples and (re)placing catheters. In other cases, district nurses monitor the care needs and give advice. During an interview, a district nurse (DN) called on the care recipient and her two LIMC workers (LIMC3, 4‐A; both present because the one was just replacing the other). The three discussed the recipient's condition (she had a cough) and a solution (what cough syrup they should buy at the drugstore). A care coordinator (CC‐C) said LIMC workers have an important signalling function. Because they spend so much time with the client, they quickly recognise changes in the client's health and can report this to the formal caregivers.

We received mixed signals concerning the communication between regular formal caregivers and LIMC workers. Various respondents (LIMC2, 6‐A; RCR1‐A; LIMC1‐B) said that the contact between LIMC workers and physiotherapists went well: LIMC workers encourage their clients to do the exercises the physiotherapist recommended, which, according to both parties, improves the client's condition. A district nurse (DN) said the two LIMC workers she works with ‘are very nice ladies who do their job well’.

However, as Carpentier and Ducharme (2003) pointed out, in heterogeneous care networks, miscommunication between different caregivers easily arises. One LIMC worker (LIMC3‐A) complained that the case manager who orders medication for her client comes to visit once a month, but is hard to contact by email or telephone. The inspector of the Dutch Health & Youth Inspectorate we spoke with said that the opposite—LIMC workers (or the organisations that employ them) failing to keep in touch with regular formal caregivers—also occurs and can lead to risky situations. According to the inspector, LIMC organisations generally want the best for their clients, but they do not always acknowledge in time that clients need more help than LIMC workers can offer.

DISCUSSION AND CONCLUSION

Precariousness: not a self‐evident outcome

LIMC work is dominantly described as precarious work (Hellgren, 2015; Salami et al., 2017). But as Antonucci (2018: 888) concluded in a cross‐country study on university students: ‘not all experiences of precarious work lead to precarity’. Although our research was not comparative, based on what is known about LIMC workers in less extensive welfare states (cf. Muehlebach, 2012), we may cautiously conclude that LIMC workers in the Netherlands encounter less precarity than in many other countries. More central to this study, we found that within a certain national context, differences in LIMC workers’ experiences exist related to the characteristics of clients and their care network actors.

Table 2 summarises how characteristics of care network actors can moderate the risk of precariousness. Our findings suggest that if mitigating factors—such as sufficient private space, good matchmaking, a supportive social network and guaranteed time off—are absent, there is a higher risk of precariousness.

Table 2

Summary of results

Actor	Factors mitigating risk of precariousness
Care recipient	No permanent attention needed Sufficient private space for LIMC worker Social contacts with relatives, friends, neighbours
Intermediary organisation	Selection of low‐risk clients and caregivers Good matchmaking Monitoring and evaluation of care process
Live‐in migrant care worker	Necessary soft skills Supportive social network in country of residence and/or country of origin
Client's relatives	Guarantee time off for LIMC worker
Formal caregivers	Guarantee time off for LIMC worker Give advice and act upon LIMC workers’ signals

The dynamics among these actors largely shape the outcomes. A care recipient needing almost constant attention does not have to lead to precarious situations when other actors in the care network alleviate the LIMC worker's workload. Vice versa, even if a care recipient is relatively independent, precarious situations can arise when the LIMC worker has personal problems but lacks a supportive social network, or when the client's relatives expect the LIMC worker to do tasks outside their job description. Care networks are not static, but always changing: changes for one actor influence the other actors. Unlike what Hellgren (2015) suggests, a ‘migrant precariat’ is not a predetermined outcome, but depends on national, organisational and personal circumstances and actions.

Limitations

Our study has some limitations. First, because of privacy issues, organisations did not allow us to observe the daily practices of LIMC workers. Although the interviews revealed what the work entails (e.g. checking on the client multiple times or talking to a district nurse), more thorough observation would have been preferable. Second, we did not speak to all the actors in a single care network. In some cases, we talked with formal caregivers, in other cases with informal caregivers, and in yet other cases, we could only interview LIMC workers. The organisations did not allow us to interview care recipients and therefore we cannot draw any definite conclusions about their experience of precariousness. Third, since LIMC workers are difficult to reach, our participants were selected through the organisations that employ them. Agencies might have selected workers whom they thought had (only) positive experiences. However, through snowball sampling, we also spoke with an LIMC worker who was not selected by the organisation and her experience was much the same as the others’. Moreover, respondents selected by their organisations talked openly about the hard parts of their job, such as feelings of isolation and having headaches. Finally, in some interviews, the language barrier made it difficult to discuss personal issues such as emotions in‐depth. An interpreter assisted in one interview, but this negatively affected the spontaneity of the interviewee. However, most LIMC workers spoke Dutch relatively well and two were fluent in English.

Implications for research and policy

For both research and practice, it is important to pay attention to ‘risk profiles’ of care networks. In addition to existing literature on risk profiles of individual care recipients and informal caregivers (e.g. Dury et al., 2017), Janse (2018: 148) recently argued for a focus on ‘dyadic risk profiles’. This focus acknowledges that risks (e.g. deteriorating health in frail older people and overburdening family caregivers) are shaped by the interactions among care recipients and caregivers. On the basis of our findings, we argue that such risk profiles should include all relevant actors involved. In determining the risk profile of care networks involving LIMC services, we should include the characteristics of clients, LIMC organisations, LIMC workers, informal caregivers and regular formal caregivers. Future research should scrutinise which types of care networks moderate the potential precariousness of LIMC work most effectively.

Rather than trying to ban this type of service, as some politicians and researchers advocate, we recommend regulating the practice effectively. The changing LTC landscape in many welfare states demands innovative solutions. Organisations that offer LIMC services can be seen as healthcare ‘rebels’ who want to offer good quality care by doing things differently (Bal, Weggelaar‐Jansen, & Wallenburg, 2017). We should acknowledge the potential risks of LIMC work, but our findings suggest that managing the care networks to which LIMC workers belong makes it possible to mitigate the associated precariousness. In line with the recommendations of Leiber and Rossow (2019), we argue that this should not depend on intermediary organisations’ risk management and self‐regulation alone, but also on other national and supranational actors in the governance of cross‐border care arrangements.