The dark side of the work of immigrant women caregivers in Spain: Qualitative interview findings.

AIM: To analyse the perception that immigrant women caregivers have of their relationship with the person receiving the care and their family and the possible impact those relationships may have on a caregiver's health.
DESIGN: A qualitative study was conducted.
METHODS: Thirty-four semi-structured interviews were applied in the Spanish city of Salamanca from November 2015-November 2016. The "interpretative hermeneutics" technique was used as the framework for the analysis.
RESULTS: The discourse studied indicated that the immigrant women's cultural background, as well as their gender's assumed stereotypes, gave rise to an emotional attachment to the person receiving the care, which could generate a burden. Most of the women interviewed reported situations of abuse from the old people under their care. The fact that increasingly more old people are being cared for by immigrant women renders it necessary for social and health policies to consider this collective.

INTRODUCTION

Globalization helps to identify the trend toward the feminization of migratory flows (Browne & Braun, 2008). Women normally encounter a segmented labour market in the host country (Ho & Chiang, 2015) where there continues to be a division of labour by sexes. In addition, immigrant women are in a more vulnerable situation than host‐country women, especially when pregnant or giving birth (Balaam et al., 2013). This approach, focusing on the theory of “care chains,” emphasizes this sexual division in the structures in both home and host countries (Hochschild, 2000).

The new demographics and advances in medicine have led to an ever‐higher percentage of older people, many of whom are dependents (Spain's Ministry of Labor & Social Affairs, 2004). This phenomenon, together with a transformation of the family structure (Roberto & Blieszner, 2015), the mass inclusion of women in the labour market and an increase in the migrant collective (Muñoz‐Comet, 2012), has changed the way Spanish households deal with caring. For some years now, the informal care provided by family members for attending to disabled older people is being “replaced” by the care provided by immigrant women caregivers. The bulk of these caregivers are Latin Americans because they are easy to hire, they speak Spanish and their collective image of warmth and an innate predisposition to care (Casado‐Mejía, Ruíz‐Arias, & Solano‐Parés, 2012).

Although a significant number of scientific papers are now beginning to appear on the subject of immigrant women caregivers, there are still specific aspects that have yet to be properly addressed, such as the direct or indirect impact the relationship between caregiver and care receiver has on an immigrant woman's health. Consequently, this study aims to analyse the perception that immigrant women caregivers have of their relationship with the person receiving the care and their family and the possible impact those relationships may have on the caregivers' health and their working conditions.

METHODS

A qualitative study was conducted using an intentional sample. This methodology was chosen because of the difficulties in obtaining a representative sample of immigrant caregivers. Spain does not have a register of immigrant caregivers, but instead only a register of immigrant domestic workers, many of whom are also caregivers (Bover & Gastaldo, 2005). The fieldwork was conducted in the Spanish city of Salamanca from November 2015–November 2016.

This study was performed using “interpretative hermeneutics” as a theoretical framework to analyse the discourse from the semi‐structured interviews. This theory takes into account the environment where the object of study is located as its main characteristic (Crist & Tanner, 2003).

Participants and context

The research has involved holding 34 semi‐structured interviews with immigrant women working as caregivers of disabled older people. The sampling strategy was theoretical and intentional. The eligibility criteria for inclusion in the study were to be a female immigrant, to work or have worked as a formal caregiver and to have been living in Spain at least for 1 year. Moreover, to provide a wide diversity of profiles, the women were selected for the interview according to country of origin, family circumstances, date of arrival, possession of residency visa, current job situation (although if unemployed, we could select the person if she had worked more than a year as a caregiver) and level of education. The interviewees were located through the employment services of Caritas and the Red Cross, the “María Inmaculada” religious community, two evangelical churches, a recruitment company for caregivers and informal networks (in this last case, using the snowball effect). In the case of Caritas and the Red Cross, a social worker in the employment services made contact with immigrant workers who cared for older people. In the “Maria Inmaculada” religious community, a nun facilitated our access to immigrant workers. The researchers involved in this study went to two evangelical churches in Salamanca where we knew there were many immigrant caregivers. Our key informant (Payne & Payne, 2004) for gaining access to immigrant women was a doctoral student who practiced the evangelical religion. Moreover, a recruitment company for immigrant caregivers arranged our interview with a female worker from Latin America. Additionally, we used our informal network (acquaintances) to complete the sample.

In the selection of the “sample,” we considered the following requirements: length of stay in Spain, country of origin, etc., to take into account their impact in the perception that immigrant women caregivers have of their relationship with the person receiving the care. According to our key informants, only two people refused to be interviewed and both came from Latin America. The qualitative method always carries a risk of bias, but in this case, we encounter the typical profile of a female immigrant caregiver in Spain: female, Latin American or from Eastern Europe, with different family circumstances (with or without children, children in or out of Spain, married, divorced…), with legally resident or not, with different job situations and levels of education (Oso, 2012).

Data collection

The interviews lasted between 30–60 min and were digitally recorded and then transcribed word‐for‐word. Each oral or verbatim segment was referenced in the text with the interview number and the interviewee's nationality (e.g., I no. 2, Russian refers to interview number 2 involving a Russian immigrant woman caregiver). Each interview's numerical code is shown in Table 1.

Table 1

Sociodemographic characteristics of the interviewed immigrant women

No.	Country	Age	Level of education	Marital status	Occupation *	Arrival in Spain	No. children
1	Colombia	31	Secondary	Single, with partner (in)	Both	2006	1 (in)
2	Russia	43	Secondary	Separated	Both	2007	1 (in)
3	Peru	35	Secondary	Single, with partner (out)	Both	2015	2 (out)
4	Ukraine	47	University	Separated	Both	2002	1 (out)
5	Honduras	38	Secondary	Separated	Both	2014	1 (out)
6	Cuba	50	Secondary	Married (in)	Both	2011	Childless
7	Bolivia	35	University	Married (in)	Both	2007	2 (in)
8	Dominican Republican	34	University	Single	Both	2008	Childless
9	Ukraine	48	University	Married (in)	Both	2000	2 (in)
10	Ecuador	59	Secondary	Divorced	Both	2003	Childless
11	Bolivia	38	University	Married (in)	Both	2004	Childless
12	Ecuador	42	Secondary	Single	Both	2007	Childless
13	Bolivia	35	Secondary	Separated, with partner (in)	Both	2001	2 (in)
14	Portugal	30	Secondary	Single	Both	1998	Childless
15	Nigeria	37	Primary	Married (in)	Both	2001	2 (in)
16	Bolivia	45	Primary	Married (in)	Both	2005	3 (in)
17	Morocco	39	Primary	Separated	Both	2000	3 (in)
18	Dominican Republican	31	Secondary	Married (in)	Both	2008	1 (in)
19	Honduras	32	Primary	Single	Both	2007	1 (in)
20	Bolivia	44	Primary	Married (in)	Both	2007	4 (in)
21	Dominican Republican	48	University	Married (in)	Both	2008	4 (in)
22	Dominican Republican	45	University	Separated	Both	2004	3 (in)
23	Romania	51	Primary	Separated	Both	2009	4 (in)
24	Mexico	43	Secondary	Married (in)	Both	2011	1 (out)
25	Morocco	37	Primary	Married (in)	Both	2005	3 (in)
26	Bolivia	53	Secondary	Married (in)	Both	2006	7 (in, out)
27	Bolivia	45	Secondary	Married (in)	Both	2000	3 (out)
28	Bolivia	30	Primary	Single	Both	2006	1 (in)
29	Bolivia	47	Primary	Married (in)	Both	2006	3 (in, out)
30	Bolivia	18	Secondary	Single	Both	2016	Childless
31	Bolivia	51	Secondary	Married (in)	Both	2002	3 (out)
32	Honduras	40	University	Separated	Both	2007	Childless
33	Romania	29	Secondary	Single	Both	2005	Childless
34	Colombia	63	Secondary	Separated	Both	2001	3 (out)

(In) Partner or child(ren) resident in Spain; (Out) Partner or child(ren) not resident in Spain.

By both we mean the combination of domestic work and caring.

We use the information saturation criterion when all the topics on the interviews' guide had been addressed by the interviewees and no new information was forthcoming (Barriball & While, 1994).

The direct interviews' quotes that appear in the text were translated from Spanish to English, with alterations made to the literal translation to preserve the intended meaning. The procedure for the translation of these quotes consisted firstly of a translation from Spanish to English by a native English speaker who is highly familiar with the Spanish language. Subsequently, a bilingual native Spanish speaker translated the text from English back to Spanish, after which another native English speaker translated the text back to English again. This text was then compared with the original text from the interviews and differences were noted among the three translators. The translators then discussed the discrepancies between the original and the translations and modifications were made wherever deemed necessary to achieve the most accurate result. This procedure was based on experts' recommendations for the transcultural translation of tests (Biering‐Sørensen et al., 2011).

Ethical considerations

The study was approved by Salamanca University's Research Bioethics Committee. A written and verbal request was sent to the agents listed above with the required profile for the interviewees, explaining the study's purpose, aims and content. Informed written consent was obtained from the people voluntarily taking part in the research, guaranteeing confidentiality and respect for the information provided.

Data analysis

The lead researcher and the co‐authors each systematically read through all transcripts and made a list of codes, sub‐categories and categories, highlighting all the main categories in the text. Qualitative sequential discourse was the method used for data analysis. This method follows three steps (Hsieh & Shannon, 2005): (a) Coding: the words or sentences that express the essence of the discourse are highlighted line by line. (b) Sub‐categorizing: once the coding process is complete, codes are regrouped into new forms, giving rise to conceptual codes or sub‐categories, which have a higher level of abstraction than the coding process. (c) Categorizing: categories have a higher level of abstraction than sub‐categories and each category includes several concepts (e.g., in this study, the category “relationship between the immigrant woman and the person receiving the care” includes concepts such as ambivalence and abuse of the immigrant caregiver by the receiver of the care). Some categories matched the core topics previously determined by the researchers, that is, problems and conflicts with the person receiving the care was one of the core topics which fits the category “relationship between the immigrant woman and the person receiving the care.”

FINDINGS

Sociodemographic data

Table 1 shows that most of the women interviewed came from Latin America; their average age was 40.9 (SD: 9.3); their educational profiles were very diverse and in terms of occupation they all combined domestic work with caring.

Categories identified

Three categories have been identified during the analysis of the content of the semi‐structured interviews, which are shown below.

Relationship between the immigrant woman and the person receiving the care

Ambivalence

The relationship with the person being cared for was defined by a great deal of ambivalence; that is, the relationship was described as having both positive and negative features. There was often a feeling of affection and responsibility, reflecting the view the immigrant women had of the older people and their assimilation of gender roles. This affection meant they were on very friendly terms with the older person. Nevertheless, on many occasions, even when there were signs of affection, there were conflicts with the people being cared for:

…Yes, once I had problems at home and I was feeling on edge and when I arrived she was rude to me, which hurt me and I started crying, but then she apologized and gave me a hug “forgive me, but that's the way I am. I'm sorry.” (I no. 6, Cuba)

Abuse of the immigrant caregiver by the receiver of the care

In many of the interviews held described cases of abuse by the people receiving the care. All these cases highlighted the existence of racism, xenophobia and classism as the main reasons for this ill‐treatment:

The lady treated me terribly (…) she used to tell me I didn't know how to speak, that I didn't know how to pronounce the s and the c, the c and the z, sorry, that this… that I was a waste of space, that I didn't know how to do anything, she told me that I was illiterate. (I no. 24, Mexico)

Relationship between the immigrant woman and the family circle of the person receiving the care

No relations

Some of our interviews revealed that the relations with the families of the persons tended to be conflict‐free because in those cases there was hardly any contact with the sons or daughters of the people being cared for:

Everything was fine with the daughter, there were no problems or anything like that, we hardly spoke, when she got home from work I left … (I no. 25, Morocco)

Conflicts arising from the exercise of authority in the family circle

Some of the caregivers in our study told us they were expected to work long hours and to do too many chores. Furthermore, they had to put up with too much control over their tasks, with threats and even instant dismissals. These dismissals were possible because there were no legal contracts, only the agreements drawn up with one of the family members of the person being cared for. All this had an ensuing effect on the physical and emotional state of the caregiver in the performance of her work:

With the perfect partner (…) but they had a daughter, the granddaughter in other words, a girl, well not really a girl as she was 33 years old and she always wanted to be like…well you know, giving orders that weren't hers to give. Because if you are doing your job right, actually no, you are doing more than you agreed to do, well she was demanding, she wanted more. (I no. 21, Dominican Republic)

The economic crisis and its impact

Immigrant caregivers living in spain before the crisis

This aspect was specifically brought up in many of the interviews and it was an important factor among the immigrant women living in Spain before the onset of the crisis for explaining the following: (a) a worsening of their contractual terms and conditions; (b) a reduction in their number of working hours; (c) greater job insecurity in this ambit; and (d) the acceptance of conditions they would have rejected in another economic scenario. The labour consequences of the crisis might have negative repercussions on their health:

It seems to me that, on the one hand, it has been affected too, because of course with the changes in the law and things like that people have got used to paying cash‐in‐hand and not paying national insurance. On the other hand, well, as each family has its resources, but there is always a demand for work… Some reduced hours, if they used to work eight hours now they work four, but they continue to work. (I no. 9, Ukraine)

Immigrant caregivers arriving in spain after the onset of the crisis

As regards the immigrant workers that started working during the crisis, not all of them were exposed to the poor conditions alluded to earlier. Some of these women had arrived through family reunification processes, under favourable conditions, while others had indeed referred to the precarious conditions they had encountered:

I only got a half‐day off on Saturday and all‐day Sunday, nothing else. On Sundays, I had to be back by nine in the evening at the latest…And all that only for 500 euros. (I no. 5, Honduras)

DISCUSSION

Most of the immigrant women caregivers were from Latin America, as we need to remember that Spanish households prefer to hire people from those countries. This choice was based on the assumption of a whole series of racial, gender and cultural stereotypes and because they spoke the same language (Rodríguez, 2009; Vega Solis, 2009).

The average age of our “sample” corresponded to a working age group, as is common in the migrant collective (Rodríguez, 2009). The educational profiles were very wide‐ranging, although the academic level tended to be higher than the Spanish nationals working in this field, as reported in the literature (Vega Solis, 2009).

The difficulty in distinguishing between domestic work and caring and the lack of continuity between one and the other, has also been widely analysed in different publications (Bover & Gastaldo, 2005).

The analysis of the discourses highlights the ambivalent nature of the relations between the immigrant women caregivers and the people being cared for. This often involved an affectionate relationship alongside cases of abuse by the receivers of the care. This affective involvement clearly breached the boundaries established in the literature between formal and informal care (Bonelli & Ulloa, 2001). The origin of this involvement lies, on the one hand, in the cultural context of the immigrant women (especially Latin Americans) and, on the other, in the desire the older adult's family circle has for the immigrant woman to play an affectionate role (normally provided by the old adult's family members, such as the children or grandchildren). It is important to note that the collective image of immigrant women is based on a stereotypical assumption of the inherent abilities immigrant women have for performing this kind of relational task (Rodríguez, 2009; Vega Solis, 2009).

Although the existing literature mentions the relationships of dominance and conflict that exist in care in a domestic setting (Casado‐Mejía et al., 2012) and even the abuse of the women by the people they care for (Bazo, 2002), there is no in‐depth analysis of this issue. This mistreatment was clearly reflected in attitudes and words with racist and xenophobic connotations. This behaviour tending toward abuse could be linked to the misgivings the older people had about letting “strangers” into their homes and their ignorance of the immigrant woman's culture (Dueñas et al., 2006). In addition, the maltreatment to the caregivers has been widely studied in the case of some diseases such as dementia (Holst & Skär, 2017).

The relations with the family circle of the people being cared for, normally involving informal caregivers, were described as largely non‐existent. In some cases, they were even conflictive and always related to the unilateral abuse of authority. The literature places special emphasis on the tense relations between the caregiver and the daughters of the persons being cared for, with this being mainly for two reasons: (a) the expectations placed on the immigrant woman as an emotional substitute for the informal caregiver and the triangle that may be formed giving rise to jealousy and veiled competition (Vega Solis, 2009) (we should make it clear that our study did not detect any such triangle); (b) the development of highly asymmetrical relationships, based on different power bases (Briones Vozmediano et al., 2014). Furthermore, job insecurity, reinforced by an irregular employment situation, informed scenarios where the immigrant woman caregiver was placed under excessive scrutiny, with threats and dismissals. This context was favoured by the economic crisis, whose consequences led to a fall in wages in the domestic sphere (precisely where the immigrant women work) (Bover et al., 2015; Muñoz‐Comet, 2012). We must take into account there are other countries and other situations where unregulated caregivers provide care to older people such as Canada (Afzal, Stolee, Heckman, Boscart, & Sanyal, 2018). In this case, the unregulated caregivers are not only immigrant women.

Study limitations

Among the study's limitations that we should single out, a major one is the inability to generalize the results, according to any qualitative research (Mays & Pope, 1995). In addition, the qualitative method in general could have a problem with the search for social acceptability by the interviewees and with recall biases. Nevertheless, the guarantee of confidentiality and anonymity and the importance of the subjective perception reduce those possible risks. Another limitation is the difficulty in recruiting immigrant women caregivers and the restricted selection sources. This limitation has meant that the discourse pertaining to the profile of immigrant woman caregivers hired through private companies was scarcely represented. Furthermore, the almost non‐existent line between domestic work and caring made it difficult to distinguish between these two spheres and the relationship with health. We should also highlight the need to interview Spanish women caregivers and their employers to further explore aspects already mentioned in this article. Finally, it would have been expedient to include a comparison group of non‐immigrant paid caregivers working in homes (who are very scarce in Spain, as most paid caregivers are immigrants) to verify whether the experiences of the study participants were any different to what is encountered by non‐immigrant paid caregivers.

CONCLUSIONS

The strength of this study is that it reveals the possible impacts that caring may have on immigrant women's health. As a major finding, we should single out the mistreatment that immigrant women often receive in this field of work.

Regarding the study's possible recommendations, we consider it essential to legally regulate the work of immigrant women caregivers, as in Spain they are governed by the same law as domestic workers. It would also be highly convenient for the legislation covering dependents and their informal caregivers to refer specifically to immigrant women caregivers, as to some extent they have become invisible for the public administration. The existence of specific laws for immigrant women caregivers might mitigate certain phenomena such as abuse. Furthermore, as we have already pointed out, we think the nursing profession should have an important role to play in the supervision and training of women immigrant caregivers.

We suggest also to study the care provided by immigrant women caregivers in rural areas which have hardly been studied. Furthermore, once initiatives have been implemented for supervising and training immigrant caregivers, studies should be conducted on evaluating these activities and for verifying whether the situation of immigrant caregivers improves.