Guardians of Health: The Dimensions of Elder Caregiving among Women in a Mexico City Neighborhood

Abstract

Little is known about the family care of older adults in Mexico and the role of women in this process. To begin to fill this knowledge gap, this paper describes how a small sample of low-income women in one Mexico City neighborhood conceptualized the caregiver role and identified the forms of assistance they gave to their older relatives on a daily basis. A grounded theory approach was used to collect and analyze the data. Forty-one semi-structured qualitative interviews were conducted with female caregivers. The age of participants was between 19 and 83 years, and care recipients between 56-92 years. The relationship of caregiver to care recipient was wife, daughter, daughter-in-law, granddaughter, sibling, and other relative. The mean length of time providing care was five years. Most participants were not employed outside the home, and the median monthly household income was 2,000 pesos. We found that caregiving was a life changing event, with 27 of 41 participants viewing themselves as guardians. Caregivers' emphasis was on care recipients' emotional needs in order to provide “the most precious gift” of “time and attention.” Two forms of assistance were ‘keeping company’ and ‘watching out’ as safeguards against poor health or further decline in health. These findings increase the cultural understanding of caregiving in Mexico. Further research is needed to test the caregiving concepts identified in this study.

Introduction

Mexico's population will rapidly age in the 21^st Century—the age 60 and older population is expected to reach over 35 million (24% of the total population) by 2050 (United Nations, 2002). Mexico's elderly population is also experiencing a morbidity and mortality shift from communicable to chronic diseases as well as increased longevity (Martínez & Leal, 2003). The number of elderly persons (age 65 and older) for every working age person (ages 15-64) is expected to increase from 7.6 in 2000 to 30.0 in 2050 (United Nations, 2002), suggesting that the social support needs of the elderly are likely to increase in the future. Given that the risk of disability and chronic disease increase with age (Pan American Health Organization, 2002, 2003), these trends suggest that the demand for formal and informal caregiving will be overwhelming as Mexico continues to gray.

Although the Mexican population is aging, the long-term care options for older adults are limited. Formal supportive services for the elderly are not included in the Mexican health care system (Barraza-Llorens, Bertozzi, Gonzalez-Pier, & Gutierrez, 2002) or covered by the Mexican Social Security Institute (Gutierrez-Robledo, Reyes-Ortega, Rocabado-Quevedo, & Lopez-Franchini, 1996). The lack of access to formal long-term care is reflected in the low rates of nursing home use by the elderly in Mexico City (Gutierrez-Robledo, 2002). These circumstances suggest that older Mexicans are either going without care or are having their daily needs met elsewhere. Research in both western and non-western societies has documented the important role that families play in the lives of their older members. Informal elder care falls within the scope of unpaid care work—activities that serve others, require time and energy, and are not financially rewarded (UNIFEM, 2000).

Throughout the world, unpaid care work has predominantly been seen as a woman's responsibility, including in Mexico. A recent Mexican study found that elder caregiving falls predominantly to women because it is considered a woman's job (Robles-Silva, 2000). The Mexican literature on women's occupations in the private domain suggests that individuals are trained or socialized into roles that define caregiving as a woman's responsibility (Finkler, 1994; Willis, 1993). Little is known about elder caregiving in Mexico and the role of women in this process, although the abundant North American caregiving literature makes an implicit assumption about the universal adaptability of caregiving concepts and measures. However, the extent to which aspects of caregiving are culturally universal is not known because of the limited research on caregiving in other contexts.

The current study was undertaken as a small but important step in filling the knowledge gap about informal elder caregiving in Mexico. The aims of this paper are to describe how a small sample of low-income women in one Mexico City neighborhood conceptualized the caregiver role and identified the forms of assistance they gave to their older loved ones on a daily basis. The overall objective of our study was to explore how Mexicanas conceptualized caregiving as a construct in terms of cultural beliefs, social norms, role functioning, and familial obligations. We operationalized this goal by developing a subset of research questions, including those which are related to aims of this paper.

Methods

This study used a grounded theory approach to explore how elder caregiving was socially constructed and its component parts imagined, among women actively engaged in caregiving in one small neighborhood of Mexico City. We were particularly interested in how women organized their behaviors and interpreted their caregiving experiences to create their social realities as caregivers. We used the qualitative method of in-depth interviews, which can be an valuable approach to understanding the experience of caregivers (Abel, 1991). In-depth interviewing techniques can be especially effective in revealing both the emotional and the symbolic meanings of elder caregiving that are not detected in typical survey approaches (Blieszner & Hamon, 1992).

Data Sources and Collection Procedures

Study Site

Mexico City, also known as the Federal District, is composed of 16 geographic units called delegaciones, and each delegación is divided further into colonias or neighborhoods. The study site was the neighborhood of Tetelpan, located in the delegación of Alvaro Obregón. We chose Tetelpan because we wanted a location where the majority of the population had limited economic resources in order to increase our chances of finding women undertaking the primary caregiver role themselves. Purchasing in-home services through the informal work sector is common among middle and high income families (Lomnitz, 1982; Lomnitz & Perez-Lizaur, 1979). Such services can include laundry, housecleaning, child care, and elder care. Although neighborhood-level data were not available, data at the delegación level showed that Alvaro Obregón had fewer middle-income and more low-income people compared to the Federal District. Almost two-thirds of Alvaro Obregón's population earned two times or less the minimum wage compared to 61% for the entire Federal District. The minimum daily wage in Alvaro Obregón was 26.45 pesos (Instituto Nacional de Estadística Geografía e Informática, 1998a), the equivalent of $2.82 US dollars.

Sample Recruitment

Women who met the following criteria at the time of interview were eligible to participate in the study: 1) minimum of 18 years old; 2) self-identified as the primary person responsible for the overall care of a dependent elderly family member; and 3) caregiver and elder lived in the same household. A dependent elderly family member was defined as a person at least 60 years old related through blood or marriage who needed help with one or more Activities of Daily Living (basic functions such as feeding, bathing, dressing, transferring, toileting and personal hygiene) (Katz, 1983), or Instrumental Activities of Daily living (Lawton & Brody, 1969) (higher-order functions such as transportation, cooking, grocery shopping, housework and financial management). We made an exception to the age requirement twice when care recipients were not yet 60 years of age but were identified by their caregivers as “old.”

Study participants were selected using a combination of snowball and purposive sampling methods. Initial study participants were solicited using life-long neighborhood residents and members of a local parish ministry group as key informants. Snowball sampling, a technique for finding study participants using referrals from other participants, was especially useful for locating hard-to-find subjects with similar characteristics (Bernard, 1995), such as being a caregiver. After initial interviews, we conducted purposive sampling to increase the number of study participants who represented a range of caregiving situations to explore relevant themes that emerged from previous interviews (Bernard, 1995). This approach was necessary for the study because low-income, informal caregivers would not have been likely to receive formal work sector or governmental benefits, and therefore would not have been readily identifiable through census or Mexican Social Security Institute data.

Data Collection

We obtained informed consent from study participants, using procedures approved by the University of California, Los Angeles (UCLA) Institutional Review Board. Data were collected through semi-structured interviews using a guide that covered the following four topics: 1) story of becoming a caregiver; 2) forms of assistance and contexts of caregiving; 3) social and cultural beliefs about aging; and, 4) beliefs about the caregiver role. The guide was initially developed in English and administered to three Mexican American women living in East Los Angeles, California. The guide was subsequently revised, translated into Spanish and pre-tested with two Mexican women, one who lived in Tetelpan and one who lived in a nearby colonia. The interview guide underwent several revisions during the data collection process as new concepts emerged from the iterative analytic process that ran concurrently with data collection activities. All interviews were tape-recorded, conducted in Spanish by a trained native speaker and lasted an average of 50 minutes. The principal investigator was present during the interviews to take observational notes. The interview tapes were transcribed verbatim in Spanish by a professional native-speaking transcriber.

Data Analyses

The interview transcripts were analyzed in Spanish. The transcripts were interpreted through a grounded theory approach (Strauss & Corbin, 1994) that involved an iterative process of taxonomic organization, content analyses, and code mapping. Each transcript was hand-coded during repeated examinations of the text. The text was first broken down into fragments and “clustered” around single words or phrases. Clusters of text were coded and organized into a hierarchy of categories at consecutively higher levels of abstraction to build thematic content.

Quotes were subsequently translated into English for use in this article. The original quotes in Spanish are available from the authors upon request. A language consultant translated five interviews into English as a verification check of the researchers' translations.

Results

Sample Characteristics

A total of 41 female caregivers participated in this study. The study sample tended to be long-time residents of Tetelpan, and to have low levels of income and education (Table 1).

Table 1. Sample Characteristics of Respondents and Care Recipients.

(N = 41)
Demographic Characteristic		SD
Mean Caregiver Age (years)	49.8	18.2
Mean Care Recipient Age (years)	75.2	10.9
Mean Years of Caregivers' Education	7.7	4.3
Median Length of Caregiving (years)	5	18.5
Median Monthly Household Income (pesos)	2000	1893.4
Mean Years of Marriage for Spousal Caregivers	51	8.4
	n	%
Caregiver Employment Status
Not employed	25	61
Part-time employed	12	29
Full-timed employed	4	10
Marital Status of Non-Spousal Caregivers
Married	15	50
Never married	9	30
Divorced or separated	4	13
Widowed	2	17
Relationship of Caregiver to Care Recipient
Wife	11	27
Daughter	17	41
Daughter-in-law	4	10
Granddaughter	6	15
Sister	2	5
Other relative	1	2

Three-quarters of participants had lived in Tetelpan for more than two decades, and over half had lived in the colonia for their entire lives. The mean age of participants was almost 50 years, with a range from 19 to 83. The study participants' educational levels ranged broadly from one year (kindergarten) to 18 years (graduate school), with an average of just under eight years, the equivalent of primary plus one year of secondary education. However, just over half of the study participants had education equivalency levels of primary school only, with 19 of them not completing their primary education. Although the majority of participants did not work outside the home, 25% had part-time jobs and 10% had full-time employment. The median monthly household income was 2,000 pesos, with a broad range from 800 to 10,000 pesos. The monthly household income was a gross estimate for the entire household, including working and nonworking persons. Therefore, when the number of working persons in the household was taken into account, the average monthly income was under 500 pesos per person or the equivalent of $53 US dollars.

Description of Caregiving Relationships

The majority of participants gave care to a non-spousal relative (Table 1). Overall, the mean age of care recipients was 75 years old, with a range from 56 years old to 92 years old. The median number of years spent caregiving was 5 years, with a broad range from two months to 65 years, reflecting a mix of short-term and long-term caregiving situations. Two women identified the length of caregiving as their “entire lives” since their family members had always lived with them.

The Caregiver Role as Guardians of Health

We asked study participants a series of open-ended questions about their likes and dislikes about being a caregiver and what they believed was the most important aspect about taking care of an older relative. We also asked each study participant about her relative's health and how it had changed over the course of her caregiving experience. In our analysis of the answers to these and other questions, the primary theme that emerged was the caregivers' view that they were guardians of their relatives' health. We found that 27 of 41 caregivers viewed themselves as guardians. Table 2 presents excerpts of interviews with study participants who viewed the caregiver role as a guardian and those who did not. These excerpts represented various caregiving situations, including spousal and non-spousal caregiving and long-term and short-term caregiving. The diseases or conditions that care recipients were reported to have had ranged from mental diseases, such as dementia, depressive symptoms, and schizophrenia, to physical ailments such as arthritis, diabetes, obesity, heart disease, stroke, and old age.

Table 2. Exemplar Quotes of Guardians of Health.

Guardians' Quotes	Context
“I think that she has improved because I believe that company is good for an older person, partly because that way they don't feel alone. They don't feel forgotten about, and I think that yes, she has improved a little [because of me].”	Twenty-three-year-old Tatiana had been caring for her 61-year-old grandmother for almost 2 ½ years at the time of interview. The grandmother required a lot of assistance with basic functions like self-grooming and mobility. However, Tatiana decided to be her caregiver because of the company and attention that she perceived her grandmother needed in her old age. This passage reflected her views on aging as well as on caregiving as a form of protection against poor health.
“I'm not bragging but the doctors have told me when I take her there [to the office]. They say ‘you're a good nurse.’ My neighbors, relatives and friends who come to visit also say that she looks good. It makes me happy that I take good care of her. She [the doctor] says ‘she's not all skinny and [sick] like she used to be.’ I'm not bragging or anything but I think that yes, she gets attention that's necessary…and I try to see that she has the things [she needs] and doesn't lack anything.”	Francisca and her mother had been living together for the last 18 years. Francisca only began caring for her mother two years prior to the interview when the 82-year-old went blind due to diabetes-related complications. In this excerpt, Francisca was commenting on how her care had made improvements in her mother's health.
“Before, she wasn't able to stand up or go to the bathroom [by herself] and [her health] was very bad, very bad. But now her health has improved, including she can walk with a cane and walker…. Yes, the care and affection we all give her…and the support I give her [has made a difference].”	Maria, a 56-year-old divorcee with no children, had been caring for her 88-year-old mother for 5 years because of her mother's heart disease, although the mother had no functional impairment and was physically active. We show in this quote Maria's response to a series of questions on the changes in her mother's health over the 5 years and the impact her care has had on her mother's health.
“They tell him, ‘Don Eugenio you're doing very well, you've conquered the sugar because of your wife's care’…. I'm satisfied to see him as he is [now]. He's conquered it…. When the doctors were seeing him, they told me, ‘How many years has Don Eugenio been sick? It's been 27 years? [He] has been sick for 27 years and look how well he is.’ I think he hasn't become an invalid over the years [because] I have been dedicated to caring for him.”	At the time of the interview, Amalia had been caring for her 64-year-old husband for the last 30 of their 45-year marriage. Don Eugenio suffered a stroke at the age of 34, leaving him with weakness in the limbs, hearing loss, and difficulty with eating and walking. Three years prior to the interview, he was diagnosed with diabetes. In this excerpt, Amalia recalled a recent conversation she had with her husband's medical providers. The excerpt highlighted her opinion of the positive effect her dedicated care has had on her husband's health.
Non-Guardians' Quotes	Context
“Well yes. Because like they [doctors] told me that her treatment—they gave her treatment for the chronic illness she has [bronchitis]. So you can say that [she has been better] for some time. It [the treatment] made her better, it alleviated her.”	We asked 55-year-old Tere if she believed her mother's health had improved over time and what she attributed to the positive change. Her 86-year-old mother did not suffer from a particular disease, but from “old age.” This quote pointed out how Tere attributed the improvement in her mother's health to receiving medical treatment.
“He was feeling bad so I told him, ‘go see the doctor. You can't be like that. So…she [daughter] did me the favor of taking him [to the doctor] and it seems that he is much, much better now…Yes [it's been about a month now that he's been better] because they gave him tests…. and gave him a prescription. They are drops that he's taking but already thanks to God he is much better.”	Guadalupe was caring for her 64-year-old husband who suffered from obesity and had trouble walking. At the time of the interview, Guadalupe had been taking “extra” care of him for three years. We asked her questions about the changes in her husband's health during that time. Although she monitored his diet and made sure “his heart didn't get agitated,” she did not take credit for her husband's improved health.

Study participants who viewed the caregiver role as a guardian saw themselves as directly responsible for the positive changes in their elderly relatives' health. Study participants in the non-guardian group also monitored the changes in their elderly relative's health, but attributed improvements in health to other factors, such as taking medication or receiving medical treatment. In examining these two groups of study participants, we found different patterns in the ways they described the effect caregiving had on their lives. Almost all study participants in the guardian group viewed caregiving as a life-changing event that included having to adapt to a different daily routine, reduce paid employment and social activities, and changed living arrangements to a shared household. However, this pattern was not found among study participants in the non-guardian group.

Emotional Dimensions of Care

The Activities of Daily Living (ADLs) (Katz, 1983) and the Instrumental Activities of Daily Living (IADLs) (Lawton & Brody, 1969) measure the extent to which care recipients cannot perform self-care tasks and are used to characterize levels of caregiving need (National Center for Health Statistics, 2006). Unlike characterizations in the literature, we found that study participants viewed caregiving within the context of their elderly relatives' emotional needs rather than strictly physical ones. Two forms of assistance emerged from the grounded analysis of caregiving discussions: keeping company and watching out. Study participants viewed these forms of assistance as safeguards against poor health or further decline in health. Keeping company and watching out did not refer to one particular care behavior but a number of behaviors that, when examined collectively, reflected the study participants' focus on avoiding setbacks in health by preventing the elder from feeling abandoned, lonely, or unloved.

Keeping Company

Keeping company (acompañar) or companionship (compañerismo) was the most common form of assistance that emerged from the study participants' discussions of caregiving. We asked participants open-ended questions to describe a typical day and the things they usually did to help their relatives. We had them recall their day from the moment they woke up to when they went to bed and describe what they did in the mornings, afternoons, and evenings. We also asked which activity or part of the day they enjoyed the most and the least. We found that 36 of 41 study participants identified keeping company as a form of care, which was referred to in terms of being together (convivencia) and/or conversing (platicar).

Being together (convivencia) had literal and figurative meanings. Being together literally referred to occupying the same physical space or performing an activity jointly, such as going for a walk, to mass, or a medical appointment. Some participants stated they “made breakfast,” “did household chores,” “watched television,” and “knitted” with their elderly relatives. Being together figuratively referred to the sense of closeness or togetherness shared by the caregiver and the care recipient. Some participants described their days with their elderly relatives as “being at peace,” “co-existing,” “sharing with each other,” and “being a family.” The following excerpts illustrated the idea of being together as a form of keeping company:

“We don't do what many people do, push him off to the side, no. We coexist (convivimos) with him very well. My daughter loves him very much.”

“What happen is that as time passes, she needs more affection, more support, more companionship (acompañia). So I have increased the time I am with her.”

“[The most important thing about being a caregiver is] to pass the most time with him, to live together with him (convivir), well, you could say his final days and to record the life he has lived, that which I have lived.”

These excerpts referred to both the literal and figurative elements of being together; study participants physically spent time with care recipients, and being together cultivated the closeness between them.

Conversing (platicar) was also a common form of keeping company that emphasized making the elder feel loved and not abandoned. It should be noted that participants chose to use platicar most often instead of other possible words such as talking, speaking, or discussing. The literal English translation of platicar is “to converse,” but participants used this word to refer to informal conversations. For example, some study participants indicated that their elders, “talk” to them, “transmitted her knowledge,” and “converse with us.” The following excerpts highlighted participants' views on conversing as another form of keeping company:

“What I like most is when she starts to chat with us (nos platica) because she makes us laugh. That's what I like the most, when all of us are together, we are all together, being a family and she makes a funny joke.”

“He was a little shy with me at first because he didn't like it very much that I had to bathe him and it made him embarrassed. But with time, he is accustomed to me, so now he confides in me and he is very comfortable with me, and me too…He tells me (me platica) his secrets, how he lived or when I bathe him, he's not embarrassed anymore, even when I see him [naked].”

These excerpts suggested that conversing represented an intimacy between the caregiver and care recipient through a sharing of stories, ideas, and time. These excerpts also suggested that being together and conversing were not associated with specific physical activities per se but underscored the importance of safeguarding care recipients' emotional health by maintaining a sense of togetherness and closeness through keeping company behaviors. Moreover, study participants were expressing the enjoyable or pleasurable aspects of giving care that stemmed from focusing on the emotional and expressive components of caregiving. Participants also seemed to give positive meaning to otherwise routine aspects of their daily lives.

Watching out

Watching out (estar al pendiente) was the second most common form of assistance that emerged from our analysis of the transcripts, where 28 of 41 study participants reported that they watched out for their older relative. While there was no literal English translation of the phrase “estar al pendiente,” watching out referred to the caregiver's anticipation that the elder would soon have a need, making the elder potentially waiting for some care. The phrase did not refer to elders' expectations or waiting for care. For example, one study participant stated, “I like to be watching out for all these little details for him, that his bed is clean, that he frequently bathes so he's clean and all that.” Two other study participants stated that, “I see how sick he gets so I have to be watching out for him,” and “[I have] to be vigilant and watching out for him [because] he needs a lot of care.” Other participants watched out to make sure that care recipients were “comfortable” and that they didn't “fall,” but slept and ate “well.” Additional behaviors included preparing meals, providing a clean change of clothes and helping with personal grooming.

However, watching out was more complex than the physical labor implied by the various forms of assistance performed by study participants. We found that watching out was related to caregivers' perceptions of elders' risk for emotional decline rather than physical frailty (Table 3).

Table 3. Exemplar Quotes of Watching Out (Estar al Pendiente).

Quote	Context
“What happens is that sometimes I am preoccupied with having to leave her for a few hours because of her old age. She goes to the bathroom and sometimes gets dizzy and could fall. That's the only thing to be watching [out for] and sometimes it's uncomfortable [for me] to leave her alone. That's exactly why I call her on the telephone and ask her “how are you? Do you need anything?” And to be watching out. That's the only worry that stays with me.”	Olga had been caring for her grandmother for 5 years at the time of interview. Five months prior, the 88-year-old grandmother suffered from an embolism. She had since recuperated completely, was physically healthy and needed help only with getting in and out of the bathtub. This excerpt illustrated that watching out had elements of worry and concern, and that watching out was due more to the grandmother's status as an old person than to the state of her health.
“He is a little better because we're here watching out for him. Before we didn't used to [live] here and I was visiting every week and he was sadder, more disheveled like his memory was going. And now he seems a little calmer.”	Thirty-one-year-old Olivia had spent the last 2 ½ years caring for her 56-year-old father who suffered from schizophrenia. This quote pointed to watching out as a way of improving the care recipient's emotional health.
“When we are here [at home], we try to let her be, to watch over her, watching her, her breakfast, that she gets up, gets out of bed…we let her [walk] so that she doesn't feel useless. We try to let her be.”	This quote from Francisca illustrated that watching out was central to keeping her grandmother active and to maintaining a sense of self-sufficiency. This quote complements the excerpt from Olivia that improving the care recipient's emotional health was tied to watching out behaviors.

Over half of study participants who identified watching out as a form of care were caring for non-impaired elders. In other words, study participants were watching out as a form of prevention for healthy elders who were perceived to be at risk for a decline in health.

Of the study participants caring for elders with at least one ADL, about half viewed watching out as caregiving. One of these study participants was 46-year-old Rosi. Rosi gave care to her very frail 72-year-old mother who was diabetic and suffered from multiple ailments. She helped her mother with bathing, walking, dressing, and toileting, and was “constantly watching out precisely because she [mother] is no longer self-sufficient.” However, Rosi was not the typical study participant. She left her husband so that she could care for her ailing mother. A recent caregiver of two months, she admitted that, “It's a daily fight (lucha diaria) of trying so that she's [still] with us and we get to keep having that relationship.” She also stated, “I'm watching out so that she [mother] doesn't have even graver consequences…. The most precious gift that we can give to another human being—be it our family or not—is time and attention.”

Although Rosi was an exceptional case, over two-thirds of study participants were constantly on watch to ensure that care recipients were well attended, from both physical and emotional points of view. From a physical point of view, participants ensured that care recipients were constantly being watched over and that their basic care needs were met on a daily basis. From an emotional point of view, they gave the kinds of assistance that they believed would maximize elders' feelings of being loved and wanted.

Discussion

The aim of this paper was to describe how a small sample of Mexican women socially constructed elder care in terms of their role as caregiver and the behaviors they defined as caregiving. Our findings suggested that caregivers viewed themselves as gatekeepers to maintaining or improving the health of their elders through watching and waiting behaviors. Women defined forms of assistance as caregiving if the assistance met what they perceived to be the emotional needs of their elderly family members. While study participants performed tasks traditionally associated with caregiving, their predominant concern was for the emotional welfare of their older loved ones.

Although there is a dearth of literature on elder caregiving in Mexico, we can place our study findings within the context of the abundant literature on gender role functioning, and particularly the female marianisma role (Bridges, 1980; Nader, 1986; Peñalosa, 1968). The marianisma role is inculcated from early childhood and is particularly influential in women's expected behaviors of femininity, submission, weakness, reservation, and virginity. This gender role is based on the emulation of the Virgin Mary in the Catholic religion and has also been referred to as la madre abnegada (Hubbell, 1993), meaning “self-sacrificing mother.” Thus, the image is that the Mexican mother sacrifices her own needs and happiness for the sake of her children and family regardless of the personal hardship (Finkler, 1994; Hubbell, 1993).

La marianisma has been examined in the literature in terms of the marital relationship and child-rearing. Although we did not specifically ask study participants for their views about la marianisma, our findings suggest that this role extends to the caring of older relatives because participants viewed themselves as protectors of elders' physical and emotional health. The level of care they provided through keeping company and watching out implied that some amount of sacrifice was involved to constantly be on call and available to care recipients. It also appeared that caregivers created meaning out of everyday behaviors, such as preparing meals and doing housework, by defining them as watching out. They created their social realities as caregivers by elevating the importance of everyday behaviors—as such, they fulfilled the marianisma role within the context of their overall lives because these behaviors benefited all members of the household, not just the elder.

We can also compare our study findings with the abundant North American literature on caregiving and social support. Although there is no universal definition and measurement of social support in the literature (Mutran, Reed, & Sudha, 2002), most studies incorporate similar conceptualizations that include elements of instrumental and affective aid (House & Kahn, 1985). Instrumental aid refers to the provision of tangible goods, services and assistance whereas affective aid refers to expressive or emotional support, involving the provision of empathy, love, and trust (Mutran et al., 2002). Our findings are particularly similar to earlier research on affective aid (Abel, 1991; Sherman, Ward, & LaGory, 1988; Spitze & Logan, 1990), which documented the diverse range of care that wives and daughters provided to the frail elderly. This early body of work suggested that affective aid may be as important to the health of older care-receiving adults as instrumental aid. Our findings that caregivers focused on forms of assistance that had emotional elements are supported by this early research.

Although Abel and others suggested a course of research in expressive support and the emotional aspects of caregiving, research has predominantly focused on quantifying the instrumental aspects of the caregiving experience. A recent meta-analyses of gender differences in caregiver stressors, social resources and health (Pinquart & Sorensen, 2006) found that caregiving was most commonly measured by the number of hours of care provision, the number of caregiving tasks, and the number of months in the caregiver role. Our findings suggest that the quantification of caregiving by type, duration, and frequencies of activities may not address the emotional aspects of care or adequately incorporate the meanings of activities that caregivers in our study identified as giving care, such as keeping company or watching out.

Most research on expressive support as a form of social support has focused on the relationship of support to caregiver well-being (Pinquart & Sorensen, 2006). Research has particularly focused on the social support that caregivers receive as one mediator of stress and other health outcomes, instead of expressive support as a form of care toward the care recipient. One study (Robles, Rizo Curiel, Camarena García, Cervantes Coles, Gómez Medrano, & Siordia González, 2000) conducted in Guadalajara, Mexico, suggests that expressive support may indeed be a form of care of older Mexicans. The study looked at the social network and social support among poor elderly in Mexico's second largest city and found that social support is more frequently in the form of emotional support than either financial or instrumental forms of support. Although Robles et al. measured social support according to a North American model (Barrera & Ainlay, 1983), the findings are nonetheless consistent with ours.

Limitations of Study

There are some important limitations to our study. First, men were purposely excluded from this study. We realize that focusing solely on the experiences of women does not advance our understanding of gender differences in the experience of caregiving. However, we focused on female caregivers as a beginning for the knowledge development of the Mexican literature in this area. We acknowledge that in response to Mexico's lack of long-term care provisions for the elderly, unskilled homemaking services, including housekeeping and elder care, are widely available in informal labor markets. However, the majority of our study sample had low household incomes and limited resources for purchasing unskilled labor. Therefore, this study did not take into account women's limited life options in caring for older relatives, but rather the meaning they gave to such endeavors within the context of their life situations. Additionally, we only interviewed each study participant once. Multiple interviews would have given study participants the opportunity to think through their experiences, providing richer data. Lastly, the generalizability of our study is limited due to the sample's characteristics. The caregivers in this study lived in an area of Mexico city insulated from transitory migration and other population changes, factors which might influence their life experiences and views on elder care. However, similar to our study participants, many Mexican families are poor and live in small isolated suburbs around the Federal District and throughout the country (Instituto Nacional de Estadística Geografía e Informática, 1998a, b).

Conclusion

Our exploratory study suggests that female caregivers view themselves as instrumental forces in the lives of their elderly family members. It could be that women care for their elderly family members out of necessity because of the country's lack of long-term care options. It could also be that women experience caregiving as a role that was cultivated during early childhood and occupied throughout the lifetime, as mothers, daughters, or wives. This explanation would be consistent with research findings on populations in developed and developing countries (Holroyd, 2001; Hsu & Shyu, 2003; UNIFEM, 2000; Wallhagen & Yamamoto-Mitani, 2006). Our findings may also reflect the broader institutionalization of cultural ideals within the Mexican family that drive women to reciprocate for assistance they received earlier in life. Research on intergenerational relationships has shown that transfers of assistance within the family indeed occur throughout the lifetime and that the exchange is often both instrumental and affective (Antonucci, Akiyama, & Birditt, 2004; Lowenstein, Katz, & Daatland, 2004). Moreover, research in rural China (Silverstein, Cong, & Li, 2006) and five countries in Europe and Israel (Lowenstein, 2007) has found that older adults benefit emotionally and physically from these transfers, consistent with the caregivers' perceptions in this study.

An alternative explanation of women's views of the caregiver role may be their exposure to conditions of poverty, dependency and powerlessness. Some literature suggests that Mexican women are a socially and economically vulnerable group and are not generally in as much control of their existence compared to men. This body of work argues that women have limited access to the political system, employment, and positions of power and authority (Chant, 1985; de Oliveira & Garcia, 1990), leading to subordination that is institutionally structured and culturally rationalized (Browner, 1986; Finkler, 1994; García & de Oliveira, 1997; Nader, 1986). Women may therefore become caregivers in response to their subordination within domestic and social spheres, leading them to attach importance to this prescribed social role to justify their actions. In other words, the women in this study may have given meaning to caregiving to elevate their otherwise compulsory situation or responsibility. Further work is warranted in this area to understand the underlying factors that motivate women to be caregivers.

Further research is needed to test the caregiving concepts identified in this study, especially in terms of an intergenerational perspective. Moreover, our findings on the emotional aspects of care suggest that caregiving is indeed socially constructed in ways that are not reflected in medical and policy literatures on the subject. Researchers in Mexico should think about how caregiving is measured on existing instruments before adapting them for widespread use because current measures may be problematic in the Mexican population. We further suggest that more research is needed to understand what exactly comprises caregiving—particularly under-recognized activities that caregivers engage in—before it can be properly measured.

Research on elder care in Mexico is in its early stages, and our study is one small but important contribution to understanding the role that Mexicanas play in the lives of their elderly family members. Women's roles as elder care providers, however, are likely to evolve with changes in social and economic trends, such as women's delayed fertility and increased entry into the workforce (Gutierrez-Robledo, 1990). A review of the research on the changes in gender role responsibilities and family functioning in Latin America over the past 100 years suggests that conflict within the home as result of these changes will impact elder care (Chant, 2002). One study suggests that working women in Mexico are becoming more reluctant to take on the responsibility of caring for older relatives (Varley & Blasco, 2000). Changes in family role functioning may therefore compound an already difficult situation created by pressing structural constraints like working outside the home and putting off motherhood. Moreover, the negative health effects of caregiving have been well established in other countries (Pinquart & Sorensen, 2003, 2006, 2007). The findings from an exploratory study suggest that Mexican women are indeed experiencing caregiver burden (Mendez-Luc, Kennedy, & WSallace, 2008). More research is needed to understand women's abilities to care for their older family members in a changing social environment and the effects of caregiving on their health. Research in this area can provide an empirical foundation for developing social policies that assist and support women's caregiving efforts as they struggle with the competing demands on their time.