When parents are not present: Decision-making dynamics for young children’s health and illness in migrant-sending households in rural Cambodia

Abstract

As migration rates increase globally, an increasing proportion of children in low- and middle-income countries live apart from their parents. In particular, skipped-generation households are becoming more common, with grandparents serving as primary caregivers for young grandchildren whose parents migrate and provide remittance support. Yet, how families living apart make decisions for young children’s healthcare is not well understood. Processes of care seeking for children’s health are examined in Cambodia, which has high rates of internal and international migration, and an increasing number of skipped-generation households. We conducted semi-structured in-depth interviews with grandparents who are primary caregivers for children under ten in skipped-generation households in rural migrant-sending areas. Respondents discussed their family’s migration history, care seeking and decision making for grandchildren, dynamics with absent migrant parents, and sources of support. Decisions for children’s care are intimately tied to the availability of migrant remittances, which are the modality of the relationship between grandparents and parents. Additional remittance support allows grandparents to make decisions for grandchildren’s health based on convenience rather than cost, given the burdens of caring grandparents have taken on a result of the parent’s migration. Instrumental supports such as in-kind assistance and informational supports further facilitated children’s access to healthcare. This qualitative study informs and explains findings from recent quantitative studies of migration and child health.

I. Introduction

Many low and middle-income countries (LMICs) are experiencing high rates of migration among young adults. Over the last several decades, this shift has contributed to a growing prevalence of young children globally living apart from their parents, in the care of grandparents, as parents migrate for work (Sadruddin et al., 2019; Zimmer and Treleaven, 2020). Although the determinants of care seeking and familial decision-making for children’s health are a common area of study across diverse settings globally, how these processes shift in the face of parents’ physical absence has received little attention. Yet, identifying how families living apart negotiate access to healthcare for young children is necessary to understand how contemporary migration, family separation, and complex living arrangements shape disparities in access to healthcare, and the ways in which families work together across distance—or fail to—to return children to health.

In this study, we aim to examine how processes of curative care seeking for young children’s acute illnesses shift in the face of parental absence, and to identify how absent parents and caregivers engage in decision-making for children across physical distances. We identify forms of support, resources, and knowledge utilized by grandparent caregivers in the altered family structures brought about by parents’ migration, and outline how various actors are involved in decisions about care, and how this differs across dimensions of children’s illnesses. We examine the specific case of skipped-generation households, that is, households comprised of grandparent(s) and grandchild(ren) with no middle generation (parents), drawing on data from Cambodia, a country with a high prevalence of skipped-generation households, a complex health system, and poor child health indicators (Meessen et al., 2011; National Institute of Statistics et al., 2015; Zimmer and Treleaven, 2020).

II. Background

1.1. Shifting roles in migrant-sending households

Familial roles, structures, and living arrangements shift with migration, impacting the daily lives of children left behind in the care of non-parental caregivers. Often, parents and family members who are not co-resident still provide financial and emotional support to one another, which reinforces familial bonds in the face of migration and physical distance, creating a “modified extended family” (Knodel and Saengtienchai, 2007). Parents who remain active in the lives of their children left behind might be considered “shadow household” members (Caces et al., 1985). However, because these parents are not physically present, caregiving tasks previously performed by migrant parents must be assumed by the child’s caregiver(s), creating a new distribution of caregiving labor within the family.

Several theoretical perspectives offer insight into post-migration decision-making processes for children and their health and well-being in migrant-sending households. Baldassar and Merla (2014) conceive of care as circulating throughout migrant families; as families maintain their bonds across physical distance, migrants and the left-behind each provide certain types of support to the other throughout the family’s migration experience and life course. These include financial, social, and instrumental supports. This characterization of care is particularly relevant for understanding the motivations, structure, and dynamics of the skipped-generation household, that is, households comprised of grandparents and grandchildren where the middle generation is absent. For example, while migrant parents provide financial support to grandparents and children left behind, grandparents provide practical care to their grandchildren, who in turn are expected to later provide financial and practical care to the older generations. Given that migration restructures roles in the family, especially in the left-behind household, shifts in familial power dynamics may follow. Migrant parents as primary earners generally control financial resources, though grandparent caregivers also hold a form of power due to their prerogative to upkeep negotiated caregiving commitments (Merla, 2014).

However, Baldassar and Merla’s framework does not sufficiently consider the social and/or structural barriers to providing care that some families left behind may face. When migrant parents fail to provide financial or instrumental support to grandparents and young children left behind, grandparents may be unable to adequately care for their grandchildren. In their framework of pathways of care linking grandparent caregivers and grandchildren, Sadruddin and colleagues (2019) underscore the importance of both interpersonal and structural context in determining the ways in which grandparental care influences grandchildren’s health outcomes. For example, grandparents’ own health status, relationship with migrant parents, and community context and availability of health services may moderate the relationship between grandparents’ caregiving and children’s health. Given that none of the studies included in their systematic review explicitly examined grandparent interactions with the health system, the ways in which these social and structural characteristics shape healthcare utilization among children left behind remain under-theorized. Moreover, considering the explicit case of migrant-sending skipped-generation households, existing gender norms, familial roles and expectations, and migrants’ experiences may shape the ways in which various families interact to “do family” across distance—creating important differences in the ways this specific type of modified extended family makes decisions for children’s health.

1.2. Decision-making processes for children’s health

When a child falls ill, parents and caregivers use multiple metrics when deciding how to address the illness. These include cost; distance; perceived quality and provider reputation, prior experiences with specific providers, facilities, and the medical system; and local understandings of disease etiologies, among others (Akin and Hutchinson, 1999; Colvin et al., 2013; Leonard, 2014). Colvin et al. (2013) describe a framework for household recognition and response to children’s illness. After recognizing illness, parents and caregivers may first pursue home- or community-based treatment options through informal drug sellers or traditional healers, and later access formal biomedical care in public or private facilities. In general, care seeking moves from within the home to outside, especially as illness severity increases. Along this trajectory, parents or caregivers may consult with others for advice, wait to see if symptoms resolve on their own, or negotiate access to care, resources, or financial or instrumental support—potentially delaying treatment. Here, we consider financial support separately from instrumental support, which we define as time, in-kind, informational, and other social support. Considering the potential array of financial, informational, and social resources necessary to respond to illness, care seeking for children is a dynamic and potentially complicated process, especially when the etiology of illness is unknown (D’Acremont et al., 2014; Scott et al., 2014).

The physical distance between migrant parents and their children, changes in caregiver roles, and availability of financial, social, and instrumental resources may shift care-seeking and decision-making processes in migrant-sending households in specific ways. Depending on kinship systems, gender norms, and other factors that confer status or a lack thereof within the family or community, different actors (i.e., mothers, grandparents, heads of household) in the process of seeking health services for young children have varying degrees of agency to make decisions for a child’s health, and may each face different constraints in accessing care for a child (Colvin et al., 2013; Richards et al., 2013; Scott et al., 2014). Thus, in the face of parental absence, the relative roles and processes through which grandparents, parents, and other actors negotiate healthcare seeking for children may vary, particularly where grandparent caregivers hold differing agency, health knowledge, and access to financial and instrumental resources than parents. However, the existing frameworks describing care-seeking process for child health, including Colvin et al. (2013), generally center a child’s mother, and fail to consider scenarios where a mother is absent.

1.3. Study setting

We examine these processes in rural Cambodia, a setting with high rates of out-migration among young adults. Internal and international migration are critical livelihood strategies for many households, driven by limited job opportunities and low agricultural incomes (National Institute of Statistics, 2013). About 35% of rural Cambodian households have at least one member who has migrated (World Food Program, 2019). Rural-rural migration accounts for 13%, rural-urban 57%, and cross-border 31% of all migration, with the majority of rural-urban migration to the capital city, Phnom Penh, and the majority of international migration to Thailand (National Institute of Statistics, 2013).

Concurrent to economic development, widespread migration has induced important changes in Cambodian family systems and living arrangements. It is estimated that nearly half of migrant-sending households include a child living without at least one parent, and almost 20% have an elderly grandparent as the primary caregiver (Zimmer and Van Natta, 2018). Since 2000, the proportion of young children residing in multigenerational or skipped-generation households has increased significantly, and currently, almost one-tenth of Cambodian children reside in skipped-generation households (Zimmer and Treleaven, 2020).

Cambodia has a pluralistic health system, ranging from public-sector primary health facilities and hospitals, pharmacies, private or NGO-managed hospitals, or medical professionals providing services from their own or at patients’ homes, to untrained drug sellers or traditional healers (Kru Khmer). Informal providers and lower-level private trained providers mostly render curative healthcare, while public sector and higher-level facilities are often farther away in more urban areas. Private and informal providers are poorly regulated. User fees in the under-funded public sector prove problematic for many Cambodian families; beyond the cost of services and medicines, parents and caregivers must consider costs associated with transportation, food and accommodations, and lost income, which are particular challenges for the poor (Khun and Manderson, 2008). However, health subsidies are in place for the poorest families through the donor-funded Health Equity Funds (HEF) program, which subsidize or eliminate public-sector user fees and transportation costs for the poorest Cambodians (Flores et al., 2013). Under the HEF program, recipients receive fee waivers and providers are reimbursed directly for their services.

While the health system and subsidies have expanded over the last two decades, and child mortality has declined, Cambodia has high rates of undernutrition, endemic dengue and malaria, and low coverage of recommended treatments for diarrhea (National Institute of Statistics et al., 2015). Given trends in migration patterns and living arrangements among young children; the complex, pluralistic health landscape; and sub-optimal health indicators and utilization rates among young children, Cambodia offers a compelling setting in which to study young children’s health and family dynamics among migrant-sending, skipped-generation households.

III. Methods

From February to June 2015, we conducted semi-structured in-depth interviews with 25 grandparents who were primary caregivers for one or more grandchildren under ten years of age in migrant-sending households. We included households where both of the grandchild’s parents migrated out of the village of origin and had a different primary residence at least half of the time. There were no age or gender restrictions for grandparents.

Data collection

This study was sited in four provinces in eastern Cambodia (Kampong Cham, Kandal, Prey Veng, and Tbong Khmum), chosen for their prevalence of migration and diversity in migrant destinations. Migration patterns in this region vary by distance to the capital city and access to major roads, though migration to Phnom Penh, the capital city, and to Thailand is common in each province. Migrants from villages closer to Phnom Penh tend to return to their households of origin more frequently, while those who migrate to Thailand may only return once per year. Kandal surrounds Phnom Penh and is the most urbanized of the provinces. Kampong Cham was the most populous province until it was split into Kampong Cham and Tbong Khmum provinces in 2014. For the purposes of this analysis, we treat the two as single province. Kampong Cham has a large provincial town and several smaller market towns along national highways. Prey Veng is the most rural of the three. Although it borders Vietnam, migration to Vietnam is rare.

Within these provinces, districts with high rates of migration were identified based on a nationally-representative survey of migration (Ministry of Planning, 2012). Where districts could not be identified due to irregular romanization of Khmer place names, alternate districts were selected from a commune database based on location. The purposive sample of districts aimed to capture areas that ranged in distance from provincial capitals and highways. In each province, two geographically distinct districts were visited, and households from at least two villages within each district were interviewed. Participants were purposefully sampled by migrant destination to capture a sample representative of the different locations to which Cambodians migrate.

Within districts, we approached officials to introduce the study and obtain assistance in identifying villages for sampling. We aimed to identify villages that were likely to include skipped-generation households and where village chiefs would be amenable to allowing research to take place. District officials recommended specific villages, and in most cases, helped approach village chiefs. An interviewer introduced the study and its objectives to village chiefs, and obtained assistance in identifying eligible households. In cases where the village chief did not help identify eligible households (approximately one-third of villages), an interviewer approached the closest shop or restaurant. In these villages, households were located with the assistance of a shop worker or customer. Interviewers then approached households, accompanied by the village chief in about one-quarter of the sample. The interviewer introduced the study to the head of household using an IRB-approved script. Participants were administered informed consent by the interviewer.

Interviews lasted approximately 60 to 90 minutes, and were conducted in or near participants’ homes in a quiet setting by trained Cambodian interviewers in the Khmer language. One grandparent was interviewed per household. In households where two grandparents were present, grandparents self-selected who would complete the interview. At the close of the interview, participants completed a short demographic questionnaire that included a household roster, children’s school attendance, and socio-demographic characteristics of the migrant parent(s). Field notes were written after each interview to document the household context and other reflections.

The interview guide used a semi-structured format, piloted iteratively in two rural districts in Phnom Penh and Kandal provinces. Topics included parents’ migration history, daily life, care seeking for grandchildren’s illnesses, financing care, and sources of support and health knowledge. Participants were asked about migrant parents’ involvement in decision-making for health and other domains, and receipt and use of remittances. Interviews were transcribed in Khmer, and then translated to English. English-language transcripts were reviewed for accuracy by the interviewer, translator, and the authors.

The study was approved by [AUTHOR INSTITUTION IRB] and the National Ethics Committee for Human Research in Cambodia. Due to the low literacy rate among the target population, verbal consent was sought. Participants received soap worth approximately $1 USD as a token of appreciation for their time.

Sample

The sample consisted of 20 women and five men. Approximately half of the households sent at least one migrant to Phnom Penh; two households to other rural areas within Cambodia; ten to Thailand, and one to Malaysia. Several households reported the parent(s) had migrated to a different destination prior to their current destination, most commonly to Thailand prior to Phnom Penh. In a majority of households, the parent(s) had migrated for more than five years. Parents from households in Prey Veng and Kampong Cham had generally migrated for longer durations than households in Kandal.

Grandparents ranged in age from 42 to 74 (median age 61.5). About half of grandparents were married, the rest widowed. Many grandparents cared for more than one grandchild under ten. In a majority of households, grandparents had more than one adult child who migrated, but only a few cared for grandchildren from more than one adult child. Almost all grandparents had provided some level of care to their grandchildren prior to their parents’ migration, and about one-quarter reported that they had cared for their grandchild(ren) from infancy. Over half of grandparents reported other adult children or relatives living in the same village. While almost all households in the sample were comprised only of the grandparents and grandchildren, two households included other adult children, that is, aunts and/or uncles of the grandchildren. Grandchildren ranged in age from a year and a half to fourteen. Based on respondents’ self-ratings compared to other households in their villages, interviewer observations regarding housing characteristics and durable goods, and respondents’ reports of food insecurity, households ranged from very poor to middle class by rural Cambodian standards, with most in the sample considered poor. A small number of households reported food insecurity. Socio-economic status (SES) was more closely tied to migrant destination than duration. Migration to Thailand represented a greater financial risk than migration to Phnom Penh, as well as greater reward: the richest and the poorest families in the sample sent migrants to Thailand.

Data analysis

The analytic approach followed the contemporary grounded theory tradition (Charmaz, 2006). English language transcripts were uploaded to Atlast.ti (Scientific Software Development). Interviews were initially coded line-by-line for emergent themes using an open coding method by the first author. An initial codebook was created and refined during this process. A sub-sample of five interviews were independently coded by the second author. Together, we assessed agreement between codes and further refined the codebook. The first author then reviewed and recoded all interviews. In addition to line-by-line coding, we created descriptive profiles for each household. These included households’ migration and financial history, a narrative of children’s illnesses, and descriptions of daily life post-migration as described by the respondent. Profiles provided context for grandparents’ attitudes towards migration and caregiving, familial motivations and decisions around migration, and the social and cultural environment in which they lived and made healthcare-related decisions for their grandchildren. Profiles aided in identifying broader commonalities across households and trajectories, the intersections of different phenomena of interest, and in highlighting differences of experience (Dubbin et al., 2016). Decision models for specific illnesses were mapped to identify care-seeking trajectories and points of differentiation for children’s illnesses (Miles et al., 2014).

Using the query tool in Atlas.ti, specific codes and code families were examined and analyzed in detail, including variation among households and relationships between codes. Code groups included barriers to care and sources of support; decisions about children’s health; financing healthcare; impacts of migration; trajectories of care; and grandparent attitudes towards raising grandchildren, grandchildren’s health, and care seeking. Throughout the data analysis, we wrote analytic memos to identify emerging themes and phenomena in the data. Analytic memos were informed by analysis using specific codes and code families based on Atlas.ti queries, as well as analysis using the descriptive household profiles and decision models. We discussed emergent themes and corroborated findings with existing literature and quantitative data to triangulate our findings and minimize personal bias. These analytic memos formed the basis of the findings presented in this paper. We present selected quotations illustrative of key themes.

IV. Results

We found that the altered caring structures in families where young children lived apart from their parents due to migration shaped decision-making and care seeking trajectories when these children fell ill. Grandparent caregivers managed these processes in consideration of parents’ physical distance, their own social and financial resources, and social and structural barriers to care for their grandchildren. In the sections below, we examine the impacts of parental absence on decision-making and care seeking for young children’s health, with attention to the roles of SES, gender, forms of support, and consequences of financial vulnerability.

4.1. Restructured roles and responsibilities in skipped-generation households

After migration, both parents’ and grandparents’ roles shifted significantly. In most households, the division of childcare labor was clearly defined, even in households where migrant parents had only been gone for a short time. While grandparents took on the vast majority of caregiving tasks, parents were responsible for securing a livelihood that provided for the three generations. Grandparents held greater responsibility for daily caregiving and decision-making for young children. With these additional daily demands, many grandparents reported “difficulties” in balancing their grandchildren’s needs with their own work and needs. These burdens took a toll on grandparents’ physical and mental health, leaving them “depressed,” “weak,” and “exhausted,” especially older grandparents within the sample.

Generally, households followed a traditional gendered division of labor. In most households with two grandparents, grandfathers were responsible for agricultural and income-generating activities, while grandmothers took on household tasks such as cooking, cleaning, and caregiving. However, gender norms were upended in single-grandparent households. In the few households where grandfathers lived alone with grandchildren, and in one household where the grandmother was blind, grandfathers assumed the domestic tasks typically performed by women. A widowed grandmother in Prey Veng caring for her two school-age grandsons reported “everything is difficult, because I’m alone being both man and woman.” Households headed by single grandparents were among the poorest in the sample, and many of these grandparents expressed greater stress related to caregiving.

4.2. Trajectories of children’s healthcare utilization in grandparent-headed households

The care-seeking trajectories described by grandparents reflect the diversity of healthcare options in Cambodia’s pluralistic health system. Grandparents sought care for their grandchildren at a variety of facilities, and tended to select sites of care by weighing the treatment needs of their grandchild and available financial and instrumental resources. Poorer and single grandparents experienced greater constraints in reaching more distant and higher quality health facilities. Grandparents described a wide range of symptoms and illnesses, from frequent cases of cough, cold, diarrhea, fever, and urinary tract infections, to rare instances of injuries such as fractures and burns or conditions requiring urgent skilled medical attention, such as seizures. Care typically began in or near the home at local pharmacies, with private doctors or public primary health facilities mentioned by almost all grandparents in the sample.

Grandparents deviated from their usual sites of care when their grandchild’s illness became more serious and required new or more advanced treatment strategies. Many grandparents described a more rapid response involving higher-level providers for care when they deemed a grandchild’s illness “serious”, such as dengue, typhoid, seizures, or acute injuries, or when symptoms like fever did not resolve. A grandmother caring for two grandchildren with her husband in rural Prey Veng described how she differentiated illnesses:

I observe his symptoms. If it is not severe, I will not bring him to the hospital. I sometimes buy from the small medicine stall in the village. I just take my grandson to visit the doctor and then the doctor provides medicine for him…If the symptoms are not severe, I buy only medicine. When he got meningitis, I took him to Phnom Penh and then his mother took him to Kantha Bopha (a national pediatric referral hospital).

In this family, where four of the grandparents’ five children migrated to Phnom Penh and the fifth to another rural province, minor illnesses, those which the grandmother felt competent to treat herself, were managed at home or in the village. Given her lower SES, the grandmother preferred to treat at home if possible to avoid user fees, but she was able to pay for treatment with a private doctor when needed and accessed the national hospital by relying on her daughter’s occasional remittances—even this small amount of financial support eased the grandmother’s access to the private doctor. When illness escalated or in more serious cases, most grandparents recognized the need for more intensive or higher-level care.

For a number of grandparents, timing of illness and competing demands on their time played important roles in determining where and how they sought care for a sick grandchild. Because almost all grandparents expressed a desire to provide some form of treatment immediately upon recognizing illness, many described alternate strategies if illness struck when a usual or preferred provider was not available. Rather than wait until morning, grandparents sought out informal or private-sector providers with evening or weekend hours, or those who would make house calls, often at greater expense. A poor, widowed grandmother had cared for two grandchildren in Kampong Cham since the eldest was three months old as their parents had migrated to another rural province. She noted that she was “always in debt” with a local doctor, even though she felt the public health center was high quality and lower cost. “Weekend, public service is not available [at the health center]. The private doctor is always there for us. Even though I have nothing, I want my grandchild to get better. So, I call him no matter what.” Like many grandparents, initiating treatment was her utmost priority, regardless of the financial consequences—a common sentiment even among low SES families who could ill afford the extra expense for timelier and/or home-based treatment. Particularly for grandparents who took on additional domestic and/or agricultural labor after their child(ren) migrated, weighing cost and convenience, home-based care was particularly attractive given the constraints on their time.

Cost was frequently mentioned as a consideration in care-seeking decisions, though it did not consistently drive or dissuade grandparents to utilize a specific site of care. Rather, payment method—and whether a specific provider accepted deferred payment—was more important to most grandparents, particularly poorer households who could not rely on regular remittances. For example, while some grandparents perceived the user fees at public health centers to be low, others avoided these centers because payment was required at the time of treatment. A poorer grandmother caring for three school-age grandsons and a three-year-old granddaughter in Kandal noted that despite continued financial challenges, the family’s livelihood had generally improved after her daughter migrated to Malaysia and more recently to Phnom Penh. She stated, “We can owe some money and repay later. We do whatever it takes to get my grandson better.” Payment flexibility allowed grandparents to avoid delaying care, making the potentially overall higher cost of private sector care acceptable.

4.3. Managing children’s health when no parent is present

Because parents’ migration shifted caring structures in skipped-generation households, grandparents were placed at the center of healthcare decisions for their grandchildren. Grandparents were responsible for recognizing illness, accruing resources for care, and ensuring their grandchild obtained the care he or she needed. With parents at a physical distance, grandparents managed their participation in care, only involving them in specific cases, such as when a child’s needs escalated, or for specific aspects of care, such as payment.

Grandparents managed parents’ knowledge of their children’s health and illness through telecommunications. For a few families, the high cost of international calls constrained communications to parents living abroad, though all but the poorest grandparents in the sample reported owning a mobile phone. In cases where the migrant parent was within Cambodia, most grandparents reported weekly or even daily contact with absent parents. Yet, when their grandchild was ill, most grandparents chose only to reach out to parents in limited situations. Some grandparents felt conflicted as to whether they should contact a migrant parent when their child had an illness they felt competent to manage on their own, while others deliberately did not tell parents about the illness until the child recovered. Grandparents more frequently reported contacting parents when the parent was within Cambodia versus abroad. A grandfather in Kampong Cham whose son went to Malaysia explained he “does not want them to worry. When they call, we just tell them that the child was sick, and now is fine.” The grandfather recognized the lack of control migrant parents yielded from a distance. Yet, a majority of grandparents did not hesitate to contact parents in serious cases, generally conditions that required higher-level care beyond the village.

Most grandparents made decisions about where or how to initiate care for a sick grandchild by themselves if widowed, or in consultation with their spouse. Occasionally, they consulted other nearby relatives or a neighbor. Very few consulted with migrant parents at the onset of illness, except for a minority of grandparents who wanted to confirm their plan for care or to ask the parent’s preferences. These grandparents reported regular communications with the migrant parent and described strong relationships despite the distance. Grandparents who undertook decisions for their grandchild’s healthcare without the parent’s input reported doing so for several reasons. Often, these grandparents felt capable of managing illness without input from the parent, especially for common illnesses such as fever, diarrhea, and cough. A grandmother and grandfather in Prey Veng had seven adult children who migrated from their village, and cared for four grandchildren from two children. When asked if she spoke with the children’s parents before buying medicines from a local pharmacy, the grandmother replied, “I never consult them because they are far away from home...I will handle it. I do not want to affect their business.” Other grandparents felt that absent migrant parents “did not know the situation well.” Grandparents drew on their own caregiving knowledge to take ownership, as many noted that providing care was their “responsibility,” regardless of the quality of their relationship with the parent and the length of time they had cared for the child.

In many cases, parents were not engaged until additional advice or resources for care were needed. In cases where a parent expressed a desire for a child to be seen by a specific provider or to receive a specific medicine, grandparents typically followed the parent’s instructions without issue. However, a few grandparents described scenarios where they actively ignored a migrant parent’s stated preferences. This generally occurred when the grandparent faced constrained agency, that is, when they did not have the financial resources at hand to carry out these requests immediately. In these few cases, grandparents expressed confidence in the trajectory they had chosen despite going against the parent’s advice.

More commonly, grandparents contacted parents specifically for financial support. A grandmother in Kampong Cham had a son who worked in construction in another rural area. She cared for his two sons with her husband, and said, “When my grandson just had a typical flu and fever, I did not call his parents. But, when he got sick and needed an injection, I called to inform them and asked them to transfer money to me because I did not have any.” For many grandparents, securing financial support was the primary goal of contacting parents, while informing them of illness was a secondary aim. Calls were also placed when a grandparent needed to decide how to proceed with the child’s healthcare, particularly in regard to paying for care at another provider. If they did not call at the time of illness, many grandparents did, however, later relay news of the child’s illness after the child recovered. For some, this was an opportunity to ask for additional money to cover treatment costs. For others, it was simply to keep parents updated on their child’s well-being. Several grandparents neither notified parents when the child was ill nor after the fact, typically those who were unable to make outgoing international calls.

4.4. Safety nets, social networks, and the role of remittances: Forms of support for children’s health

Grandparents made use of various types of support to care for their grandchildren during times of illness. These included instrumental support from other family members and neighbors in their social networks as well as institutionalized forms of support such as subsidized user fees at public sector facilities. Remittances from migrant parents played a particularly important role in care-seeking trajectories, and facilitated new possibilities for care. These sources of support aided grandparents in navigating and managing their grandchildren’s illnesses, and shaped the ways children utilized healthcare. However, while many grandparents described experiencing “stress” and “difficulties” in the process of seeking healthcare for their grandchildren, very few mentioned sources of socio-emotional support, and only one reported receiving such support from a migrant child.

About half of grandparents relied on other nearby relatives or neighbors for instrumental support in caring for grandchildren, most often other adult children. In two households, other co-resident adult children provided regular assistance with caregiving and household expenses. Widowed grandparents were especially likely to report instrumental assistance from other relatives and neighbors. Relatives and neighbors assisted grandparents by watching grandchildren so grandparents could work, attend religious ceremonies, or rest; by providing food, reducing food insecurity; or by consulting with grandparents to make healthcare or other decisions related to the grandchild. In some families, especially where the grandparent was in poor health, other relatives or neighbors took children to a health facility or procured medicine from a nearby market or pharmacy to treat the child at home. For example, two grandmothers in Kampong Cham reported that a neighbor would take their grandchild to the health center via motorbike because they did not have one; the widowed grandmother in Prey Veng relied on her other children living nearby to do the same. Though the level and specific types of support varied widely across families, such support shifted the possible trajectories of care for sick grandchildren by facilitating children’s timely entry to care at better equipped facilities outside the village, such as primary health centers and district hospitals.

For some grandparents, trusted neighbors provided informational support when they needed to make decisions about where to seek care, reducing their emotional burden. Specifically, neighbors offered information about the quality of facilities and services. A grandmother from a middle-class family in Prey Veng was raising her three-year-old grandson with her husband while her daughter worked in Thailand. She recalled, “I went to the hospital based on a recommendation from my neighbors. I asked them which hospital is good and which hospital is bad. I asked them and if they said that this place is good, I would follow their recommendation to take my grandson to get treatment there.” She previously had a poor experience at a pediatric hospital in Phnom Penh, and wanted to avoid being disrespected by providers. Learning about others’ experiences with providers at different facilities allowed her to make an informed decision about the care she might receive at a facility she had not yet attended.

Financial support was the most influential form of support in terms of shaping children’s healthcare utilization. About three-quarters of grandparents reported receiving some form of financial support from migrant parents, though the level and regularity of remittances varied widely among these families. While some grandparents reported they received remittances “every month” or “around one time per month, sometimes two times per month,” others reported their children remitted “one time per year,” “rarely,” or “[not] since last year.” The ability to depend on a migrant parent to cover the costs of their children’s healthcare provided a financial safety net that allowed families to avoid health-related debts. The elderly grandmother of two in Kandal relied on monthly remittances from two migrant daughters working in Thailand; the family’s financial position had improved after her daughters left. She reported, “Whenever my granddaughter needs medicine, [her mother] always send the money. I don’t need to worry...I don’t need to wait until the end of the month.” Families with regular remittances or where migrant parents reliably and quickly sent money when their children incurred health expenditures were able fund care more flexibly than grandparents with no or unreliable remittance support, about half the sample.

Families used remittances to finance grandchildren’s healthcare in two ways. In some families, particularly those with higher SES within the sample, grandparents paid for health expenditures using regular remittance income previously received from the migrant parent. In other families, grandparents notified the child’s parent about health expenditures at the time of treatment, and parents sent money to cover costs of care. Among families where regular remittances were used to cover care or where parents “give money immediately,” grandchildren enjoyed access to a greater number of facilities because grandparents could pay at the time of service. Some grandparents used remittance money to access providers that reduced burdens on their time, such as a private doctor who made house calls and “shows up right away”. Others sought to avoid places with services or medications they deemed “ineffective” or “hard to be trusted”. Reliable remittances allowed grandparents to make decisions about where to seek care based convenience and perceived quality.

Where grandparents relied on migrant parents to send funds for health expenditures after treatment, their options were more limited. Some migrant parents reimbursed grandparents who paid treatment costs up front; others paid directly to providers who accepted deferred payments after treatment. Some grandparents were forced to incur debts with providers or borrow from neighbors, engendering stress and uncertainty. The grandmother in Prey Veng with five adult migrant children reported, “I wait for money from my children, but sometimes I wait such a long time. When I have money, I pay back the doctor.” A grandmother in Kandal caring for a granddaughter whose parents worked in a garment factory in Phnom Penh stated, “When I take her to the hospital, I borrow money from neighbors. Sometimes, I owe the doctor about 60,000 or 70,000 [about $14.50 to $17] until I can find the money to pay back. It is like that.” Another grandmother in Kandal lived with her husband, her own elderly father, and two grandchildren; the household regularly experienced food insecurity. As their migrant daughter did not earn enough working in a restaurant in Phnom Penh to regularly remit, the grandmother faced financial challenges when her grandchildren fell ill:

I cannot owe money to the pharmacist for so long. If the pharmacist injects medicine today, he will come to get money tomorrow about 7 or 8 am… I borrow money from my neighbor and when I have money, I will pay them back…Sometimes, my daughter asks her boss to get her salary advanced. It is so difficult for my family.

Given these financial constraints, these grandmothers were limited in the providers they could access for their grandchildren if they could not secure a loan elsewhere, and accruing debts with providers or neighbors created anxiety.

Grandparents with no financial support from migrant parents were forced to navigate paying for grandchildren’s healthcare on their own. In a few cases, they delayed or avoided seeking treatment altogether. A grandfather in Kandal caring for two grandchildren with his disabled wife was in a particularly precarious financial situation. His daughter migrated to Thailand a year prior because of the family’s debts; they were food insecure and often harassed by lenders. Asked about his eight-year-old grandson’s recent illness, he said,

I treated him myself first. I bathed him with warm water and gave him paracetamol. If he felt better, I would not have to ask the doctor to treat him because I didn’t have any money. But I would ask the doctor to treat him if he did not feel better…the doctor needs the money and if I don’t have it, the doctor will not inject medicine for us.

By waiting to see if his grandson still needed care after first-line treatments at home, he could potentially avoid paying for healthcare he could not afford. Several of the poorest families received subsidized healthcare in the public sector through a government support scheme, Cambodia’s health equity funds (HEF). These families noted the financial benefits of free care, and a few reported that they used this benefit for their grandchildren. However, HEF subsidies were not sufficient to address two critical barriers: transportation and competing domestic burdens, which created opportunity costs to seeking care. While HEF is designed to include transportation subsidies, many families pay transportation costs up front and wait for reimbursement. Reports of transportation barriers were consistently related to ability to pay, rather than grandparents’ own physical health. For example, the grandfather in Kampong Cham whose son migrated to Malaysia reported that he was often unable to pay for the cost of gasoline to get to the primary health center where they were eligible for free care. Instead, he paid for his grandchildren to attend a nearby private provider, despite being unable to afford their school fees. For the grandmother in Kandal caring for four grandchildren, the distance to the primary health center outweighed the benefits of subsidized care given the additional labor she and her husband took on after their children migrated. Instead, these grandparents preferred to pay out-of-pocket to remain closer to home given the demands of caregiving on their time—despite the difficulties in affording such treatment.

IV. Discussion

This study describes how families navigate care seeking and decision-making processes for young grandchildren’s health in the face of parents’ physical absence. While an increasing proportion of young children globally reside apart from their parents and with grandparents, the processes of how families living apart seek healthcare for children has received little attention in the literature, and is under-theorized in existing frameworks of care-seeking processes for child health. We examined this phenomenon in Cambodia using the case of grandparent caregivers and young children in skipped-generation migrant-sending households, where parents have migrated for work and family members left behind have varying financial and instrumental support. We argue that physical absence limits parents’ ability to participate in daily decisions for their children, as well as many healthcare-related decisions. Given their daily, proximal involvement in young children’s lives, it is grandparent caregivers who are at the nexus of care seeking and decision-making processes when parents are absent, controlling parents’ knowledge of and involvement in their children’s illness. However, with this control comes the responsibility of mobilizing resources for treatment and managing children’s return to wellness, and for many grandparents, additional stress and hardships. In spite of the ways in which care-seeking processes change in parents’ absence, grandparent caregivers use similar metrics when making decisions about where to seek care, and broadly follow similar trajectories for seeking higher-level or more advanced care as young children’s’ illnesses escalate or fail to resolve (Colvin et al., 2013).

To navigate financial, logistical, and informational challenges, grandparents relied on a combination of available support resources. As families that previously incurred debts to pay for medical expenses gained reliable remittance income, migration created increasing levels of financial security in times of children’s illness. Yet even among families that felt their livelihood had improved over time, financial difficulties described by many grandparents suggests these benefits often were not as frequent or far-reaching as families imagined at the outset of migration. Even in the absence of financial support, instrumental support facilitated children’s access to care. The ability to consult with family or neighbors for information about child health treatments, facilities, and providers served to reduce stress among grandparent caregivers. Instrumental support to grandparent caregivers augmented their ability to manage care seeking for grandchildren in tandem with other daily tasks, reducing barriers to care in the face of the additional domestic burdens they assumed after migration. The benefits of instrumental support observed in this study are similar to relationships between instrumental support and improved children’s health outcomes observed in other settings (Clark et al., 2018; Turney, 2013).

The hierarchy of families’ well-being by the type of support they received—whether financial or instrumental—was clear. Those with both financial and instrumental support were least likely to report that caring for their grandchildren was a difficult undertaking, and expressed greater confidence in their ability to navigate options for care. In comparison, the small number of families without financial or instrumental support experienced a number of challenges. These families were the most vulnerable; when grandparents could not rely on financial support from migrant parents, they accrued further debts, and often navigated care-seeking decisions alone under stress and uncertainty.

This analysis highlights the limitations of physically absent parents’ roles in the left-behind household. After migration, shifts in familial power dynamics and flows of care vary along several axes, including distance (Baldassar and Merla, 2014). Because grandparents in skipped-generation households do not merely support their children as parents, but rather absorb many of the daily tasks of a parent, their roles within the family more closely resemble that of parent than grandparent after migration. Yet, they still hold the higher status of a grandparent within the family, creating role conflict. These grandparents cannot adopt a strategy of “non-interference”, boundary setting that helps parents and grandparents maintain clear, prescribed roles within the family (Thang, 2012). Rather, as their roles shift after migration, decision-making processes are also upended. Grandparents maintained greater autonomy than might be predicted by a modified extended family perspective, especially as many managed parents’ involvement in their children’s illness by controlling the flow of information. After parents’ migration, grandparents gained power by nature of their ability to carry out or ignore absent parents’ expressed preferences—additional dimensions of power beyond those predicted by Baldassar and Merla. In this case, differences in households that sent internal and international migrants were evident. Compared to migrant parents in Thailand or other countries, migrant parents in Phnom Penh remained more actively involved in their children’s illness trajectory. Conversely, grandparents rarely involved international migrant parents, even internalizing their physical distance to create emotional separation.

The care-seeking patterns and preferences described by grandparent caregivers were broadly similar to those reported by mothers in other low- and middle-income settings. Grandparents generally sought the highest quality healthcare they could afford, balancing perceptions of quality and need (Leonard, 2014). For example, while cost and distance were important drivers of care seeking, perceived quality of care and prior experience mattered as well (Leonard, 2014; Scott et al., 2014). Grandparents were willing to bypass closer facilities when illnesses were severe, or required more advanced care (Akin and Hutchinson, 1999; Kahabuka et al., 2011). Grandparent caregivers did not express preferences for seeking care that were inherently different from parents, but rather, financial dynamics and informational constraints played an important role in determining grandchildren’s utilization of care. Our empirical analysis contextualizes the strategies grandparent caregivers utilize to navigate structural constraints to their grandchildren’s healthcare utilization, underscoring the importance of social and structural context outlined in Sadruddin et al.’s framework of grandparent influences on grandchild health (2019). Our specific case centers grandparents in the careseeking process, providing an important counterpoint to existing child healthcare utilization frameworks that center the mother (Colvin et al., 2013).

Our findings have several implications for child health policies and programs. Health equity schemes were explicitly mentioned by several grandparents as enabling their access to public sector facilities. However, others still faced additional barriers related to transportation and other indirect or opportunity costs that prevented them from using these benefits. In areas with high rates of out migration, subsidy schemes might explicitly consider the types of economic stresses and debts frequently experienced by migrant-sending households (Lam et al., 2013). To ensure the most vulnerable households are able to make use of subsidized care, subsidy programs should consider alternatives to the current reimbursement model, which requires families to have cash on hand. Grandparents’ reliance on kin and social networks for decision-making support and information about services suggests they might also benefit from targeted child health communications.

Several limitations should be considered in the interpretation of these results. While the sample was theoretically driven to mirror the diversity of Cambodian migration patterns, these results are not generalizable to other settings. The experience of the Khmer Rouge genocide in the 1970s has specific implications: all grandparent caregivers in this setting and sample are survivors of the genocide, though its intensity varied across the country (Strangio, 2014). Many in this generation have very limited formal education (Zimmer, 2008), and post-traumatic stress disorder and other anxiety disorders are common (De Jong et al., 2003). Therefore, the quality of care provided and psychosocial well-being among grandparent caregivers may vary more in Cambodia than in other settings. Because most migrant parents maintained some level of communication and/or financial ties with their children left behind, the health-related consequences of residing apart from their parents likely vary for children whose parents have migrated versus those in skipped-generation households formed for other reasons, such as parental mortality or incarceration (Ice et al., 2012; Lam et al., 2013; Sadruddin et al., 2019). Given its sensitive nature, we did not explicitly ask about legal status of international migrants, which might further inform this analysis. We were unable to analyze differences in illness management and decision-making by children’s ages given insufficient data. We also lacked data on chronic illness management, and had insufficient data to explore differences in processes and resources among single grandfather-headed households, which would further inform analyses related to grandparent gender. Finally, exploring the perspectives of absent parents is necessary in future research to further understand care-seeking decisions, negotiations, and power dynamics in families where parents live apart from their children.

Conclusions

This qualitative study provides novel insights into how care-seeking and decision-making processes for children’s health shift in the face of parental absence, using the case of skipped-generation migrant-sending households, a family form where parents live apart from their children but remain engaged in various ways from a distance. Grandparent caregivers who provide daily care for young children are centered in care-seeking and decision-making processes when parents are absent. As grandparents direct and control care-seeking for their grandchildren while parents’ involvement is limited often to financial support, efforts to improve child health in migrant-sending areas should explicitly target grandparent caregivers, and identify and implement ways to support these types of families that address the specific vulnerabilities they face.

Table 1.

Characteristics of grandparent caregivers interviewed.

	N
Province
Kampong Cham	8
Kandal	7
Prey Veng	10
Grandparent sex
Male	5
Female	20
Marital status
Married	15
Widowed	10
Number of co-resident grandchildren under 10
1	13
2	9
3	3
4+	1
Any other adults in household	2
Other relatives in village or nearby	14
Median size of household left behind	5 members
Migrant destination
Phnom Penh	12
Rural Cambodia	2
Thailand	10
Malaysia	1
Remittances
Regular	11
Periodic or infrequent	8
None	6
Financial situation
Stable	8
Poor	14
Food insecure	3

Notes:

1. Other adults in household include adult children of the grandparent respondent who are not the parent of any grandchild in the household, or an adult grandchild (age 16 or older).

2. Most grandparents mentioned whether they had other relatives or extended kin in their village or in a nearby village, though not all respondents explicitly reported this information.

3. Households’ financial situation is inferred based on three types of data: respondents’ self-ratings compared to other households in their villages; interviewer observations regarding housing characteristics and durable goods; and respondents’ self-reports of food insecurity.

Highlights:

• Demands of daily caregiving on grandparents create barriers to care for children

• Many grandparents have autonomy in treating grandchildren’s illnesses

• Grandparents shape migrant parents’ involvement in child healthcare decisions

• Beyond financial support, instrumental support facilitates children’s healthcare