Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The dramatic increases in life expectancy that have occurred throughout the twentieth century have had a profound impact on both individuals and families. In 1900, the average life expectancy was 47 years and people over 65 years of age accounted for approximately only 4% of the U.S. population—or less than one in 25 persons. Today, life expectancy at birth has risen to 72.5 years for men and 79.3 years for women and people over the age of 65 represent 12.8% of the total U.S. population (U.S. Census Bureau, 2000). The democratization of the aging experience or the longevity revolution has led to a life course revolution (Skolnick, 1991). The changes in mortality have greatly influenced the concepts of young, middle, and older adulthood. Moreover, as a result of longer life expectancy, many of today’s families are multigenerational. Indeed, the modal family structure for American adults aged 50–54 is a three-generation family of adults who have one or more parents or parents-in-law and one or more children.

In this chapter, we explore the impacts of families experiencing more years of “shared living between generations” on middle-aged adults. Our focus is on the role of the middle generation as the “kinkeepers” with a special emphasis on the adult child and aging parent caregiving relationship. As part of our analysis we explore how gender, race, ethnicity, and culture may influence appraisal of the parent care role. We also examine the effectiveness of caregiver interventions in enhancing family members’ well-being. For most individuals, the midlife experience revolves around family and work. Thus, we examine how structural transformations in these two domains affect how men and women negotiate their caregiving responsibilities as well as the responsiveness of public and private sector policies to the changing realities confronting families.

Middle Adulthood

The middle years of adulthood have been referred to as the “last uncharted territory in human development” (Brim, 1992, p. 171). Concern about the welfare of children and the elderly contributed to the scientific study of these two vulnerable populations. In contrast, few resources were directed towards the study of middle adulthood because, from a public policy perspective, adults were not viewed as a vulnerable population requiring protection of “their best interests” by the state. The aging of the baby boomer generation and the sheer number of this cohort entering midlife, however, has generated new interest in this life stage. In 2000, over 80 million members of the baby boomer generation were between the ages of 35 and 54 (U.S. Census Bureau, 2000). This interest spurred, for example, the MacArthur Foundation Research Network on Successful Midlife Development (MIDMAC), one of the most significant interdisciplinary research endeavors devoted to the study of midlife. Indeed, a primary objective of MIDMAC is to support the development of national representative databases that allow the identification of “major biological, psychological well-being, and social responsibility during their adult years” (http://midmac.med.harvard.edu).

It has been noted that the period called middle age lacks well-defined boundaries. Farrell and Rosenberg (1981, p.16) suggest “like defining a period of history, no one quite agrees when middle age begins or ends.” Not surprisingly, the subjective boundaries of middle age vary positively with age (Lachman, Lewkowicz, Marcus, & Peng, 1994). The older an individual is, the more likely s/he will identify later entry and exit years as demarcating middle age. Although the ages of 40–60 are typically considered to be middle-aged, for some individuals middle age starts as young as 30 and for others middle age is not perceived as ending until age 75. As life expectancy increases, the boundaries of middle age may continue to be extended upwards. In fact, one-third of Americans who are currently in their 1970s describe themselves as being middle-aged (National Council on Aging Survey, 2000).

The research of the past decade has dispelled many of the myths and negative stereotypes of middle age. Although they express concerns about weight gain, future declining health and mortality, most middle-aged adults enjoy good physical health (American Board of Family Practice, 1990). Only 7% of adults in their early 40s, 16% of adults in their early 50s, and 30% of adults in their early 60s have a disabling health condition. Despite the persistent societal view of menopause as a stressful life experience, research has consistently documented that most women pass through menopause with little difficulty (Avis & McKinley, 1991; Dillaway, 2005; Gonyea, 1996).

While childhood and adolescent transitions are often marked by rites of formal passage, the transition from young adult to middle aged adult is marked neither by special rites of passage nor by predictable chronological events. Rather the transition from young adulthood to middle adulthood is often a gradual one and social cues, especially changes in family and work domains, may be better indicators of developmental change than chronological age. Increases in life expectancy have led to middle age becoming the normative life stage in which adult children typically confront parental declining health and death. About 40% of Americans enter midlife with both parents alive, while 77% leave middle adulthood with no living parents.

Demographic Trends Impacting the Family

A number of demographic changes that occurred over the twentieth century have profoundly affected American families. Increased life expectancy accompanied by decreased fertility means that contemporary American families have more generations alive, but fewer members in each generation. The U.S. birthrate has declined from 4.1 in 1900 to 2.0 in 2000 (Fields & Casper, 2001). This shift from “high mortality-high fertility” to “low mortality-low fertility” means that for most families the age structure has changed from a “pyramid” to a “beanpole” (Bengston, Rosenthal, & Burton, 1990). Exploring the effects of the past century’s mortality changes on the supply of kin to provide support to family members, Uhlenberg (1996) found that for those born in 1900, only 21%—about one in five—had any grandparent living by the time they reached age 30. In contrast, he noted that for those individuals born in 2000, by age 30, 76% will have at least one grandparent alive. Bengston (2001, p.5) comments that “what might be lost in a review of macrosocial demographic trends are the consequences for individual family members and their chances of receiving family support.” He proceeds to identify both potentially positive and negative consequences of more years of shared living across generations. For example, greater years of “cosurvivorship between generations” may offer a multigenerational kinship network to provide family continuity and stability across time as well as instrumental or emotional support in times of need. Yet, longer years of shared living may also mean extended years of caregiving for frail or disabled elders or family conflict (Bengston, 2001). Parents and children now share five decades of life, siblings may share eight decades of life, and the grandparent–grandchild bond may last three or four decades.

In addition to the verticalization of the family, other demographic and social changes of the latter half of the twentieth century have affected American family life, including increased educational and labor force opportunities for women, technological advances in reproductive choice, and greater public acceptance of diverse lifestyle and family choices. Contemporary adults—both men and women—face unprecedented choices about whether and when to marry, whether to remain married, divorce, or remarry, and whether and when to have children. Phenomena that were once clear markers of young adulthood, such as marriage and parenthood, are less predictable and there is greater diversity in the structure of families. Between 1970 and 2000, the median age at first marriage for women increased by 4.3 years to 25.1 years of age; for men, the increase was 3.6 years to 26.8 years of age. One in five women in the U.S. now has her first child after age 35. There has also been a dramatic change in the numbers of American women who have entered the paid labor force. In 1950, about one in three women participated in the labor force whereas today nearly three out of every five women of working age are actively engaged in the labor force. This increase reflects the dramatic rise in working mothers; the labor force participation rate for American working mothers has grown from slightly less than half (47.5%) in 1974 to 70.7% in 2004 (U.S. Bureau of Labor Statistics, 2005).

It is important to note that not only has America’s population grown older, it has also become more ethnically diverse. Persons of color are the fastest growing segment of the U.S. older population. Whereas elders of color currently comprise only 16% of the aging population, they will account for 25%, or one in every four elders, by 2030. Immigration has played an increasingly important role in reshaping the makeup of the aged population. The March 2000 Current Population Survey revealed that the number of foreign-born residents and children of immigrants in the United States has reached the highest level in history. In 2000, 56 million Americans, or 20% of the population, had either been born in a foreign country or had at least one parent who was foreign born, with the largest immigrant group being Mexican. Although the immigrant population is comprised disproportionately of working-age adults, these individuals will “age in place” and bring their parents as “invited elders” to this country. It is increasingly important to understand the possible consequences of these demographic shifts on the ability of current and future cohorts of adult children to provide care for their aging parents.

The Adult Child-Aging Parent Caregiving Experience

Prior to the 1970s and 1980s, family caregiving for older dependent members was largely invisible and often assumed to be nonexistent. The field of caregiver research owes much to early pioneers such as Ethel Shanas whose groundbreaking research refuted the prevailing social myth of families’ abandonment of their elders and Elaine Brody who introduced the notion of parent care as a normative family stress (Brody, 1985; Shanas, 1979). In fact, during the past three decades, caregiver research has burgeoned. Today, the central role that families play in the lives of older frail and/or disabled members is widely recognized and the term “family caregiver” has entered American lexicon. According to the 1989 National Long-Term Care Survey, among community-dwelling elders with disabilities, over 90% of this population receives care informally from family, friends, and neighbors, approximately 25% use a combination of informal and paid care, and only 9% rely exclusively on formal care.

Much of the early research on elder care sought to describe who were these family members engaged in caring, the type of supports provided as well as the intensity and duration of this caring, and the consequences of taking on the caregiving role. Estimates of the numbers of families actively engaged in elder care vary widely based on the definition of “caregiving” used. The 2005 National Alliance for Caregiving (NAC)/AARP report, for example, estimated that 16% (or 33.9 million) of American adults are involved in caregiving for an elder. In the survey, the estimate included those caring for any friend or relative age 50 or older, including those free of a chronic disability, in the prior 12 months. In contrast, the 1994 National Long Term Care Survey (NLTCS) defined an informal caregiver as someone providing help with one or more activities of daily living (ADLs) or instrumental activities of daily living (IADLs) because of the elder’s health or disability for at least 3 months. Based on this more restrictive definition, slightly less than 2% of the U.S. population aged 15 and older were viewed as actively engaged in elder care and an additional 7% were labeled potential family caregivers (Spillman & Pezzin, 2000).

The importance of the parent–child bond across the life span is underscored by the fact that adult children represent the largest group of active carers for disabled elders. Two-thirds of the active informal caregivers, according to the 1994 NLTCS, were spouses or children of the older care recipient. However, given both high rates of widowhood among the older population and the sharing of caregiving responsibilities by multiple siblings, children greatly outnumbered spouses as active carers (Spillman & Pezzin, 2000). Among adult children there is a significant gender difference as to who takes on the primary caregiver role. Although the participation of sons as primary caregivers rose by 50% between 1984 and 1994, they still accounted for only 15% of the primary caregivers in 1994. Daughters remain twice as likely to become the primary caregiver than are sons (Spillman & Pezzin, 2000).

In fact, one of the most consistent findings in family research is that the vast majority of carers for relatives with chronic disabilities are women. Although male and female roles within the American families have changed over the past few decades, women are still the primary nurturers, kinkeepers, and carers of family members. Moreover, it is women in the middle generation who are most centrally involved in maintaining family communication and cohesion across generations. Of all family caregivers to the elderly, 23% are wives, 29% are daughters and daughters-in-law, and 20% are more distant female relatives (Older Women’s League, 1989). The crucial role of gender in the hierarchy of obligation to elderly family members is reflected in the fact that after spouses and daughters, it is daughters-in-law and not sons who are the next lines of resort (Qureshi & Walker, 1989). It has been argued that social scientists’ continued use of gender-neutral terms such as parent, spouse, sibling, and family caregiver has obscured differences in men’s and women’s roles and led to gender insensitivity in the development of social policies and programs (Traustadottir, 1991). Family caregiver has become a euphemism for one primary caregiver, typically female (Hooyman & Gonyea, 1995).

Even when men provide care to elders, there appears to be a gender difference in the type of support rendered. Men are more likely to help with IADLs such as yard care, financial management, and transportation rather than perform ADLs such as bathing, dressing, and feeding. Females are more likely than males to perform tasks that are physically draining, involve daily interruptions, and entail intimate or bodily contact (Delgado & Tennstedt, 1997; Matthew, Mattocks, & Slatt, 1990). Increasingly, researchers have explored not only how men and women differ in their enactment of the caregiving role, but also the way gender influences the meaning of caring, the social context of providing care, and the consequences of performing the carer role. Women, for example, generally have more extensive social networks than do men. Yet, as Antonucci (1990) notes, while women’s more extensive family networks and involvements may be useful resources in adaptation to life’s stresses, the networks themselves may also be sources of stress. Russel’s (2001) qualitative study of male caregivers revealed that men employ a more managerial approach that potentially offers them a “greater perceived control, the sense of being in charge, feelings of self-efficacy, as well as the ability to choose to act or not to act” (p. 355).

In addition to gender, variations in the caregiving experience also exist due to factors such as race, ethnicity, sexual orientation and marital history as well as the structural availability of family, contact between family members, the type of support exchanged, norms of filial obligation and the quality of these relationships. Personal experiences, social location and membership in social groups, birth cohort and social context influence the nature of the adult child-aging parent bond. In fact, diversity and complexity characterize families’ patterns of support. For example, distance—both geographic and emotional in nature—often inhibits family members’ provision of care. In the 1994 NLTCS, 97% of primary caregivers and 93% of secondary caregivers lived within an hour’s distance of the elder (Spillman & Pezzin, 2000). Divorce has also been shown to have a negative effect on interaction with children—especially for men. Cooney and Uhlenberg (1990), for instance, found that divorced fathers were less likely to consider their adult child as a potential source of support in times of need.

Hierarchies of privilege—race, ethnicity, social class, gender, and sexual orientation—reveal how accumulated advantages and disadvantages across the life course differentially equip families with a set of skills, liabilities, and resources that affect how they are able to care for older dependent members. The question of who will care for frail and disabled older parents is a particularly salient question among ethnic minority communities. Due to greater exposure to risk factors such as inadequate housing, poverty, poor nutrition, smoking, manual jobs that are potentially physically debilitating, and lack of adequate health care, elders of color have higher rates of morbidity and mortality than do White elders. The higher rates of health problems do not, however, translate into higher rates of institutional care; a smaller percentage of elders of color (3%) than White elders (5.8%) live in nursing homes. Yet, as research in the emerging field of ethnogerontology reveals, even among ethnic groups who have strong norms of filial piety and familism, economic and social forces are increasingly impacting their abilities to care for aging family members (Angel & Angel, 1997; Olson, 2001).

The Phenomenon of the Sandwich Generation

One phenomenon of the changing age structure of families that has received growing attention is the sandwich generation. As Ward and Spitze (1998) note there are two meanings to this term:

Structurally, it refers to middle-generation cohorts sandwiched between older and younger cohorts in the population. Individually, it refers to persons in middle adulthood who simultaneously have relations with their adult children, as they enter and adjust to adulthood, and their parents, as they deal with issues of later life. Members of this sandwich generation are presumed to face potential stresses from the combined and perhaps competing demands of their intergenerational roles as parents and children. (p. 647)

Recent research has raised questions, however, about the size of this phenomenon. Spillman and Pezzin (2000) analysis of the 1994 NLTC data set reveals that among women with children under the age of 15 only 9.1% were the primary caregiver and 25.5% were the secondary caregiver for either a disabled elderly spouse or parent. Among men with children less than age 15 only 5.6% were primary caregivers and 21.1% were secondary caregivers for a disabled elder. Similarly, a study of 12 European Union countries found that only 4% of men and 10% of women aged 45–54 had overlapping responsibilities for children and older adults who required care (Hagestad, 2000). A much larger proportion of American and European midlife men and women faced the competing demands of caregiving responsibilities and paid employment than the dual responsibilities of child and elder care.

Although it appears that there are fewer members of the sandwich generation who are simultaneously engaged in caring for children and aging parents, there may be substantial burdens for those who are in fact providing intergenerational assistance. Moreover, others stress that rather than using a narrow definition of caregiving we should look more broadly at the effects of dual occupancy of the parent and adult child roles. For example, in industrialized societies young adults from their late teenage years through their twenties are increasingly allowed a prolonged period of independent role exploration which has correspondingly led to a prolonged parenting phase for many midlife parents. Arnett (2000) suggests that this period of “emerging adulthood” should be viewed as a new life stage. He argues that emerging adulthood is the only period of life in which nothing is normative demographically. Almost all American adolescents from 12 to 17 years of age live at home, are enrolled in school, and are unmarried and childless. In contrast, emerging adults’ lives are characterized by diversity. About one-third of young persons in the United States go off to college, another 40% move out of their parental home for independent living and work, and about 10% of men and 30% of women remain at home until marriage (Arnett, 2000). It is estimated that about 40% of recent cohorts of young adults have returned to their parents’ home after moving away (Goldschieder & Goldscheider, 1994). A significant proportion of midlife parents are actively engaged in helping their children (i.e., advising, guiding, worrying) as they explore choices in love and work and transition to young adulthood. In an era of rising costs, particularly for younger families attempting to buy a first home and/or as new parents coping with child care costs, there may be growing expectations for assistance from the middle generation (Goldschieder, Thornton, & Yang, 2001).

Grandparenting as a Midlife Role

Although the popular image of grandparents is as frail older adults, most people become grandparents for the first time between the ages of 49 and 53. The transition to the role of grandparent as well as the experience of grandparenting itself varies by life circumstances. For example, factors that have been identified as significant predictors of grandparent contact with grandchild include: geographic distance, quality of relationship between grandparent and parent, number of grandchild sets, gender of grandparent, lineage of the grandchild set, and marital status of the grandparent (Uhlenberg & Hammill, 1998).

Utilizing a family life course perspective, Silverstein and Marenco (2001) found that younger grandparents tended to live closer to and have greater contact with grandchildren, often babysitting and sharing recreational activities. In contrast, older grandparents tended to provide greater financial assistance, assume a more formal relationship with grandchildren, and more strongly identify with the grandparent role. Many of the age difference effects in grandparent roles may also be related to the grandchildren’s age and the type of interaction that the youth are seeking. By the time grandparents approach their late 60s or early 70s, most of their grandchildren have entered adolescence or young adulthood.

Parents are a bridge between the two generations. They often set the tone for grandparent and grandchild relations by how they function as gatekeepers between the two generations. Generally, greater closeness and contact between parents and grandparents results in greater closeness and contact between grandchildren and grandparents. Most studies suggest that adults derive considerable pleasure from the grandparent role. The two exceptions are when an individual takes this role on at a relatively early age due to a teenage birth or when a grandparent becomes the primary caregiver due to events in the parent generation such as divorce, drug addiction, incarceration, illness, and death. In recent decades, the crack cocaine and HIV/AIDs epidemics have contributed to a dramatic rise in the number of grandparents who are surrogate parents for their grandchildren (Casper & Bryson, 1998).

The Consequences of Elder Care

Recognition of the family as the primary source of long-term care for the elderly has led to the proliferation of research studies exploring the consequences of assuming the caregiving role. Before discussing the findings, however, it is important to point out several limitations in the existing literature. First, our understanding of caregiver is primarily derived from information gathered from one member of the family, typically a woman. The concept of caregiving as a “woman’s role” has led to the respondents to surveys and interviews being overwhelmingly women. Second, while supporting a family member can be a rich and rewarding experience, the literature has often focused on the negative effects of taking on the caring role. Although recognizing that this emphasis on the costs of caregiving is an attempt “to argue that families cannot absorb additional obligations and that government must devote adequate resources to support them.” Abel (1991, p. 8) criticizes the current caregiving research agenda “which has focused almost exclusively on the issue of stress.” Third, most studies of caregiving are based on nonprobability samples and, at times, lack a comparison group of noncaregivers. Fourth, the findings on the nature of caregiving experience are, at times, inconsistent. These inconsistencies may partly reflect variations in the populations studied, such as type of impairment as well as the caregiver’s gender, race and ethnicity, and the nature of the relationship to the care recipient, whether the bond is as a spouse, an adult child or a sibling. Finally, longitudinal studies remain relatively sparse. Most of the conclusions regarding the long-term effects of providing care are based on cross-sectional studies and must be viewed as tentative.

The Concept of Caregiver Burden

Although it is often assumed that the burdens of caregiving are fairly obvious, developing a concise definition of “caregiver burden has proven difficult because it is a multidimensional concept that encompasses a wide range of stressors, including “physical, psychological or emotional, social and financial problems” (George & Gwyther, 1986, p. 253). In her review of the caregiving literature, Braithwaite (1992) identifies two major theoretical efforts to define the burden concept more precisely. The first effort was the theoretical distinction drawn between objective and subjective burden. This differentiation was first noted by Hoenig and Hamilton (1966) in their research on the effects of an adult with schizophrenia on the household and was subsequently adopted by the mental health, gerontology, and developmental disabilities fields (Marsh, 1992a, 1992b; Montgomery, Gonyea, & Hooyman, 1985; Thompson & Doll, 1982).

Objective burden refers to the real demands that confront the family member who assumes the carer role. The dimensions of objective burden can include symptomatic behaviors of the impaired relative, disruptions of family life in areas such as domestic routines, leisure activities, and opportunity to socialize, problems with health and legal systems, and alterations in family roles (Bulger, Wandersman, & Goldman, 1993; Marsh, 1992a, 1992b). Whereas objective burden focuses on behavioral phenomena associated with performing the caring role, subjective burden refers to the feelings and emotions aroused in family members as they fulfill their caregiving functions (Braithwaite, 1992). These emotional reactions may include anger, guilt, worry, tension, loneliness, sadness, depression, difficulty sleeping, withdrawal, and empathic suffering (Bulger et al., 1993; Marsh, 1992a, 1992b). The second major theoretical effort noted by Braithwaite (1992) was the recognition that burden is a subjective experience—that is, what is difficult for one caregiver need not be difficult for another. This phenomenon, first identified by Poulshock and Deimling (1984), helped to explain the confusion as to why family members did not experience the same or similar caregiving experiences as “burdensome.” For example, when two adult daughters are providing the same levels of personal care assistance, such as dressing, feeding, and bathing, one might report feeling “very burdened” while the other might identify “very little stress.” Variation in the experience of caregiver burden may also reflect differences in midlife expectations and priorities. As parenting demands diminish, some begin to anticipate a greater sense of freedom relative to their own time and the ability to pursue new or postponed opportunities. The onset of parent care responsibilities may therefore be viewed by some as a burden. The frailties of aging parents also may underscore one’s own mortality and the finite nature of the lifespan. For those in midlife, decisions about parent care are thus increasingly made within a framework of having a limited amount of time available to achieve one’s life goals, creating for some a stronger sense of urgency.

In fact, much of the caregiving literature has attempted to understand the nature, prevalence, and predictors of carer burden. And, the research to date does suggest that taking on the caregiver role is not without risks. A number of studies have found that family members who provide care to disabled family members experience increased health problems, including physical exhaustion, poorer immune responses, and deterioration of their own health status (Biegel, Sales, & Schulz, 1991; Kiecolt-Glaser et al., 1987; Marks, 1996). Moreover, poorer physical health or lower physical stamina has been found to be associated with emotional distress and psychiatric disturbances (Clyburn, Stones, Hadjistavropoulos, & Tuokko, 2000; Hooker et al., 2002; Schulz & Martire, 2004; Schulz, O’Brien, Bookwalla, & Flesissner, 1995).

Gender and Caregiver Outcomes

A growing number of researchers are investigating how gender, race, ethnicity, and culture may shape the caregiving experience (Dilworth-Anderson, Williams, & Gibson, 2002; Harrington Meyer, 2000; Hooyman & Gonyea, 1995; Campbell & Martin-Matthews, 2003). Miller and Cafasso (1992) meta-analysis of 14 studies of gender differences in caregiving conducted from 1980 to 1990 revealed that female and male carers did not differ either in terms of the severity of the elder’s functional impairment or the level of total care involvement, but a significant difference was found in terms of the type of task involvement and the level of caregiver burden. While the size of the gender effect was small, female caregivers were more likely to perform personal care and household tasks and were more likely to report greater burden than male caregivers do. Based on their findings, Miller and Cafasso (1992, p.152) suggest that “what is needed may be less documentation of specific gender differences in isolated components of caregiving and more attention to the role that gender-role expectations play in assigning meaning to the caregiving experience.”

Updating and further expanding upon Miller and Cafasso meta-analysis, Yee and Schulz (2000) synthesis of 30 research reports examining gender differences in psychiatric morbidity among caregivers published between 1985 and 1998 revealed that women caregivers reported higher levels of depression, anxiety, psychiatric symptomatology, and lower life satisfaction than did men caregivers. Yee and Schulz note that several studies found that male caregivers were more likely to obtain informal assistance, to relinquish the caregiver role, and to engage in preventative health behaviors than female caregivers.

Raschick and Ingersoll-Dayton (2004) found that gender, rather than relationship status, was a predictor of caregiving costs; women, whether wives or daughters, reported experiencing greater caregiving costs than did men (husbands or sons). The authors attribute this difference to not only gender differences in caregiving styles but also dissimilar internalized norms about nurturing responsibilities. Women may be more likely to view caregiving as the expected mode of behavior or a moral responsibility while men may be more prone to feel abandoned or alone when cast into the caregiving role (Harris, 2002).

The role that gender may play is also reflected in Strawbridge and Wallhagen (1991) study of family conflict in caring for frail elders. Their data not only underscored that one family member’s disability reshapes each family member’s life, but that there may also be significant gender differences in caregiving expectations. Over 40% of 100 adult children caring for a frail parent or parent-in-law in their sample reported a serious conflict with another family member, usually a sibling who was unwilling to provide the amount of assistance the caregiver expected. Moreover, a greater proportion of female caregivers experienced conflict than did their male counterparts. Although this gender difference may reflect that fewer male caregivers had a living sibling available, it may also reflect gender-based assumptions regarding the appropriate roles of men and women in family care. A married working daughter with young children may resent the fact that her brother with fewer responsibilities (simply as a function of gender) is not asked to do more, or volunteers to do so little.

Parks and Pilisuk (1991) study of the psychological costs and coping strategies used by adult children caring for a parent with Alzheimer’s disease suggests that men and women approach the caregiving role from different perspectives and that it takes on very different meanings for the two groups. The researchers discovered that while daughters and sons did not differ in terms of depression or anxiety, daughters did have a higher rate of self-reported stress than did the sons. They also found that daughters predominantly used fantasy to cope, whereas withdrawal was the most common technique employed by sons. Perhaps what is most intriguing was the differential effectiveness of these coping strategies for men and women assuming the caring role. Significant predictors of anxiety for daughters were an external sense of control and the use of fantasy as coping mechanisms. For sons, anxiety was associated with lack of social support coupled with the use of either fantasy or withdrawal as a dominant coping style.

Race, Ethnicity, Culture, and Caregiver Outcomes

As Dilworth-Anderson and her colleagues’ analysis of 20 years (1980–2000) of caregiving research exploring issues of race, ethnicity, and class reveals, it is difficult to draw definitive conclusions as methodological issues such as nonprobability samples and noncomparable measures limit our understanding of caregiving among diverse populations. Although outcomes over a range of constructs (i.e., depression, burden, role strain) do vary across racial and ethnic groups, the findings are inconsistent (Dilworth-Anderson et al., 2002). For example, while several studies suggest a lack of significant difference in depression among African American and White caregivers (Cox, 1999; Knight & McCallum, 1998; White, Townsend, & Stephens, 2000), other research has found that African American caregivers report significantly lower levels of depressive symptomology (Farran, Miller, Kaufman, & Davis, 1997; Haley et al., 1995). Janevic and Connell (2001), in their review of 21 articles published between 1996 and 2000 comparing two or more racial, national, or cultural groups on the dementia caregiving experience, also found inconsistencies in the findings. They note a general trend toward White caregivers reporting caregiving as more stressful and experiencing greater depression than do African American caregivers; however, findings were mixed regarding differences in coping and social support.

Both sets of authors—Dilworth-Anderson and her colleagues and Janevic and Connell—identify a number of the same methodological limitations in the current body of research including the appropriateness of using the “grouping variable”—race, ethnicity, culture, or national origin—to attribute differences between groups of caregivers. Janevic and Connell comment:

When grouping caregivers by any of these categories, researchers should have a clear idea about the hypothesized mechanism by which membership in this category can affect the caregiving experience. In general, effects of the grouping may be due to cultural factors (the symbolic and normative aspects of social life, such as language, values, beliefs or norms); or minority status, with the latter implying the effects of inequality and discrimination, factors that continue to play a major role in the lives of minority group members in the United States and affect the psychological outcomes in these groups. (2001, p. 344)

Montgomery and Kosloski (1994) similarly caution that despite social scientists’ fondness of measuring sociodemographic variables (i.e., gender, race, marital status), these variables are not the causes of caregiver outcomes; rather, sociodemographic variables vary with the underlying causes. Through their articulation of a sociocultural stress and coping model, Knight, Silverstein, McCallum, and Fox (2000) make an important contribution to the caregiving field. In this model, ethnicity is viewed as a cultural variable that influences how individuals are socialized to view caregiving. They suggest that African American families’ religious beliefs, traditions, and social support bolster the value of family caregiving and caregiver rewards.

Indeed, there is a growing call for the conduct of more qualitative or ethnographic studies to understand the meaning of the findings of significant cultural differences in caregiver outcomes emerging from the quantitative studies. For example, through open-ended ethnographic interviews with African American, Chinese American, Irish American and Latino family caregivers for elders with dementia, Levkoff and her colleagues explored cultural differences in symptom appraisal (biomedical model versus folk model of attribution), family management of the disease, and help-seeking behaviors (Fox, Hinton, & Levkoff, 1999; Hinton & Levkoff, 1999; Levkoff, Levey, & Weitzman, 1999). Such research offers greater insights into how caregivers from different cultural groups perceive the nature and cause of their family member’s illness as well as normative expectations about the caregiver role. The growing body of ethnographic studies suggest that to gain a better understanding of the diversity in the caregiving experience, we must also increase our understanding of cultural variations in the meaning and significance of dependency, autonomy, family, community, health, illness, medicine, and death and dying.

Positive Aspects of Caregiving

The dominance of caregiver stress or burden research in the caregiving literature has often overshadowed other aspects of the caregiving experience. The literature increasingly suggests that family members caring for relatives with a cognitive, physical, or psychological impairment can experience both burden and satisfaction from their caregiving roles. Similarly, feelings of conflict and intimacy can coexist for the carer. There is growing evidence that feelings of caregiver satisfaction or gratification may be linked to the subjective meanings attributed to the caregiver role. In their study of role engulfment or loss of self in the caregiving role, Skaff and Pearlin (1992) report that one of their more intriguing findings is the lack of relationship between loss of self and self-gain:

The independence of these two indicators of the impact of caregiving on the individual suggests that loss and gain are not opposite points on the same continuum… Some caregivers may feel that they have grown as a result of their experiences, but whatever personal enrichment they might experience does not protect them from suffering a loss of identity. (p. 659)

Kramer (1997) argues for research focused on the positive aspects of caregiving experience emphasizing that: it is an area which caregivers want to talk about, understanding the positive gains may enable professionals to work more effectively with families; and it may enhance theories of caregiver adaptation and well-being. Based on her analysis of 29 studies focused on positive gains published through 1996, Kramer suggests the adoption of a conceptual framework in which the appraisal of positive role gains and role strains are viewed as intervening processes in understanding caregiver well-being outcomes.

Recent research has, in fact, sought to elaborate on our understanding of the “meaning of caregiving” in individuals’ lives as well as how positive appraisals may mediate the negative effects of the caregiver role. Based on a qualitative study of 48 caregivers, for example, Noonan, Tennstedt, and Rebelsky (1996) found the predominant themes of “caregiving meaning” to include: gratification and satisfaction; family responsibility and reciprocity, and friendship and company. Several researchers have sought to develop a quantitative measure tapping the positive aspects of caring. Picot, Youngblut, and Zeller (1997), for instance, constructed and tested the “perceived caregiver rewards scale” while Farran and her colleagues, using an existential framework, developed the “finding meaning checklist scale” (Farran, Miller, Kaufman, Donner, & Fogg, 1999). Recent research also suggests the potential importance of caregiving meaning to the stress process. Noonan and Tennstedt (1997) found that meaning in caregiving explained a significant portion of the variation in depression and self-esteem even after controlling for demographic and stressor variables. Similarly, Cohen, Colantonio, and Vernich (2002) study of 289 Canadian caregivers revealed that almost three-quarters of the individuals could identify at least one positive aspect of caregiving; and that positive feelings of caregiving were associated with lower levels of depression and burden and better self-assessed health.

There is also a growing exploration of how culture or ethnicity influences caregiving appraisal. Picot and her colleagues (1997), for example, found that race had both direct and indirect effects on perceived caregiver rewards. Black caregivers in general, and Black caregivers who received comfort from religion and prayer, reported greater rewards in the caregiving role than did White caregivers. Picot suggests that for many African American caregivers faith and prayer may act as a buffer to caregiving stresses and may be associated with perceptions of rewards, such as being blessed by God for their efforts. They note that many African American families have an expectation of caring for an aging relative and that there is often a sense of pride in being able to fulfill this role. Similarly, White et al. (2000) found that African American adult daughters reported less stress and more rewards in the parent care role than White adult daughters did. They suggest that these racial differences may reflect a normative expectation among African American families of impairment as a part of the aging process as well as more positive and respectful views of elders. More recently, Roff and her colleagues, in a study of 275 African American and 343 Caucasian caregivers of individuals with Alzheimer’s disease, also found that African Americans appraised their caregiver role more positively than Caucasians (Roff et al., 2004). Consistent with the Picot et al. (1997) study, they found higher religiosity among the African Americans; and, this higher religiosity partially mediated the relationship between race and positive appraisal. Their data also revealed that for African Americans, higher positive appraisals were associated with lower anxiety, lower bother by care recipient’s behavior, and lower socioeconomic status. Together, these studies underscore the importance of understanding how sociocultural variables influence both caregiver appraisals and outcomes.

Caregiving as a Process: Conceptual Frameworks

Although social scientists recognize that caregiving is a dynamic process that unfolds over time there are still relatively few longitudinal studies of caregiving outcomes. Moreover, given that these studies are characterized by a variety of measures and variability in samples, the results are inconsistent and difficult to compare. Yet, despite the continued reliance on cross-sectional or panel studies, researchers have increasingly adopted process models as an analytic tool to understand the caregiving experience. One of the most frequently used conceptual frameworks is the stress and coping process model (Lazarus & Folkman, 1984; Pearlin, Mullan, Semple, & Skaff, 1990; Pearlin & Schooler, 1978). This model differentiates between the occurrence of stressful events and how individuals react to and appraise them. In fact, recent research suggests that subjective appraisals of caregiver stressors may be better predictors of caregiver outcomes than objective stressors (Gonyea, O’Connor, Carruth, & Boyle, 2005; Pot, Deeg, van Dyck, & Jonker, 1998). Schulz, Gallagher-Thompson, Haley, and Czaja (2000), therefore, propose ways in which this theoretical model may be used by professionals to guide the development, implementation and evaluation of caregiver interventions whether throughout targeting the actual stressors, strengthening social supports and/or altering caregivers’ appraisals of behaviors and events.

Research on the caregiving experience has often focused almost exclusively on the help given to vulnerable family members and ignored the contributions of the care recipients. The use of social exchange theory, which emphasizes the interdependence in dyadic relationships, offers us a greater understanding of bidirectional or mutual exchanges between older parents and their adult children (Dwyer, Lee, & Jankowski, 1994; Walker, Martin, & Jones, 1992). In fact, there is a growing body of research focused on caregiving as a dyadic process, which incorporates the perspectives of both the caregiver and receiver (Lyons, Zarit, Sayer, & Whitlatch, 2002). Much of this literature focuses on the attachment bonds between adult daughters and older mothers and the ways in which they negotiate care (Carpenter, 2001; Hollis-Sawyer, 2003; McGraw & Walker, 2004).

A number of social scientists have used role strain theory—felt difficulty in fulfilling role obligations—to explore the dynamics of elder care. Role strain theory argues that individuals have a limited amount of time and energy and that social organizations demand most of that energy (Marks, 1977). Barnett and Baruch (1985) identified two types of role strain that impact women who are simultaneously occupying various roles such as spouse, parent, employee, and/or elder caregiver: role demand overload and role conflict. Role demand overload is having so many demands related to one’s roles that satisfactory performance is improbable. Role conflict emerges when the demands from multiple roles are such that adequate performance of one role jeopardizes adequate performance in another role. Role strain theory has been employed as a conceptual framework in the burgeoning field of work-family research. Recognizing that the worlds of family and work are not separate or parallel domains, researchers are increasingly exploring the specific intersections, transactions, and spillovers between family and work (Marks, 1998).

Emphasizing the concept of “caregiving as a career,” an increasing number of researchers are employing a life course-role identity perspective to understand the caregiving role and its consequences. Drawing on the principles of the life course perspective (Elders, 1992), Moen and her colleagues propose that to understand the consequences of caregiving role on well-being, one must consider the developmental timing of transitions to the caregiver role, the intersection of other roles in relation to the carer role, and the historical context of performing the caregiver role (Moen, Robison, & Fields, 1994). Whereas most researchers study the work-family interface in terms of individuals at one point in time, Han and Moen (1999, p. 98) propose “a coupled-careers model directly addressing the multiple interfaces between work and family and between men and women as they unfold overtime.” Central to Moen’s work is the concept of “linked lives”—that is, “individuals’ life paths are played in tandem with the life courses of parents, spouses, children, friends, and co-workers” as well as a recognition that there are clear gendered cohort differences in normative expectations for later life (Moen, 2001, p. 181). Thus, Moen stresses “that men and women frequently experience different transitions and trajectories in later adult years, but even the same trajectories can be vastly different by gender” (Moen, 2001, p. 184). Based on her work, Moen concludes that we must create more flexible and open arrangements to support both men and women in balancing work and family at all stages of their life course.

Caregiver Interventions

During the past two decades interventions or services to support caregivers—psychoeducational interventions, support or mutual help groups, and respite and adult day care—have proliferated. Whereas the primary objective of psychoeducational interventions and support groups is typically to enhance caregiver competence, coping, and/or well-being, the primary aim of respite care and adult day care is to reduce the amount of care provided by the family member. Both types of interventions, however, are typically motivated by an overarching goal of prolonging the family’s ability to provide care, reducing institutionalization, and thereby saving public monies (Hooyman & Gonyea, 1995). Early studies on the effects of caregiver interventions, which relied primarily upon the clinical impressions of the group leaders or professionals or satisfaction surveys of small, select samples of caregiver consumers, often reported strong positive effects (Toseland & Rossiter, 1989). However, more recent studies that use standardized measures of caregiver distress and more rigorous research designs have found, on average, only small to moderate effect size on caregiver outcomes (Knight, Lutzky, & Macofsky-Urban, 1993). Based on their review of 29 studies of respite care, for example, McNally, Ben-Shlomo, and Newman (1999) conclude that respite care produced neither consistent nor enduring positive effects on the carer.

Arguing that the geropsychotherapy field is still in need of better process and outcome measures, Arean and her colleagues offer guidelines for the optimal conduct of intervention research that include the selection of age-appropriate therapies and control conditions as well as treatment outcomes, the use of consumer-based methods for recruitment, and adjusting the research design to accommodate age-specific life events (Arean et al., 2003). There is also growing emphasis on the importance of culturally-sensitive or appropriate interventions and the need to assess possible differential effects of interventions by race and ethnicity (Burgio, Stevens, Guy, Roth, & Haley, 2003; Gallagher-Thompson et al., 2003).

Based on their analysis of the body of dementia caregiver research published between 1996 and 2001, Schulz and colleagues report that “although many studies have reported small to moderate statistically significant effects on a broad range of outcomes, only a small proportion of these studies achieved clinically meaningful outcomes” (Schulz et al., 2002, p. 589). Noting that the caregiver intervention research is now more than a decade old, they urge researchers to focus on demonstrating clinical significance—that is, “the practical value of the effects of an intervention, or the extent to which it makes a ‘real’ difference in the everyday life of the individual” (Schulz et al., p. 590).

Sorensen, Pinquart, and Duberstein (2002) meta-analysis of 78 intervention studies offer important insights into the effectiveness of six types of programs—psychoeducational interventions, supportive interventions, respite and adult day care, psychotherapy, care receiver competence interventions, and multicomponent or combination intervention—on the outcomes of caregiver burden, depression, coping, sense of well-being and knowledge as well as care recipients’ symptoms. Moreover, the researchers investigated the moderating influence the sets of variables—intervention characteristics, caregiver characteristics, and care recipient characteristics—on the targeted outcomes. Based on their analysis, Sorensen and her colleagues conclude that: (a) psychoeducational and psychotherapeutic interventions produce the most consistent short-term effects; (b) while group interventions were less effective at improving caregiver burden and well-being than individual or combined individual and group interventions, group interventions were more effective with regard to improving caregiver knowledge and ability and care recipient symptoms; (c) length of treatment was important in terms of alleviating caregiver depression and care receiver symptoms; (d) adult children appeared to benefit more from interventions than spousal caregivers; and (e) dementia caregivers experienced fewer benefits from interventions than did carers for elders with other disabling types of conditions.

Together, these meta-analyses identify the existing challenges and opportunities in conducting caregiver intervention research and offer creative strategies for producing scientifically rigorous and clinically meaningful evaluations of practice interventions for older adults and their families. They also suggest that professionals’ choice of intervention strategies should be guided by the primary treatment goals as well as the characteristics of the caregiver and receiver (i.e., gender, age, ethnicity, functional status); and, professionals should maintain realistic expectations with regards to the targeted caregiver and receiver outcomes.

Supports for Family Caregivers

Surveys consistently reveal that Americans are experiencing stress in balancing their work and family responsibilities and wish that both government and businesses would take a more active role in seeking solutions to resolve this tension. For example, the 1998 National Partnership for Women and Family Survey found that two-thirds of Americans say that time pressures on working families are getting worse, not better, and that most want employers (90%) and government (72%) to do more to help working families. The United States’ lack of an explicit and coherent national family policy, however, has led to the creation of ambiguous and sometimes conflicting policies for specific programs to address the long-term care needs of chronically ill or disabled older adults. While few disagree with the abstract statement that the state should facilitate families’ ability to care for dependent members, translating that philosophy into specific public policies and programs has proven challenging.

America’s domestic policies are based on an ideology of familism—that is, the assumption of the primacy of families in meeting the care needs of their members. Predicated on the notion of “families first,” formal alternative arrangements are perceived as necessary only when families are unavailable or unable to perform their responsibilities. Clark (1993) notes that both the United States and Canada embrace an ideology of familism; however, Canada tempers familism through an emphasis on collectiveness or sense of community whereas the U.S. stress on individualism heightens familism:

[The United States] particularizes and compartmentalizes social policies along the lines of individual or static group-based need, rather than seeing public programs as responding to changing life course needs across the entire society. (p. 34)

These values of familism and individualism have formed the basis for residual approach to public policy in the United States. Residualism—meaning that the state becomes involved only after the family has assumed as much responsibility as possible—serves the federal government’s goal of cost containment of public funds. Research by Binney, Estes, and Humphers (1993) and Glazer (1990) found that efforts to contain Medicare costs, through such mechanisms as the 1983 Medicare Prospective Payment system, have transferred work (and costs) from the formal to the informal care system. Because of the shifting of highly technical work to unpaid family members, Glazer estimates that the medical industry has saved $10 billion annually in costs. More recent limits on Medicare home health visits resulting from the 1997 Balanced Budget Act have placed even greater demands on family caregivers. Focusing on Medicaid (the primary method through which the United States distributes long-term care), Harrington Meyer and Storbakken (2000, p. 217) demonstrate how recent Medicaid cost-containment efforts are reshaping frail and disabled elders’ access to long-term care and “shifting the burden back to families.”

One of the primary barriers to the expansion of formal (paid) home care services for chronically ill or disabled elders is the fear that policymakers have that it will cause families to reduce or curtail their caregiving efforts (Binney et al., 1993; Hooyman & Gonyea, 1995). This concern or fear continues to exist despite almost two decades of research revealing that formal in-home care services do not appear to substitute for informal care provided by family, friends, and neighbors. Penning (2002) study, for example, provides convincing evidence refuting the “substitution hypothesis.” She found that the extent of self-care and informal care received was most strongly related to the level of health care need (i.e., chronic conditions, functional impairment; cognitive impairment) and, in the case of informal care, the availability of informal resources. Summarizing her findings, Penning notes that contrary to expectations, there was no evidence that as older adults’ health declines, “self-care gradually gives way first to informal care and ultimately to formal care. Rather they seem to suggest that as health declines, all forms of care (self, informal, and formal) may increase” (Penning, 2002, p. 14).

While policymakers continue to remain concerned with the willingness of families to provide care, two recent significant—albeit modest—federal policy initiatives have been directed toward supporting families’ ability to perform caregiving activities: The Family and Medical Leave Act of 1993 (FMLA) and the National Family Caregiver Support Program of 2000 (NFCSP). Having been debated in the U.S. Congress since the mid 1980s, the Family and Medical Leave Act (Public Law 103–3) was signed into law by President Clinton on February 5, 1993. The private sector generally opposed the passage of the FMLA perceiving it as costly and not beneficial to businesses; whereas many women’s organizations stressed the necessity for a federally-mandated leave policy noting the difficulties that American workers were confronting in their efforts to balance their work and family lives. Women’s organizations further buttressed their argument by underscoring that the United States was among the last industrialized countries to offer a leave policy.

Under the FMLA, businesses with 50 or more employees are required to grant up to 12 weeks of unpaid leave annually when a child is born or adopted, when an immediate family member with a serious health condition needs care, or when the employee is unable to work because of a serious health condition. The FMLA defines a serious health condition as an illness, injury, impairment or physical or mental condition that requires inpatient care in a hospital, hospice, or residential medical facility or continuing treatment by a health care worker. The worker has the right to take the leave intermittently or on a reduced schedule basis. The employer is required to maintain any preexisting health coverage during the leave period, and once the leave was completed, reinstate the employee to the same or equivalent job. For retirement and pension plans, FMLA leave is counted as continuous service for purposes of eligibility and vesting.

As the FMLA only applies to employers with 50 or more employees, it is estimated that slightly more than 10% of private sector worksites are covered under the Act. This relatively small percentage of U.S. businesses, however, employs almost 60% of American workers. Yet, not all workers of covered employers are eligible for FMLA benefits. The Act only provides benefits to employees who have worked for an employer for at least 12 months and who, during that 12-month period, worked for at least 1,250 h (an average of 25 h per week). Based on these criteria it is estimated that less than half (46.5%) of private sector employees are eligible for leave under FMLA (Commission on Family and Medical Leave, 1996).

As mandated by federal law in 2000, the U.S. Department of Labor (DOL) commissioned a survey of usage of FMLA. The DOL survey revealed that since its enactment, usage has been fairly limited; approximately 35 million American workers have taken leave under the FMLA with the median length of leave being 10 days. Among leave takers, approximately half (52%) did so because of their own health; about 39% was taken by relatively young parents to care for children at birth, adoption, or during a serious illness; and about 19% was taken by somewhat older employees to care for ill parents or spouses. Although latent demand for leave may be high, the fact that it is unpaid is a major barrier to utilization. Approximately three-quarters of workers who desired to take a leave stated that they could not do so because of financial reasons.

Proposals to reform FMLA continue to be debated in U.S. Congress. On the right, legislators continue to question the basic premise of the law. Republican members of Congress, in an effort to limit the FMLA, are pressing for clarifying amendments to restore the definition of “serious medical condition” and “intermittent leave” to the original intent of the Act. In contrast, Democratic proposals build upon the ideology that government has a legitimate role in addressing employment-related tensions between employer and employee. Their most fundamental proposal focuses on the importance of wage replacement and the need for paid leave. Democrats have also proposed to extend coverage to: (a) employers with at least 25 employees; (b) beyond “immediate family members” to include a parent-in-law, adult children, sibling, grandchild, grandparent, and domestic partner; and (c) allow workers to take up to 4 h in any 30-day period to accompany children to school or extracurricular activities or accompany ill relatives to medical appointments. In the Republican and Democratic proposals, we saw different visions about personal, private, and public responsibilities for care of dependent members (Hudson & Gonyea, 2000).

The National Family Caregiver Support Program (NFCSP) was established through enactment of the Older Americans Act Amendments of 2000 (Public Law 106–501) and reauthorized in 2006. The federal share of the funding is 75% with the remaining 25% provided by the state and local sources. The NFCSP gives a higher priority to services to older caregivers having the greatest social and economic needs (emphasizing lower-income older individuals), and to older caregivers of persons with mental retardation and developmental disabilities. Recognizing the growing phenomenon of grandparents who are the sole caregivers for grandchildren, the Act also allows each state to use some funds to support grandparents who are aged 55 or older and relative caregivers of children who are not more than 18 years of age or handicapped children of any age. The principal component of the NFCSP calls for all states to work in partnership with the local Area Agencies on Aging (AAAs) and other service providers to establish a network of services. The five service areas outlined under the NFCSP are: (1) information for caregivers about available services; (2) assistance to caregivers in gaining access to services; (3) individual counseling, organizing of support groups, and caregiver training; (4) respite care; and (5) supplemental services, on a limited basis, to complement the care provided by caregivers. A number of current NFCSP activities implemented at the state level should greatly expand the development and testing of evidence-based approaches to support, complement, and sustain the efforts of families to care for dependent relatives.

Conclusion

The demographic changes of the aging population mean more years of “shared living” between generations. Family relationships are of unprecedented duration as parents and children now share five decades of life, siblings may share eight decades of life, and the grandparent–grandchild bond may last two or even three decades. We might therefore anticipate that multigenerational bonds will be of greater importance in the upcoming decades. Although there is great heterogeneity in the time and sequencing of adult life transitions, increases in life expectancy have resulted in middle age becoming the normative life stage in which adult children confront parental declining health and death. Despite the popular rhetoric that the American family is in decline, research has consistently demonstrated the strength and resilience of family members’ bonds across the generations. The vast majority of long-term care to older frail and disabled relatives is provided by families—typically spouses, daughters, and daughters-in-law. While male and female roles within American families have changed over the past few decades, women remain the primary nurturers, kinkeepers, and caregivers.

The caregiving experience is diverse, complex, and dynamic. Appraisal of the parent care role is shaped by our family experience, history, and values. A growing number of researchers are exploring gender, race, and cultural differences in symptom appraisal, family management of the disease and illness, caregiver stress and burden, coping strategies, and help-seeking behaviors. Although much of the caregiving literature has focused on caregiver stressors and role strain, attention is now also being directed toward positive gains in assuming the caregiving role. Conceptualizing caregiving as a dynamic process has led to a growing number of researchers adopting a life course perspective to understand the nature of the work-family interface as it unfolds over time.

Although most women, men, and children now live in households that bear little resemblance to the male breadwinner and female homemaker families that symbolized the 1950s, both public and private sector policies have not kept pace with the changing realities of work and family. The challenge is to understand the impact of our social policies on the quality of life for both the caregiver and recipient. Madonna Harrington Meyer (2000, p. 2) emphasizes that as we search for policy reforms we must assess: “What are the complexities, strengths, and weaknesses of emphasizing families, market-based solutions, or welfare state programs?” and “How might we create a mixture of options that balance the burden across all three spheres?” Policies such as the Family and Medical Leave Act of 1993 and the National Family Caregiver Support Program of 2000 represent efforts to address this question of what is society’s shared responsibility for care of dependent members.