| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Douglas A. Wolf, Center for Policy Research, Syracuse University, 426 Eggers Hall, Syracuse, NY 13244-1020. E-mail: dawolf{at}maxwell.syr.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Parental care needs • Informal care • Positive and negative consequences of care • Caregiver's family
The theories that have been developed to understand caregiver stress and its consequences indicate both that stressors include factors other than caring activities and that the stress process affects persons other than caregivers. Yet the complexities implied by these theoretical frameworks have not fully informed empirical research on the consequences of informal caregiving. For example, among the studies reviewed in three published surveys (Schulz, O'Brien, Bookwala, & Fleissner, 1995; Schulz, Visintainer, & Williamson, 1990; Yee & Schulz, 2000), a total of 80 studies inferred an association between caregiving and mental health when they compared mental health indicators in caregiver-only samples to population-based norms for those indicators. Only 13 studies drew conclusions from comparisons of caregiver samples to matched comparison groups of noncaregivers.
Both the one-sample and the two-sample approaches mentioned herein suffer from methodological problems that complicate the task of identifying impacts uniquely attributable to caregiver activities. First, the individuals in a sample of caregivers differ from either the population at large or a matched comparison group in ways other than the fact that they are all providing care. Perhaps the most important difference is that they have a close family member whose illness or disability creates a need for care. Thus, differences in outcomes across groups cannot be uniquely attributed to caregiving activities. Second, individuals in a matched comparison group of noncaregivers, and certainly those in the population at large, may experience stressors associated with informal caregiving even if they are not themselves providing care. Finally, caregivers experience the rewards as well as the stressors associated with caring, whereas noncaregivers do not. It is also worth noting that most caregiver research focuses on persons caring for individuals with substantial care needs, often victims of Alzheimer's disease or related dementias. Caring associated with moderate levels of need has received less attention. Thus, for several reasons, it may be inappropriate to base conclusions about the magnitude of mental health impacts of caregiving on comparisons of caregivers to noncaregivers, without further controls for other care-related stressors.
Particularly interesting is the proposition that a family member's needs for assistance are an independent source of stress, because this raises the possibility of parallel mental health consequences among noncaregivers as well as caregivers. In this regard, Schulz and colleagues (1995) noted that
Caregiving researchers ... assume that the majority of the negative effects of caregiving are attributable to the tasks of providing care. ... An alternative attribution is that the distress among caregivers is caused by the decline and impending death of someone they love. (p. 788)Our research proceeds from this observation and attempts to move beyond a simple "caregiver–noncaregiver" distinction, recognizing a variety of relevant factors that influence mental health outcomes. We focus on adults' caring activities directed at their older parents. Following the suggestion from Schulz and colleagues just quoted, we pay particular attention to the distinction between the parents' care needs and their child's care activities. We also take into account the family setting, including the possibility that the care activities of one's spouses and siblings may produce stresses that diffuse throughout the family network.
Theoretical Framework
Current approaches to stress found in the research on informal caregiving originate in a general stress and adaptation literature, where stress has been viewed as a complex process with interrelated domains of "sources," "mediators," and "manifestations" (Pearlin, Lieberman, Menaghan, & Mullan, 1981; Pearlin, 1989). In that literature, stress is defined as an organism's response to a variety of fundamentally intolerable situations: discrete life events, possibly converging with chronic life strains (such as interpersonal conflicts, interrole conflicts, or role captivity), and stimulated by certain self-concepts and values. The outcomes of stress can manifest themselves on multiple levels of physical and mental health.
The conceptual schema typically used in studies of caregiving stress is grounded in this general framework (Biegel, Sales, & Schultz, 1990). Caregiving stressors are the difficulties created by the demands for care, which are disruptive for a caregiver insofar as they exceed an individual's ability to adapt (Aneshensel, Pearlin, Mullan, Zarit, & Whitlach, 1995). Stress outcomes (certain fears, tensions, and worries) are mediated by a complex web of background characteristics, such as the caregiver's physical, social, and economic status, and available internal and external resources. More specifically, researchers differentiate between primary and secondary stressors. Primary stressors are closely related to the care recipient's impaired condition. They include objective conditions related to the patient's disability (such as cognitive impairment, problematic behavior, and ADL, or activities of daily living, dependencies), as well as subjective reactions (such as feelings of overload and captivity) to those conditions. Secondary stressors include problems that arise as a result of caregiving but outside of caregiving roles, for example in the family or at work (Aneshensel et al., 1995). Gatz, Bengtson, and Blum (1990) noted that such frameworks in general have been developed with an individual caregiver, rather than a family, context in mind. The usual frameworks represent a model of stress proliferation for one caregiver and for one caregiver only. They have generally not been used to explore how the patient's disability becomes problematic for persons not directly involved in the provision of care.
Although most research on the psychological consequences of caregiving emphasizes its problematic aspects, several researchers have recognized that being a caregiver can produce psychological benefits as well as costs, or "rewards" as well as "hassles." Enjoyable aspects of caregiving reported by many caregivers include an enhanced caregiver–care recipient relationship, personal growth, caring self-competence, satisfaction with social involvement, increased understanding of the aging process, ability to provide good care, and other specific uplifts in the caregiving process. Although the evidence concerning the association of these aspects of caregiving with caregiver well-being is mixed (Kramer, 1997), one study found that the number of positive aspects of caring was associated with significant decreases in burden and depressive symptom scores, and with better self-assessed health (Cohen et al., 2002), and another found that the number of rewards reported by socially disadvantaged caregivers was associated with lower burden scores (Riedel, Fredman, & Langenberg, 1998). Similarly, Martire, Parris Stephens, and Atienza (1997) found a negative correlation between caregiving satisfaction and depression symptoms scores. An important aspect of our work is the fact that noncaregivers are unable to experience the rewarding aspects of caring activities, but they still have the possibly stressful experience of having a family member with care needs.
Viewing caregiving issues as a family-system problem stimulated the emergence of a third strand of caregiving research, one that emphasizes the impacts of care needs and care activities on the family, which is to say on noncaregivers as well as caregivers. This approach is exemplified by Brody, Hoffman, Kleban, and Schoonover (1989):
[T]he homeostasis of the family as a whole is affected by a disturbance in any of its parts. An elderly parent's disability and need for care is such a disturbance; each person in the family experiences the repercussions, and changes occur in the balancing of individual roles and responsibilities. (p. 529)Family members other than the primary caregivers frequently provide care and experience a certain amount of stress (Cicirelli, 1992; Zarit, Reever, & Bach-Peterson, 1980). However, noncaregiving family members can also be adversely affected by caregiving situations. First, close relatives of a caregiver may suffer from the diversion of the caregiver's time, attention, and care away from them and toward the care recipient, as well as from accompanying changes in the family's lifestyle and conditions (Bass 1990; Brody 1985). Second, the caregiving situation can produce intense family conflicts over the caregiving arrangements in many caregiving families, when the primary caregiver's strain is "radiated" to other family members (Gatz et al., 1990; Strawbridge & Wallhagen, 1991; Toseland, Smith, & McCallion, 1995). Finally, noncaregiving relatives can be affected simply by the primary "disturbance," that is, the patient's disability, and therefore experience some negative outcomes, such as chronic grieving (Bass, 1990) or guilt (Brody et al., 1989).
Our analysis rests on a model of the stress process that reflects these strands of past literature. The essential elements of our model are illustrated in Figure 1. In key respects the model is adapted from past work, such as the model shown in Lawton, Moss, Kleban, Glicksman, and Rovine (1991). The model pathway most often featured in past research is represented by the sequence care needs 
care by self distress psychiatric morbidity. In this sequence distress is an inner state, not directly measurable, whereas psychiatric morbidity is a measurable manifestation of the underlying distress. Among caregivers, observed indicators of psychological health are influenced by both the stressful and the rewarding consequences of caring. We suggest, however, that there is a distinct version of the stress process that applies to noncaregivers, which is represented in our model by the sequence care needs distress psychiatric morbidity. Noncaregivers experience the stress of having a parent that needs care, but they do not experience the consequences of meeting those needs in part or in full. The noncaregiver's perceived stress is not mediated by caring behavior, and therefore the observable manifestations of the care needs distress relationship for noncaregivers and the care needs care by self distress relationship for caregivers may be quite different. Finally, the stress associated with having a spouse or sibling who is providing care is represented by the care needs care by others distress psychiatric morbidity pathway.
|
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Dependent Variable
Our indicator of psychiatric morbidity is a shortened version of the Center for Epidemiological Studies Depression (CES-D) scale administered in the HRS as a part of its Emotional Health Measures module. The original scale was designed to measure the current level of depressive symptoms in the general population (Radloff, 1977). Whereas a formal diagnosis for depression requires a clinical examination, the self-reported CES-D emphasizes subjective and affective elements of depressive symptoms (McDowell & Newell, 1996). The shortened version of the CES-D used in the HRS included 11 out of the 20 original items. The response categories are "none or almost none of the time," "some of the time," "most of the time," and "all or almost all of the time," coded 0 through 3, respectively. An overall score was obtained by adding the 11 items, producing a scale that ranges from 0 (least depressed) to 33 (most depressed).
Independent Variables
Need
We created a dummy variable that indicates whether at least one of the respondent's parents or (for married respondents) parents-in-law "needs any help with basic personal needs like dressing, eating, or bathing," or, as a result of a health condition, cannot "be left alone for an hour or more," which in the original HRS format represented two separate questions with "yes" and "no" response categories. Because the first question refers specifically to ADLs, whereas the second one may imply a severe cognitive limitation, our "need" variable identifies parents with fairly severe disabilities. Because there are no measures of parents' limitations in instrumental ADLs (IADLs; e.g., housework, shopping, meal preparation, and money management) in the first wave of the HRS, we are mainly focused on the analyses of outcomes of relatively extensive personal care needs.
Caregiver
We constructed a dummy variable indicating a person's caregiver status. The following question was used by the HRS to obtain information about personal care provisions: "Have you [or your (husband or partner)] spent 100 or more hours in the past 12 months helping your parent(s) (or stepparents) with basic personal needs like dressing, eating, and bathing?" Respondents were specifically instructed to exclude time spent on transport, shopping, cooking, and paying bills. The 100-hr threshold was adopted in the HRS in order to filter out low-intensity (1–2 hr per week) caring activities (Soldo & Hill, 1995). Those responding positively to this question were asked to specify (for themselves and their spouses separately) the number of hours spent helping each parent. A separate set of questions about help given to the respondent's spouse's or partner's parent(s) (or stepparents) for more than 100 hr in the past 12 months, as well as the subsequent questions specifying the hours of help, was asked in the same order. Single persons who spent 100 hr or more caregiving, as well as married persons who met the 100-hr criterion as a couple (while also individually contributing more than 0 hr) by helping either spouse's parents, were coded as caregivers. As a result, whereas most caregivers personally spent 100 hr or more helping their own or spouse's parents, only 6 married females (and 41 married males) coded as caregivers personally spent more than 0 but less than 100 hr on caregiving yet met this requirement as a couple.
Because the HRS obtains separate information about caregiving activities of husbands and wives in married couples, we were able to define the caregiver variable for each spouse and create a spouse caregiver variable for married people. The latter variable is zero by definition for single respondents. The dummy variable, sibling caregiver, indicates whether any of the respondent's (or the spouse's) siblings has helped their parents with basic personal needs in the past 12 months. In contrast to the HRS question directed at the respondent's and spouse's care activities, the question about sibling help to parents specifies no minimum number of hours of help given. Finally, we created a variable, parent institutionalized, indicating whether any one of the respondent's or the spouse's parents (or stepparents) was living in a nursing home at the time of the interview.
When the need, caregiver, and parent institutionalized dummy variables were constructed, the needs, care, and living arrangements of the spouse's parents were not considered separately from those of the respondents' own parents. Similarly, both husband's and wife's siblings' care activities were included in the sibling caregiver variable. A more detailed set of distinctions, recognizing the gender of caregiver and care recipient, as well as own-parent versus parent-in-law care, is possible with these data. However, the complexity introduced through these distinctions, particularly in combination with the Need x Caregiver interactions that are important for the hypotheses we test, quickly overwhelms the analysis. In the present article the effects of key variables among married couples are their average effects; that is, all possible intrafamily patterns of care needs and caring activities have been averaged over.
Additional Variables
We created three dummy variables reflecting respondents' current health (poor health), change in health compared with 1 year ago (health decline), and current disability status (disabled). The poor health variable is coded as 1 if a respondent rated his or her health as "fair" or "poor." The health decline variable highlights cases in which the respondent's health is self-described as "somewhat worse" or "much worse" than last year. The third variable indicates whether the respondent, in answering a question about employment status, states that he or she is "disabled and unable to work." The same variables are used to define health and disability measures for married respondents' spouses. Additional individual-level variables measure marital status (i.e., divorced or separated, widowed, and never married, with married or cohabiting respondents comprising the omitted category), and two indicators of educational attainment. The first is a dummy variable indicating that a person has received a high school diploma or passed an equivalency test; the second indicates that the respondent has earned a college degree or higher.
Wealth (which is a couple-level variable for married respondents) measures total net worth, including total housing and nonhousing equity. Family context is further represented by a dummy variable indicating whether the respondent lives with a child who is younger than 18 years old.
Several possible covariates, such as respondent's employment status or proximity to the parents' residence, were not included in our model in order to avoid the possibility of endogeneity bias; that is, a person's work can contribute to caregiver stress in different ways, and the excessive caregiving burden can in turn affect working arrangements, in order to fit the responsibilities within the family.
Multiple Regression Analysis
We present results from four alternative forms of a multiple regression analysis of the correlates of depressive symptoms. The simplest form of the regression corresponds to the caregiver–noncaregiver distinction found in the literature, namely
|
|
1 is the regression-adjusted difference in the mean depressive-symptoms score between caregivers and noncaregivers.
As already noted, much past research on caregiver stress has focused on care in the presence of severe care needs, comparing outcomes to population norms or to groups of noncaregiving adult children of relatively healthy parents. For our sample, which includes children of parents with a range of care needs, the most severe of which are captured by the need variable, a model likely to produce results more comparable with the conventional ones takes the form
|
|
2 is the regression-adjusted difference in the mean depressive-symptoms score between caregivers of parents with severe care needs and all others.
Our third model allows us to estimate stress outcomes associated with various pathways, discussed in the theoretical framework, by separating the effects of parental needs for care and the child's caring activity:
|
|
3 represents a regression-adjusted effect on the mean depressive-symptoms score attributable to having a parent with severe personal care needs, among noncaregivers, whereas 4 is the effect of providing care in the absence of severe care needs. The effect uniquely attributable to care provision while having a parent with care needs is calculated by adding coefficients 4 and 5. The fourth and final variant on our model adds to Model 3 the "family" variables of spouse caregiver and sibling caregiver as well as the variable indicating the use of institutional care. Because one's own caregiving stress process may vary depending on a spouse's or sibling's participation in the care network, we also add interactions of caregiving by self and caregiving by a spouse or sibling.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
The regression results displayed in Tables 4 and 5 include only the key variables of interest; results for other variables included in the regressions are presented in the appendix. Similar to the findings from previous studies, in Models 1 and 2 the caregiving activities are associated with significant increases in depressive-symptoms scores for female caregivers: female caregivers in general, Model 1, as well as female caregivers dealing with a parent's severe care needs, Model 2, had significantly higher CES-D scores than noncaregivers (estimate of 1 = 0.497 with SE = 0.235; estimate of 2 = 0.576 with SE = 0.260). Caregiving is not, however, significantly related to the outcomes for male caregivers.
|
|
|
4 + 5) = 0. The test fails to reject the null hypothesis (t =.208; SE = 0.754), suggesting that caregiving activities do not lead to higher levels of depressive symptoms once parental needs have had their effects. Indeed, the needs variable alone (independent of care provision) is associated with a significant increase in CES-D scores (estimate of 3 = 0.687; SE = 0.159). Similar results are obtained after adding the family context variables spouse caregiver, sibling caregiver, and the corresponding interactions with the caregiver variable. As Table 4 reports, both the caregiver dummy and the interaction of caregiving with severe parental needs have an insignificant effect on the dependent variable, whereas the variable need by itself continues to produce a significant increase in the depression symptoms score. In Model 4, having a caregiver sibling significantly increases a noncaregiver's CES-D score, a finding that appears for both men and women. However, having a caregiving spouse shows no significant relationship with the depressive symptoms, regardless of the respondent's caregiving status.
Although we failed, in general, to find significant effects of caregiver activities on men's CES-D scores, it can be seen that the estimates of effects in men's Models 1 and 2 are comparable in size with their counterparts in the regressions for women. The similarity of these estimates indicates that increasing the sample size for male caregivers might produce significant differences that are comparable for both groups. The results in the men's version of Model 3 were similar to those in the women's version: parental needs produced a significant increase in the CES-D scores (estimate of 3 =.283; SE = 0.138), whereas no effect of care activities was found. This pattern of similarity of effect sizes and strengths does not hold, however, for Model 4.
In addition to the models using the caregiver dummy variable, we estimated Model 4 by using, in its place, a continuous variable measuring hours of care. Selected results are shown in Table 6. For women, and consistent with the previous findings, parental need significantly increases CES-D scores. However, in this case caregiving hours in the absence of parental needs (i.e., the main effects of hours of care) produce significant increases in depressive symptoms scores. Nevertheless, care hours in the presence of severe needs have essentially no effect on the dependent variable: the main effect and the interaction effect cancel each other out. For men, no significant effects of parental needs or own caring activities are found. Finally, in these regressions we again find significant adverse effects of having a sibling caregiver, for both men and women (sibling caring activity is represented here by a dummy variable, because the HRS data do not include hours spent caring by siblings).
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
When the impact of caregiving in the absence of severe care needs was examined, no significant association between caregiving and depressive symptoms score was found. In such circumstances, care arrangements may not be absolutely necessary, or care needs may be minimal, and therefore observed caring activity is likely to be more voluntary in nature, prompted by the caregiver's filial affections, and consequently more pleasant or rewarding than stressful. Although most past caregiving research overlooks this type of care arrangement, focusing mainly on the care provided as a result of a severe disability, some researchers find that caregiving in the absence of a disability is not a rare phenomenon. Bootsma-van der Wiel and colleagues (2001) examined the differences between the concepts of competence (the ability to perform an activity without help, with or without difficulty) and the actual performance of basic ADLs and IADLs. The study found that 22% of women (10% of men) lacked the ability to perform one or more ADLs without assistance, yet 35% (22% of men) received help with one or more activities. In other words, a substantial proportion of men and women who receive ADL assistance could perform the ADL task[s] by themselves, albeit in some cases only "with difficulty." Receipt of assistance in these circumstances is not, of course, necessarily inappropriate and almost certainly improves the recipient's quality of life. The study's authors argued that these discrepancies can result from various causes: as a way to optimize the use of time; as a result of long-term division of roles (especially for men); or as a result of personal circumstances. In addition, Bootsma-van der Wiel (2001) noted that "inactivity can be a negative side effect of receiving help. Caregivers may stimulate inactivity and learned helplessness by taking over more activities than necessary" (p. 913). In our study, which measures parental disability through a child's report of parental competence (i.e., "can do" questions), it is interesting to find that care that is delivered despite parental competence is not associated with additional depressive symptoms.
Perhaps the most important finding of this study pertains to the impact of severe parental care needs on the depression score. We found strong support for the hypothesis that parental care need is associated with depressive symptoms among noncaregivers. In past studies in which caregiving was accompanied by severe care needs, the effects of parental needs were obscured by the pervasive focus of research on caregivers, and, as a result, the observed adverse consequences were mostly attributed to the provision of care. Meanwhile, a separate examination of parental care needs for noncaregivers allows us to see that this distinct stressor may in fact produce a more negative effect than caregiving itself.
We found no evidence of increased manifestations of stress among persons caring for severely disabled parents, beyond the stressors associated with parental care needs. The estimated effect of caregiving, conditional on the presence of severe care needs, had a negative sign in our regressions. Because the CES-D score used in our analysis represents a measure of net impact of caregiving, combining both negative and positive affects, we believe that the negative effect of the primary stressors, such as parental care needs and caregiving burdens, is mediated (and substantially reduced) by caregiving rewards and uplifts, previously found to be associated with the decrease in burden and depressive symptoms (Cohen et al., 2002; Riedel et al., 1998). This interpretation is supported by our finding that the negative impact on noncaregivers, who do not experience rewards associated with the provision of care but are stressed as a result of parental disability, appears to be stronger than on caregivers. Unfortunately, we do not have an independent measure of positive aspects of caregiving, which would clarify this finding.
The comparison of depressive symptoms among caregivers' family members supports the idea that having a caregiving relative can become a separate pathway of stress for noncaregivers. Although we have found no evidence of increased depressive symptoms among caregivers' spouses, the impact of having a caregiver sibling was significant for both men and women. One possible interpretation for finding a strong negative impact of having a sibling caregiver and no such effect for caregivers' spouses can be considered here. We would expect to observe a greater consensus regarding caregiving arrangements within a couple, where discussions of the issue of parental care responsibilities are more likely to lead to a mutually satisfactory arrangement. This could take the form of assigning the responsibility to the wife, in accordance with traditional (if inequitable) social norms, or an assignment based on the closeness of the relationship with the needy parent. In contrast, care arrangements often exacerbate serious unresolved tensions between adult siblings, who are equally related to the disabled parent, have certain mutual expectations and appraisals, but are less able to develop fair and convenient care models. Most literature on family caregiving addresses the issue of conflicts, specifically highlighting the examples of conflicts between siblings, often caused by perceptions of insufficient mutual participation or disagreement about the quality of provided care (Strawbridge & Wallhagen, 1991; Toseland et al., 1995).
A colleague has suggested that the large negative effect of the sibling caregiver dummy variable reflects reverse causality: perhaps excessive stress produces situations in which caregivers "burn out," stop providing care, and transfer their responsibilities to their siblings. In such situations the sibling's care activities may be the result of, rather than the cause of, the respondent's elevated depression score. However, the way caregiving and depression are measured in our study makes this interpretation improbable. In our analysis, the respondents' stress outcomes are measured at the time of the interview, whereas the question on caregiving activities refers to the 12 months preceding the interview. Therefore, it is unlikely that caregiving that occurred well before the survey would produce a negative impact that manifests itself at the time of the interview. A simpler, and more likely, form of reverse causality might arise if someone experiencing poor emotional health resists taking on the caregiver role, encouraging (possibly by default) his or her siblings to do so. According to this interpretation, caregivers may not necessarily be the most emotionally disadvantaged members of the family. Finally, it must also be recalled that there is some asymmetry in the wordings of the questions on primary respondents' and their siblings' care provisions; that is, although there was a minimum of 100 hr specified for the primary respondent, no minimum was specified for the contributions of siblings.
As early as 1981, Pearlin and colleagues noted that "little is known of the manner in which the various components of stress are interconnected to form a process" (Pearlin et al., 1981, p. 337). The complexity of the stress process still remains one of the obstacles in examining the sources and pathways of caregiver's stress. Our findings demonstrate that different combinations of care provisions and parental care needs within the family setting produce statistically distinguishable levels of depression manifestations. The usual caregiver–noncaregiver distinction is too simple to capture all of these outcomes. Accounting for care needs and viewing parental care from a family perspective is important in order to understand the consequences of informal caregiving and to design the appropriate public policies.
Several limitations of this study can be identified. Given our cross-sectional design, the findings of our research are particularly vulnerable to omitted-variable threats. For instance, measures of assistance received from nonfamilial sources, including formal home- and community-based services, were not included in our analyses because of the absence of data. Furthermore, because HRS questions about care needs refer to the time of interview, whereas the questions on care activities refer to the past year, it was not possible to observe any acute, short-term changes in parental needs that might have resulted in certain care arrangements during the past year. Finally, some of those who appear to be providing care in the absence of severe care needs may have provided care earlier in the year to a parent who was then relatively more disabled than he or she was at the time of interview. In this study we have also disregarded the distinctive effects among various caregiver–care recipient gender combinations, some of which can be expected to have a particularly strong effect on stress, such as in the case of the "male caregiver–female care recipient" dyad.
As already mentioned, we do not have a separate measure of IADL limitations. Therefore, parents with IADL needs only are included in the "no severe needs" category of our analysis and are not distinguished from parents who need no assistance at all. Similarly, respondents providing assistance to a parent or parent-in-law in the performance of IADL tasks only should not have reported any hours of care, and therefore are not coded as a caregiver in our analysis. Assuming that IADL needs and IADL caregiving responses evoke stress responses that are analogous to the ADL need and care-response patterns we have found, the consequences of our lack of information on the more moderate care needs is to understate our findings. In other words, we compare indicators of stress among persons whose parents exhibit severe needs to a composite group consisting of persons whose parents have moderate care needs and those whose parents can and do perform all tasks without difficulty. The presence of those with moderate care needs in our comparison group biases toward zero our estimates of the consequences of parental needs and of caring given parental needs. Because the regression coefficient on the need variable is consistently positive, and in nearly every variant of our model statistically significant, we conclude that having a parent (or parent-in-law) with substantial personal-care needs is detrimental to one's psychological well-being.
|
Footnotes |
|---|
1 Center for Policy Research, Syracuse University, NY.
Decision Editor: Linda S. Noelker, PhD
Received for publication November 15, 2002. Accepted for publication March 6, 2003.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. W. Williams, C. S. Williams, S. Zimmerman, J. Munn, D. Dobbs, and P. D. Sloane Emotional and Physical Health of Informal Caregivers of Residents at the End of Life: The Role of Social Support J. Gerontol. B. Psychol. Sci. Soc. Sci., May 1, 2008; 63(3): S171 - S183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Schulz, R. S. Hebert, M. A. Dew, S. L. Brown, M. F. Scheier, S. R. Beach, S. J. Czaja, L. M. Martire, D. Coon, K. M. Langa, et al. Patient Suffering and Caregiver Compassion: New Opportunities for Research, Practice, and Policy Gerontologist, February 1, 2007; 47(1): 4 - 13. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. Amirkhanyan and D. A. Wolf Parent care and the stress process: findings from panel data. J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2006; 61(5): S248 - S255. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. K. Pavalko and K. A. Henderson Combining Care Work and Paid Work: Do Workplace Policies Make a Difference? Research on Aging, May 1, 2006; 28(3): 359 - 374. [Abstract] [PDF]
|
|
|
|
|
|
M. Pinquart and S. Sorensen Gender Differences in Caregiver Stressors, Social Resources, and Health: An Updated Meta-Analysis J. Gerontol. B. Psychol. Sci. Soc. Sci., January 1, 2006; 61(1): P33 - P45. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B. Adams, K. A. Smyth, and M. J. McClendon Psychosocial Resources as Moderators of the Impact of Spousal Dementia Caregiving on Depression Journal of Applied Gerontology, November 1, 2005; 24(5): 475 - 489. [Abstract] [PDF]
|
|
|
|
 |
|
 Y. Yueh-Feng Lu and M. G. Austrom Distress Responses and Self-Care Behaviors in Dementia Family Caregivers With High and Low Depressed Mood Journal of the American Psychiatric Nurses Association, August 1, 2005; 11(4): 231 - 240. [Abstract] [PDF]
|
|
|
|
|
|
L. W. Li From Caregiving to Bereavement: Trajectories of Depressive Symptoms Among Wife and Daughter Caregivers J. Gerontol. B. Psychol. Sci. Soc. Sci., July 1, 2005; 60(4): P190 - P198. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
