| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
a Department of Sociology and Centre on Aging, University of Victoria, British Columbia, Canada
Correspondence: Margaret J. Penning, PhD, Centre on Aging, P.O. Box 1700, University of Victoria, Victoria, British Columbia, Canada V8W 2Y2. E-mail: mpenning{at}uvic.ca.
Decision Editor: Laurence G. Branch, PhD
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Substitution • Self-care • Informal care • Home care
Twenty years ago, Shanas 1979
referred to Hydra, the multiheaded monster of Greek mythology, as a metaphor to characterize what she saw as the then already persistent and widely held but empirically false belief that in modern industrialized societies, older adults were alienated from their families and, consequently, found themselves abandoned, isolated, and unable to count on their families for help or support. This led, in her view, to the hypothesis that introducing formal services into the mix somehow discouraged or undermined these efforts, thereby substituting formal bureaucratic care for informal care.
Two decades later, despite the continued accumulation of disconfirming evidence and albeit somewhat transformed, Hydra appears largely intact. Although evidence indicating that families are primary providers of care to older relatives now appears widely accepted, concern that enhancing public services will lead to a reduction in such care has remained. As noted by Hanley, Wiener, and Harris 1991(p. 507), for example, "One of the main barriers to the expansion of paid home care for the chronically disabled is the fear that policymakers have that it will cause friends or relatives to curtail their informal caregiving efforts" (also see Binney, Estes, and Humphers 1993; England, Keigher, Miller, and Linsk 1991; Penrod, Harris, and Kane 1994, pp. 21–22). According to Binney and colleagues 1993, not only is there a reticence on the part of policymakers to introducing or increasing formal services, but increasingly one also sees long-term care policies and practices being introduced that see informal care being drawn on as a mechanism to reduce the publicly financed costs of care. Thus, existing long-term care policy is not only "built on the premise that family is and should be the primary party responsible for impaired older persons" (Montgomery 1999, p. 383) but that state intervention should be avoided so as to encourage such efforts. As a result, "Care that was once provided by paid professional providers has shifted to unpaid lay providers, typically women" (Binney et al. 1993, p. 193).
Although the primary focus of literature in this area has been on the relationship between formal and informal care, newer versions of this orthodoxy also appear to include the companion belief that formal care will not only discourage informal community-based care but will simultaneously undermine self-reliance and personal responsibilities for care. As Hooyman and Gonyea 1995 as well as others (e.g., Clark 1993) have pointed out, an assumption of private responsibility underlies the emphasis on family responsibilities for long-term care. This assumption is consistent with an ideology of individualism that values independence and self-reliance. A major component of this ideology is the obligation that individuals have to avoid being a burden to others. People are expected to look after themselves and depend only on those who voluntarily assume an obligation to support them—that is, family and friends (Winston and Bare 1993, p. 393). Therefore, as noted by Montgomery 1999(p. 394), "Together the ideologies of ‘individualism’ and ‘familism’ legitimate the belief that for most purposes, claims on the family should have priority, with claims on the state only coming into play when family resources are exhausted."
According to Penrod and associates 1994(pp. 21–22), given conceptual and methodological limitations, a review of empirical research on the relationship between formal and informal home care does little to minimize policymakers' concerns that increasing access to publicly funded home care services will lead to a withdrawal of informal support. Weissert, Cready, and Pawelak 1988 reviewed articles published from 1960 through the mid-1980s detailing the results of studies designed to assess the impact of formal community-based home care programs and demonstration projects on informal care and other services. They concluded that informal support did in fact tend to decline with increased use of home- and community-based care, based on findings indicating that of the 53 specific empirical relationships examined within the eight studies they reviewed, the majority of relationships (n = 30, 56.6%) were negative in direction. However, reassessment of their findings indicates that when only statistically significant relationships are included, their evidence seems to suggest the opposite conclusion—that is, that community-based home care programs have little impact on informally provided care. Only 11 (20.7%) of the 53 relationships were statistically significant; 5 revealed a negative impact of formal on informal care, and 6 revealed a positive impact (see Penning and Keating 2000).
Table 1 outlines the major studies published on this issue since 1985. They include cross-sectional as well as longitudinal studies, as both tend to be widely used and cited as addressing such issues. Once again, the findings appear mixed. Of the 10 studies included in this review, 6 (including 2 drawing from the same general data set; see Christianson 1988; Kemper 1992) reported at least some evidence of substitution, and 4 did not. However, of the 48-plus specific relationships tested in these studies, the majority (30, or 62.5%) once again suggested no significant impact of formal on informal care. Moreover, where significant findings were evident, they tended to be relatively small in magnitude and limited to the short term. For example, Christianson 1988 reported evidence of a reduction in the receipt of informal care following the introduction of publicly funded home care services. However, the reduction was noted to be modest in magnitude, evident only in terms of the care received from friends and neighbors (but not family members), and only over the short term (i.e., at 6 and 12 months but not 18 months following the introduction of formal services). Similarly, Edelman and Hughes 1990 reported finding reductions in informal care during the first 9 months but increases in informal care over the course of 48 months following the introduction of formal services, and Tennstedt, Crawford, and McKinlay 1993; Tennstedt, Harrow, and Crawford 1996 reported that substitution, although evident, tends to be temporary and to occur in such situations as the loss of a primary caregiver—that is, when substitution is in fact intended.
|
). Yet, self-care is rarely considered a component of the overall health care system (Kart and Engler 1994). Although some studies have focused on the relationship between self-care and both informal and formal (medical) care in response to symptoms (particularly acute symptoms) of illness (see, e.g., Chappell and Penning 1990; Fleming, Giachello, Andersen, and Andrade 1984; Kart and Engler 1994), fewer studies have addressed the relationship between formal care and self-care in relation to the long-term management of the functional disabilities associated with chronic illness (Norburn et al. 1995). Yet, as recent literature has acknowledged, self-care not only includes actions taken to detect, diagnose, prevent, or treat illness, but also a broad range of behaviors undertaken with the intent of maintaining and promoting health and functional independence. According to Ory, DeFriese, and Duncker 1998, for example, self-care encompasses three general categories of behavior: (a) steps taken by lay persons to compensate for functional limitations affecting everyday activities of daily living (e.g., steps taken to modify the environment, change behavior, and/or use special equipment and devices to reduce the gap between personal capabilities and environmental demand; see also Norburn et al. 1995); (b) actions taken to prevent disease or promote health (e.g., lifestyle modifications); and (c) medical self-care aimed at the diagnosis and treatment of symptoms of illness or the self-management of chronic conditions (e.g., self-medication).
Furthermore, to the extent that relationships between self-care and either or both informal and formal in-home care have been addressed, research findings remain equivocal. Findings reported by Struyk and Katsura 1987, indicating that older adults with greater access to support and assistance from others were less likely to report making structural changes to their homes to accommodate the needs of disabled family members, suggest that informal and formal care may in fact limit and thereby substitute for self-care. In contrast, findings reported by Manton, Corder, and Stallard 1993 indicate that although most older adults with severe impairments receive assistance from others, this tends to be supplemented by the use of special equipment. Similarly, findings reported by Norburn and associates 1995 concerning relationships between self-care and assistance from others (formal or informal) in coping with functional limitations reveal that the likelihood of engaging in self-care increases as the severity of disability increases (except among the most severely disabled persons). Those receiving assistance from others (especially with independent activities of daily living tasks) were more likely than others to engage in self-care activities. Thus, the receipt of assistance from others was found to supplement and/or complement rather than substitute for self-care coping strategies. Similar findings are reported in a more recent study conducted by Agree and Freedman 2000 to explore the potential for self-care through the use of assistive technology to substitute for or supplement personal care among disabled older Americans. They reported evidence of greater use of assistive devices among those receiving informal and especially formal care when compared with those not receiving such care, regardless of disability level.
In general, based on the findings available to date, the evidence for substitution (i.e., of formal for informal or self-care) would seem limited. Few studies have shown that informal care declines significantly (in scope or intensity) and over the long term with the introduction of formal home care services. Nevertheless, the need for further research evidence remains. For example, aside from addressing the already well-documented limitations of existing research, studies have yet to examine relationships between formal in-home care and both informal and self-care among those receiving publicly funded versus privately paid care. Previous studies have either focused only on publicly funded services or have not differentiated one type of formal care from the other. Yet, according to Penrod and colleagues 1994(p. 29), "There are two main groups of formal home care users: those who finance home care privately from family resources and those who are beneficiaries of public programs. Decisions about formal and informal care use arise through very different decisionmaking processes for each of these two groups." We would add a third: those who combine privately paid services with those that are publicly subsidized. The concern among policymakers would seem to lie with publicly provided services. Yet, the possibility of substitution is not limited to such services. Also, increasingly, those receiving publicly funded services are having to share in the costs of care as well as to supplement the care received with care from private sources. The paucity of evidence available with respect to self-care also suggests a need for further investigation in this area.
In response to these needs, the focus of the current study is on the implications of the receipt of formal in-home care for self- and informal care among older adults coping with long-term illnesses and disabilities in the community setting. In particular, it examines whether increases in the extent of formal in-home care received are associated with decreases in the extent of self- and informal care received. The implications of both publicly subsidized and privately paid services are addressed.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
In 1997, all those who took part in the initial interviews were recontacted and asked to participate in a follow-up interview. The refusal rate for this set of interviews was just under 10%. Others were lost to follow-up because they were deceased, too ill to be interviewed, could not be located, or had been institutionalized (28%). Comparison of the characteristics (baseline) of those who did and did not participate in the follow-up interview revealed that those lost to follow-up were more likely to be male and to have lower incomes, poorer perceived health, and poorer cognitive functioning and to be receiving publicly subsidized home care services. Significant differences were not evident with regard to age, education, marital status, living arrangements, number of chronic conditions, or depression.
Among those reinterviewed in 1997, 301 (45%) received publicly subsidized home support services and 360 (54%) did not. Of those who received publicly subsidized services, most (n = 222; 73%) received both publicly funded and privately paid care; 22% (n = 79) received only publicly subsidized care. Of those who did not receive publicly funded services, 78% (n = 281) received privately paid care and 22% (n = 79) did not receive any formally provided care.
Measurement
Means, standard deviations, and the coding schemes used for all variables included in the analyses are outlined in Table 2 .
|
To measure the extent of formal care received, two indicators were used. The first assesses the scope of assistance received and is measured using the total number of basic, instrumental, and other tasks (ADL, IADL, and AADL) for which respondents reported receiving formal assistance. The second measure assesses the intensity of care provided and is operationalized using the total number of minutes per month of assistance provided across the 15 task areas by formal sources of care. Both measures include publicly subsidized and privately funded care. The overall number of task areas for which respondents reported receiving formal care ranged from 0 to 12 (M = 2.12, Mdn = 2.00, SD = 1.88). The number of minutes of formal service received ranged from 0 to 26,810 min per month (446.83 hr, almost 15 hours per day), with the average being 1041.35 min per month (17.35 hr, just over one half hour per day; Mdn = 420.00, SD = 2,069.86 min). To correct for their skewed distributions, we used logarithmic transformations of these variables in the analyses (see Table 2 ).
We used similar measures to assess informal care. The scope of assistance received was assessed using the total number of ADL, IADL, and AADL tasks for which informal care was received, and the intensity of informal care received was assessed using total minutes per month of assistance provided across the 15 task areas by all unpaid sources of care (i.e., family members, friends, etc.). Overall, 51% of those in the combined sample reported that they received informal care. The actual number of task areas for which respondents reported receiving such care ranged from 0 to 11 (M = 1.26, Mdn = 1.00, SD = 1.88), with the extent of care received during the past month ranging from 0 to 21,780 min (i.e., the equivalent of 363 hr or approximately 12 hr per day; M = 884.35 min per month, or approximately 14.7 hr per month or 0.5 hr per day; Mdn = 25.00, SD = 2,206.61 min). Once again, we used logarithmic transformations.
To assess self-care within the same domains, we drew on respondents' answers to the question "Are there things that you currently do or have done for yourself in order to help with these activities" of daily living. Responses to this open-ended question varied depending on the particular task involved and included the rearrangement of furniture and the installation and use of assistive aids and devices (for personal care and mobility-related tasks) as well as modifying one's clothing or eating habits (for problems dressing or with meal preparation tasks) and taking frequent rests or cutting back on the frequency of activities (for difficulties with shopping, housework, etc.). Although some researchers (see, e.g., Dean 1989; Dill, Brown, Ciambrone, and Rakowski 1995; Hickey, Akiyama, and Rakowski 1991) have also conceptualized such things as doing nothing or securing assistance from others (either informal or formal) as components of self-care, they are not included here as we were unable to determine whether doing nothing represents a consciously chosen course of action in response to the difficulties imposed by chronic illness or indicates a lack of awareness of or attention to the difficulties involved. Also, including assistance from others as a component of self-care would confound our measure of self-care with our measures of informal and formal care.
In this study, the scope of self-care was measured using the total number of activities of daily living (ADLs, IADLs, and AADLs) for which one or more self-care activity was reported. Possible responses once again ranged from 0 to 15, and actual responses ranged from 0 to 12 (M = 3.14, Mdn = 3.00, SD = 2.40). The intensity of self-care was not assessed because it was not possible to develop a measure comparable to the actual time spent by others (i.e., informal or formal) in the provision of care; self-care included home modifications and other activities for which time estimates are not applicable.
Controls.
Other variables included in the study were selected in accordance with the behavioral model of service use (Andersen and Newman 1973), the most widely used conceptual framework for addressing the determinants of health service use, as well as on the basis of previous research findings indicating their relevance to the receipt of care (self-, informal, and formal). They include the respondents' age, gender, education, income, marital status, living arrangements, informal network size, and health-related needs.
Age and education are continuous measures, with age ranging from 66 to 93 or older (M = 82.38, SD = 6.88) and education ranging from 0 to 23 years (M = 11.53, SD = 3.41). Income is ordinal and measured on the basis of the respondents' reports concerning the total average monthly incomes for their households. Responses were coded into 24 categories, ranging from 1 (no income) to 24 ($5,500 or more), with intermediate intervals of $250. Those lacking income data (n = 163; 25%), primarily because they did not know the incomes for their households, were assigned the mean (8.56; SD = 3.93). We used logarithmic transformation to correct for skewness in the income distribution.
Marital status and living arrangements were found to be highly correlated with one another (r = .74, p < .001). Consequently, the two variables were combined and two dummy variables created. The first contrasted those who were not married and living alone (coded as 1) with others (coded as 0), and the second contrasted those who were married and living with a spouse (coded as 1) with others (coded as 0). In total, 65% of those in the study sample were not married and living alone, and 21% were married and living with a spouse. Network size was measured by using the total number of parents, siblings, and children available; the number of other relatives that they had weekly contact with; and the number of neighbors they reported knowing. Responses ranged from 0 to 101 (M = 13.91, SD = 11.20). Once again, logged responses were used to correct for a skewed distribution.
To assess health-related need for care, we used several measures: number of chronic conditions, level of functional impairment, level of cognitive impairment, and perceived health status. The total number of chronic conditions was computed on the basis of the respondents' reports concerning whether they suffered from each of 33 different conditions, including high blood pressure, heart disease, stroke, arthritis, Parkinson's disease, and so on. Functional impairment was assessed on the basis of the respondents' reports concerning the level of difficulty they experienced with various ADLs. Following Wolinsky, Callahan, Fitzgerald, and Johnson 1993, we subsequently classified it into three domains: impairment with regard to basic ADLs (including dressing, using the toilet, getting about the house, getting in and out of bed, and bathing/showering), IADLs (including using the telephone, shopping for clothes or groceries, preparing meals, light housework, heavy housework/laundry), and, finally, AADLs (including eating, yardwork, taking medications, taking care of money, and handling long-term finances). For each activity, responses were coded on a 4-point scale ranging from 0 (no difficulty) to 3 (completely unable to perform the task). Within each domain, component item scores were then summed to obtain the overall summary measure. Because each included five items, possible scores initially ranged from 0 to 15 for each of the three measures, with higher scores indicating a higher level of impairment. Interitem reliability coefficients were .75 for ADL impairment, .85 for IADL impairment, and .66 for AADL impairment. Given somewhat skewed distributions, scores for ADL and AADL impairment were subsequently truncated (at 8 and 9, respectively) before the analyses.
The level of cognitive impairment was assessed with a modified version of the Mini-Mental State Examination (i.e., the ALFI–MMSE; see Roccaforte, Burke, Bayer, and Wengel 1992), a 14-item measure designed to assess several different aspects of cognitive functioning. Respondents were asked a series of questions (e.g., "What year is this?" "What is today's date?" "What country are we in?") and given a number of tasks (e.g., word recall, number recall), with points given for each correct response. Possible scores therefore ranged from 0 to 22, with higher scores indicative of better cognitive functioning. Because the majority of respondents were not cognitively impaired and therefore obtained higher rather than lower scores on this measure, it was necessary to square the scores for purposes of analysis. Finally, perceived health status was operationalized on the basis of the respondents' assessments of their health relative to others of the same age. Responses ranged from 1 (excellent) to 4 (poor/very poor).
Very few (4%) of those in the study sample reported 0–2 chronic conditions, and the average number of conditions reported was 9.32 (SD = 4.09). Similarly, more than two thirds (68%) of those in our sample reported experiencing at least some difficulty with ADLs and AADLs (70%), and almost everyone (95%) experienced difficulty with one or more IADLs. In terms of cognitive impairment, scores on the ALFI–MMSE ranged from 3.0 to 22.0, with 16% receiving scores of 16 or below, a cut-off frequently used to indicate cognitive impairment. In terms of perceived health, almost two thirds (65%) assessed their health as excellent (17%) or very good (48%) compared to others of their same age. The remainder assessed their health as either good (26%) or poor/very poor (9%).
Methods of Analysis
In ordinary least squares (OLS) regression, it is assumed that there is no reciprocal causation between any of the variables included in the model; thus, it ignores the feedback component of such relationships. Consequently, if such relationships do exist, the result will be biased estimates of the model's parameters due to the correlation of the error term(s) and the predictor variable(s) (Norusis 1993). In this study, the possibility of reciprocal relationships between self-, informal, and formal care is acknowledged, and consequently the relationships were examined using two-stage least squares (2SLS) regression analyses. Self-, informal, and formal care were treated as endogenous variables. Exogenous variables included age, gender, education, income, living arrangements, informal network size, and health status, factors that existing theory and research have identified as important determinants of the receipt of care (self, informal, and formal). In analyses addressing the substitution of formal for informal care, two of these variables (age and informal network size) were included in the equation predicting formal care but not in the equation predicting informal care, based on the findings obtained from initial bivariate and multivariate analyses revealing that both are significantly correlated only with formal care. Similarly, age and education were included in the equation predicting formal care but not in the equation for self-care in the analyses assessing the substitution of formal for self-care. The exclusion of these variables as predictors in these equations served to identify the regression models (Kline 1998).
In 2SLS regression, two equations are simultaneously determined—one predicting each of the two endogenous variables (i.e., formal care and either self- or informal care as appropriate). The procedure is comparable analytically to an OLS regression analysis conducted in two stages: (a) the regression of formal care on instrumental (exogenous) variables presumed to be related to levels of formal care, followed by (b) the regression of self (or informal) care on explanatory variables that include the predicted values of formal care obtained in the previous equation as well as on instrumental factors presumed/known to be related to each of the two forms of care (i.e., self or informal).
Here, we conducted these analyses first using the sample as a whole (i.e., both users and nonusers of formal in-home care services). They were then repeated among four subgroups: (a) all users of formal in-home care services (i.e., users of either or both publicly subsidized and privately paid care), (b) users of publicly subsidized formal in-home care services only, (c) users of privately paid formal in-home care services only, and finally (d) users of both publicly subsidized and privately paid in-home care services.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
We obtained similar findings when separate analyses were conducted among all those receiving formal services (public and/or private) as well as among subgroups including those receiving publicly subsidized services only, privately paid services only, or a combination of both publicly and privately paid services. For none of the subsamples did the intensity of formal care emerge as a significant predictor of the intensity of informal care.
Table 6 reports the results of similar analyses but with the scope of self-care included as the dependent variable and predisposing, enabling, and need factors included as predictors along with the scope of formal care. For the sample as a whole as well as for most of the four subgroups, it was those older adults reporting more chronic conditions and greater ADL impairment who reported more extensive self-care. Other factors that emerged as significantly related for the sample as a whole were the level of AADL impairment (with those reporting greater impairment also reporting more extensive use of self-care) and cognitive impairment (with those reporting better functioning also reporting greater self-care). Once again, however, with these factors and the reciprocal influence of self-care on formal care taken into account, the extent of formal care received did not, in general, emerge as having a significant impact on self-care. The sole exception to this trend was evident within the subgroup receiving publicly subsidized formal services only. However, within this relatively small subgroup, the impact of formal care was positive in direction, indicating that those who receive greater formal care also practice more extensive self-care.
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
With regard to the relationship between formal and informal care, the findings revealed no evidence to indicate that greater use of formal services is associated with less extensive informal care. They therefore lend support to the conclusions arrived at by several prior investigations into this area; that is, that formal in-home care services do not appear to substitute for informal care provided by family members, friends, and neighbors. This is the case for the sample as a whole as well as for users and nonusers of publicly subsidized services and those receiving either or both publicly and privately paid care. Nor do the results appear to support the substitution hypothesis when applied to self-care either for the sample as a whole or for specific subgroups defined in terms of the use of publicly subsidized and/or privately paid services. Given similar levels of health-related need and so forth, there is no evidence to indicate that those who receive more extensive formal in-home care services practice less extensive self-care.
Instead, our findings suggest that for older individuals with relatively high levels of illness and disability and living in the community setting, the extent of self-care practiced and informal care received are most strongly related to the overall need for care as indexed by the number of chronic conditions and levels of functional impairment. They are also related to levels of cognitive impairment; those with higher levels of impairment (or need) receive more extensive informal care but also practice somewhat less extensive self-care. Thus, in general, our findings concur with previous reports (e.g., Norburn et al. 1995) suggesting that self- and informal care, like formal care, are directly and positively related to health and therefore increase as health-related needs for care increase. In the case of informal care, they are also responsive to the availability of informal resources for providing such care (e.g., as indexed by living arrangements).
Taken together, findings indicating that self- and informal care are influenced by needs for care and by the availability of informal resources for providing such care but not by the receipt of formal care have important implications for theory, research, and policy. At the theoretical level, they have implications for ongoing debate concerning relationships between various sources of care within the overall helping network. They also suggest that conceptualizing long-term care as a unidimensional process (or continuum of care) is problematic. Specifically, the findings appear inconsistent with the notion that as 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. Thus, rather than equating self-care with independence and seeing the introduction of informal and then formal sources of care as reflecting increases in dependency, these findings suggest it may be more appropriate to conceptualize self-, informal, and formal care as resources used by individuals to cope with their conditions. As the need to cope with the difficulties imposed by chronic illness and disability increases, so too do all of the resources that are accessed in order to meet this need.
Second, the findings also point to the need to acknowledge self-care as an important component of the overall system of care and the need for additional research to be directed toward examining the relationships among self-, informal, and formal care as well as how these relationships change in conjunction with changes in chronic illness and disability. Finally, with regard to health policy and service delivery, our findings suggest that there is little need for concern that increases in the availability of formal in-home care services to older chronically ill and disabled adults will be associated with decreases in levels of self- or informal care. Consequently, policies that restrict formal services in order to ensure that they do not undermine self- or informal care or as a means to enhance levels of self- or informal care appear unnecessary and are likely to be ineffective. Indeed, our findings suggest something of the opposite may be true, indicating that for individuals receiving publicly funded home care services only, more extensive formal care is associated with more extensive self-care. Similar findings are reported by others (e.g., Agree and Freedman 2000).
The major strengths of this study were its ability to address relationships involving self-care as well as formal and informal care, its inclusion of all informal care providers (and not just primary or coresident caregivers) as well as both publicly subsidized and privately paid formal services, and its use of 2SLS regression in order to take into account reciprocal relationships between the various forms of care. However, several limitations should also be noted. First, although the question of substitution is, by definition, a temporal one, the data used here were cross-sectional in nature. As a result, any conclusions regarding causality are necessarily tentative and should be treated with caution.
Also, the study sample consisted of older elderly adults with relatively high levels of chronic illness and disability, including some degree of cognitive impairment. It could be argued that substitution is likely to be less of an issue when needs for care are relatively high. Yet, those individuals who were interviewed represented survivors from a sample originally contacted 1–1.5 years previously. Thus, their health was significantly better than that of those who died, were institutionalized, or were too sick to participate in the second interview, and they were less likely to be receiving formal care. Consequently, one could also argue that substitution is primarily an issue among those with the poorest health and that the lack of support found for the substitution hypothesis in this study reflected the fact that survivors were better able to look after their own needs or had better access to supportive social ties than were those lost to follow-up. The potential significance of selection effects at both ends of the continuum in producing the findings observed suggests a need to address the issue of substitution among those with varying levels of illness and disability.
The influence of cognitive impairment may also warrant further investigation. A significant proportion of those included in our study was assessed as having some level of cognitive impairment. They were retained in the sample on the basis of the fact that they maintained residence in the community setting, with or without informal and formal supports. Not only are cognitively impaired individuals increasingly likely to maintain residence in the community setting, but they also represent an increasing proportion of those receiving home care services. Yet, their inclusion in the sample also raises concerns regarding the validity of the findings. To address the impact that including these individuals had on our findings, we compared findings obtained for the sample as a whole to those obtained for a subsample of those without significant cognitive impairment (i.e., scores of 17 or greater on the ALFI–MMSE). The findings (not reported here) revealed only minor differences; none indicated a significant impact of formal care on self- or informal care. Nevertheless, given the increasing importance of addressing the needs of cognitively impaired people within the community setting, greater attention to this issue and its research implications appears warranted.
Also, this sample was obtained in one region of one province at one point in time. Consequently, the generalizability of the findings to other locales (geographic or policy relevant) or other time periods is limited. In Canada, home care and other long-term care services are not part of the national publicly funded health care system and therefore are not universally accessible but, rather, vary considerably (in terms of accessibility, type of services, cost to the user, etc.) from one province and region of the country to the next.
Finally, limitations are also evident with regard to the measurement of self-, informal, and formal care. For example, most studies of self-care tend to focus on specific health practices (e.g., self-medication, use of devices; see Dean 1992) and rely on categorical measures to assess use of various types of self-care (e.g., Manton et al. 1993). Rarely are attempts made to focus on broader levels of self-care, particularly in relation to chronic illness and disability. Yet, given our interest in assessing the ability of formal care to substitute for self-care more generally, we developed a measure to assess the scope of self-care across the same domains as used for informal and formal care. Thus, we used the total number of ADL, IADL, and AADL tasks for which respondents reported having done something to help themselves with the particular activity. However, the questions on which this measure was based asked respondents to list things "that you currently do" or "have done" in order to help with these activities. Ideally, our measure would assess the scope of the self-care strategies that are in fact currently being used by the respondents, regardless of when the strategies, devices, or technologies were first installed or used. Although formal care that is currently being received cannot substitute for past self-care, it can displace current use of these self-care strategies (e.g., a handrail that is no longer used once care providers assist). The fact that these questions followed others that asked respondents about current activity limitations and use of informal and formal services suggests that the focus was likely on current strategies for self-care. Nevertheless, the possibility exists that some of the activities reported by the respondents took place well before the interview (e.g., grab bars that had been installed 5 years previously).
An additional limitation stems from the fact that respondents were not asked whether the self-care strategies they used were recommended or financed (either in part or in whole) by other sources of care (formal or informal), raising the possibility of some confounding of self-care with either or both informal and formal care. This issue is not, of course, unique to self-care. The initiative behind using formal services or seeking informal assistance (as well as payment for these services) may also lie with the individual care recipient, and consequently care conceptualized as exclusively informal or formal may contain an element of self-care as well. A related issue concerns the difficulty of comparing various types of self-care strategies. For example, it is possible that specific self-care strategies (e.g., rearrangement of furniture) can be used to make a difference to several tasks simultaneously, whereas others (e.g., installation of an elevator) may make a difference to one task only (e.g., going up and down stairs). Yet, the effort involved on the part of the individual may be equal or even greater in situations where only one task is involved. Along similar lines, it is unclear whether the rearrangement of furniture or installation of grab bars in different rooms should be considered separate self-care strategies or just one. These and other difficulties in the measurement of self-care attest to the need for greater attention to be directed to the development of operational definitions, particularly the level or extent of self-care used in relation to chronic illness and disability where informal and formal care are also likely to be involved.
Other limitations of measurement should also be noted. For example, to measure the intensity of the informal and formal care received, respondents were asked to estimate the average length of time spent on each ADL, IADL, or AADL task during each visit. Although the total amount of time spent across tasks appears fairly consistent with official estimates, the accuracy of such estimates is not clear. Not all tasks are performed during each visit, and the amount of time spent on each task may be difficult for respondents to assess as distinct tasks (e.g., housekeeping, meal preparation) may be performed in conjunction with one another rather than in discrete units, respondents may not be in a position to observe the amount of time spent on all tasks, and so forth. We were also unable to assess the intensity of self-care.
To some degree, these and other limitations of this study attest to the need for further research to be conducted concerning the relationships among self-care and informal and formal community-based care. In particular, they suggest a need for research that carefully examines relationships between self-, informal and formal sources of care and how these relationships change over time with changes in health-related needs. The need to consider self-care in conjunction with informal and formal care appears particularly acute. There is also a need for more comparative studies to be conducted to assess how these relationships vary within different policy and service-delivery environments. Yet, despite these limitations, our findings also attest to the need for research to go beyond repeated attempts to assess and affirm the involvement of private resources (i.e., older individuals, their families and friends) in the provision of care and assistance to those with chronic illness and disability. Thus, as suggested by Shanas 1979 so many years ago, it is time to abandon the myth that old people are alienated from their families (including its various corollaries) and to direct our research and policy efforts elsewhere.
Practice Concepts
The Forum
Book Reviews
|
Acknowledgments |
|---|
Received for publication March 1, 2001. Accepted for publication October 9, 2001.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
