INTRODUCTION

Family support involves complex relations of giving and receiving material assistance, instrumental support aimed at behavior or the context of care and emotional support expressed through presence, companionship, affection, empathy, listening, and confirmation¹. Several questions permeate the theoretical study of such assistance and the intervention itself, aimed at understanding and managing support and relationships with elderly persons within the family. What is more important for the elderly person: providing or receiving support? There is no single answer to this question, which is related to age, gender, the size of the closest social network of the individual and the motives of those involved^1-3. Who helps more in family relationships, elderly women or men? In general, women offer more instrumental and affective support to their peers and descendants than men, who in turn offer more material support to their children². By behaving in this way, both respond to social norms associated with gender and specific conditions of the possession of material goods^4,5. What offers the greater assurance of help to the elderly, home-based intergenerational arrangements or marriage? The answer is that it depends. Where there is marked physical and cognitive dependence of a spouse, when their partner is no longer able to offer the necessary instrumental assistance, co-dependence with descendants may favor protection^3,4. Often, co-residence functions as an arrangement of survival or convenience for members of two or more impoverished generations, with care provided to the elderly within this context of needs⁴. Intergenerationality is not necessarily a guarantee of the adequate supply of care^2,4,5.

The majority of elderly persons value the support they receive as a reinforcement of the affective ties constructed within the family. Others perceive the risk of appearing dependent and incompetent in the support received^1,6. Providing support in a positive emotional setting, combined with personal characteristics among the recipients of care that predispose them to accepting help increases the chance of the subjective well-being of such individuals⁶. However, support that is considered exaggerated or disruptive and a lack of contact with partners of their choosing can generate more stress and harm to the well-being of the elderly than a lack of support⁷.

Smilkstein^8,9 describes family functionality in terms of five resources: adaptability, partnership, growth, affectivity, and resolutive capacity. In the daily routine of caring for the elderly, the functionality of family relations is constantly subjected to cognitive evaluation, a process that consists of comparing what is observed with individual and group norms, values and expectations. Satisfaction with the dynamics of family relations is a strong determinant of subjective well-being which, in turn, is related to other positive outcomes in the physical and mental health of caregivers^8,10.

Knowing how elderly caregivers of other elderly persons perceives the dynamics of family functionality and the exchanges of support in the context of care is fundamental to understanding how families organize themselves to meet the demands of everyday life and provide the necessary resources for the well-being of family members.

The present study aimed to investigate associations between the satisfaction of family caregivers of the elderly with family relations; sociodemographic variables; the type, direction, adequacy and burden of family support; and the number of social partners involved.

METHOD

Adopting a descriptive and analytical cross-sectional perspective, the present study used the database of the "Psychological Well-Being of Elderly Persons who Care for Other Elderly Persons in a Family Context" study, from which the variables of interest were extracted. The sample size was calculated as 148 individuals, based on the correlations between quality of life measures^11,12 and perceived burden^13,14, which were part of the main survey protocol (significance level 1%, test power 90% and minimum correlations of 0.40).

The eligibility criteria for the sample were: aged 60 or older, have cared for a sick elderly relative with some degree of dependence for six months or more, and score above the cut-off point in the CASI-S (Cognitive Abilities Screening Instrument - mini-test)^15,16 at an initial interview. Based on these criteria, 148 participants were selected, 48.0% of whom were indicated by public services; 8.8% by private home care services; 39.9% by geriatricians or related specialties where the care recipients were elderly; and professionals of the Family Health Program (3.4%) in the cities of Jundiaí (38.5%), Indaiatuba (29 , 1%), Campinas (18.2%) and Vinhedo (14.2%) in the state of São Paulo.

In order to evaluate satisfaction with family relations, the family APGAR measure was adopted^8,9, an acronym that corresponds to adaptability, partnership, growth, affectivity, and resolutive capacity. These resources of family functionality were evaluated by caregivers through five three-point scales (0=never, 1=sometimes, or 2=always). Total scores from 0 to 4 indicated low levels of satisfaction; 5 and 6 intermediate satisfaction, and 7 to 10 high levels of satisfaction.

The sociodemographic variables considered were gender, based on a yes or no answer to the alternatives male and female; age, which asked for the number of years lived since the date of birth; living arrangements, which included the questions “do you live alone?” (yes x no) and "who do you live with?" for those who answered no, with the alternatives being husband, wife or partner, father or mother, father-in-law or mother-in-law, daughter or son, husband or wife of daughter or son, grandchildren, great-grandchildren, other relatives and non-family members, all with yes or no answers. Cohabiting was evaluated by the question "do you live with the person you care for?" with a yes or no answer.

The variables dynamic of material, instrumental and emotional support in the family and nature of the link with the social partners involved were evaluated through two questions. The first focused on whether the caregiver received and provided material, emotional and instrumental support for Instrumental Activities of Daily Living (IADL) and Basic Activities of Daily Living (BADL) in the context of the family. The second identified the person with whom the individual exchanged support. The options were: spouse or partner; parents and in-laws; children, son-in-law/daughter-in-law, grandchild and great-grandchild; other relatives; friends or neighbors; volunteers; domestic employees; health professionals and the person cared for. The answers were recorded in a matrix derived from the study of Allen and Wiles¹, together with the results of the application of the assessment items described below.

The evaluation of the adequacy of the support received and the burden caused by the provision of support was made shortly after the introduction of the items on the support received and provided. Those who received support were asked if it met their needs or expectations. In order to identify whether each type of support offered was a generator of burden for the caregiver, direct questions were asked, with yes or no alternatives.

Seven trained interviewers carried out interviews in homes (61.5%), private doctor's offices (25.0%) and the geriatric outpatient clinics of a university hospital (13.5%), duly authorized by those responsible for the services and according to the availability of the caregivers. The mean duration of the single session in which each caregiver participated was 56.0 (+12.2) minutes, which included the items of interest of the present research and the other items of the study on the psychological well-being of caregivers, of which it is part. The project was approved by the Research Ethics Committee of the Universidade Estadual de Campinas (CAAE N^O 35868514.8.0000.5404) which also approved the contents of the Free and Informed Consent Form, which was read, discussed and signed by all caregivers.

To analyze the data, two sets of points in the family APGAR were created: 7 to 10 - high levels of satisfaction, and 0 to 6 - low and intermediate levels of satisfaction. The variable direction of support was derived from the answers yes or no to the alternatives I receive and I provide, with the variations: unidirectional (only receive or only provide), reciprocal (receive and provide) and absence (neither receive nor provide). For each type of support the number of social partners which only provided, only received, or which exchanged reciprocal support was counted. The nature of the bond was not considered. The alternatives for housing arrangements were reduced to five: spouse; spouse and ascendants; spouse, ascendants and descendants; ascendants and descendants, and others (other relatives, friends, people from outside the family, neighbors).

The Shapiro-Wilk test was used to analyze the suitability of the distributions for the use of parametric tests. Based on their non-parametric nature the chi-squared and Fisher exact tests were used to compare the frequencies of the scores of the participants for the two levels of satisfaction (0 to 6 and 7 or more). The Mann-Whitney test was used to compare the distributions of the ordinal variables, according to the two levels of satisfaction.

To analyze the relationship between high levels of satisfaction with family relations and the other variables, univariate logistic regression analyzes were performed, based on which the variables with an association with the dependent variable were selected, with statistical significance indicated by p <0.30. These variables were organized in a hierarchical multivariate regression model, with three blocks of variables, which were introduced consecutively. Block 1 included the types and directions of supports variables; Block 2 the number of social partners available for the exchange of support, and Block 3 the adequacy of the support received and the sense of burden of the support provided. The multivariate analysis data were adjusted for gender and age.

RESULTS

The caregivers had an average age of 69.8 (+7.1) years and the recipients of care an average age of 81.2 (+9.9). The average duration of care, from the outset, was 4.5 (+4.1) years. Of the care recipients, 31.1% had severe dementia, 24.2% mild or moderate dementia, 23.6% questionable dementia and 21.1% an absence of dementia, according to the scores attributed by caregivers using the Clinical Dementia Rating instrument^17,18; 21.0% were described by their caregivers as having limited mobility; 44.3% as incapable of carrying out five or six BADL without help and 66.2% as incapable of carrying out between five and seven IADL without help.

The majority of caregivers were women (77.0%). There was a predominance of married families (39.9%) and those formed by a pair of elderly persons, their ascendants (parents or in-laws) and descendants (children and grandchildren) (27.7%). The majority resided with the elderly recipient of care (85.7%).

Most caregivers scored high in satisfaction with family relations (68.1%). Among both highly satisfied caregivers and those with low or intermediate satisfaction, there was a significantly higher percentage of people who only received or provided care than there was of those who did not provide and did not receive instrumental support for BADL.

There was a higher percentage of caregivers with high levels of satisfaction among those who reported experiencing reciprocal emotional support, and a higher percentage of caregivers who only received emotional support than those who did not receive and did not provide such support among caregivers with low and intermediate satisfaction with family relations. A higher percentage of elderly persons with high levels of satisfaction was observed among those who had at least one social partner involved in the exchange of support and a higher percentage of caregivers with intermediate and low satisfaction was found among those who had none. There was a higher frequency of caregivers with high levels of satisfaction among those who judged the emotional support they received to be adequate and of low or intermediate satisfaction among those who considered such support to be inadequate. The non-provision or provision of support for IADL, without burden, was related to high satisfaction, while burden associated with the provision of support for IADL was related to low or intermediate satisfaction (Table 1).

Univariate regression analyzes were performed to investigate the associations between high levels of satisfaction with family relations and the independent variables. A p value <0.030 was used as the selection criterion. With respect to the type of support and the nature of the family exchanges, the following were related to high levels of satisfaction with family relations: neither provide nor receive material support, in comparison with only offering and only receiving material support; neither provide nor receive instrumental support for BADL and IADL, in comparison with only offering and only receiving such support; neither provide nor receive emotional support, in comparison with only offering and receiving such support; one or more social partners from whom caregivers receive instrumental support for BADLs and emotional support, in comparison with none; one or more social partners to provide material and instrumental support for BADLs and IADLs, in comparison with no partners (Table 2).

For the association between high levels of satisfaction and the number of social partners involved in the support the following variables were selected: have one or more social partners from whom the individual can receive emotional support versus none, and have one or more social partners to whom the individual can provide instrumental support for BADL and IADL versus none (Table 3).

For the association between high levels of satisfaction and evaluation of quality of the support, it was selected to evaluate the support for BADL and the emotional support as adequate for the support received, and non-burdensome for the support given support (Table 4).

Using multivariate hierarchical regression analysis, the associations between high levels of satisfaction with family relations and the independent variables with p<0.30 in univariate analysis were assessed, which studied the associations between the same and the satisfaction of caregivers with family relations. A model with three blocks, which were included successively, were constructed. Adjustments were made for gender and age. Block 1 included variables that represented the nature of the material and instrumental support for BADL and IADL and the emotional support. In Block 2, the number of social partners involved in the receiving of emotional support and the providing of support for BADL, IADL and emotional support were included. Block 3 included the adequacy of the material and the emotional support and the instrumental support for BADL and IADL received and the sense of burden stemming from the provision of instrumental support for IADL and from emotional support were included.

From Block 1, the analysis identified emotional support and instrumental support for BADL as being significantly associated with high levels of satisfaction with family relations. These variables remained in the model after the 2nd block test, but not in the final result (block 3), where only adequacy of the emotional support received was significantly associated with high levels of satisfaction with family relations. In other words, irrespective of the variables gender and age, the association between high levels of satisfaction with family relations and the evaluation of the emotional support received as adequate was the most robust of the associations tested [PR= 3.8 (CI95% 1.34-11.18); p-value = 0.010] (Table 5).

DISCUSSION

The most important finding of the present study was that, in the opinion of caregivers, the quality of the support exchanged is more important than the number of social partners involved. The second most important finding is the primacy of emotional support over other types of support when determining the satisfaction of caregivers with family relations. Both these results are in line with theoretical and empirical literature on social and family support and subjective well-being among the elderly and caregivers^1-3.

No statistically significant relationships were found between the level of satisfaction of the participants with family relations and the variables gender, age, household arrangements and cohabiting, suggesting that the influence of support dynamics and the functionality of family relations override sociodemographic variables. Another possible interpretation is that these results were affected by the small sample size.

Most of the participants were daughters or spouses who were a little younger than the recipients of care. These data are comparable to those of other studies^4,5,19 and relate to social norms of gender, income and solidarity^4,5,18, as well as the high frequency of cohabiting observed, evidencing the search for facilitative arrangements of care^4,5. When desired by the elderly, cohabiting favors family functionality^2,4.

Bi-generational or tri-generational household arrangements prevailed, representing functional solutions for the distribution of goods and support^4,18-20. In families with good levels of adaptative and resolutive capacity these translate into more efficient distribution of tasks, effective crisis management, the objective adequacy of support, and social support from other relatives and friends^10,14. Having one or more social partners who provide emotional and instrumental support for BADL was related to greater satisfaction with family relations, possibly facilitated by intergenerational living arrangements and cohabiting.

Age brings a reduction in the size of social networks, which tends to affect the peripheral affective relationships of older adults and the elderly more than their close relationships²¹. Influenced by the restrictions of the temporal perspective that characterizes old age, elderly or almost elderly people tend to prioritize relationships that are affectively significant to them and discard those that are not²². They also tend to invest in emotional comfort rather than in the pursuit of information and status²². With a minimum number of social partners, the quality of support and the affective links between such partners are therefore more conducive to the well-being of older people and the elderly than the number of such individuals^23,24.

The quality of social interactions and affective bonds is more related to reciprocal than to unidirectional support^22-24. In other words, emotional support is mediated by mechanisms of social-emotional selectivity that favor the sense of adequacy of the same, when it comes from significant people who, for this reason, are selected by the elderly to form part of their close social network^21-24. These relationships are the main reason why emotional support occupied a privileged place in the satisfaction of caregivers with family relations.

Because of their physical limitations the elderly are not always able to provide instrumental support. At the same time, receiving instrumental assistance can give rise to feelings of ineffectiveness and dependence². This explains the finding of greater satisfaction with family relations among those who neither provided nor received instrumental assistance for IADL and BADL and among those who reported reciprocity in such assistance. To receive instrumental help when needed but also be able to provide it is a critical aspect for the definition of the elderly as autonomous individuals who are able to participate in exchanges of family support. Equally important is the possibility of offering support without a sense of burden, probably associated with the presence of more physical and emotional skills, and more functional and pleasurable family relations^2,5,25-27.

The provision of instrumental and emotional support without feelings of burden is an indicator of solidarity, demystifying negative stereotypes about old age, according to which the elderly are selfish and self-centered. On the other hand, when social support comes from the closest social relationships, feelings of burden among caregivers tends to be reduced^27,28. When the elderly are part of a dysfunctional family, however, offering instrumental or emotional support may predispose them to the risk of excessive physical and emotional exhaustion^5,28. The possibility of offering emotional and instrumental support in IADLs contributes to a sense of autonomy and control, and thus to satisfaction with family relations among caregivers²⁹.

In the sample investigated, the evaluation of the adequacy of emotional support was equivalent to high satisfaction with the functionality of family relations, suggesting the presence of resources to cope with stressors and affective attachment to significant others.

Replicating the design of the present study with a larger sample is one suggestion for further studies. Another is to include the level of dependence of recipients of care and the level of caregiver burden. To control the effects of variables of context, the influence of socioeconomic status and ethnicity as well as the gender and age of caregivers could be included. The complex statistical designs used in comparative and prospective studies with large and probabilistic samples may help to clarify the relationship between satisfaction with relations and family functionality, as well as the objective and subjective quality of support.

CONCLUSION

The most important finding of the present study is that caregivers believe that the quality of the support exchanged is more important than the number of social partners involved. The second most important result is the primacy of emotional support when determining the satisfaction of caregivers with family relations.

Satisfaction with family dynamics among elderly caregivers who care for other elderly persons involves important relationships with the exchange of the emotional, instrumental and material support that occurs among family members in response to social norms of solidarity and earnings.

The adaptive capacity, companionship and strength of affective bonds of the family, as well as the opportunities for personal growth it provides to its members and its ability to solve problems are central elements to good family functionality.

The results obtained are of considerable importance, both in terms of theory and application, in the context of providing information and emotional support to elderly caregivers, especially when the caregivers themselves are both frail and burdened by the effects of age and caring.