Background and Objectives
Family caregivers provide help to a wide range of family members or significant others who need care resulting from long-term health conditions or age-related needs (Family Caregiver Alliance, 2016). In Canada, roughly 46 per cent of individuals aged 15 years and older have provided various types of care and support to their loved ones at home or in the community (Statistics Canada, 2015). Almost half of family caregivers support their parents and/or parents-in-law; other groups of care receivers include spouses (8%), children (5%), grandparents and other relatives (23%), and non-kin individuals such as friends and neighbors (16%) (Statistics Canada, 2015). For the purposes of this study, caregiving for children differs from parenting, and specifically refers to parents supporting children with disabilities, neurological health issues, or other long-term health conditions.
A substantial number of studies have examined caregiving outcomes based on the different types of relations between caregiver and care receiver, including parent, spouse/partner, siblings, children, grandchildren, or friends (Himes & Reidy, Reference Himes and Reidy2000; Namkung, Greenberg, & Mailick, Reference Namkung, Greenberg and Mailick2017; Petrová Kafková, Reference Petrová Kafková2015). Previous studies have confirmed that the association between caregiving and resulting psychological well-being outcomes may be correlated with the nature of caregivers’ relationships to their care receivers (Pinquart & Sörensen, Reference Pinquart and Sörensen2011). This correlation suggests that caregivers who provide care for spouses and children are more likely to experience decreased psychological well-being than those caring for parents or siblings (Litwin, Stoeckel, & Roll, Reference Litwin, Stoeckel and Roll2014; Penning & Wu, Reference Penning and Wu2016).
The psychological well-being of family caregivers is best accounted for by taking the interaction between the caregiver-receiver relation and other contextual factors into consideration (NamKung et al., Reference Namkung, Greenberg and Mailick2017). Some of the factors that influence the association between the relationship of the caregiver to the recipient and the psychological outcome of caregivers include caregiving intensity, living arrangements, quality of relationships between caregiver and care receiver, and the caregiver’s age and ethnicity (e.g., Lee & Farran, Reference Lee and Farran2004). However, relatively few studies have examined the effects of caregiving choice on the caregivers’ psychological well-being (Quinn, Clare, & Woods, Reference Quinn, Clare and Woods2009), and no study has addressed the interaction between the perceived choice in caregiving and the caregiver-receiver relation, or the effects of such interaction on caregivers’ psychological well-being.
This study investigated the influence of individuals’ perceived choice in becoming caregivers, and the effect of interaction between caregiving choice and caregiver-receiver relations on caregivers’ psychological well-being. This study explored two research questions: whether having a choice in becoming caregivers is associated with caregivers’ psychological well-being; and what effects the interaction of perceived caregiving choice and caregiver-receiver relations have on caregivers’ psychological well-being.
Caregiver-Receiver Relations and Psychological Well-being
The negative association between caregiving and psychological well-being has been extensively documented, as have the varying psychological well-being outcomes that depend on caregivers’ relationships with their care receivers (Anderson et al., Reference Anderson, Edwards, Pearson, Talley, McGuire and Andresen2013; Mausbach, Chattillion, Roepke, Patterson, & Grant, Reference Mausbach, Chattillion, Roepke, Patterson and Grant2013). In general, caregivers providing support to spouses or children report higher levels of psychological distress and life stress than caregivers for other relatives or non-kin care receivers (Penning & Wu, Reference Penning and Wu2016). Furthermore, evidence has demonstrated the complex relationship between the connection of caregiver to recipient and the caregiver’s psychological well-being.
The quality of the relationship between caregiver and receiver helps to explain the differing psychological outcomes of the caregiving process, due to the fundamental differences in relationships such as those between spouses, between parents and children, and between friends and neighbours (Quinn et al., Reference Quinn, Clare and Woods2009; Williams, Morrison, & Robinson, Reference Williams, Morrison and Robinson2014). For example, the effect of relationship quality on the psychological well-being of caregivers differs between caregiving daughters and caregiving sons: Caregiving daughters who have a lower relationship quality with their parents report better mental health, whereas caregiving sons with lower relationship quality report worse psychological well-being than those with higher relationship quality (Marks, Lambert, Jun, & Song, Reference Marks, Lambert, Jun and Song2008).
Some other aspects of the connection of caregiver to receiver influence the psychological well-being outcomes through their association with varying levels of caregiving intensity and strain. Spousal caregivers are most likely to become depressed, compared to other types of relationship, such as daughters or daughters-in-law (Lee & Farran, Reference Lee and Farran2004). The intensity and strain of caregiving among spousal caregivers are likely to increase because the caregivers and recipients typically live together, which further affects the caregivers’ mental health and quality of life. In addition, disparity in psychological well-being outcomes between different age groups has been observed (Ha, Hong, Seltzer, & Greenberg, Reference Ha, Hong, Seltzer and Greenberg2008; Nijboer et al., Reference Nijboer, Triemstra, Tempelaar, Mulder, Sanderman and van den Bos2000). Compared to older spousal caregivers, younger adult children caring for aging parents tend to experience more negative effects due to the stress and strain resulting from multiple roles, including the disruption to their work schedule and social activities (Nijboer et al., Reference Nijboer, Triemstra, Tempelaar, Mulder, Sanderman and van den Bos2000). Therefore, the complexity of caregiver-receiver relations and their influence on caregivers’ psychological well-being underscores the importance of identifying more factors in the caregiving context to provide a better understanding of family caregivers’ psychological well-being.
Caregiving as Choice in Caregiving Context
Not every family caregiver takes on caregiving responsibility by choice. Previous studies have shown that about 54 per cent to 69 per cent of family caregivers indicated they had a choice in taking care of others (Longacre, Ross, & Fang, Reference Longacre, Ross and Fang2014; Winter, Bouldin, & Andresen, Reference Winter, Bouldin and Andresen2010). Family relationship is one important factor related to caregiving choice. For example, individuals are less likely to have a choice in caregiving for parents or spouses than for non-family members or relatives (Schulz et al., Reference Schulz, Beach, Cook, Martire, Tomlinson and Monin2012). Various other factors also influence the decision to become a caregiver, such as the availability of social services for the care receiver, and the nature of the relationship between caregiver and receiver, which is associated with obligation, affection, and cultural norms (Stajduhar & Davies, Reference Stajduhar and Davies2005). For instance, Arksey and Glendinning (Reference Arksey and Glendinning2007) have pointed out that available social service options are closely related to family caregivers’ sense of choice in the caregiving situation. In some cases, it is assumed that a certain family member has more time or capability for caregiving responsibilities, and so that family member becomes the caregiver. Another study found that some people choose to become family caregivers because they do not trust their siblings’ capability in taking care of aging parents (Tatangelo, McCabe, Macleod, & Konis, Reference Tatangelo, McCabe, Macleod and Konis2018). Therefore, an individual’s perception of caregiving choice is affected by many relevant factors in the caregiving context.
Some studies have sought to explore the function of caregiving choice, and previous studies have confirmed that lack of choice can be a risk factor in family caregivers’ psychological well-being. For example, Winter et al. (Reference Winter, Bouldin and Andresen2010) have reported that, compared to caregivers who had a choice, those who did not become caregivers by choice were three times more likely to experience caregiving stress. Other studies indicate that having more flexible options in caring for family members is associated with higher life satisfaction, more positive affect, and lower distress, whereas lack of choice in caregiving is correlated with caregivers’ emotional stress, physical strain, and other negative health effects (Longacre et al., Reference Longacre, Ross and Fang2014; Pakenham, Chiu, Bursnall, & Cannon, Reference Pakenham, Chiu, Bursnall and Cannon2007). A lack of choice in the decision to become a caregiver further affects caregivers’ willingness and motivation to take care of family members in need (Burridge, Winch, & Clavarino, Reference Burridge, Winch and Clavarino2007); such caregivers also tend to resent their caregiving role, and feel less rewarded and quite dissatisfied with caregiving (Williams et al., Reference Williams, Morrison and Robinson2014). Some studies show that when individuals are forced to become caregivers, they may feel reluctant to perform caregiving tasks or to learn the skills and knowledge to be effective caregivers, which can further affect their psychological and mental well-being (Burridge et al., Reference Burridge, Winch and Clavarino2007; Williams et al., Reference Williams, Morrison and Robinson2014).
Theoretical Framework
This study was guided by three streams of thought related to caregiving choice. First, according to Dowding’s (Reference Dowding1992) model of extending (or increasing) choice, making a choice becomes meaningful when at least two alternative options are available. When individuals are provided with options to support family members in need, such as family caregiving and/or services such as adult day care or long-term care facilities, making a caregiving choice is meaningful. However, choosing to become a family caregiver instead of using public social services could be out of the family’s control: Necessary services might be unavailable, or budgets for public services might be limited; information or knowledge about services could be lacking, or the services themselves could be inflexible (Arksey & Glendinning, Reference Arksey and Glendinning2007; Wiles, Reference Wiles2003). In this case, taking care of family members or friends in need at home becomes the only option for many family caregivers. In such cases of limited choice without alternative options, the choice is not meaningful.
In addition, individuals’ psychological well-being can be affected by their sense of control (Schulz et al., Reference Schulz, Beach, Cook, Martire, Tomlinson and Monin2012). Making a choice is a way of practicing that sense of control; consequently, the absence of choice can be a stressor (Leotti, Iyengar, & Ochsner, Reference Leotti, Iyengar and Ochsner2010). When choices are minimal to nonexistent, caregivers are likely to lose control over the decision to become family caregivers and find themselves forced into the role. Having control over becoming a caregiver closely relates to how an individual manages caregiving responsibilities and the associated stress and burden (Szabo & Strang, Reference Szabo and Strang1999). In addition, the stress process model of caregiving (Pearlin, Mullan, Semple, & Skaff, Reference Pearlin, Mullan, Semple and Skaff1990) emphasises how family caregivers understand the caregiving context affecting their psychological well-being. In other words, a positive appraisal of the caregiving situation tends to result in a positive caregiving outcome. Thus, psychological well-being is influenced by how caregivers perceive the choice to take on the responsibility of care (Pakenham & Bursnall, Reference Pakenham and Bursnall2006).
The role of perceived choice in the caregiving context can thereby be explained in terms of the stress process model, the caregivers’ sense of control, and meaningful choice. Therefore, when examining family caregivers’ experiences and particularly with regard to caregivers’ psychological well-being, these three elements underlie the importance of caregiving choice in caregiving situations. However, there has been relatively little research on the effects of caregiver choice on the association between caregiver-receiver relation and caregivers’ psychological well-being. Consequently, we undertook this study to examine whether the caregiver’s perceived choice moderates the association between caregiver-receiver relations and psychological well-being.
Research Design and Methods
Data and Sample
The study sample for this project is selected from the Public Use Microdata File (PUMF) of the 2012 Canada General Social Survey (GSS, Cycle 26): Caregiving and Care Receiving. GSS 26 is a national survey focusing on Canadians who provide help or care to others as a result of long-term health issues, disability, or aging-related needs. A total of 23,093 respondents aged 15 years and older and living in private households in Canada completed the survey. We selected participants for this study according to the following criteria: those who had an identified primary care receiver and were still providing care during the survey; and those who completed the survey in a non-proxy interview. This study focused on a group of 5,285 participants who met these two criteria.
Focal Measures
Psychological well-being of caregivers is measured in terms of two main variables: life satisfaction and psychological symptoms.
Life Satisfaction
Life satisfaction, one of the most frequently used indicators of psychological well-being, refers to the evaluation of one’s quality of life (Winefield, Gill, Taylor, & Pilkington, Reference Winefield, Gill, Taylor and Pilkington2012). This study measured life satisfaction by asking participants, “How do you feel about your life as a whole right now?” Responses were represented on a scale of 0 to 10, where 0 meant “very dissatisfied” and 10 meant “very satisfied”.
Psychological Symptoms
Inviting participants to report whether they experienced any psychological symptom to indicate their psychological wellness is common in studies examining psychological well-being (Gaugler, Eppinger, King, Sandberg, & Regine, Reference Gaugler, Eppinger, King, Sandberg and Regine2013), as well as in family caregiving studies (Monin, Schulz, & Kershaw, Reference Monin, Schulz and Kershaw2013). One example is the Brief Symptom Inventory scale to assess psychiatric symptoms such as somatic symptoms, anxiety, and depression (Brown et al., Reference Brown, Gan, Jeffress, Mullen-Gingrich, Khan, Wood and Rush2006). In the current study, psychological symptoms were an aggregated variable computed from 10 different symptoms that participants experienced due to caregiving, including tiredness, worry or anxiety, feeling overwhelmed, loneliness, isolation, short temper or irritability, resentment, depression, appetite loss, or disturbed sleep. This 10-item scale has a high level of internal consistency with the Cronbach’s alpha at 0.838. Each reported symptom was counted and this variable rated on a scale from 0 to 10, with 0 indicating no symptoms and 10 indicating a maximum number of symptoms.
Caregiver-Receiver Relation
The nature of the relationship between caregivers and their care receivers was measured according to 25 categories, which were further grouped into five main categories: spouse (spouse/partner of respondent), children (son and daughter), parents (father and mother), relatives (ex-spouse, brother, sister, grandson, granddaughter, and so on), and non-kin others (close friend, neighbor, and co-worker). This five-category classification of caregiver-receiver relations was similar to that used by Penning and Wu (Reference Penning and Wu2016), and we only modified their categories by grouping siblings as relatives because there were no significant differences between the two groups in terms of caregiving stress and mental health.
Caregiving Choice
Participants were asked, “Do you feel you had a choice in taking on your caregiving responsibilities in the past 12 months?”, and the answers were recorded as “yes” (1) or “no” (0).
Caregivers’ Characteristics
The background information of participants included their age, gender, marital status, highest education attainment, employment status, country of birth, and personal annual income. Age was divided into three levels: younger than age 45 years, aged 45 to 64 years, and aged 64 years and older. Gender was classified as male or female. Marital status information was originally collected within eight categories, and further re-grouped into two levels (married/common law and not married). Education level was divided into four levels, from less than high school to university degree received. Employment status was divided into three categories, including employed, retired, and other (such as household work or going to school) based on participants’ main activity during the 12 months prior to the survey. The country of birth was classified as Canada or elsewhere. Personal annual income was originally grouped into 13 categories, which were later re-grouped into three levels with cutoffs at $30,000 and $60,000.
Care Receivers’ Condition
The characteristics of primary care receivers in this study include gender (female or male), age (younger than age 65 or aged 65 and older), and major health conditions. The main health condition information was originally collected according to more than 20 categories. For the purpose of data analysis, we organised these into four main categories: (a) aging/old age/frailty; (b) chronic disease and disability such as arthritis, cardiovascular disease, developmental disability or disorder; (c) mental illness and neurological issues such as Alzheimer’s disease or dementia; and (d) other debilitating conditions such as injury resulting from an accident or surgery.
Caregiving Context
Another five variables related to caregiving context include (a) caregiving demand, (b) living arrangements between caregiver and primary care receiver, (c) primary caregiver status, (d) availability of formal social support, and (e) use of coping strategies. Previous studies have confirmed the relationships of these variables to family caregivers’ health and well-being outcomes (e.g., Kim, Carver, Shaffer, Gansler, & Cannady, Reference Kim, Carver, Shaffer, Gansler and Cannady2015; Mackay & Pakenham, Reference Mackay and Pakenham2012). Participants were asked whether they provided eight different types of caregiving tasks to other individuals, including transportation, meal preparation, house maintenance or outdoor work, personal care, medical treatment, and so on. The more tasks in which participants were engaged, the higher the level of caregiving demanded of them.
The living arrangements of participants and primary care receivers were classified into two categories: living in the same household or not living together. Primary caregiver status was determined according to whether the care receiver considered the participant to be the primary caregiver. In the data file, the participants were originally asked whether they had occasional relief or respite care; received money from government programs; or received any federal tax credits for eligible caregivers. If the participants answered “yes” to any of these, they were considered to have had available formal support. Coping with family caregiving was measured according to the question, “Have you used any coping methods to help with your caregiving responsibilities in the past 12 months?” with the possible answers “yes” or “no”.
Data Analysis Procedure
We first performed a descriptive analysis to explore the characteristics of the study sample. Then, we conducted two sets of bivariate analysis: one on five different types of caregiver-receiver relations between caregivers and care receivers, and the other on participants’ perceived caregiving choice. For the numeric variables, an analysis of variance (ANOVA) post hoc test was also performed to classify the multiple group differences. Two sets of hierarchical linear regression were conducted to examine, first, the associations between the caregiver-receiver relation and psychological well-being (life satisfaction and psychological symptoms), and, second, the moderating effect of the perceived caregiving choice on these associations. In model a, we included all the relevant variables; in model b, we added the interaction between caregiver-receiver relation and perceived caregiving choice. We also performed a two-way analysis of co-variance (ANCOVA) to examine the main effect of the interaction between caregiving choice and caregiver-receiver relations in predicting the life satisfaction and psychological symptoms of the caregivers, and the simple effects of those two potential interacting factors.
We conducted a preliminary analysis to detect the missing values in the data set to determine the nature of missing values. Missing values from most variables are less than 1.6 per cent, and the nature of missing values is completely at random (MCAR) (Little’s MCAR test, p > .05). Only the personal income variable has about 12.7 per cent missing value. Because of the large proportion of missing data, this study indicated missing data as “not stated”. In addition, the multicollinearity test indicated that the variation inflation factor (VIF) is less than 2, and the tolerance is more than 0.5. Both indicators meet the recommendation for further regression data analysis (Braun & Oswald, Reference Braun and Oswald2011). Also, for all the data analyses we conducted, the sampling weight was applied for descriptive analysis, and standardised weight was performed for all bivariate and multivariate analyses.
Results
Almost 42.3 per cent of caregivers provided care for parents, followed by 28.5 per cent of caregivers caring for other family members, 9.4 per cent, for spouses, 12.7 per cent, for non-kin others such as friends, and 7.2 per cent, for children (Table 1). Roughly 55.3 per cent of participants reported that they had a choice to be caregivers. Family caregivers showed significantly different levels of perceived life satisfaction and psychological symptoms depending on their relationship to their primary care receivers (Table 2). The results of the post hoc test indicated that family caregivers who were taking care of relatives and non-kin individuals reported the highest level of life satisfaction; those who were supporting their spouses or children reported the lowest level of life satisfaction. A similar pattern was identified for psychological symptoms: family caregivers caring for their spouses and/or children reported the most psychological symptoms, followed by those taking care of parents, other relatives, and non-kin individuals. Roughly 80.4 per cent of family caregivers for non-kin others had the choice to become caregivers, followed by 67.6 per cent for relatives and 49.9 per cent for parents, as compared to 23.9 per cent of caregivers for spouses and 30.7 per cent for children.
All the results are based on weighted data.
All the results are based on weighted data.
*p < .05. **p < .01. ***p < .001.
As illustrated in Table 3, the differences between caregivers with and without choice in caregiving showed that a higher proportion of participants who were female, married/common law, and living with their care receiver tended to have less choice about their caregiving roles. Family caregivers with choice in caregiving reported significantly higher levels of life satisfaction and fewer psychological symptoms when compared to those without choice.
All the results are based on weighted data.
*p < .05. **p < .01. ***p < .001.
The results of hierarchical regression are presented in Table 4. Model 1a shows that family caregivers who took care of children, parents, other relatives, and non-kin others were more likely to report higher levels of life satisfaction than spousal caregivers. In addition, family caregivers’ perception of caregiving choice was significantly associated with higher levels of life satisfaction. In Model 1b, the interaction terms of caregiver-receiver relations and caregiving choice were added, and the results showed the significant moderating effect of caregiving choice on caregiver-receiver relations in predicting the caregivers’ level of life satisfaction.
All the results are based on weighted data. Reference is indicated in the (--).
*p < .05. **p < .01. ***p < .001.
Model 2a shows that family caregivers who cared for parents, other relatives, and non-kin individuals were more likely to report fewer psychological symptoms than spousal caregivers. No significant differences were identified between caregivers for children and for spouses. Perception of the choice to be a caregiver was significantly associated with fewer psychological symptoms. With the interaction terms added (Model 2b), the results showed the significant moderating effect between caregiving choice and caregiver-receiver relations, particularly for caregivers for children, parents, and other relatives when compared to caregivers for spouses.
Figures 1 and 2 show the significant interaction effect between caregiving choice and caregiver-receiver relations in predicting life satisfaction (F = 5.287, p < .001) and psychological symptoms (F = 41.557, p < .001). As indicated in Figure 1, when family caregivers have no choice, the lowest level of life satisfaction (M = 6.569) was reported by spousal caregivers, followed by caregivers for children (M = 7.023) and caregivers for parents (M = 7.370). Caregivers for relatives and non-kin others reported the highest levels of life satisfaction, and there was no difference between these two groups of caregivers. In addition, when caregivers had the choice to become caregivers to family members, no significant difference was observed between spousal caregivers and caregivers for children or parents, whereas caregivers for relatives and non-kin others reported higher levels of life satisfaction than spousal caregivers (p = .005 and p = .039 respectively). When caregiving was a choice, family caregivers for children reported significantly lower levels of satisfaction than caregivers for parents (p = .001), relatives (p < .001), and non-kin others (p = .001). This was also the case for family caregivers for parents when compared to relatives (p = .012).
Furthermore, family caregivers for spouses, parents, other relatives, and non-kin others reported higher levels of life satisfaction when they had a choice in caregiving, as compared to when they did not, but no difference was identified between caregivers for children who had a choice to do so and those who did not.
Figure 2 indicates the interaction effect of one’s perceived caregiving choice and the caregiver-receiver relation in predicting the number of psychological symptoms. When family caregivers had no choice, caregivers for spouses (M = 4.179) and children (M = 4.058) reported more psychological symptoms than caregivers for parents (M = 3.606), for other relatives (M = 3.130), and for non-kin others (M = 3.390), and no significant difference was identified between caregivers for spouses and children, or between those for relatives and non-kin others. When family caregivers had choice in their caregiving roles and responsibilities, caregivers for children reported more psychological symptoms (M = 3.372) than all the other types of caregiver-receiver relationships. Spousal caregivers showed no significant difference when compared to caregivers for parents, and showed more symptoms than caregivers for relatives (M = 1.958, p = .039) and for non-kin others (M = 1.651, p = .001). In addition, family caregivers for parents (M = 2.319) reported significantly more psychological symptoms than caregivers for relatives (p < .001) and for non-kin others (p < .001). Also, caregivers for relatives reported more symptoms than those for non-kin others (p = .012).
In addition, family caregivers among all relations with care receivers showed significant differences in psychological symptoms between those with and without choice, and caregivers reported fewer psychological symptoms if they had the choice to be caregivers.
Discussion and Implications
This study suggests that family caregivers’ psychological well-being is influenced by their relationships with care receivers, consistent with previous findings (Pinquart & Sörensen, Reference Pinquart and Sörensen2011). The results also support theoretical assumptions on the basis of caregiving stress process models, sense of control, and meaningful choice (Dowding, Reference Dowding1992; Leotti et al., Reference Leotti, Iyengar and Ochsner2010; Pearlin et al., Reference Pearlin, Mullan, Semple and Skaff1990), as well as previous studies (Longacre et al., Reference Longacre, Ross and Fang2014; Tatangelo et al., Reference Tatangelo, McCabe, Macleod and Konis2018) showing that family caregivers with perceived choices in caregiving reported better psychological well-being than those without choice.
The findings of this study also show that family caregivers for spouses and children report significantly lower levels of life satisfaction and more perceived psychological symptoms, consistent with the previous studies (Young & Kahana, Reference Young and Kahana1989). The reason for this may be that family caregivers for children and/or spouses tend to take more caregiving responsibilities and perform more caregiving tasks than caregivers for other relatives or non-kin care recipients (Pakenham, Reference Pakenham2012). Caregivers for spouses and children are more likely to live with the care recipients, and thus spend more time performing caregiving tasks, leaving less time for taking care of themselves (Bertrand, Fredman, & Saczynski, Reference Bertrand, Fredman and Saczynski2006), which would further influence their psychological well-being. In addition, Pashos and Mcburney (Reference Pashos and McBurney2008) tried to explain the caregiving situation based on the concept of biased kin investment, suggesting a caregiver’s investment in the care receiver, and the emotional closeness between caregiver and receiver varies due to different kinship (e.g., maternal relative vs. paternal relative). Therefore, it is also highly possible that family caregivers experience more toll when helping core family members, such as children and spouse, resulting in worse psychological well-being.
This study also supports previous conclusions which found that family caregivers who had no choice in caregiving tend to report worse psychological well-being than those who had a choice (Longacre et al., Reference Longacre, Ross and Fang2014). When family caregivers realise that they have no alternative or other support in providing care, they tend to compromise their own lives over their role as caregiver, and to have fewer positive experiences as a family caregiver than those who chose to become caregivers (Pakenham et al., Reference Pakenham, Chiu, Bursnall and Cannon2007). In general, family caregivers for spouses or children are more likely to have limited choice in becoming caregivers, and to be seen as primary caregiver, compared to the other three types of caregivers. Likewise, family caregivers who live with their care receiver and support them as primary caregivers are more likely to have no choice in becoming family caregivers at the beginning than those who do not. Therefore, caregiving choice plays an essential role in predicting family caregivers’ psychological well-being.
Furthermore, this study also points out the significant interaction effect between caregiving choice and caregiver-receiver relationship on family caregivers’ psychological well-being. It is notable that the moderating effect of caregiving choice is limited for caregivers for children, since caregivers for children are the only group to report the same level of life satisfaction regardless of caregiving choice. Also, when they have a choice in becoming a caregiver, caregivers for children experience the greatest toll on their psychological well-being among all relations. This is because caregivers for children usually deal with long-term health conditions including intellectual and developmental disabilities and other neurological health issues (Miodrag & Hodapp, Reference Miodrag and Hodapp2010), and they are likely to report significantly greater distress, chronicity of distress, and more emotional and cognitive problems (Brehaut et al., Reference Brehaut, Kohen, Raina, Walter, Russell, Swinton and Rosenbaum2004), regardless of whether or not they had the choice to become caregivers.
Another potential reason for these results is that caregivers for children tend to feel a lack of control over their own lives. As reported by Murphy, Christian, Caplin, and Young (Reference Murphy, Christian, Caplin and Young2007), caregivers for children with disabilities tend to receive unexpected calls from school as a result of their children’s health conditions, or they need to spend extra time dealing with teachers, doctors, or other agents to advocate for their children’s welfare. In addition, caregivers for children tend to worry about their children’s future (Knock, Kline, Schiffman, Maynard, & Reeves, Reference Knock, Kline, Schiffman, Maynard and Reeves2011). Caregivers might experience concern or even suffer from long-term anxiety when they plan for their children’s future, including life arrangements for the children when they live independently, medical expenditures, and available services (Murphy et al., Reference Murphy, Christian, Caplin and Young2007). One additional reason could be related to employment and income loss. About 60 per cent of the caregivers for children in this study were employed. Earle and Heymann (Reference Earle and Heymann2012) have pointed out that caregiving for children with special health needs increases the possibility of lost income by 48 per cent, which is significantly higher than that of caregivers for adults with health issues (29%). Therefore, family caregivers for children may need to worry more about negative consequences for employment and income in their daily lives.
Another notable result is that when there is no caregiving choice, family caregivers to non-kin others share similar psychological well-being traits with those individuals caring for parents or relatives. As a result of demographic trends in the population, people increasingly provide support for unrelated care receivers (Nocon & Pearson, Reference Nocon and Pearson2000). This type of care receiver is likely to be single, without children, and/or without active social networks. Therefore, they would heavily rely on their caregivers’ support (Barker, Reference Barker2002), even though caregiving tasks might be simply instrumental and practical tasks such as grocery shopping or transportation (Himes & Reidy, Reference Himes and Reidy2000). However, this would still increase the family caregiver’s emotional burden and the intensity of caregiving, resulting in a feeling of strain (Nocon & Pearson, Reference Nocon and Pearson2000).
The findings of this study have implications for public policy and community service organisations to provide more tailored service and support to family caregivers for children who have intellectual and developmental disabilities. Family caregiving for children can be a lifelong commitment that takes a great toll on health and well-being, whether the children live with their parents or somewhere else (Seltzer, Floyd, Song, Greenberg, & Hong, Reference Seltzer, Floyd, Song, Greenberg and Hong2011). More respite services and intervention programs would help parents better handle the caregiving situation and balance their family responsibilities and other social roles. It is expected that more support and care would be provided by non-kin individuals, such as friends and neighbors, because increasingly fewer family members can provide care and support for individuals who require care (Pleschberger & Wosko, Reference Pleschberger and Wosko2017; Redfoot, Feinberg, & Houser, Reference Redfoot, Feinberg and Houser2013). Our findings suggest that caregivers who provide help for non-kin others experience similar psychological symptoms regardless of their choice in becoming caregivers. Therefore, community services can provide support to individuals without family members or caregivers who are supporting their friends, neighbors, and co-workers.
There are limitations in this study. Participants in the GSS 26 were asked to identify one primary care receiver if they provided care to more than one person in the past 12 months. Thus, only the information relating to the primary care receiver was collected, and we were unable to account for the effects of other care receivers on the participants’ psychological well-being. In addition, the question about caregiver choice is limited to the caregivers’ feelings over the past 12 months. However, there is a possibility that family caregivers might have initiated care earlier than 12 months, and the feeling of caregiving choice might have changed over time, which is not indicated in the data set. Future studies should explore the role of perceived choice in caregiving, and how it would affect caregivers’ health and well-being from different perspectives.