Caring for others without neglecting oneself? Grandparent caregivers, gender, and perceived health-related quality of life

Depósito de Investigación de la Universidad de Sevilla https://idus.us.es/ This is an Accepted Manuscript version of the following article, accepted for publication in Health Care for Women International: García-Jiménez, M., Torres- Enamorado, D., Sánchez-Sánchez, E., García-Gil, C., & Casado-Mejía, R. (2024). Caring for others without neglecting oneself? Grandparent caregivers, gender, and perceived health-related quality of life. Health Care for Women International. https://doi.org/10.1080/07399332.2024.2316142. It is deposited under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. 1 Caring for others without neglecting oneself? Grandparent caregivers, gender, and perceived health-related quality of life Abstract In this study, we explored the impact of caregiving in quality of life and health perceptions and outlined the profile of grandparent caregivers in Andalusia (Spain) in terms of a range of sociodemographic variables related to the care of their grandchildren. A sample of 171 participants (21.6% men) completed the Health-Related Quality of Life (HRQoL) questionnaire and another ad hoc one providing sociodemographic and caregiving data. We studied the relationships between these variables and HRQoL using ANOVA, chi-square and Multiple Linear Regression. We found a mainly female profile for the care of grandchildren and interesting relationships for the physical and mental components of HRQoL. Some relationships were marked by gender: caregiving for pleasure was more often the motive for men while by imposition was more common among women. We discuss the impact of caregiving on health according to the Self-Determination Theory and suggest practical implications derived from the main findings. Keywords: Grandmothers, caregiving, quality of life, health, Self-Determination Theory The improvement in life expectancy has led to a structural change in the population with the so-called "ageing of the elderly" (World Health Organization, 2015). This raises challenging social, cultural, and health issues and requires research to be performed in this area (Freixas, 2013): the study of grandparents has been overlooked consistently compared to other phenomena in the social sciences (Tarrant, 2010). The European population is steadily ageing and there is growing political and social recognition that public policies need to be more finely tuned to the characteristics and requirements of older people today (Leyra & Roldán, 2013). 2 Not all elderly people are dependent, although obviously ageing does increase situations of dependency (Casado-Mejía & Ruiz-Arias, 2016; Del Río-Lozano et al., 2013). Many elderly people play an active socio-family role, very often in caregiving, an essential economic activity that is fundamental to the maintenance of life (Colliére, 1999). Traditionally, the gender mandate has ensured that the onus of the task of care falls predominantly on women (Horsfall & Dempsey, 2015). Time use surveys in Spain show that care is done by families, but especially by the women in the family (Carrasco, 2016). In Europe, gender inequalities in the use of time allocated to caring for people and domestic tasks as opposed to that allocated to social activities are actually on the rise, with women increasingly having to bear the burden of caring for children, the elderly and/or the disabled (European Institute for Gender Equality [EIGE], 2017). According to this report, time spent doing domestic and care work has a major negative effect on women's ability to work, be financially independent or have leisure time. This gender gap is focusing debate in European Union policy circles, due to the challenge it poses to equality and the achievement of work-life balance (EIGE, 2017). Women entering the labour market not matched by a corresponding influx of men into the domestic sphere creates the conditions for absent parenting of varying severity that may imply a care crisis ensues, and across Europe, different approaches are being taken to tackle the problem. In some societies, the integration of women into the labour market has been successful thanks to the network of social services that make it possible to conciliate leave and care with work outside the home. In others, women's participation in the labour market is growing alongside family structures with strong ties of dependency and solidarity, as well as a sexist and patriarchal division of family work that reflects gender inequality. This model generates strong intergenerational solidarity 3 (Brenna & Di-Novi, 2016), as was shown in the crisis that struck particularly severely in Spain in 2008 (Caïs & Folguera, 2013). Thus, grandparents have been the source of intergenerational financial, instrumental, and emotional solidarity which has given them a pivotal role in the 20th and 21st centuries (Freixas, 2013; García-Díaz et al., 2014). This support provided by grandparents is widely prevalent and enduring over time. On average in Europe, 58% of grandmothers and 49% of grandfathers have offered assistance of some kind in caring for their grandchildren. The lowest percentages of grandchild care are recorded in Spain, Italy, and Switzerland (slightly over 50% of grandmothers and 40% of grandfathers), while this intergenerational support is more extensively prevalent in Sweden, France, the Netherlands, and Denmark a (Zanasi et al, 2023, Börsch-Supan et al, 2013; Hank & Buber, 2009). The profile of grandparent caregivers is generally described in terms of the occupational situation of their children. Grandparents look after their grandchildren when their daughters and sons cannot (Xu & Chi, 2018). In 90% of the cases, the care is required because their time at work does not allow them to reconcile work and family life (Caïs & Folgueras, 2013). To be precise, 44% of grandparents in Europe take care of their grandchildren when parents are absent, but in general this is a task that falls to women (Glaser et al., 2013; Hank & Buber, 2009). Grundy et al. (2012) explain the gender bias between grandparent child-carers based on this gender mandate. Horsfall and Dempsey (2015) argued that for grandfathers it is a personal choice, while for grandmothers it is more of a standard response, with less leeway to choose freely about care. The temporal stability observed in the provision of grandchild care in Europe in recent years is also evident when analyzing gender-specific trends. It is more likely for grandmothers to care for their grandchildren than grandfathers (Zanasi et al., 2023). 4 Whether something is done voluntarily or not, and the consequences of this, have been widely studied under the precepts of the Self-Determination Theory (SDT, Ryan & Deci, 2017). Through this theory, Ryan and Deci (e.g., 2017) explain the motivational processes in human beings in their interaction with the social context and the degree to which behaviours are volitional or self-determined. According to this theory, motivation represents a continuum consisting of different levels of selfdetermination that give rise to different types of motivation. When social factors lead to a perception of the feeling of autonomy, competence, and social relation, selfdetermined forms of motivation will develop, associated with positive consequences such as improved mental health and well-being. In contrast, the frustration of these perceptions will lead to non-self-determined or controlled motivational forms and reduced well-being and worse mental health (Ryan & Deci, 2000). While it is true that the role of grandparent caregivers varies depending on each family's circumstances and needs (Freixas, 2003), the questions remain: what consequences does this role have on the ageing process? Does it help or hinder healthy ageing as perceived by grandparents? To what extent do the reasons for caring - perceived as a voluntary choice or as an obligationaffect grandparents’ health perception? Several authors have highlighted the detrimental impact of care work on the health of caregivers (e.g., Flores et al., 2014; Larrañaga et al. 2008), particularly mental health (Fortinsky et al., 2007), and on social relations (Van-den-Berg et al., 2014). Caring for grandchildren has serious implications in terms of depressive symptomatology and psychological distress (Musil et al., 2017). The consequences for grandmothers tend to be more severe (Navaie-Waliser et al., 2002) as, more often than not, the caring for relatives comes down to women (Luna et al., 2016). On the other 5 hand, Arpino et al. (2018), who examined the role of grandparental caregiving in various aging European societies, pointed out that in a context where these caregiving roles are beneficial for society (as they replace high-cost services), research on the relationship between such care and the subjective well-being of grandparents is limited. Providing care to their grandchildren leads to higher levels of subjective well-being compared to non-caregiver people, and this is a trend observed in several countries studied, and more pronounced among women (Arpino et al., 2018). Therefore, it appears that caregiving might also have a positive impact on subjective well-being, as measured in terms of life satisfaction as a key component of health. According to this, the general view by other researchers in the topic is that caring favours active ageing, but is this really the case or do grandparents become trapped by the sense of responsibility? According to the review of the literature, gender seems to be a variable to be considered to understand the phenomenon of grandparental caregiving and health. However, researchers on this topic in the Spanish literature (e.g., Martínez-Martínez et al., 2019; Meil et al., 2018; Noriega et al., 2022) seemed to reduce the focus on gender analysis to showing differences between men and women without delving into the impact these mean on different areas of life (e.g., Martínez-Martínez et al., 2019; Noriega et al., 2022). The failure to include gender as a category of analysis is also reflected in systematic reviews on the subject (e.g., Danielsbacka et al., 2022). Additionally, no previous studies with Spanish participants have included analyses, within the framework of the SDT, of the relevance that caring for the grandchildren might have on health and quality of life depending on the perception of free election to care versus feeling it as an obligation. 6 Given the ambiguity and paradoxes involved in the ageing process, further research is required (Freixas, 2013) to provide a more accurate picture of the population that concerns us in this study in Spain. Thus, on the basis of the research questions and the limitations identified in the Spanish research on the topic, we addressed two objectives. The first is to identify the profile of the grandparents who care for their grandchildren, identifying the tasks and types of care they perform. The second objective is to analyse the link between the variables in this profile and their perceptions of health and quality of life, analysing them from a gender perspective and under the precepts of the SDT. Following a review of the literature, we have drawn up one main initial hypothesis: we expect that when care is not provided voluntarily or for pleasure, but perceived out of obligation, subjects will be predicted to have worse scores in the perception of health-related quality of life (HRQoL), both physical and mental, and controlling this effect by gender. In addition, we expect that women, compared to men: a) will be the most likely to combine care with other household tasks, and b) will show that caring is usually experienced as an obligation. We shall also conduct exploratory analyses of variables related to caregiving both in relation to the descriptive analysis of the profile of grandparent caregivers and the relationship between variables and participants' perceived HRQoL scores. These variables will be specified in the Method section. Methods Participants We established the following inclusion criteria in order to test the main hypothesis: that the person had been caring for one or more grandchildren for at least 6 months, and for at least 10 hours a week. All participants cared for grandchildren whose families had at 7 least one parental figure to whom they provided support with care. The sample of this study was selected by convenience because there is not any registration, census or list about grandparents in Andalusia (Spain). Finally, our sample consisted of 171 people, with 134 women (78.4%) and 37 men (21.6%). They all had at least one grandchild, lived in the province of Seville, and spoke Spanish as their first language. Their average age was 64.80 (SD = 7.09). The majority had completed primary education (49.7%), while 19.3% had not completed any formal education and 31% had secondary or university studies. Of the sample, 75.4% did no remunerated work (retired, housewives or unemployed) and only 27.5% did not live with a partner (single, widowed or separated). Instruments Given the overall objectives, we administered the following instruments: (1) A sociodemographic questionnaire to determine gender, age, occupation, level of education, their allocated health centre, and marital status. These variables appear in Table 1 of the Results section. (2) A questionnaire to collect caregiving information, such as how they were related to the person they care for; whether they had been to a workshop for grandparent carers; whether they lived with their children and where they care for their grandchildren; the number of hours and days dedicated to care; extra tasks performed while caring for the grandchildren; the reasons for deciding to care for their grandchildren; the benefits and limitations (from 0 = none to 5 = many) that they perceive from their caregiving. (3) The Spanish 4-week version of the SF12v2 questionnaire (Alonso et al., 1998) gives an overall score in the mental health component (MHC) and physical health component (PHC), from the scores in 8 dimensions (Physical Function, Physical 8 Role, Body Pain, General Health, Vitality, Social Function, Emotional Role and Mental Health). It is a Likert-type scale with 3 or 5 options with a total of 12 items allowing a maximum score of 100 points. The higher the MHC or PHC score, the better the perceived health status in each area. Procedure First of all, we identified possible variables linked to caregiving that could define the main characteristics of the study sample. For that reason, we reviewed the bibliography and prepared a draft of the questionnaire with the potentially relevant variables in relation to care. We added some other variables for exploratory analysis. This version of the questionnaire was evaluated by 13 experts and we did not reach the definitive version until there was consensus on each question. Once the Andalusian Commission of Ethics and Health Research had approved the study, the data was collected in the province of Seville (Andalusia, Spain) from April 2016 to July 2018. We gained access to the sample in an incidental manner in several Health Centres (HC) in Seville (n = 118) and then followed a snowball procedure to reach other grandparents (n = 53). The information was collected in one single interview either by members of the research team who accessed subjects directly in the HCs, or by health centre staff who were instructed on how to administer the instruments. The questionnaires were individual and took less than 20 minutes. Before they were given the questionnaires, participants were told about the purpose of the study and given guarantees regarding anonymity and confidentiality in the treatment of their data. They gave their consent to participate voluntarily in the study and understood their right to withdraw from it. 15 Perception of Health-Related Quality of Life: variables related to the Physical Health Component In Table 3, the authors present the results for the MLR models we developed to analyse PHC scores in relation to the reason for care controlling by gender, as stated in the main hypotheses. We showed that neither the first model including gender and the reason for care (R2 = .02), nor the second step when adding the interaction between both predictors (R2 = .01) resulted to be statistically significant. TABLE 3 AROUND HERE As can be seen in Table 3, we found indeed a significant principal effect regarding the reason for care, but with small effect size (ΔR2 = .02), meaning that participants in the sample had higher scores in PHC when they perceived that they care for pleasure or voluntarily. Discussion In this study we identified some interesting characteristics in terms of a profile of grandmothers and grandfathers who take care of their grandchildren in the region in question. First, the average profile from our study sample was a 65-year-old grandparent who had completed primary education and was not in paid employment. Overall, the sample showed average indices in both components of the HRQoL, with no significant differences between men and women. On average, the participants looked after two grandchildren for around six hours a day on five days of the week. The perception of the benefits resulting from the care was higher and almost doubled the average value of the perceived limitations resulting from caregiving. The profile consisted of mainly (70% of the sample) maternal female grandparents, maybe because of the criteria exclusion to select our sample. Few people 16 had attended grandparent carer workshops or received any kind of financial compensation for their caring role, which was mainly performed in their home and not in their children's, fitting into what Jendrek (1994) and Musil and Ahmad (2002) called day care providers. More than half of the sample stated that they looked after their grandchildren because they felt they had no choice or out of obligation, and gender differences showed higher percentage of women caring because they felt out of obligation, compared to men. Another characteristic was having to do other tasks while looking after the children, again done more by women. The most frequent were having to take their grandchildren to school or other activities, followed by extra-caring tasks including preparing meals, doing the shopping, and housework. The data confirmed what we had hypothesized: women spent more time than men performing extra tasks while they were looking after their grandchildren. At a descriptive level, the percentage of women who performed each type of extra task was greater than the percentage of men, especially in the above-mentioned tasks. This could be evidence of the gender gap regarding housework and informal care (García-Calvente et al., 2011). Not only did women spend a far greater number of hours and days caring than men as other studies have found (e.g., Glaser et al., 2013; Hank & Buber, 2009; Horsfall & Dempsey, 2015), but also, while pleasure was grandfathers' motive for caring for grandchildren, grandmothers perceived it as an obligation. As our hypothesis was met, this may reflect the socio-cultural influence on caregiving: for grandmothers it is imposed or a perceived duty whereas for grandfathers it is linked to pleasure, they decide to do it voluntarily (Grundy et al, 2012; Horsfall & Dempsey, 2015). When this is combined with the unequal gender distribution of care-related tasks and time (Pinquart & Sorensen, 2006), it illustrates the social construction of gender roles. Gender roles have been shown to have a profound and comprehensive impact on women's health 17 (Valls-Llobet, 2001). By this, we mean the set of expectations that patriarchal society establishes about behaviours, where the feminization of care is taken for granted as women are considered natural caregivers (Del Río-Lozano et al., 2013; Horsfall & Dempsey, 2015). Our findings might indicate that the cultural contexts of southern Spain might still be enduring this gender gap. While the benefits of caregiving are perceived and valued in a similar way by grandmothers and grandfathers, the former place greater emphasis on the drawbacks associated with care which could be explained by the excessive burden traditionally placed on women's shoulders (Luna et al., 2016). In fact, we did find that the perception of limitations was related to HRQoL in the PHC score, while it has been also shown to be lower when caring is done for pleasure. The perception of limitations may be a good indicator of the well-being of caregivers (e.g., Van-der- Berg et al., 2014). Furthermore, we found that maternal grandmothers, rather than paternal ones do most of the caring, which could be an example of caregiving being seen as an extension of the maternal role (Lee et al., 2015) and, therefore, as a female role. This should not conceal the fact that there is an upward trend in men becoming caregivers (Casado-Mejía & Ruiz-Arias, 2016; Torres-Enamorado, 2017). Given the profile of grandparents caregivers in our sample, let us now turn to the most interesting relationships found between our study variables. According to our first hypothesis, the MLR models showed that there was a better perception of mental health among those who were caregivers for pleasure than in those who indicated that they cared because they had no choice, while gender does not moderate this relationship. This prediction was not found regarding the physical component, meaning that the hypothesis was not fully fulfilled. In any case, our results regarding the MHC score are 18 coherent with the premises of the Self-Determination Theory (Ryan & Deci, 2017): when care is seen as an obligation it indicates an externally-imposed form of regulation associated to care behaviours. This is related to less self-determined motivational forms, which trigger off associated negative consequences, such as poorer well-being and mental health (Deci & Ryan, 2000). When there is a form of internal regulation or intrinsic motivation, such as caring because one wants to: these are associated to positive consequences for mental health and well-being (e.g. Rahman et al., 2015). Our results align with existing research on the impact of health on caregiving in general (e.g., Flores et al., 2014; Larrañaga et al., 2008). Helping others is beneficial for the health of the caregiver, as demonstrated by Hilbrand et al. (2017). Grandparents providing care had a 37% lower risk of mortality compared to grandparents who did not provide childcare, as well as non-grandparents. There is also an impact on mental health according to Fortinsky et al. (2007) and Musil et al. (2017), who emphasised the importance of autonomy or sensing the voluntary nature of caring as factors that protect one's mental health rather than acting as risk factors for carers. Knowing that the reason to care predicts the MHC score might have practical implications for primary care professionals, allowing them to refer grandparents to mental health prevention programmes. Indeed, Bordone et al. (2017) emphasized the importance of family policies (public childcare services and effective parental leave) and the improvement of the conditions of the female labor market regarding grandparental childcare patterns across Europe. Like these authors suggested, we also believe that grandparental childcare is not solely motivated by the need to address policy shortcomings, but it may also stem from an intrinsic motivation to maintain intergenerational connections (Bordone et al., 2017). 19 The differences we found along gender lines whereby men care out of choice and for pleasure whereas women care out of imposition, tie in with previous studies (e.g., Grundy et al., 2012; Horsfall & Dempsey, 2015), even though this relationship does not affect the relationship between the reason for care and MHC. These gender differences also reinforce the above-mentioned arguments concerning the gender gap in domestic life (García-Calvente et al., 2011). Given that the health consequences for caring grandmothers are more serious, particularly in the case of those that suffers more physical and emotional burden, it seems more pressing than ever to bolster and reinforce the ethics of care, as Carol Gilligan (1982) argued, as a moral value and virtue in men. As for the physical component of HRQoL, we did not find it related with the reason for care, maybe because of the effect of other non-studied variables in this work, such us the strength character (Noriega et al., 2022) or being a custodial grandparent (Danielsbacka et al., 2022). Even though, other relationships deserve to be mentioned. We did find worse PHC score for those with lower education level, in line with studies with the general population in Spain (e.g., Regidor et al., 1999) and research with grandmother caregivers (Yalcin et al., 2018). Education level correlates with many other socio-economic elements, such as income levels, which in turn tend to be related not just to better access to private health resources and medicines, but also to greater sustainability of healthy habits throughout life and greater well-being (Van-der-Berg et al., 2014). The place where they care raises interesting points. Although there is evidence regarding the relationship between this variable and emotional health (e.g., Berg-Weger et al., 2000), we are unaware of research that has found a similar relationship regarding the PHC as we did. Maybe grandmothers were living with their children and grandchildren due to poor health. In this situation of dependency, we believe that 20 grandmothers may themselves be called on to care for other dependent people, which could, in turn, explain their feeling of caring out of obligation. Although this interpretation would require specific research to substantiate it, if this were to be the case, it would put grandmothers seriously their psychical health at risk. The authors’ aim was not just to illustrate the impact of care on grandparents but also to understand that its effect depends on whether the care is perceived as voluntary and not imposed. Thus, the findings in this study should not be used against intergenerational care. Instead, given the gender differences found that seem to confirm the perpetuation of the feminization of care in the older generations, they should be used to develop social and care policies based on an effective and equal social redistribution, leading to improvements in mental health and well-being. This will only be possible if it is accompanied by greater participation of grandfathers in care work, which could make caregiving a more gratifying activity providing a sense of connection and belonging (Freixas, 2013). We would now like to draw attention to some limitations. The first to be highlighted is the absence of sample contextualization regarding the real situations that were being considered by each participant as caring voluntarily or as a pleasure versus caring because they felt it as an obligation. This could have implied different groups of participants: those helping with single parenthood, parents’ workload, or other situations of serious vulnerability –child abandonment, parent drug abuse. The data collection questionnaire did not include these categories, so we call for further research on this aspect, since this potential bias has neither been considered in other studies in Spain, to our knowledge. Notwithstanding, while interpreting our findings with caution, interesting results were evidenced regarding the participants’ perception of their caregiving motives according to the SDT precepts. We recognize the significance of the 21 presence or absence of motivation for caregiving, and therefore, we acknowledge it as a limitation in our methodology and emphasize the need for further exploration of the topic using different methodological approaches. Nevertheless, it is worth noting that there is a dearth of studies on this phenomenon in our country. Despite being aware of this limitation, we believe that with our findings we provide with an initial exploration of the phenomenon. Furthermore, we can consider our results as an initial step in the clinical context, as the negative impact on health of caregivers without motivation is evident, even if the exact reasons are yet unknown. Future studies should be conducted to classify and delineate the impact of various situations or motivations that lead grandparents to provide care when they have no other alternatives. Another limitation has to do with other non-controlled variables that would help understand the phenomenon of interest. For instance, the regularity of care beyond the number of hours per day or per week, and the extent to which parents were involved in such care, i.e., differentiating whether grandparental care is primary or secondary to parental care. Regarding the characteristics of the sample, it should be noted the small number of men in the sample, which may be related to the exclusion criteria and the sample´s characteristics: most grandfathers to be surveyed for this study had not been caring their grandchildren for a minimum of 10 hours per week and at least for the previous 6 months before they were contacted. These criteria were established with the purpose of ensuring proper analyses of the effect of care on health. Notwithstanding, this is a result by itself as it shows the feminization of care when outlining the profile of grandparents who care in southern Spain. In addition, the fact that these criteria hindered a larger group of men may be a limitation when analysing differences between women and men. However, the differences highlighted from our results are based on statistical differences that reached medium effect sizes, indicating that the effect does probably exist at a population level. 22 Second, these results come from one sample from a region in southern Spain. Hence, similar studies are needed in others before we consider them as representative of Spain. Of course, generalising the results to other European countries should be done with even greater caution, bearing in mind that family and care models vary greatly from one country to another (Brenna & Di-Novi, 2016). Another point concerns access to the sample, which was primarily through health centres and may have meant that the participants were generally in poorer health than if the sample had been selected through other community or recreational services, affecting the sample representativeness. We should also remember that researchers assumed participants' perceived quality of life, with the corresponding limitation of self-report questionnaires in terms of subjectivity when assessing one's own health. Finally, quality of life can be affected by a wide range of factors that can act as covariates and these have not been studied here. Despite these limitations, we found mainly grandmothers' role in caregiving in Spain. Although there have been several studies in Spain looking into health-related aspects of caregiving grandmothers, we have identified a profile and the most important characteristics in the sample that seemed to still reinforce gender roles in the south of Spain. Also, underlining not only gender differences but their implications has often been omitted from studies in our country. Although the role of caregiver continues to be undertaken primarily by women, the inclusion of men in our sample does suggest that men are becoming increasingly involved in care. This also enabled us to study the role of motivation, following the precepts of the SDT in the importance of the voluntary nature of caregiving, and the impact this has on the health of the caregiver. The focus on perceiving care as voluntary or not is a novelty with respect to previous studies with Spanish participants on the topic, especially because we evidenced the impact of gender 23 in the sense that the voluntary nature and pleasure associated to caring were limited almost exclusively to men. Disclosure of potential conflicts of interest The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. References Alonso, J., Regidor, E., Barrio, G., Prieto, L., Rodríguez, C., and De La Fuente, L. (1998). Valores poblacionales de referencia de la versión española del Cuestionario de Salud SF-36. Med Clin Barc, 111, 410-416. Arpino, B., Bordone, V. & Balbo, N. (2018). Grandparenting, education and subjective well-being of older Europeans. European Journal of Ageing, 15, 251–263. DOI: 10.1007/s10433-018-0467-2 Berg-Weger, M., Rubio, D. M., and Tebb, S. S. (2000). Living in and caring for older family members: Issues related to caregiver well-being. Journals of Gerontological Social Work, 33(2), 47-62. DOI: 10.1300/J083v33n02_04 Bordone, V., Arpino, B., & Aassve, A. (2017). Patterns of grandparental childcare across Europe: The role of the policy context and working mothers' need. Ageing & Society, 37(4), 845-873. doi:10.1017/S0144686X1600009X Börsch-Supan, A., Brandt, M., Hunkler, C., Kneip, T., Korbmacher, J., Malter, F., Schaan, B., Stuck, S. and Zuber, S., on behalf of the SHARE (2013). Data Resource Profile: The Survey of Health, Ageing and Retirement in Europe (SHARE). International Journal of Epidemiology, 42, 992–1001. DOI:10.1093/ije/dyt088 24 Brenna, E., and Di-Novi, C. (2016). Is caring for older parents detrimental to women´s mental health? The role of European North-South gradient. Review of Economics of the Household, 14, 745-778. DOI: 10.1007/s11150-015-9296-7 Caïs, J., and Folguera, L. (2013). Redefining the dynamics of intergenerational family solidarity in Spain. European Societies, 15(4), 557-576. DOI: 10.1080/14616696.2013.836404 Carrasco, C. (2016). El tiempo más allá del reloj: las encuestas de uso del tiempo revisitadas. Cuadernos de relaciones laborables, 34(2), 357-383. DOI: 10.5209/CRLA.53433 Casado-Mejía, R., and Ruiz-Arias, E. (2016). Influence of Gender and Care Strategy in Family Caregiver´s strain: a cross-sectional study. Journal of Nursing Scholarship, 48(6), 587-597. DOI: 10.1111/jnu.12256 Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. DOI: 10.4324/9780203771587 Colliére, M. F. (1999). Encontrar el sentido original a los cuidados enfermeros. Revista Rol de Enfermería, 22(1), 303-307. Danielsbacka, M., Křenková, L., and Tanskanen, A. O. (2022). Grandparenting, health, and well-being: a systematic literature review. European Journal of Ageing, 19, 341–368. DOI: 10.1007/s10433-021-00674-y Del Río-Lozano, M., García-Calvente, M. M., Marcos-Marcos, J., Entrena-Durán, F., and Maroto-Navarro, G. (2013). Gender Identity in Informal Care. Impact on Health in Spanish Caregivers. Qualitative Health Research, 23(11), 1506-1520. DOI: 10.1177/1049732313507144 Table 2. Data for the two-step hierarchical Multiple Linear Regression model to predict the mental health component (MHC) scores β SE F/t(gl)a p R2/∆R2 b Model 1 5.28(2,167) .006* .05 Gender -.10 1.67 1.27(167) .206 .01 Reason for care .19 1.42 2.43(167) .016* .03 Model 2 3.59 (3,166) .015* .04 Gender -.15 2.53 -1.22(166) .226 .01 Reason for care .12 2.94 0.5(166) .452 < .01 Interaction .08 3.38 0.08(166) .618 < .01 Notes: *p < .05 aF for the complete model and t for each predictor; bR2 for the complete model and ∆R2 for the increase in R2 caused by each predictor. Table 3. Data for the two-step hierarchical Multiple Linear Regression model to predict the physical health component (PHC) scores β SE F/t(gl)a p R2/∆R2 b Model 1 2.39(2,167) .095 .02 Gender -.01 1.92 -0.13(167) .899 < .01 Reason for care .16 1.65 2.43(167) .044* .02 Model 2 1.79(3,166) .152 .01 Gender -.08 2.86 -0.66(166) .509 < .01 Reason for care .05 3.37 0.32(166) .053 < .01 Interaction .13 3.87 0.77(166) .442 < .01 Notes: *p < .05 aF for the complete model and t for each predictor; bR2 for the complete model and ∆R2 for the increase in R2 caused by each predictor. Supplemental material Table A. ANOVA and correlations between all the independent variables and the MHC Scores M SD F(gl) p R2 Occupation (working) 49.88 7.56 0.89(1,169) .348 < .01 (not working) 48.75 9.06 Blood relation 0.13(1,161) .724 < .01 Maternal grandparent 49.04 7.84 Paternal grandparent 49.57 10.29 Attended a workshop for grandparent carers (Yes) 50.20 5.31 0.08(1,165) .780 < .01 (No) 49.10 8.73 Receives support 1.19(2,164) .307 .01 No support 49.19 8.70 Financial/institutional support 43.90 6.07 Family support 50.44 8.37 Lives with grandchildren 0.92(1,63.22) .342 < .01 Yes 49.49 7.63 No 47.80 11.09 Where they care 0.17(2,164) .844 < .01 In their home 49.10 9.01 At their children’s homes 49.26 7.06 Both 50.76 7.44 Why they care 8.92(1,168) .003* .05 For pleasure 51.64 6.78 They have no choice 47.56 9.38 Tasks they do while caring Take them to school/ activities (Yes) 49.14 9.24 0.01(1,145) .941 < .01 (No) 49.27 7.22 Prepare meals (yes) 48.85 9.21 1.76(1,152) .280 .01 (No) 50.73 6.43 Give them baths (Yes) 48.27 8.97 1.16(1,139) .284 .01 (No) 49.90 9.05 Put them to bed (Yes) 48.56 10.33 1.10(1,133) .295 .01 (No) 50.15 7.23 Housework (Yes) 48.47 9.25 1.07(1,148) .304 .01 (No) 50.05 7.92 Shopping (Yes) 48.53 9.430 0.98(1,148) .326 < .01 (No) 50.10 7.22 Care for other adults (Yes) 46.72 11.20 4.84(1,127) .030* .04 (No) 50.43 7.46 N r p Age 171 .14 .070 N. of grandchildren cared for 161 .04 .589 N. hours/day of care 150 -.15 .073 N. days/week caring 158 -.06 .429 Perceived benefits (1 to 5) 171 .10 .195 Perceived limitations (1 to 5) 168 .12 .111 Notes: *p < .05 Table B ANOVA and correlations between all the independent variables and the PHC Scores M SD F(gl) p R2 Occupation (working) 48.01 9.88 0.53(1,169) .467 < .01 (not working) 46.36 9.84 Blood relation 0.01(1,161) .925 < .01 Maternal grandparent 46.86 9.35 Paternal grandparent 46.70 10.49 Attended a workshop for grandparent carers (Yes) 51.44 6.21 1.29(1,165) .258 .01 (No) 46.38 9.84 Receives support 0.96(2,164) .384 .01 No support 46.23 9.95 1.29(1,165) .258 < .01 Financial/institutional support 48.85 8.27 Family support 49.12 9.69 Lives with grandchildren 0.05(1,169) .832 < .01 Yes 46.86 9.12 No 46.50 11.66 Where they care 3.83(2,21.60) .038* .03 In their home 45.85 10.23 At their children’s homes 50.06 6.53 Both 50.32 7.60 Why they care 4.79(1,168) .030* .03 For pleasure 48.91 8.86 They have no choice 45.50 10.22 Tasks they do while caring Take them to school/ activities (Yes) 49.14 9.24 0.00(1,145) .987 < .01 (No) 49.27 7.22 Prepare meals (yes) 45.75 10.17 2.10(1,152) .150 .01 (No) 48.59 8.55 Give them baths (Yes) 46.18 9.55 0.01(1,139) .908 < .01 (No) 45.98 10.67 Put them to bed (Yes) 45.61 11.01 0.36(1,133) .466 < .01 (No) 46.87 9.10 Housework (Yes) 45.63 10.38 1.17(1,148) .281 .01 (No) 47.51 9.09 Shopping (Yes) 46.10 9.60 0.04(1,148) .852 < .01 (No) 46.43 10.41 Care for other adults (Yes) 43.03 11.11 5.15(1,127) .025* .04 (No) 47.36 9.30 N r p Age 171 -.13 .098 N. of grandchildren cared for 161 .04 .604 N. hours/day of care 150 .06 .429 N. days/week caring 158 -.08 .319 Perceived benefits (1 to 5) 163 .12 .121 Perceived limitations (1 to 5) 160 -.26 .001* Notes: *p < .05