DiscussionAs people live longer in our time, successful ageing is one of the most important research topics in gerontology with enormous ongoing research (Bowling & Dieppe 2005; Blazer 2006). Assessing the predictors of elders’ PWB in Greece is important as Greece is mainly a collectivistic culture that has also certain individualistic values (Kafetsios ; Nezlek, 2004). Heintzelman ; Bacon (2015) have shown that maintaining supportive relationships has been positively associated with subjective well-being and life satisfaction, especially for individuals with high interdependency.
Consistent with the expectations, social support/relationships correlated and were important predictors of PWB, and in fact they were stronger predictors than other health-related or sociodemographic variables, which is in line other findings (e.g., Robinson-Dooley, 2012). In particular, social support from partners was positively correlated with elders’ PWB and negative relating with the oldest child was negatively correlated with PWB. Unexpectedly, social support from friends neither correlated nor predicted PWB, whereas, partners’ support did. However, partners’ support failed to retain its significance in the multivariate regression model; only negative relating (NR) with the oldest child did. Furthermore, it was demonstrated that the NR with the child had a detrimental impact upon the elders’ PWB, far beyond the positive influence of other supportive relationships (i.e.
, partners and friends), which is in keeping with other research findings (e.g., Lee ; Szinovacz, 2016).
This finding also provides additional support of the importance of family values in Greece. We are not sure whether this contradicts Carstensen’s theory of socioemotional selectivity (Carstensen & Isaacowitz, 1999), which assumes that as people grow older they value most the qualitative social interactions (i.e., those that are emotionally meaningful), rather than interactions with other purposes.
Taking into consideration the familial values in Greece and the elders’ expressed desire to meet more often their child(ren) and grandchildren than other social contacts, it appears that these are the contacts they value most. Although it would seem likely to assume that well-being decreases with age because of activity limitations (Asakawa, Koyano, Ando, ; Shibata, 2000), i was depression that predicted lower PWB in the non-demented, but also non-physically healthy elders of our sample and not any health-related variables. Our findings support a growing body of evidence that this is the case among older adults (Ghubach, Zoubeidi, Sabri, Yousif, ; Moselhy, 2009; Meyer, Rumpf, Hapke, ; John, 2004) and suggest that the mental status of geriatric individuals may be more detrimental to PWB than physical frailty or comorbid illness. The elders in our sample had a moderate to high level of resilience, which might be a potential moderator in the association between the negative effects of functional decline and PWB. As anticipated, resilience was the most significant positive predictors of well-being, which is in line with previous research findings (Wu et al., 2017). Elders have to cope with multiple stressors, including frailty, illness, disability, limited independence, bereavement, negative emotions, decline in cognitive functions and poor living conditions (Ryff, Singer, Love, ; Essex, 1998).
The more resilient an elder would be the better coping with such stressors, and in turn, the better well-being. Resilience was also found to be a mediating factor in the relationship between NR with the oldest child and elders’ PWB. Since NR with oldest child is significantly valued by the elders in our sample, NR predicted reduced resilience and PWB. Our sample was relatively highly resilient. We cannot, however, know if this was due to the lack of conflicting parent-child relationships or to aggregated resilient resources because of age. Parent-child negative relatingThe elders perceived their child’s relating towards them more negative than what they believe of themselves, as their child’s perceived relating towards them did not correspond with their self-assessed relating in five out of the eight scores, and also the total score.
The elders reportedly avoid being overinvolved with their children, although they are vulnerable and needful of others’ help (LD). They believe that their relating needs are not being fully acknowledged, as their children perceive them as enjoying distance (ND), and at the same time trying to attract their attention (NC), also being insecure of being ignored and thus, repeatedly seeking assurance and approval (LN), and fearing of being pushed away and not being respected (UN). It is also noteworthy that in the regression analysis it was the elders’ beliefs about their children’s negative relating to them that predicted their PWB (three other-ratings), and not their actual relating towards them (one self-rating). Interestingly, it is not the way that the elders relate to their children, but the way that they consider that their children relate to them that impacts upon their PWB. We can assume that elderly parents actually blame their children for negative relating to them, rather than themselves for negative relating towards their children, which –if the latter is true- they were not able to perceive in themselves or they were reluctant to admit. Although various types of social biases may have contributed towards these perceptions, an accurate understanding of their child’s relating to them could also be true. Since we have only the elders’ ratings we cannot know of which is true. A possible extension of this work would have been the administration of the FMIQ to the elders’ children and the comparison of their between scores.
It was also interesting that the elders’ self-perceptions and their perceptions of their child’s relating to them coincided in only one case, that of the LD. LD includes acquiescent, subservient, compliant, obedient, timid, withdrawn, and resigned relating. Perhaps the most striking finding of this study is that, although elders’ self-assessed negative LD relating decreased their well-being, the other-assessed negative LD relating them increased their well-being.
In line with the Self-Determination Theory (Deci ; Ryan, 2000), it seems likely that elder parents might be reluctant to request or accept big amounts of support, such as instrumental, from their children in order to maintain a sense of self-competence and autonomy, which in turn, increases their well-being (Grundy, 2010; Gur-Yaish, Zisberg, Sinoff, ; Shadmi, 2013). Although Greek elder parents dislike being LD with their children, it might be that they like being perceived by them as such (i.e., withdrawn and resigned), which in turn resulted in higher levels of well-being. A possible explanation would be that they have high expectations about receiving support and care by their children in their old age.
Although famililism norms are gradually changing in Greece, the elders in our sample seem to be still reluctant to limit their expectation of filial piety. LimitationsThere are several shortcomings in the present study that deserve acknowledgement. The lack of a random sample limits the generalizability of the findings,. The findings may not be generalisable to samples from other cultures too. It would be informative to repeat the study in a different culture in which famililism is not so strong. Elders residing in other parts of Greece with different characteristics (e.g.
, lower expectations or adherence to filial piety) might have provided different results. The cross-sectional nature of this study makes impossible causal inferences. Intergenerational interactions may influence SWB or vise versa. However, our findings appear to support the conclusion that SWB is strongly affected by NR with the oldest child.
Self-report bias may have occurred, as respondents provided their subjective evaluation of well-being. It would have been clinically useful to have had qualitative accounts of the elders’ relating with their children. Useful insights could have been occurred should we have the adult children’s perceptions of their relating with their parents (both self- and other-assessed). Should the sample size was larger the findings might have been different. Most of the goodness-of-fit statistics were adequate, though the high value of RMSEA might have been artificially, due to the small df and low N (Kenny, Kaniskan, ; McCoach, 2014). Future studies with larger random samples are needed to replicate these findings. The findings might have also been different in highly conflicting parent-child relationships. The elders in our sample reported a relatively good relationship with their child.
It should also be acknowledged that the characteristics of the oldest children and of their relationship were not known (e.g., proportion of those who were unmarried and/or living with their parents, frequency and kind of support they provided), which, however, could have contributed to different findings.ImplicationsDespite the limitations, the implications of these findings are important. The findings add further support to the growing body of evidence and also call aging research to examine the role of the key determinants of well-being among older adults in order to provide important insights about the pathways by which elder populations can achieve better mental health.Factors such as depression, resilience, and negative interactions with children all represent modifiable factors, which should be considered by any interventions targeted in improving SWB and enhancing successful aging of elders.
Geriatric practitioners should assess these factors regularly and attention should be given to negative parent-child relationships as key targets, since evidence suggests that they are critical to the elders. They should also address elders’ mental health needs and direct their efforts at buffering them from the negative effects of depression, while at the same time, strengthening their resilience in adversity. The importance of an interdisciplinary approach to the care of geriatric population is highlighted, focusing not only on their medical problems, but also on any underlying emotional and social issues among elderly people. A comprehensive perspective on the factors contributing to SWB should form the basis for preventive actions too. Policies should also address resilience which could have long-term and positive effects on the well-being and longevity for senior population.