Claudine Burton-Jeangros & Dorith Zimmermann-Sloutskis
Abstract
Introduction
Old age is a priori a vulnerable stage of the life course. Empirical research shows contradictory trends with regard to the relationship between quality of life and aging.
On one hand, intuitively one expects quality of life to decline in older age due to loss, isolation, declining cognitive and physical abilities. On the other, some studies report a relative stability of quality of life over old age based on the theory of ‘satisfaction paradox’ or the ‘response shift’, which highlights the adaptation mechanisms used by individuals to review their own standards and values along the evolution of their individual situation. In the first view, external factors are expected to influence life course trajectory through a changing society and environment influencing different cohorts as they grow older through different periods. In the second view, psychosocial adjustment mechanisms, based on social and temporal comparisons, would allow individuals to cope with a progressively declining situation. Living conditions have changed in the last decades, these changes have particularly affected baby boomers cohorts in general, women’s life conditions in particular.
The rare studies, which differentiate age, cohort and period effect, conclude that once the cohort-effect is controlled for, quality of life is declining among the elderly.
Most of these studies use repeated cross-sectional surveys reporting the aging process for synthetic cohorts. To our knowledge only few studies use individual repeated data across several cohorts.
The purpose of this study was to investigate the distribution, prevalence and trajectory of Life Satisfaction (LS) as a measure of quality of life in the process of
37
aging among Swiss elderly women. We aimed at studying the effect of education, satisfaction with household income, social support, living arrangements (alone or with a partner) and health on LS and its trajectory within the age categories of the ‘young old’ and the ‘old old’ across different cohorts. The question was whether those determinants act differently on LS level or trajectory knowing that women’s social and economic conditions have changed over the recent period.
Method and statistical analysis
We used the Swiss Household Panel (SHP) survey data to address the effect of social inequalities on LS. The SHP is a longitudinal survey on social change within the non-institutionalized Swiss population running since 1999. We selected 1,402 women aged 65 to 84 years, providing 6,387 observations between 2000 and 2010.
We compared cohorts born between 1936-1945 and 1926-35 aged 65-74 years and cohorts born between 1926-1935 and 1916-1925 aged 75-84 years at different periods. The prevalence of life satisfaction and its cross-sectional association with social determinants in terms of odds ratios was estimated using the Generalized Estimation Equation models with conservative standard errors for repeated observations. LS trajectory was estimated with new developed multilevel mixed Hierarchical Age-Period-Cohort (HAPC) model to investigate the effect of age and covariates net of period variation. Birth year was used as a stratification dummy variable to compare former versus recent cohorts across identical age category net of period effect. Most studies using HAPC models are repeated cross-sectional using age for synthetic cohorts. With the SHP we followed repeatedly “cohorts within a cohort” and the within individual variance was added to the level-2 model in addition to the random period effect. Age and covariates were introduced as fixed in the level-1 model.
Results
Overall, despite a decline with age, about 50 percent of elderly women in Switzerland were very satisfied with their lives. However, the mean LS score and the prevalence of satisfied women were lower in more recent cohorts of identical ages. Although former cohorts had a higher LS score, their LS trajectory (slope) was significantly
38
negative with age. More recent cohorts displayed stability in life satisfaction with aging. These results hold equally for the ‘young’ and the ‘old’ elderly women.
High education, satisfaction with income, social support, living with a partner and good self-perceived health were all positive and significant predictors of LS (all OR>1, p<0.05) and these associations were similar across cohorts of similar ages (all interaction terms p>0.05). In earlier, generally more satisfied, cohorts most of the greater decline was due to lower education, dissatisfaction with income, low social support, living alone and being unhealthy. In more recent cohorts the stable trajectory was roughly similar at both levels of each covariate. Controlling simultaneously for these determinants, our results indicate that the life satisfaction trajectory remains flat with age. These results show that covariates have not only an impact on the overall level of life satisfaction but also on its trajectory with age.
Discussion
Interpretations: Our results suggest that the elderly born in the early 20th century (years 1916-1925) and consequently exposed during their early adult lives to the harsh social conditions of the years 1925 to 1945 appear to cope less well with the aging process. More recent and better educated cohorts, though more demanding and less satisfied with life, do indeed adapt better to the disadvantages of old age.
The stability of life satisfaction among more recent cohorts might be a result of their greater resources to deal with their own aging. In short, there is a significant difference in life satisfaction between cohorts of identical ages, with earlier cohorts being more satisfied than more recent cohorts, though the later ones are indeed better off in terms of socio-economic status and health at identical ages. Results further suggest that between-individual socio-economic resources may not be the only determinants of life satisfaction. Within-individual characteristics like psychological adaptation, levels of aspirations and comparisons with peers, are probably as important in the level of life satisfaction.
Limitations: The present study is limited to elderly women born between 1916 and 1945 who in the years 2000-2010 were anywhere from 65 to 84 years old.
Consequently, all the three observed birth cohorts (1916-1925, 1926-1935 and 1936-1945) might as children and young adults have experienced relatively hard economic
39
conditions, but enjoyed during most of their adults life the economic upturn of the sixties and thereafter. Still the findings may not be applicable to the upcoming “baby boomers” generation just now entering the life stage of the “young old”. However, things being equal by else, this upcoming highly demanding generation of elderly might experience lower levels of life satisfaction despite their enhanced capacity – better educated, higher income and increased relational resources - to cope with old age adversities.
Conclusion
While aging is actually encompassing a biological process of progressive decline, it is also essential to take into account social and cultural factors in the experience of the elderly during their life course. Previous results emphasized that there is not one single predictor of quality of life and its trajectory among the elderly, but that the quality of life is influenced by a combination of determinants. Quality of life should not be reduced to health. Though good health is a significant determinant of a good quality of life, so are personal and social resources. Consequently, assigning primarily medical answers to social needs appears inappropriate. Longitudinal analyses allowed us to disentangle the aging process net of cohort and period effects and to distinguish between external to the individual social determinants and within-individual psychological traits influencing the self-evaluation of LS.
40
Introduction
Quality of life among the elderly
Old age is a priori a vulnerable stage of the life course: over that period, individuals are potentially more exposed to loss, isolation, declining cognitive and physical abilities than in younger ages (Netuveli & Blane 2008, p. 123; Grundy 2006;
Kunzmann et al. 2000, p. 511). This ‘paradigm of decline’ used to predominate in gerontological research (Higgs et al. 2003). Theories of ‘disengagement’ and
‘structured dependency’ emphasized the marginalization from society of retired people (Baltes & Lang 1997, p.422; Blane et al. 2004, p. 2172). In this perspective, aging is negative and old age is seen as a problem (Walker 2005, p. 7). Quality of life of the elderly is expected to be low and measuring it is of little interest.
Recent perspectives developed more positive images of aging, as a response to increased longevity and the changing characteristics of the elderly. The distinction between a third age (‘young old’) and a fourth age (‘old old’) (attributed to Laslett 1996) is emblematic of this shift (Blane et al. 2004, p. 2172; Baltes & Smith 2003). A population-based definition situates the transition from the third to the fourth age at the age at which 50% of the birth cohort are no longer alive (around 74-80 in developed countries); a person-based definition associates the transition to the individual terminal decline towards death (Baltes & Smith 2003, p. 125). This suggests that there are multiple ages in old age. The fourth age is characterized by losses in cognitive ability, an increasing prevalence of dementia, and dysfunctionality;
some studies are now showing the negative consequences of living longer. By contrast, the ‘young old’ have been gaining in ‘good’ years of life (as observed through cohort comparisons), as a result of economic, social and cultural forces and their adaption capacities (self-plasticity). In this context, notions of positive or successful aging have developed in recent years (Von dem Knesebeck et al. 2007, p.
270; Hambleton et al. 2009, p. 4). These developments are associated with an increasing interest for the quality of life of the elderly (Zaninotto et al. 2009, p. 1301), especially since it is expected to remain high for some part of old age.
Indeed, empirical research suggests that there is no age-related decline in quality of life among the elderly (Kunzmann et al. 2000, p. 511; Gwozdz & Sousa-Poza 2010,
41
p. 399, Von dem Knesebeck et al. 2007). However, the rare studies which differentiate aging and cohort effects conclude that once cohort effects are controlled, quality of life is declining among the elderly (Gwozdz & Sousa-Poza 2010, p. 399).
This German study indicated a decline in later life, i.e. after the age of 85 (Gwozdz &
Sousa-Poza 2010). A longitudinal analysis in the United Kingdom observed an earlier decline, in the late 60s, but emphasized that the increased interval of confidence in later years suggest greater individual variations among the older (Netuveli et al.
2006; Fig 1 p. 361).
The relative continuity of quality of life over old age has been associated with the
‘satisfaction paradox’ (Walker 2005, p. 4) or ‘response shift’ (Wilhelmson et al. 2005, p. 597) which highlight the adaptation mechanisms used by individuals to review their own standards and values as their individual situation evolves. Psychosocial adjustment mechanisms, based on social and temporal comparisons, would allow individuals to cope with a progressively declining situation (Henchoz et al. 2008).
This indicates how much the measure of quality of life is complex. Traditionally, quality of life was approached in a rather limited perspective, closely associated with health-related measures and assessed by professionals (Walker 2005, p. 6;
Hambleton et al. 2009, p. 5). But the gap observed between objective measures and subjective evaluations suggests that it is necessary to measure quality of life as it is defined by the elderly themselves, which gives access to how they feel about their situation (Walker 2005). Differences between objective and subjective indicators are of interest since ‘QOL reflects the gap between individuals’ actual situation and that to which they aspire’ (Fry 2000, p. 246). It is now considered that the quality of life among older people is ‘a complex and multifaceted phenomenon that requires greater understanding’ (Higgs et al. 2003). While aging is actually encompassing a biological process of progressive decline, it is also essential to take into account social and cultural factors in the experience of the elderly.
Determinants of the quality of life
References to ‘quality of life’ are not recent, traditionally they were associated with notions of happiness and the ‘good life’ (Schuessler & Fisher 1985). Research on quality of life burgeoned after the 1960s along the developments of research on
42
social indicators (Hambleton et al. 2009). Growing interest in quality of life can be associated to sociological writings (Inglehart, Giddens, Beck) describing social change in general and transformations of values in particular, among which ‘issues of quality of life have replaced concerns of economic survival’ (Higgs et al. 2003, p.
240). Today the term is widely used, in several disciplines but also as a popular term in daily life and the media (Hambleton et al. 2009). This large use is however associated with a lack of consensus on its definition (Hambleton et al. 2009).
Measuring quality of life among the elderly is considered especially important since results are expected to help define orientations in health, social and economic policy (Hambleton et al. 2009, Grundy 2006). Recent developments, in particular around the notion of a third age, have emphasized the importance of extending quality of life measures beyond aging and health-related indicators (Wilhelmson et al. 2005, Netuveli & Blane 2008, p. 117, Netuveli et al. 2006). Furthermore, previous research has indicated that quality of life measures are often inconsistent with objective parameters (Allison et al. 1997, p. 222). It is therefore important not to restrict measurement to objective indicators but to integrate some subjective assessment, of quality of life but also of other domains. For example, Gwozdz & Sousa-Poza (2010, p. 408) observed that a decline in life satisfaction is associated with low levels of self-assessed health but not with objective health measures. Webb et al. (2011, p. 546) noted that the perceived financial situation was more important than income or wealth in itself.
In addition, several studies have identified various important determinants of quality of life, they include: social participation, family and social support, activities, psychosocial circumstances and socio-economic circumstances (Wilhelmson et al.
2005, Grundy 2006; Netuveli et al. 2006, p 361; Zaninotto et al. 2009, p. 1302;
Walker 2005). Previous results emphasized that there is not one single predictor of quality of life among the elderly, rather it is influenced by a combination of determinants. The same holds true for determinants of changes in quality of life (Webb et al. 2011, p. 544).
43 Quality of life dynamics and its determinants
Research has often considered elderly people as an homogeneous category, foremost defined by its age (Kahn & Rower 1987, p. 1; Walker 2005, p. 6). However, when they turn old people convey with them the social positioning they experienced throughout their life. It is therefore important to consider how their socioeconomic status influences their evaluations of quality of life. So far, only few studies have looked at this association (Von dem Knesebeck et al. 2007, p. 271). These social positioning is susceptible to determine contrasting levels of vulnerability (Grundy 2006 p. 129).
Furthermore, differentiating factors – social and economic resources – must be considered in a life course perspective. In that respect also, old age has typically been seen as a distinct phase of the life course (Walker 2005, p. 6), i.e. disconnected from what people had lived before the age of 65. However, ‘Life course ideas focus on the changing contexts of lives and their consequences for human development and aging.’ (Elder 2003, p. 52). This perspective takes into account individual choices over time but also the social influences and constraints that shape their opportunities (Elder 2003, p. 52; Walker 2005).
The dynamics of quality of life should then be observed at two different levels (Allison et al. 1997, p. 222): as ‘within-subject change’ it refers to the evolution of the individual’s quality of life individual over his/her life course; as ‘between-subject difference’, it considers differences among social groups, but also among different birth cohorts who are born and age in specific social contexts. Such an approach, combining individual agency and structural constraints, allows taking into account changing expectations. As suggested by Walker (2005, p. 4), the rather low expectations of the older generation could lead them to rate their lives as better than members of more recent cohorts will. In his analysis of the living conditions of successive cohorts, Chauvel (1998) emphasizes how much the cohort born just after 1945 has benefited of a very positive socio-economic context in their young age and as they entered the workforce in the late 1960s. Previous cohorts, born in the period covering both wars clearly had less favourable trajectories.
44
The articulation of social inequalities and the life course perspective has generated different theoretical developments (von dem Knesebeck et al. 2007, p. 280).
According to the cumulative advantage theory, inequalities increase with age and larger social differences are expected among older than younger people (Dannefer 2003). The convergence theory claims on the other hand that socio-economic variations are greatest during the working ages, therefore less inequalities should be observed after retirement age, also considering that in old age biological parameters gain more importance. Besides, social and health policy would ensure a reduction of differences among the elderly (von dem Knesebeck et al. 2007, p. 280). Furthermore, expected convergence could also result from a selection effect, through which more disadvantaged people die earlier. A British study confirmed the importance of life course processes: disadvantaged people reported lower quality of life in early old age, however current socio-economic status and current health problems had a stronger influence in quality of life variations (Blane et al. 2004, p. 2177). These results confirm the importance to consider the combined role of life course factors and current determinants.
Gender also constitutes an important determinant of social positions, health trajectories and life course trajectories. Women tend to be less educated and to have lower economic resources; their professional trajectories are not equivalent to those of men, especially due to their greater involvement in family tasks. Furthermore, while women live longer they tend to be in poorer health than men (Annandale &
Hunt 2000). It is therefore particularly important to take into account gender differences when studying old age, while also considering the social heterogeneity existing among both men and women (Arber 1999). Gender is expected to influence the men’s and women’s evaluation of their quality of life during old age (Netuveli 2006, Walker 2005 p. 8). It has been observed that women tend to report lower quality of life than men, with differences in income and marital status acting as major causes of gender differences (Cherepanov et al 2010).
The Swiss context
Comparatively to other countries, the Swiss population benefits from good social conditions and a high life expectancy. However social inequalities do exist and tend to be increasing (Levy 2010, p. 64). The Swiss welfare state cannot be easily
45
classified in the Esping-Andersen typology describing contrasting regimes of welfare in Europe, for two reasons: the Swiss social security system was put in place fairly recently (after the 2d world war) and it combines features of different typical models.
Today it can be considered as a mix of the conservative model (focused on the protection of work, generous regimes in terms of old age, unemployment and disability, a model which includes France and Germany) and the liberal model (emphasizing the role of the private sector while limiting its help to the most poor people, like it is the case in the United Kingdom and the United States) (Bertozzi et al. 2008, p. 20).
The cohorts now being over 65 have all experienced during most of their adult lives the benefits associated with the economic boom of the three decades following 1945.
A recent analysis concluded that the financial situation of the elderly in Switzerland is fairly good, even though their income is lower than the one of the working population.
Furthermore, retired people benefit from important levels of wealth (higher than among those younger than 65). However, there also exist some groups of poor elderly, especially those receiving only the national retirement pension (AVS) and no professional pension (2e pilier) (Wanner & Gabadinho 2008). Recent comparative analysis of quality of life among the elderly in Europe noted that inequalities in quality of life along socioeconomic indicators were limited in Switzerland (von dem Kneseback et al. 2007, p. 278).