O2.2 Social inequalities in ageing
Chair: Marja Aartsen
O2.2.1 Life-course socioeconomic differences in preventable and non-preventable mortality
Malin Ericsson1 , Anna Johansson1,2, Nancy Pedersen1, Stefan Fors1,3 Anna Dahl Aslan1,4
1 Karoliska Institutet, Sweden, 2 Cancer Registry of Norway, 3 Stockholm University, Sweden, 4 Jönköping University, Sweden
Background: This study was conducted to investigate socioeconomic differences in mortality by comparing preventable with non-preventable causes of death.
Methods: We used data from the Swedish Twin Registry born before 1935 (n=36 248). Parental social class, own education, occupation and social mobility were used as separate measures of socioeconomic status. These data were linked to the Swedish Cause of Death Register. Preventable and non-preventable mortality was categorized by cause of death according to indicators presented in the Avoidable Mortality in the European Union atlas. Using Cox proportional hazard models, we tested the association between the socioeconomic measures and total mortality, preventable mortality, and non-preventable mortality. Data were split at age 70 to investigate differences in mid- and late-life mortality. By additionally performing a co-twin control analysis, the association was adjusted for possible genetic confounding.
Results: The social gradient for all-cause mortality was most prominent for the adult socioeconomic measures and stronger in preventable mortality before age 70. In analyses of social mobility, life-time low SES and downward social mobility had an increased mortality compared to life-time high SES. Adjustments for genetic confounding did not change the observed associations.
Conclusions: We found that mortality risk followed a social gradient with an increased risk for lower socioeconomic groups. The social gradient was evident in both preventable and non-preventable mortality although the association was more pronounced for preventable mortality. Additional analyses adjusting for familial confounding did not explain the observed association, which further confirms the relationship between low socioeconomic status and increased mortality.
O2.2.2 Patterns of cognitive decline at higher ages: Are more educated persons better off?
Jonathan Wörn1, Marja Aartsen2, Hannie Comijs3, Martijn Huisman3
1 University of Cologne, Germany, 2 NOVA, OsloMet – Oslo Metropolitan University, Norway, 3 VU University Medical Center, The Netherlands
Background: Inequalities in health persist at higher ages and might also affect cognitive functioning. It is a common finding that persons with higher educational achievement show higher levels of cognitive functioning and a lower risk of dementia. Cognitive reserve hypothesis additionally suggests educational differences in cognitive decline, but longitudinal studies altogether do not support this idea. One reason for this discrepancy might be the common application of growth curve models, assuming that all persons follow the same type of cognitive trajectory. However, this approach is not ideally suited to test the predictions of cognitive reserve hypothesis, according to which cognitive decline is initially slower but accelerates later among more educated persons, while it is initially faster and levels off in the longer run among less educated persons.
Methods: Testing the existence of such qualitative differences in trajectories of cognitive decline calls for a methodology that can depict different trajectories of cognitive decline. We apply such a methodology (growth mixture modeling) and argue that the effects of education become visible only in the long run, i.e. about 20 years. Using data from the Longitudinal Aging Study Amsterdam, we identify different types of cognitive trajectories in multiple domains of cognitive functioning over a period of about 20 years and examine whether higher educational achievement is associated with more favorable trajectories of cognitive decline.
Results/Conclusions: We present results of a precise test of cognitive reserve hypothesis. The findings will give important insights into more complex patterns of health inequalities at higher ages
O2.2.3 Socioeconomic Differences in Health Depreciation over the Life Course among the Norwegian Elderly
Maja Weemes Grøtting1, Astrid Grasdal2
1 NOVA, OsloMet – Oslo Metropolitan University, Norway, 2 University of Bergen, Norway
Comparing mean health at different ages based on cross sectional survey data typically suggests that health deteriorates faster with age among low- compared with high socioeconomic status groups, but that differences level off and even become smaller at older ages. The observed patterns could be explained by socioeconomic differences in mortality, but could also be driven by health related non-response common in surveys. The relationships between aging and health can hardly be measured without longitudinal data, but also then non-response and sample attrition among individuals in the targeted population rise selection issues. Here, we argue that detailed information about use of primary health care services can be used as an alternative indicator of presence and severity of health problems. We have this information available over time (eight years) in register data covering the entire Norwegian population. Combined with mortality data, institutionalization, and a rich set of socioeconomic- and family characteristics we focus on the population aged 50+ and analyze “health trajectories” for men and women at different ages and for different socioeconomic groups. Our results based on panel data fixed effects show that health inequalities are most profound in the late 60s, after which they level off. Mortality profiles suggests that the leveling is driven by the low educated healthy survivors (mortality selection). We compare our results with results from corresponding analyses based on the longitudinal NOR-LAG data and conclude that non-response does not bias survey estimates.
O2.2.4 Divergence and convergence: How do income inequalities in mortality change over the life-course?
Johan Rehnberg1,2, Johan Fritzel1, Stefan Fors1
1 Karolinska Institutet, Sweden, 2 Stockholm University, Sweden
Do income inequalities in health increase or decrease with age? Evidence is not conclusive and competing theories arrive at different conclusions. This study aim to analyse inequality in mortality by income in a synthetic cohort aged 30 to age 99 between the years 1990 to 2009, following each individual for 19 years. We use Swedish total population data with 4 772 044 individual observations. We calculate the accumulated probability of death for ages between 31 and 99. This approach takes into account selective mortality during the study period.
The results show that the highest relative income inequality in mortality is found at age 56 for men (RR: 4.7) and at age 40 for women (RR: 4.1) with differing patterns across the younger age categories between the sexes. The highest absolute income inequality in mortality is found at age 78 for men (19 % difference) and at age 89 for women (14 % difference) with similar patterns for both sexes. Both measures of inequality decrease after the peak, with very small or no inequalities above age 95. The results reveal that income inequality in mortality remain in advanced age, with larger absolute inequalities in older ages and larger relative inequalities at younger age.
In addition to examining age patterns in absolute and relative inequalities, this study show the importance of discussing and making an active choice in the type of inequality measure that is being used as the results tend to differ significantly between different type of measures.
O2.2.5 Choice and need: Care managers’ views of inequality in elder care in three Nordic cities
Sara Erlandsson1, Helene Brodin1, Lea Graff2, Olli Karsio 3, Elin Peterson1
1 Stockholm University, Sweden, 2 VIVE, Denmark, 3 University of Tampere, Finland
Background: Implementation of choice models in eldercare raises concerns about increased inequalities among older persons, since skills and resources required for making informed choices are not equally distributed. Care managers have a unique insight in implications of choice models on different groups of older people, as they inform and guide care users to enable informed choices. This study analyses care mangers’ perceptions of the interplay between choice and inequalities among older persons in three Nordic cities.
Methods: 30 interviews with care mangers were conducted in Copenhagen, Stockholm and Tampere. These cities were selected, as they are fore-runners in marketization but have implemented choice models differently. The interviews covered themes such as benefits and risks of choice models. The analyses focused on how benefits and risks for various groups interplay with different choice models.
Results: Despite differences in the construction of choice, care managers in all cities pointed out inequalities regarding financial resources, ethnicity, social problems and disability. Some inequalities were related to the ability to evaluate different choices (cognitive impairment, poor language skills), others were related to possibilities to find available choices, for example lack of suitable services for care users with addiction- or mental health problems, and lesser options for persons with scarce financial resources.
Conclusions: The paper adds an account to the growing international literature on how market practices in eldercare enhance inequalities among the older population. In particular, the paper shows how choice models interplay with inequalities by raising new barriers for groups, who were already disadvantaged.