Ageing strand abstracts

STRAND ORGANISERS: DR ATHINA VLACHANTONI, UNIVERSITY OF SOUTHAMPTON & DR STEPHEN JIVRAJ, UNIVERSITY COLLEGE LONDON 

Ageing & the life course: applications using the English Longitudinal Study of Ageing: Monday 12 September 1:30pm

Quality of life and loneliness among people who reported visual impairment: findings from English Longitudinal Study of Ageing
Jitka Pikhartova, Professor Christina Victor, Brunel University London   

With the growing proportion of an ageing population comes an increase in interest in the quality of life (QoL) of those with visual impairments. Visual impairment reduces possibility of active participation in daily living activities and therefore alters quality of life. The aim of the study was to investigate the impact of visual impairment reported between waves 1 and 5 on QoL and loneliness reported in wave 6 of English Longitudinal Study of Ageing (ELSA). Longitudinal analyses on data from over 5,500 ELSA respondents participating in all 6 waves aim to estimate prevalence of QoL measures and loneliness to assess how visual impairment affects them later on. Impairment was based on self-reported information; QoL was measured by CASP-19 scale and life satisfaction score (LSS) by 5 questions; loneliness was measured using R-UCLA scale. Visual impairment at least once within the study was reported by 19% of respondents. Quality of life throughout the study decreased more rapidly among those with vision impairment. Negative and significant association between reported visual impairment in waves 1-5 and CASP-19 and LSS in wave 6 was found for both outcomes and remained significant for CASP-19 in fully adjusted analyses accounting for social and psychosocial characteristics, age and gender and reported loneliness (coefficient -0.28, 95%CI -0.54 to -0.02) but was almost entirely explained for LSS. Results suggest that aspects of quality of life other than just life satisfaction may be affected by visual impairment and could be targeted by interventions. 

jitka.pikhartova@brunel.ac.uk 

Biomarkers of late-life depression and its health consequences
Livia A Carvalho, University College London 

Background: Evidence supporting immune dysfunction as a causal factor for depression is inconsistent, possibly due to a lack of regard for inflammation chronicity. We aimed to examine whether chronic levels of low-grade systemic inflammation lead to future depressive symptoms. Methods: Prospective data from a population-based sample of older adults were analysed (the English Longitudinal Study of Ageing). The odds of incident depression (having ≥4 of 8 symptoms based on the 8-item Centre for Epidemiological Studies Depression (CES-D) scale) after 4-years follow-up (in 2012/13) were examined among 1963 participants free of depression on 2 baseline occasions (in 2004/05 and 2008/09), according to the number of baseline occasions (0, 1, or 2) with systemic inflammation (C-reactive protein ≥3 mg/L). Incident depression was also examined according to group combinations of chronic disease status (in 2012/13) and occasions with inflammation. Results: Compared with those inflamed on 0 occasions, those inflamed on 2 occasions showed 1.96 (95% CI=1.27, 3.02) times higher odds of becoming depressed adjusting for demographic factors; these odds remained significantly elevated after additional adjustment for antidepressant drug use, chronic disease, and health behaviours (OR=1.84, 95% CI=1.14, 2.98). Inflammation on 1 occasion was not associated with incident depression. The highest odds of becoming depressed were seen among those who had both a chronic disease and 2 occasions of inflammation (multivariable-adjusted OR=2.10, 95% CI=1.17, 3.76). Conclusions: Sustained low-grade systemic inflammation increases the risk of becoming depressed among older adults; this risk is highest among those with an existing chronic disease. 

livia.a.carvalho@ucl.ac.uk 

Is the relationship between self-perceived age and emotional distress bi-directional?
Isla Rippon, Andrew Steptoe, Department of Epidemiology and Public Health, University College London    

Background: Longitudinal population studies have demonstrated that older people who feel younger than their age have more favourable health outcomes, including reduced emotional distress. These associations may be bidirectional, but very few longitudinal studies have attempted to test the direction of the relationship between self-perceived age and emotional health. Methods: We used a national sample 5,191 older adults who completed the self-perceived age measure in the fourth (2008-09) and sixth (2012-13) waves of the English Longitudinal Study of Ageing (ELSA). We fitted multiple logistic regression models to analyse the relationship between self-perceived age and elevated depressive symptoms (defined as scores of four or more on the eight-item Center for Epidemiological Studies Depression Scale). Results: Participants were divided into three groups: those whose self-perceived age was close to their chronological age (one year older to two years younger, 74.2%), those who felt more than one year older than their chronological age (20.6%) and those who felt three or more years younger than their actual age (5.2%). After adjustment for covariates, we found that feeling older than actual age remained a significant independent predictor of elevated depressive symptoms (OR 1.31, 95% CI 0.95-1.80) in comparison with individuals who felt younger than their age. Conversely, the association between depressive symptoms and future self-perceived age was reduced to non-significance in the fully-adjusted model. Conclusions: These findings indicate that an older subjective age is associated with future emotional distress, but that the reverse pattern is less consistent. 

i.rippon.12@ucl.ac.uk   

Investigating the Bidirectional Association between Body Mass Index (BMI) and Sleep Duration in Older Adults: The English Longitudinal Study of Ageing (ELSA)Victoria Garfield 1, Clare Llewellyn 1, Andrew Steptoe 1, Meena Kumari 1, 2, 1 University College London, 2 University of Essex   

Cross-sectional analyses of the association between body mass index (BMI) and sleep duration in adults suggest that increased BMI is associated with shorter sleep duration. However these findings are equivocal in older adults. Our main objectives were to examine within a community sample of older adults, i) the cross-sectional relation between BMI and sleep duration, and ii) whether BMI is associated with changes in sleep duration, or whether sleep duration is associated with changes in BMI. Body mass index and sleep duration and co-variates were assessed at wave 4 (baseline 2008-9) and wave 6 (follow-up, 2012-13) in 5 113 respondents from the English Longitudinal Study of Ageing (ELSA). Associations were analysed by linear regression. Body mass index and sleep duration and co-variates were assessed at wave 4 (baseline 2008-9) and wave 6 (follow-up, 2012-13) in 5 113 respondents from the English Longitudinal Study of Ageing (ELSA). Associations were analysed by linear regression. 

v.garfield.13@ucl.ac.uk 

Ageing: Intergenerational exchange & social activity: Monday 12 September 4:45pm 

The dynamics of social care and paid work in mid-life
Madelin Gomez-Leon, Maria Evandrou, Jane Falkingham and Athina Vlachantoni, ESRC Centre for Population Change and Centre for Research on Ageing Faculty of Social, Human and Mathematical Sciences, University of Southampton 

This paper investigates the impact of informal care provision to older parents/parents-in-law on the employment status of adult-children in mid-life. The study analyses unique longitudinal data of a cohort of individuals born in Britain in 1958. The analytical sample comprises all respondents who were at risk of providing care (i.e. with at least one surviving parent/parent-in-law) and who were employed at age 46. Logistic regression then investigates the impact of caring on changing employment status between 46 and 50, controlling for a range of socio-demographic characteristics, respondent’s health status and partner’s employment status. Continuous caring was significantly associated with reducing or stopping work, with the effects mediated by the carers’ own occupational category and health, and by their partner’s employment status. Amongst those caring at age 50, those providing intense care (10+ hours a week) were more likely to have adjusted their employment patterns between 46 and 50 than those in less intense roles and male carers were more likely to have adjusted their employment patterns than female carers. Duration and intensity of care matter. The ability to combine paid work and parental care in mid-life will be increasingly important in the context of rising longevity. 

M.Gomez-Leon@soton.ac.uk 

The Factors That Have Impact on Older People’s Mental Well-being: An Analysis of Older People’s Needs for, Receipt of and Unmet Needs for Care in England
Jingjing Wan 1, Raphael Wittenberg 2   1University of Warwick, 2London School of Economics and Political Science   

Meeting older people’s needs for social care has never been so important to public policy as this time, when the proportion of older people is increasing while the budgets of healthcare service are facing cuts. The dataset used was the Health Survey England 2011-2012. A total of 4233 older people were selected from the general household population. Positive mental well-being was measured using the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS). The associations between mental well-being and background characteristics, health related factors, functional disability, receipts of assistance, care matrix were investigated in a correlation coefficient analysis and a linear regression model. WEMWBS (Number of obs: 1858) is negatively correlated to: Self-assessed General Health, Activities of Daily Living and Instrumental Activities of Daily Living, Receipt of care in last month and Unmet needs(-0.42 (95% CI [-0.59, -0.24]) and positively correlated to marriage and cohabitation. The Unmet needs from: getting up and down stairs; having a bath or a shower; and getting in and out of bed had greater impact on WEMWBS. Separate regressions showed gender disparities: unmet needs for care and recent employment activities had greater impact on women than on men; marital status and cut of physical activity had greater impact on men than on women. The types of carer that provided help did not have significant effect on older people’s WEMWBS, as long as there were no unmet needs, older people’s mental well-being could be ensured; however, the balance between the care-providing and the care-financing still left great policy concerns. 

A.Wang.4@warwick.ac.uk 

Does intergenerational social mobility affect general health, oral health and physical function among older adults in England?
Alejandra Letelier, Anja Heilmann, Richard G Watt, Stephen Jivraj, Georgios Tsakos, Department of Epidemiology and Public Health, University College London   

Background: Socioeconomic position (SEP) influences adult health. People who experienced disadvantages in childhood or adulthood tend to have higher adult disease levels. Even so, life is a dynamic process leading people through different socioeconomic paths. Research on social mobility takes this into account, providing a long-term view of the effect of SEP on health. The aim of this research is examine the effects of intergenerational social mobility on adult general health, oral health and functioning in a population aged 50 and over in England. Methods: This study is based on the secondary analysis of data from the English Longitudinal Study of Ageing (ELSA). Using cross-sectional data, nine social trajectories were created based on parental and adult occupational SEP. Regression models were used to estimate the associations between social trajectories and the following outcomes: adult self-rated health, self-rated oral health, oral health related quality of life, total tooth loss and grip strength; controlling for socio-economic background and health-related behaviours. Results: Associations with adult SEP were stronger than with childhood SEP. Compared to the stable high group, being in the low SEP groups in childhood and adulthood was associated with poorer health and oral health for all examined outcome measures. For adult self-rated health and edentulousness, graded associations with social mobility trajectories were observed. Conclusion: Intergenerational social mobility was associated with self-rated health and total tooth loss. Compared to only those who remained in low SEP over time reported worse health and function measurements. Potential limitations will be discussed. 

alejandra.letelier.12@ucl.ac.uk 

Digital inclusion in later-life: Cohort changes in internet use over a 10 year period in England
Katey Matthews 1 , James Nazroo  1 ,  Alan Marshall  2  , 1 University of Manchester, 2 University of St. Andrews 

Research question: How does internet use change in later life, and are these changes cohort-specific?  Methods:  We model cohort-specific patterns of frequent internet use for people aged 50 and over. Multilevel growth models are used to observe trajectories of internet use over the ten year period. Firstly, analyses are stratified by gender and wealth, and secondly we additionally test for health effects.  Data sources:  We use waves 1 to 6 of the English Longitudinal Study of Ageing (ELSA), spanning the period 2002-2012.

Potential applications:  With UK policy currently aiming to ensure everyone can access the internet frequently by 2020, this study highlights some groups of society who are most likely to remain digitally excluded. 

Results: The study finds cohort-specific differences in patterns of internet use. Rates of internet use increase faster among younger cohorts yet, despite initially increasing, begin to decline among older cohorts. Poor health is shown to be a key factor in shaping the trajectory of internet use over time. Rates of internet use are consistently lower for women than men and for those in poorer financial circumstances, independently of age cohort. 

katey.matthews@manchester.ac.uk   

Ageing:  Older age population estimates & service provision. Tuesday 13 September 11:00am

Estimating the population aged 90+ in England and Wales
Johannes Hechler, Angele Storey, Office for National Statistics

As the UK population ages, the very old are making up an increasing share. But just how many people aged 90 or over are there living in England and Wales? Robust population estimates for these oldest ages are essential for service provision, future planning, (in particular health and social care planning) and policy. They are also used in the calculation of life expectancy and are therefore important for life insurance and pensions. The presentation will describe the method currently used by ONS to estimate the very old and some of the inherent challenges. It will outline analysis on the quality of these estimates and present results from one strand of the research, designed to test the robustness of the methodology. Research question: How robust is the methodology used to produce population estimates of the population aged 90 and over in England and Wales? Methods: Survivor ratio methodology: Kannisto-Thatcher (KT) method Data sources: ONS population estimates for England and Wales, ONS deaths data for England and Wales, Deaths data for Finland and Sweden, Population Register data for Finland and Sweden. Results: Assessment of robustness of the KT method by comparison of population estimates obtained for Finland and Sweden by KT method with their respective country’s population register data. Potential applications of the results: Recommendations on whether any improvements could be made to the method in the short-term but also to provide evidence to feed into population estimation for the oldest ages in England and Wales up to 2021 and beyond. 

 johannes.hechler@ons.gsi.gov.uk 

Ageing of the ageing population, causes, consequences and models
Mike Murphy, Moshi Optat Herman, London School of Economics and Political Science 

Population ageing is one of the key issues in contemporary societies in both high income and becoming more so in middle and low income countries. While the older population is increasing both numerically and as a proportion of the total population, ageing is also occurring within the older population itself. The older the age, the faster the rate of increase with, for example, the number of centenarians is expected to increase by a factor of 13 in England & Wales between 2015 and 2050. Data and methods: We use a database of demographic trend and projections data over the period 1841 to 2114 by merging historical data from the Human Mortality Database (HMD) together with the latest ONS 2014 based projections for England and Wales. This provides a consistent 250 year period for assessing long-term ageing patterns. Results: We use the decomposition method of Preston, Himes and Eggers (1989) to separate out the (direct) effects of fertility, mortality and migration on population ageing within the older population. We discuss the appropriate indicator of ageing for such populations (conventional indices of ageing, such as the proportion of population aged 65 and over are obviously inappropriate in this case), and how the results relate to debates about the demographic determinants of population ageing. Use of services increases rapidly with age at these age groups. We therefore consider the impact of ageing within the older population on likely demand for services. 

m.murphy@lse.ac.uk 

Evaluating the effect on mortality of living in a UK retirement village
Les Mayhew, David Smith, Ben Rickayzen, Cass Business School, City University London   

Background: Whiteley Village, in Surrey, was established 100 years ago. It is a charitable retirement community of approximately 500 older people all of whom have limited financial means, which is managed by The Whiteley Homes Trust. Villagers are expected to be to be as independent as possible and to contribute as much as they can to village life. Objective: People from lower socio-economic groups tend to experience higher mortality rates than the general population. However, there is anecdotal evidence that older people who have an active lifestyle tend to enjoy longer and healthier lives. Villagers are subject to both countervailing factors since a condition of admission is that they are of poor or modest means. Our research compares the mortality rates for the Whiteley residents with those of the population of England and Wales to see which factor appears to dominate, and to analyse changes over time. Methodology: The data are from hand written ledgers, and include items such as date of birth, gender, date of entry, and date and method of exit from the village (e.g. death, transfer to a hospice and so on). From these we ascertain the exposure to risk and therefore the mortality rates over the investigation period (100 years). By grouping data by age and period, we investigate the extent to which the mortality rates diverge from those of the general population to show whether village life is beneficial or not. Our paper will report how the study was undertaken, the results obtained, and policy implications. 

lesmayhew@googlemail.com 

Evaluating the value of health insurance as a strategy to move towards universal health coverage in the context of population ageing: Evidence from Ghana
Nele van der Wielen, Jane Falkingham, Amos Channon, University of Southampton 

Universal Health Coverage (UHC) aims to provide quality healthcare to all without financial hardship. Ghana is currently undergoing a profound demographic transition, with large increases in the number of older adults. These adults require greater levels of healthcare due to increasing illness and disability as ageing occurs, which complicates progress to UHC. Ghana presents a crucial case study on the implementation of policies aimed at UHC in the context of population ageing in low and middle-income countries. This paper examines the success of Ghana in implementing their national health insurance scheme (NHIS), designed to help achieve UHC, amongst older adults. Previous studies have mainly focused on enrolment of young and middle aged adults. Using the Ghanaian Living Standards Survey from 2012-2013 this paper aims to estimate the causal effect of NHIS membership on healthcare utilisation for those aged over 50. The paper does this by analysing the usage of out- and in-patient care among NHIS members and non-members, applying propensity score matching. Understanding how health insurance is related to healthcare utilisation amongst older adults in Ghana informs us whether current strategies to achieve UHC are influencing individuals in getting the care that is needed. From a policy perspective, results of this paper indicate that NHIS coverage does indeed increase the usage of healthcare among older adults, but that geographical inequalities remain. 

n.van-der-wielen@soton.ac.uk 

Ageing:  Ageing & the life course: Life history analysis. Tuesday 13 September 1:30pm

The timing of life events and stopping smoking: How do English men and women behave?
Maria Herica La Valle, University of Southampton and ESRC Centre for Population Change    

This work focuses on health behaviours as a cause of health outcomes in late life, and investigates the factors influencing them. In particular, it aims at verifying and identifying the life events that may lead or prevent people to stop smoking and at exploring the relationship between the timing of these life events and the decision to stop smoking at a certain point of the life course. The expectation is that a particular life event, such as cohabitation, marriage, childbearing, abortion, interruption of work, job and residential mobility, kinship ties’ rupture, affects the decision to stop smoking differently on the basis of the time when it happens. By drawing data from the third wave of English Longitudinal Study of Ageing (ELSA), which provides retrospective information on individual health, employment and family and fertility histories, the purpose is to conduct a cohort analysis by a gender perspective. The work will use Survival Analysis approach (with constant time), in order to analyse the risk to stop smoking at a certain age, considering specific life course events that are thought of potentially having an effect on it depending on the age when men and women experience them. 

Herica.LaValle@soton.ac.uk 

Timing, duration and order: The influence of residential histories on later life wellbeing
Bram Vanhoutte 1, Morten Wahrendorf 2, James Nazroo 1, 1University of Manchester,  2 University of Duesseldorf 

The life course perspective holds a great promise for analytical research into the mechanisms that shape people’s lives. The long arm of childhood, and the exposure to risk that accumulative (dis)advantage entails, have received most attention as two interrelated mechanisms that determine later life to a large extent, but are often tested using relatively crude empirical support. This contribution wants to go one step beyond, by highlighting the possibilities of life history data in grasping the importance of timing, duration and order of home ownership over the life course. Home ownership is an important proxy for life course socio-economic status, as it is the most basic form of wealth accumulation in the UK. This contribution makes use of the life history data collected in wave 3 of the English longitudinal study of ageing (ELSA), in a combination of sequence analysis, cluster analysis and regression techniques. Analysis of the residential histories from birth up until the age of 50 illustrates 9 distinct residential pathways, illustrating the importance of accommodating heterogeneity in the population. Preliminary analysis shows that later timing of home ownership and a longer duration of renting accommodation are related to worse later life wellbeing. The specific timing and order of transitions in residential histories trace how period and cohort affect later life. 

bram.vanhoutte@manchester.ac.uk 

Childhood circumstances, adult family life courses and depression among older parents in England
Emily Grundy 1, Sanna Read 1, Heini Vaisanen ,2 1 London School of Economics and Political Science, 2 University of Southampton   

Previous studies have identified life course factors associated with depression including childhood circumstances, education, adult socio-economic resources, marital status and social support, and physical health status. Family life course stressors including early parenthood, short birth intervals, high parity and experience of divorce have also been associated with either depression or physical health status in later life. We use data from waves 3-5 of the English Longitudinal Study of Ageing (ELSA) to investigate direct and indirect associations between these life course factors and depression. Analyses were restricted to parents of two or more children. Logistic regression was used to test direct effects between the hypothesised stressors and depressive symptoms. A range of regression models were then undertaken to test associations between stressors and relevant intermediate variables. Finally we undertook path analysis within the structural equation framework to examine relevant pathways. Missing data were handled using the full information maximum likelihood method. All analyses were undertaken separately for men and women. Results showed that for women associations between family life course stressors (early motherhood, short birth interval and high parity) and depression were mediated through lower labour force participation and wealth, lack of a partner in later life and poor physical function. Adverse childhood events were also directly associated with depression. In men the linkage between family life course events and depression was mediated by experience of divorce. Results indicate the relevance of fertility and family events, as well as early and later socio-economic and health circumstances, to later life depression. 

e.m.grundy@lse.ac.uk 

Assessing the differential association of prospective and retrospective life course variables on health and economic wellbeing in later life
Stephen Jivraj 1 ,  Alissa Goodman 2 ,  George Ploubidis 2 ,  Cesar de Oliveira 1 , 1 UCL Institute of Epidemiology and Health Care, 2  UCL Institute of Education 

This study determines whether comparable prospective and retrospective data present the same direction and magnitude of association between childhood and life course characteristics and later life health and economic wellbeing. We use prospective data taken from the 1958 National Child Development Study at age 50 in 2008 and earlier sweeps at age 7, 11, 16, 23, 33, 42, 46 (n=8,033). Retrospective data is taken from the English Longitudinal Study of Ageing at age 50 from a life history interview in 2007 (n=921). There is a high degree of similarity in the magnitude of regression coefficients of certain childhood circumstances predicting health and economic wellbeing outcomes, suggesting that although there may be issues of recall error in retrospective data, it is not biased for the variables used in the analysis, and leads to similar associations when taking into account contemporaneous and life course characteristics. The findings provide reassurance to the growing literature using life history data to determine life course associations with later life wellbeing outcomes. 

stephen.jivraj@ucl.ac.uk 

Ageing:  Later life health. Tuesday 13 September 4:45pm

Women’s mid-life ageing in low and middle income countries
Tiziana Leone, London School of Economics   

To date there is little evidence on how the health needs and the ageing process of women aged 45-65 in Low and Middle Income Countries (LMICs), in particular when compared to other age groups as well as men. This is particularly significant given that this is the period when menopause occurs and the cumulative effects of multiple births or birth injuries can cause health problems across women’s life course. The aim of the project is to analyse inequalities across socio-economic groups in the ageing process of women between the ages of 45-65 and compare it to that of men within and between countries. Using data from wave 1 of SAGE in 4 countries (Ghana, Mexico, Russia and India) as well as longitudinal data from the Indonesian Longitudinal Family Survey, this study looks at the age pattern of physical and mental decline, looking at objective measures of health such as grip strength, cognitive functions and walking speed as well as chronic diseases. Ultimately the study aims to understand what policy implications there are for the pace of ageing in LMICs. Results show a clear pattern of deterioration of health in the middle-age group, significantly different from men, who show a more linear pattern. This pattern is particularly prominent when we look at physical health rather than mental health. The study highlights the importance to shed more light onto this field, as women have health care needs beyond their reproductive life which are potentially neglected. 

t.leone@lse.ac.uk 

Number of children and later-life depression in Eastern and Western Europe
Emily Grundy, Thijs van den Broek, Katherine Keenan, London School of Economics and Political Science   

We use cross-sectional data from the Gender and Generations Surveys to investigate associations between number of children and depressive symptoms among adults aged 65-80 in five Eastern and three Western European countries. We also investigate whether exchanges of emotional and financial support mediate links between number of children and depression and analyse changes in depression in a subset of countries with longitudinal data. We hypothesised that links between having children and depression might be stronger in the Eastern compared with the Western European countries we considered, because of higher mortality (and so higher prevalence of widowhood in particular) and the reduction of state support for older people following the collapse of the Soviet Union. Preliminary results lend some support to this hypothesis as we found that having no or only one child was associated with higher risks of depression for men in Eastern, but not Western, European countries. However there were no significant differences in associations for women. As in previous studies, we found that long-term illness, low education and difficulties making ends meet were associated with higher chances of depression. Results from the longitudinal analysis suggested that involvement in caring for children was protective against depression among men. 

e.m.grundy@lse.ac.uk 

Marital status and subsequent changes in physical capability in England among men and women aged 60+
Natasha Wood 1,  Anne McMunn 2, Elizabeth Webb 2, Mai Stafford 1, 1 MRC Unit for Lifelong Health and Ageing at University College London, 2  Department of Epidemiology and Public Health, University College London 

Maintaining physical capability at older ages is important given the ageing population. Evidence suggests that married people have better physical capability than their unmarried counterparts, however most evidence is cross-sectional with little research on subsequent changes in physical capability. This paper aims to investigate the association between marital status and changes in physical capability over a ten year period among people aged 60+ in England. Using six waves of the English Longitudinal Study of Ageing, marital status was captured in 2002 and physical capability was measured at six time points using the measure of walking speed, in metres per second (m/s). Growth curve models were run using STATA. Overall married men and women had the slowest decline in walking speed over the 10 year period. After adjusting for demographic, socio-economic and health characteristics widowed and never married men had a walking speed which declined by 0.05 m/s more over the period than men who were in their first marriage, whilst widowed and never married women had a walking speed which declined 0.07 m/s and 0.05 m/s, respectively, more than women in their first marriage. Divorced men also had greater declines in walking speed over the period than married men, which was explained by their demographic and socio-economic characteristics. Findings suggest that marriage may protect long term physical capability, although much of the association remained unexplained and it’s possible that circumstances in earlier life may select and retain men and women in marriage, resulting in a slower decline in physical capability. 

natasha.wood.12@ucl.ac.uk 

Is being in paid work beyond state pension age associated with depression, sleep disturbance, and physical health? Evidence from ELSA.
Giorgio Di Gessa 1, Corna Laurie 1, Loretta Platts 2  Diana Worts 3 Peggy McDonough 3  Debora Price 4 Karen Glaser 1, 1 King’s College London,  2 Stockholm University, Stockholm,3 Dalla Lana School of Public Health, University of Toronto, 4 University of Manchester    

Background: Given the current policy emphasis on longer working lives we investigated the health effects of being in paid work beyond state pension age (SPA). To date, work has largely focused on the health of those who exited the labour force early. Methods: We used nationally representative data from Waves 2-4 of the English Longitudinal Study of Ageing. Using multivariate analysis we investigated the longitudinal associations between being in paid work beyond SPA and three measures of health (depression, sleep disturbance, and physical health) among men aged 65-74 and women aged 60-69 (N~2000). In our analysis we not only controlled for baseline health and socio-economic characteristics but also for work histories, and health in childhood and adulthood. Results: Approximately a quarter of women and 15% of men were in paid work beyond SPA. Descriptive bivariate analyses suggest that both men and women in paid work are more likely to report better health at follow-up, particularly if they held part-time jobs, and worked in managerial and sedentary occupations. However, once baseline socio-economic characteristics as well as adulthood and baseline health were accounted for, the health benefits of working beyond SPA are no longer significant. Conclusions: Potential health benefits of working beyond SPA need to be considered in light of the fact that it is those who report good health and who are more socio-economically advantaged who are more likely to be working beyond SPA. 

giorgio.di_gessa@kcl.ac.uk 

Ageing: Grandparenting. Wednesday 14 September 11:30am

Ethnicity and grandparental childcare in the UK
Valeria Bordone 1, 2, Maria Evandrou 1, 2, Athina Vlachantoni 1, 2,  1 Centre for Research on Ageing, University of Southampton,  2 ESRC Centre for Population Change, University of Southampton   

Grandparental childcare is nowadays benefiting from longer overlapping lives of (healthy) grandparents and their grandchildren. Wide country-differences exist in patterns of grandparental childcare, often due to different needs of parents according to (un)availability of services in the country. Indeed, a European North-South divide has been shown, with Nordic grandparents more likely to engage in childcare and Mediterranean grandparents more likely to engage on a daily basis. Recent studies considering the migrant background of older people showed that grandparental childcare in Europe occurs more often among the majority population than is the case for parent-child dyads of migrant origin. In order to examine the effect of migration history and ethnicity on the receipt of grandparental childcare in the British context, data from Understanding Society allow us to examine the factors associated with grandparental childcare among separate Black and Minority Ethnic groups. The results of a series of logistic regressions on the use of (grandparental) childcare show that individuals from Indian, Pakistani and Bangladeshi communities are less likely to use childcare than White British individuals, while the opposite is true for Caribbean individuals. However, among those using childcare, individuals from the Other White, African and Caribbean groups are less likely than the White British majority to be using grandparental care as a supplemental forms of childcare, or on its own. Such findings indicate important ethnic differentials in the use of childcare per se and grandparental care in particular, which have significant policy implications, and which may mask other kinds of ethnic differences. 

V.Bordone@soton.ac.uk 

Grandparental divorce and care for grandchildren: variations across contexts Zuzana Zilincikova, Martin Kreidl, Masaryk University   

The authors analyze the effect of grandparental divorce on the odds and frequency of grandparental childcare and its variation by gender, divorce rate, and the strength of familialistic values in each country. The analysis is based on three-level hierarchical linear models, using data collected between 2004 and 2011 in 18 European countries as a part of the SHARE project. The results show significant variations in the divorce effect across countries and by gender. The negative effect of divorce is weaker at higher divorce rates, both among men and among women. The authors propose that social institutions related to divorce are more developed with a higher prevalence of divorce and thus mitigate the negative effect of divorce. The negative effect of divorce is also weaker at higher levels of familialism, at least among grandfathers, as stronger familialistic norms might discourage grandfathers from losing contact with their children and grandchildren after breakups. 

333003@mail.muni.cz 

Productive ageing in familialistic welfare regimes: a longitudinal analysis of activity determinants for Italy and South Korea
Ginevra Floridi, London School of Economics and Political Science 

Overwhelming concerns with prolonging working lives as a response to population ageing discount the fact that older people remain economically productive beyond paid work. This is especially true in familialistic welfare regimes, where women’s and older people’s contributions to informal service provision are substantial, and longer working lives may undermine the basis for family support to dependent members.

Broadening the definition of productivity to include work, volunteering, grandchild and adult care, this study investigates the individual and family circumstances determining participation in such activities in old age in Italy and South Korea. These rapidly ageing familialistic welfare states have been chosen because of economic and cultural differences that have led older people to assume family roles as ‘breadwinners’ in Italy and as ‘dependents’ in Korea.

The data consist in four waves each of two highly comparable surveys, the Italian sample of the Survey of Health, Ageing and Retirement in Europe and the Korean Longitudinal Study of Aging. Longitudinal analyses using random-effects models suggest that older people’s productive contributions respond to familial need in both settings, and that they are
importantly shaped by differentiated gender roles. Interesting differences between the two countries are found in the way socioeconomic status shapes productivity, with Italians participating mainly out of opportunity and altruistic exchange, and Koreans participating mainly in response to individual and familial need.

The study fills a gap in the longitudinal study of later-life productivity determinants, and suggests interesting ways in which the relative position of older people shapes participation in familialistic settings. 

g.floridi@lse.ac.uk 

The role of life histories on the probability of being grandparent at different ages across European countries
Bruno Arpino, Jordi Gumà, Albert Julià, Department of Political and Social Sciences and Research and Expertise Centre on Survey Methodology (RECSM), Universitat Pompeu Fabra, Barcelona 

The role of grandparents is increasingly important in ageing European countries. We use a life course approach aiming at assessing the role of individual life histories (in particular, marital and fertility histories) on the probability of being grandparent and on the number of grandchildren people have at different ages. We use retrospective data collected in the third wave of the Survey of Health, Ageing and Retirement in Europe (SHARE), called SHARELIFE, and sequence analyses to cluster individuals according to similar patterns of life histories. Then we use the cluster membership as predictors of being a grandparent, using information from the first two waves of SHARE. We find that family histories have a high explanatory power in explaining the probability of being a grandparent. The highest probabilities are found for both men and women among those who entered the first union relatively early and showed a high completed fertility. We plan to analyze also other grandparenthood-related outcome such as number of grandchildren and age of the youngest grandchild. We also plan to implement more detailed analysis by age and across countries. 

bruno.arpino@upf.edu

Share:Facebook|Twitter|LinkedIn|