Intergenerational transfers are a key concern in an ageing society especially when one considers the potential assistance they can offer younger generations in times of high unemployment, high house prices and a high cost of education (Zissimopoulos & Smith, 2010). Yet the decision to help can often be complicated when there is more than one child to consider. Using an altruistic model of transfer behaviour (Berry, 2008; Cox, 1987), this paper seeks to explore how such behaviour varies between families with different numbers of children. To do this data from the Survey for Health, Ageing and Retirement in Europe and the English Longitudinal Study of Ageing are used to construct a multilevel model of transfer behaviour (i.e. payment size). At the first level the unit of analysis is the parent-child dyad which is then grouped into family form reflecting the number of children within the family. This design allows for the change in dynamics to be observed between families with different numbers of children. The results demonstrate that the altruistic utility function is significantly different in nature when comparing families of different size (i.e. number of children). These results suggest that access to family assistance is not just stratified by family income group but also the structure of the family itself. The results also give a new insight into the altruistic model and the extent to which it varies.

Email: temery86@gmail.com

Gender, grandparenthood and psychosocial health in Santiago de Chile
Emily Grundy, LSHTM; Cecilia Albala, University of Chile; Elizabeth Allen, LSHTM; Alan Dangour, LSHTM; Diana Elbourne, LSHTM; Ricardo Uauy, LSHTM

Background: Older people make substantial contributions to the provision of services for family members and care ofgrandchildren is an important part of this. Apart from benefits to children and their parents, it has been suggested that older people benefit from this form of family participation. Aims:1)To quantify grandparenthood in a sample of 66-68 old Chileans by examining the proportions with grandchildren, numbers and ages of grandchildren and help grandparents provide to them. 2)To analyse factors associated with the provision of time and money help to grandchildren. 3)To analyse associations between help to grandchildren and psychological well-being of grandparents two years later. Data and Methods: Data were drawn from a representative sample of 2,000 66-68 year olds resident in low or middle income areas of Santiago who were enrolled in a randomised controlled trial in 2005. Sample members were followed up two years later. Results: Nearly all sample members were grandparents and over half of grandparents provided more than 4 hours per week of help to grandchildren. For grandfathers providing help to grandchildren was associated with better life satisfaction at follow-up – in the case of time help even when baseline life satisfaction was controlled. For grandmothers there were some indications of a reverse association. Conclusions: Grandparenthood is an important role for older people in Chile; implications for psychosocial health vary by gender. Perhaps for men caring for grandchildren may be a fun activity chosen voluntarily, whereas for women it may sometimes represent an addition to household and family work.

Email: emily.grundy@lshtm.ac.uk

Du ut des: do elderly benefit from grandparenting in terms of cognitive abilities?
Bruno Arpino, Bocconi University; Valeria Bordone, Vienna University of Economics & Business

Mental abilities decline as people become older. However, maintaining good cognitive skills is vital for elderly quality of life. Howthis could be done is a topic of high interest in ageing societies. In this paper, we explore to what extent taking care of grandchildren acts as a protective factor, maintaining elderly cognitive abilities, despite the ageing process. Evidence shows that elderly people involved in social activities tend to have better mental well-being because they benefit from being part of supportive social networks and they feel useful. Grandparenting is a common social activity for elderly people, which provides them with a sense of responsibility. The literature on intergenerational transfers has mostly focused on the downward effects of grandparenting, both towards children (especially daughters) and grandchildren. Less studied are the effects of grandparenting on grandparents. This paper aims to contribute filling this gap by focusing on the effect of a change in grandparenting frequency on the level of cognitive abilities for people aged 50+ in Europe. The analyses are based on the two waves of the Survey of Health, Ageing and Retirement in Europe (SHARE). We deal with the difficult task of establishing a causal link between grandparenting and grandparents' cognitive skills by implementing matching techniques. We assess the robustness of the estimates to possible unobserved time-varying confounders by means of sensitivity analyses.

Email: bruno.arpino@unibocconi.it

Work, formal and informal participation as longitudinal predictors of depression and subjective health: A European comparison
Giorgio Di Gessa, LSHTM; Emily Grundy, LSHTM; Anne Jamieson, Birkbeck, University of London

Background The World Health Organisation (WHO) are promoting the concept of "active ageing": it is believed that involvement inwork and activities either formal (e.g. volunteering) or informal (e.g. looking after grandchildren) promotes healthy ageing [1]. Several cross-sectional studies suggest that older people who are active have better health than those less engaged [2]. However, health status may affect participation in various activities and this, rather than beneficial effects of activity, may account for reported associations between activity and health [3]. Objectives This study aims to investigate whether engagement predicts physical health and how any association between engagement and health outcomes is mediated by other factors in four European countries (Denmark, Italy, France and England) selected to represent different welfare regimes. Methods We use data drawn from two parallel European surveys, the Surveys of Health and Retirement in Europe (SHARE) and the English Longitudinal Study of Ageing (ELSA). Measures of engagement (work, formal and informal social involvement), health, personal and familial socio-economic and demographic characteristics were taken at two points in time, two years apart. Cohort participants aged 50-69 at baseline were included. Fully ordinal logistic regression models were examined. Results (in progress) Large differences exist between countries and by demographic, socio-economic and health related characteristics in participation in the activities examined. Participation in paid work, moreover, competes with the engagement in other activities among the populations under study, excepting Italy. Longitudinal results show a complex interplay mediated by socio-economic factors between health and engagement as people age. Conclusions The reciprocal influence of activity and health in ageing populations remains as under-studied. Future analyses are warranted to better understand the potential impact of the 'active ageing' approach to policy.

Email: giorgio.digessa@lshtm.ac.uk

Exploring the Civic Engagement of Older Migrants in Old Immigration Countries in Europe
Albert Sabater, Autonomous University of Barcelona

The civic engagement of older migrants in Europe is an emerging field of academic interest in many European countries, partly as a consequence of the fast growing 50+ population. After the first significant waves of international migration to Europe between the end of WW2 and the economic expansion years (up to 1973), older labour migrants and refugees who became permanent citizens of the 'old' immigration countries in Europe have begun to enter older ages. Whilst recent research suggests that older adults who participate in volunteer activities have higher levels of well-being and life-satisfaction than non-volunteers, recent debates have also suggested that immigrants in general are minimally engaged in their host communities. However, little is known about the experiences of foreign-born elders and the influence of sociodemographics, age at migration and period of arrival on the presence or absence of engagement. The aim of this research is to gain a better understanding on how civic engagement operates on nativity (native and foreign-born), origin, age at arrival, period of arrival and combined age at arrival-period categories. For this purpose we use data collected during the period 2004-2005 from the initial wave of the Survey of Health, Ageing and Retirement (SHARE) in Europe for 6 'old' immigration countries (France, Germany, the Netherlands, Austria, Switzerland and Belgium).

Email: asabater@ced.uab.es

Social isolation and cognition in old age
Valeria Bordone, Vienna University of Economics & Business

As people age, social isolation and loneliness will increase with smaller family networks and fewer friends. Social relationshipsare known to be an important form of support and to affect health outcomes especially for elderly people. While it has been already largely shown the effect of social networks (and in particular marriage) on health and healthy behaviours, this paper provides a further understanding of the role of the partner in old age by looking at the causal effect of widowhood on cognitive abilities over time. These abilities are not only important to maintain the body fit, but also to age actively. Widowhood is expected to negatively affect social cognition. A likely counterbalancing effect of other forms of social integration (i.e. meeting with adult children or with friends, social participation) is also explored. This study focuses on the age group above 65 years old. I apply fixed-effects methods on the data from the English Longitudinal Study of Ageing (ELSA), which allows following individuals over four waves, between 2002 and 2008. Preliminary results suggest a general decline of cognitive skills over time, after the age of 65. However, such a decline is steeper for people experiencing widowhood rather than for their married counterparts. Not only a clear effect of widowed on cognitive abilities is evident in the year following marital disruption, but a significant decline in mental cognition seems to accompany the death of the partner, most likely because of a difficult period before the event itself.

Email: valeria.bordone@wu.ac.at

The impact of retirement on cognitive function and wellbeing: who benefits?
James Nazroo; Amanda Connolly; Nitin Purandare; University of Manchester

Introduction There are inevitable concerns over the impact on health, economic and social systems of growing numbers of olderpeople; and a desire to maximise the opportunities afforded by healthy, engaged older people – so-called third agers. A broad approach to ageing, however, ignores the significance of marked socioeconomic inequalities at older ages, which extend to cultural, social and civic activities. Of importance here are the uneven, and unequal, routes that are taken into retirement and the implications of this for a rewarding post-retirement life. In this study we have set out to understand the impact of retirement on cognitive function and wellbeing, and how this relates to socioeconomic position, route through retirement, and post-retirement activity. Methods The study uses three waves of the multidisciplinary English Longitudinal Study of Ageing. We observe the retirement of 600 respondents aged 56 to 70 over a four year period and contrast changes in cognitive function and wellbeing for this group with 670 respondents in the same age group who remained working over this period. Results Route into retirement (routine, voluntary and involuntary) is an important determinant of outcomes, as are socioeconomic position and participation in social, cultural and civic activities post-retirement. We show how these factors operate jointly to impact on post-retirement outcomes. Conclusion Findings to illustrate how social inequalities play out over the retirement process and discuss the need to consider inequalities within the older population, rather than considering outcomes to be a simple consequence of age-related transitions, such as retirement.

Email: james.nazroo@manchester.ac.uk

Understanding the dynamics of living arrangements for older people in a slum setting in Nairobi.
Jennifer Baird, University of Southampton

There has been little investigation into the living arrangements of older people in slum environments in sub-Saharan Africa. Inthis region, there is an assumption that older people live in the same household as their family with care and support for older people being closely interlinked with this coresidence with kin. Thus the living arrangements of older people are important indicators of their welfare and have the potential to highlight existing vulnerability among this group, with the decision to live alone having potentially adverse consequences for an older person. This research aims to explore this further by looking at the changes in living arrangements of older people in this environment over time and the events connected to this. The study settings are two slums in Nairobi which are demographic surveillance systems and have yielded panel data on the household types of older people from 2002 through to 2006. The research uses event history analysis to explore the transitions from one type of household to another over this period, with particular emphasis on the circumstances which lead to an older person moving from a multiple person household to a single person household. The research aims not just to understand the dynamics of living arrangements for older people in slum settings but to establish whether these moves can be viewed as positive or negative and if the latter is the case, whether there is need for a strengthening of both formal and informal types of support for older people.

Email: j.s.baird@soton.ac.uk

Developing health and welfare policies for the elderly in Ghana
Gill Miller, University of Chester

This paper discusses the issues facing Ghana in addressing the health and welfare needs of its ageing population. In developing African economies such as Ghana, there is evidence that traditional inter-generational support and care for the older generation is breaking down. Elderly people in the informal sector have to continue working because they have low incomes and no pension provision. Consequently there is an increase in the vulnerability of elderly people to chronic poverty and poor health because they find it more difficult to access health facilities and cannot afford health treatments. Despite broad recognition by a range of stakeholders of persistent poverty and neglect policies which improve personal health and health provision for the aged., Ghana's second formulation of a National Ageing Policy has yet to be ratified. The national health infrastructure is supported by the private sector and NGOs but few, such as HelpAge Ghana, are focused on the health and welfare of the elderly. In Ghana some of the aged benefit from recent government initiatives such as the National Health Insurance Scheme and National Social Protection Policy, but a comprehensive and coordinated policy for elderly health and welfare is lacking. There is significant funding support for health programmes from international health donors but the international health agenda appears not to prioritise health care of the elderly. The result is very limited progress towards reducing the health vulnerabilities of the increasingly aged population. A further barrier to improvements is the paucity of coordinated evidence to inform policy and planning. Although a number of databases capture aspects of elderly health and welfare, their disparate nature limits the formulation and implementation of appropriate policies to meet the growing needs of Ghanaian elderly.

Email: g.miller@chester.ac.uk

Perception of vulnerability to HIV infection among older people living in the slums of Nairobi, Kenya
Gloria Chepngeno-Langat, University of Southampton

Adolescents, people in the reproductive ages, and high risk groups have been the focus in HIV or AIDS discourse and policyinterventions. Older people have only featured in their role as carers for orphans and people living with AIDS and hardly as people at risk of HIV infection. It is evident that sexual activity tends to decrease with age nonetheless it is still prevalent enough to be considered a risk factor for the spread of HIV among older people. This study therefore examines the factors associated with self-perceived risk to HIV infection among older people. Survey data for 2,053 individuals aged 50 years an older drawn from a larger longitudinal study conducted in two slum areas of Nairobi, Kenya are used. A large majority of older people do not consider themselves at risk of infection. Of those who felt at risk, a greater proportion sensed only a small chance of contracting HIV/AIDS. Reasons for perceiving no risk and a small chance of contracting HIV are gendered but not for perceiving moderate to great risk. Perception of risk was associated with slum, participation in social groups or voluntary community activities, religiosity, and HIV testing. Surprisingly, age, marital status and education did not independently predict risk perception. This paper recommends evaluation of older people's self perception of vulnerability in order to best inform interventions aimed at minimising risks to HIV infection.

Email: tete@soton.ac.uk

The Pay-as-you-go vs Funded pension system in low-mortality countries: a demographic perspective
Dalkhat M. Ediev, Vienna Institute of Demography

Comparisons of the alternative pension systems, so far, were based on simplified demographic models such as Samuelson'stwo-age overlapping-generations model. At best, the systems were contrasted using a stable population model. Here, we study pension schemes in a more realistic population with changing longevity and show dramatic advantage to the PAYG system. Increasing longevity benefits the PAYG scheme as compared to the FS because current workers will live longer at retirement than current retirees and, therefore, find it easier to fund current pensions rather than saving for their own future benefits. The paper presents methodology and empirical findings. The demographic advantage to the PAYG system as compared to the FS mounts currently to 50% in terms of the benefit/contribution ratio or to 1% p.a. in terms of the real rate of return in low-mortality countries. In terms of age at retirement, a fully funded pension system would currently require people to retire about five years later as compared to a PAYG system because of the differential mortality effects.

Email: dalkhat.ediev@oeaw.ac.at

Rethinking ageing
Sergei Scherbov, Wittgenstein Centre for Demography and Global Human Capital (IIASA, VID); Warren C. Sanderson, SUNY, IIASA

Population aging continues to be a matter of significant international concern, in part because of the burdens that the elderly areexpected to impose on others. This expectation is based on conventional measures of aging that rely on fixed chronological ages. As life expectancies increase and people remain healthy longer, policy-relevant age-specific characteristics of people change. Because of this, measures based solely on fixed chronological ages can be misleading. There are now ways to measure the extent of aging that take improvements in health and longevity into account. In this paper we apply two new measures of aging introduced in our previous studies. We show, using historical data and forecasts that although populations will be growing older in many countries as measured by their median ages and proportions of population above age 65, they will probably experience much slower aging and in some cases even grow younger, as measured by alternative measures of aging that take into account the changes in life expectancies. Population aging will certainly provide challenges, but there is no reason to exaggerate those challenges through mismeasurement.

Email: sergei.serbov@oeaw.ac.at

Ageing of the indigenous populations of Australia, Canada, and New Zealand
Helen Ware, University of New England

Until very recently, most of the emphasis in the study of the demography of indigenous groups in the former dominions has beenupon the low levels of life expectancy at birth and the gaps between indigenous and non-indigenous expectations. In contrast, this paper looks at trends in indigenous ageing and more specifically the experience of the growing proportions of members of the indigenous populations of Australia, Canada and New Zealand who have already survived to the age of 50. A particular focus is upon the chronic disease patterns which are distinctive features of the lives of these groups as well as the risk factors with which these diseases are associated. The impact of aspects of indigenous socio-economic disadvantage upon the lives of the elders is also explored through a range of indices. The conclusion briefly addresses the issue of the provision of culturally appropriate services.

Email: hware@une.edu.au

Planning for a longer life: extending our horizons
Sandra Vegeris, Policy Studies Institute; David Martin, Independent Consultant

Broadened longevity and the tendency to retire earlier have lengthened the 'retirement years'. In response, there has been muchplanning of state resources to accommodate the anticipated demand for public services in later life. Much of this attention is on health, pensions and employment policy interventions. There has been relatively less attention given to day-to-day living in later life. This paper explores the need for a more holistic approach to planning for later life – both at the state and individual levels. How might we maximize the opportunities of this new old age for individuals and society? There seems to be little research or interventions on planning and preparing for other aspects of what could be a quarter to a third of a person's years of life. Drawing from a cross-national literature base, this paper argues that the cessation of paid work and the take-up of certain retirement roles at a prescribed age is a result of social conditioning and less to do with individual wishes or empirical facts about capabilities. Rather, to recognise the significance of this extended period of life, interventions are needed that enable planning for later life which embrace the concept of a more informed and goal oriented horizon.

Email: s.vegeris@psi.org.uk

Transitions into residential care in later life: evidence from the British Household Panel Survey, 1991-2009
Maria Evandrou, Jane Falkingham, Olga Maslovskaya, Athina Vlachantoni, University of Southampton

The UK's population is ageing. In 2001, people aged 65 and over constituted 16% of the total UK population; by 2031 this figureis projected to rise to 22% (ONS 2008). The majority of social care provided in later life is by family, particularly co-residential kin. The dynamics of living arrangements in later life are a crucial determinant of the level and nature of informal support an older person can expect to receive in the future. Living arrangements also affect the probability of receiving formal support, with those living alone being more likely to receive services, and a change in living arrangements may be associated with a subsequent move into residential care. This paper is part of a broader EPSRC-funded project on the Care Life Cycle, which explores the demand for and supply of social care in the UK. The paper aims to shed light on the dynamics of living arrangements in later life in the UK, with a particular focus on the patterns and determinants of transitions into residential care. This research extends and updates previous analysis by Evandrou et al. (2001) and uses 18 waves of the British Household Panel Survey (BHPS) (1991-2009). Multi-variate analysis investigates changes in living arrangements among people aged 65 and over, and explores how events such as widowhood and episodes of ill health are related to the probability of entering residential care. The results will highlight areas for further research and will inform the policy debate regarding long term care in the UK.

Email: o.maslovskaya@soton.ac.uk

Housing an ageing population with social care needs: investigating the role of extra care housing
Dylan Kneale, International Longevity Centre

Housing an ageing population with care needs is becoming an ever pressing concern for policy makers. Residential care is often viewed as a housing solution for older people with mild to moderate activity limitation, although residential care is often unattractive to residents, families and policy makers, being associated with depersonalisation and high cost. The extra care model may represent a partial solution to this dilemma for policy-makers, through keeping older people with mild to moderate activity limitation independent and away from more formal institutional care. Extra Care housing is described as ergonomically designed independent housing units that feature common spaces, facilities and care services. Residence in extra care housing is also thought to be associated with better health outcomes and lower dependence on health services such as hospital admissions. However, there is a dearth of evidence the examines these issues, including research that quantifies how long older people stay resident and the added value of extra care housing. Using longitudinal data from residents of three extra care housing providers, in this paper we present results that aim to establish baseline information on the characteristics of extra care residents, how these may differ from the general population, as well as presenting results from event history models on the outcomes of extra care residents. We also present the results from matching exercises that aim to establish whether or not extra care residents do enjoy better outcomes, and discuss some of the challenges in doing so.

Email: dylankneale@ilcuk.org.uk