Will Future Elderly Have Any Relative Available to Care for Them? a Study Based on a Mixed Micro-Simulation/ABM Model

Jeroen Spijker, Autonomous University of Barcelona (UAB)
Daniel Devolder, Centre of Demographic Studies (Barcelona)
Pilar Zueras, Centre d''Estudis Demogràfics (CED)
Nuria Calduch Verdiell, Centre d''Estudis Demogràfics (CED)

In this paper we introduce a mixed micro-simulation AMB model called DEMOCARE designed to model care supply-demand dynamics for older dependent persons in function of demographic change. The model simulates the lifecycle of a cohort of individuals (called EGOs) and of their close relatives until their death. Each year they are at risk of falling into different states of dependency and as a result require a certain number of hours of care. The model explores the EGO’s family network to ascertain whether there are relatives who can fulfil this demand. Specifically, DEMOCARE allows more detailed examination of the effects of low birth rates, increasing life expectancy and changing family structures on old-age care supply for Spanish cohorts born in 1928, 1938, 1948, 1958 and 1968, distinguishing between formal (professional) and informal (performed by family members) care. The modelling of these social changes and their use in our model will allow constructing possible future scenarios and will also permit going through what-if questions. For example, what if there was an improvement of 2%, 5% and 10% in disability prevalence (across all levels of dependency)?; or what if there was an improvement only in the prevalence of mild and moderate dependency but prevalence of severe dependency remained constant? This should have implications on demand for formal care as, according to previous studies, informal care substitutes formal care if disability of the elderly is not severe. Likewise, a transition of a parent into a dependency level could affect their children’s labour trajectory. Hence, we also consider the probability of an agent to maintain or reduce their time in paid work when one of her/his parents falls into a dependency state.

Rapid population ageing has raised concernfor policy makers on how the future provision of elderly care can be met byboth informal care (unpaid and mainly family-based) and formal care (home-basedprofessional care and institutions). In Spain, informal care is the most commonsource of support received by older people1, although it is often combinedwith formal care 2. However, the supply of informal caregivers may decline in future dueto changes in, for instance, the population’s age composition, family size, the rising incidence of childlessness and female employment.And any reduction in informal care could have a substantial influence on thedemand for formal care. A better understanding of formal and informal caredynamics is, therefore, clearly an issue of great importance.

One tool that has been used to examine thisis agent-based modelling (ABM). Although the expanding interest in ABM in socialsciences is relatively recent, in demography it has been applied in spatialdemography, transition to marriage and parenthood, fertility policies,migration, health and social care as it is a powerful technique to understand howthe dynamics of biological, social, and other complex systems arise from thecharacteristics and behaviors of the agents making up these systems. In thispaper we introduce a mixed micro-simulation AMB model called DEMOCARE designedto model care supply-demand dynamics for older dependent persons in function ofdemographic change. DEMOCARE is implemented with the software package NetLogo6, a Logo language with extensions intended for ABM programming3. Figure 1 provides an exampleof the interface.

DEMOCARE is cohort based. Each model runstarts with a cohort of 10,000 agents called EGO, born in 1928, 1938, 1948,1958 and 1968, and whose life is simulated until death from age 50. Wereconstruct the immediate family of the initial population of EGOs using amicrosimulation kinship model4. Based on nuptiality, fertilityand mortality average levels of past decades (Table 1), each EGOs has, at theirrespective ages, a certain probability of having a living spouse, and if it thecase, the corresponding probabilities of having living children, sons- ordaughters-in-law and grandchildren. For each individual, the micro-simulationmodel determines the age relative to that of the EGO, the age at union,childbirth and death to determine whether and how many relatives are alive andthe potential caregivers for EGO. Each year they are at risk of falling intodifferent states of dependency and as a result require a certain number ofhours of care. The model explores the EGO’s family network to ascertain whetherthere are relatives who can fulfil this demand. The 2008 Survey on Disability,Personal Autonomy and Dependency Situations (EDAD08) is the main data sourceused to obtain the EGO’s level of disability, their educational and employmentstatus profiles and the hours of care they require and potential carers canprovide. All individuals have a specific educational level assigned to them,according to distributions observed in EDAD08. The constraints introduced by the labour forceparticipation of our agents (full-time, part-time and not in paid work) ontheir availability as carers have also been modelled, based on the survey dataand transitions into and out of either full- or part-time paid work are alsocalculated.

In the interplay of these four domains(demographic change, education, disability and employment statuses), DEMOCARE thensimulates the demand (in hours) of elderly care and the need of formal careafter considering the supply of informal care by using children characteristics.The distribution of care among family members follows basic rules thatestablish priorities in a three-generational context. The modelling of these demographicchanges and their use in our model will allow constructing possible future scenariosand will also permit going through what-if questions, such as what if there wasan improvement of 2%, 5% and 10% in disability prevalence (across all levels ofdependency)?; or only an improvement in the prevalence of mild and moderatedependency but prevalence of severe dependency remained constant? This shouldhave implications on demand for formal care as, according to previous studies,informal care substitutes formal care if disability of the elderly is notsevere.

1.  Spijker J, Zueras P. Care for the elderly in a context of aging,social, political and economic change. Panorama Social. 2016;23:167-82.

2.  Rogero-García et al. Health services use by older people withdisabilities in Spain: do formal and informal care matter? Ageing &Society. 2008;28(7):959-78.

3.  Calduch N et al. Application of the Overview, Design conceptsand Details (ODD) Protocol to describe the DEMOCARE Agent Based Model. Papersde Demografia. 2017;456.

4.  Devolder D. Effects of the European late marriage pattern onkinship. A study using a microsimulation model. In: Derosas R, Oris M, editors.When dad died Individuals and Families Coping with Family Stress in PastSocieties. Bern: Peter Lang; 2002. p. 325-50.

 

Presented in Session 1127: Ageing and Intergenerational Relations