Page 86 - IJPS-3-1

Modeling trajectories of long-term care needs at old age: A Japanese-Swedish comparison

in coming years if they are to provide care at the present level. However, the projected
cost trends are much steeper in Japan than in Sweden.
This long-term cost analysis is very important from a governmental perspective.
However, the longitudinal survey data also allows a look at the future from an
individual perspective. By looking at longitudinal survey data from a certain age, we
can explore life chances after that age over time in terms of mortality, ill health and
dependency trends, and use of LTC. By correlating these life chances to initial health
conditions and access to LTC, we can identify the positive health and needs factors and
thus acquire a scientific basis for health promotion and ill-health prevention. For both
Japan and Sweden there exist reliable, nationally representative population surveys
providing data that can be used for this type of analysis. In both cases, the analysis
involves demography, ill health and needs trends, and the provision of services. The
idea is to compare synthesized old-age trajectories in Japan and Sweden using similar
simulation models. This also allows us to explore the effect of alternative scenarios in
terms of dependency development and LTC provision.
The exploration of individual life chances in this way is not common, but there
are some studies with similar aims. The popular death-calculator approach estimates
remaining life years based upon answers to an array of questions on life circumstances
1
that have been identified in studies as influencing mortality and lifespan More serious
are a host of studies looking at mortality or life expectancy for different subgroups—
often with a clinical aim. Keeler et al. studied the impact of functional status on life
expectancy in older persons. Among other things, they found that the life expectancy
of an ADL-disabled 75-year-old is similar to that of an 85-year-old independent person
(Keeler et al., 2010). This relationship of mortality to ADL limitations has also been
studied by Stineman et al., who divided participants into five stages of performing
activities of daily living (ADL) (0, I, II, III, and IV) and found that the risk of dying
was five times greater at stage IV than at stage 0. Some authors have developed indices
intended to predict mortality with a clinical perspective, often—but not always—
limited to frail persons (Carey et al., 2008; Klein et al., 2005; Zhang et al., 2012).
Chan, Zimmer, and Saito (2011) and Chan et al. (2016) studied gender, educational,
and ethnic differences in active life expectancy in Singapore and concluded that, unlike
Western nations, there was no gender difference.
Some studies have looked into individual prospects in LTC; for example, Kemper
et al. used microsimulation to estimate the amount of time a 65-year-old could expect
to need LTC (three years on average) and what kind of private expenditure that would
involve (Kemper, Komisar, and Alecxih, 2005). Ernsth Bravell et al. investigated how
health, ADL, and use of LTC affected survival among very old people. They concluded
that, in Sweden, the use of formal LTC increased with age and that, once the oldest
people started to receive LTC, they seldom returned to living without it. In a Cox
regression, health and ADL-dependency significantly predicted survival but not age as
such (ErnsthBravell, Berg, and Malmberg, 2008).
Other studies have calculated transition rates for level of dependency over shorter
or longer periods of time. A pioneering study was made by Manton: using the U.S.
National Long Term Care Surveys of 1982 and 1984, he observed that a significant
number of persons showed improvements even at a high level of impairment. (Manton,
1988). Transition rates and rates of institutionalization were also calculated by
Branch and Ku using Massachusetts Health Care Panel Study data. According to that
study, the best predictors of ADL status were initial ADL status, hospitalization, and
institutionalization (Branch and Ku 1989).
Béland and Zunzunegui later calculated two-year transition probabilities for
functional status (functional limitations, activities of daily living (ADL), and
instrumental activities of daily living (IADL)) by age group and gender. Like Manton,
they found that some improved functionally—especially the younger old people—
but among the older old people, deterioration was more common (Béland and

1. www.death-clock.org
80 International Journal of Population Studies 2017, Volume 3, Issue 1