Page 44 - IJPS-1-1

Concordance and discordance in successful aging measures and mortality

also for those who are not-OSA or not-SSA. Our findings are novel in that older adults
who are not-OSA or not-SSA constitute meaningful subtypes in distinguishing subsequent
mortality risk that previous studies have inadequately examined. We argue that the com-
bined use of both types of objectively measured successful aging and subjective (self-rated)
successful aging is preferable, which echoes the recent calls on the understanding of suc-
cessful aging by some researchers (Rowe and Kahn, 2015).
Successful aging, especially SSA, is an adaptive process that results in older adults
maintaining a perception of well-being and life satisfaction (Romo, Wallhagen, Yourman
et al., 2013); our approach of exploratory classification of successful aging and findings
on the associations between subtypes of successful aging and subsequent mortality may be
useful for care and service provisions in both clinical settings and community service pro-
grams, policy-making, and the individual's better adaptation. For example, the joint classi-
fication of OSA and SSA may benefit clinicians by encouraging a patient-centered defini-
tion of successful aging; clinicians could develop intervention programs or treatment plans
that help older adults to develop positive adaptations (Phelan, Anderson, Lacroix et al.,
2004). Integration of older adults’ subjective rating of successful aging may improve the
ability of health service providers to identify appropriate types of interventions and per-
son-centered service programs to improve the experience of successful aging. From a po-
licymaking perspective, although prevention of functional/capacity decline and interven-
tions are still important to achieve successful aging, more resources should be invested in
understanding and supporting those who live in poor objectively measured health to help
them adapt and maintain a positive feeling of satisfaction. From older adults' perspective,
the joint classification enables them to know their own state condition in terms of OSA
and SSA so that they can make better adaptations by themselves to maintain a subjective
feeling of satisfaction, to optimize functional domains that are appropriate and important
to them, and thus age successfully (Baltes and Baltes, 1990).
The strength of the present study is the joint classification of OSA and SSA into differ-
ent subtypes, that is, concordance and discordance of OSA and SSA based on a large na-
tionwide sample. Previous efforts at characterizing specific successful aging subtypes are
mainly limited by sample sizes and cross-sectional nature (Cernin, Lysack, and Lichten-
berg, 2011; Pruchno, Wilson-Genderson, Rose et al., 2010), which prevent researchers
from fully distinguishing different successful aging subtypes by age and sex, and examin-
ing the associations between these subtypes and mortality. With the world's largest sample
of very old adults, we are fortunate to be able to conduct such research and provide empir-
ical results for the first time. Furthermore, we have extended previous work (Cernin, Ly-
sack, and Lichtenberg, 2011; Pruchno, Wilson-Genderson, Rose et al., 2010) by using a
longitudinal dataset to examine associations between successful aging subtypes and sub-
sequent mortality in several different age groups of older adults for both men and women.
However, whether our findings about the associations between concordance and discor-
dance of OSA and SSA and subsequent mortality still hold or whether these findings are
robust across different populations deserves further examinations. As individuals with dif-
ferent health conditions may value components of successful aging differently and the
subjective ratings of successful aging can also vary by culture, further studies including
self-rated priorities of different domains of successful aging as well as differences in prior-
ities in dissimilar cultural settings are clearly needed to shed some light on the psycholog-
ical process of subjectively measured successful aging of respondents (Carr, Gibson, and
Robinson, 2001; Feng, Son, and Zeng, 2015; Phelan, Anderson, Lacroix et al., 2004; von
Faber, Bootsma-van der Wiel, van Exel et al., 2001).
The following limitations should be taken into account when interpreting our results.
Firstly, due to unavailable information on subjectively defined domains of successful ag-

International Journal of Population Studies | 2015, Volume 1, Issue 1 38