A study on estimation of subjective probabilities of survival among the elderly in Korea

A Study on Estimation of Srubjective Probabilities of Survival Among the Elderly in Korea

Young-Hwa Lee a) , Hyun-Kyung Moon b), Seung-Wook Lee a)c)
a) School of public Health, Institute of Health and Environment, Seoul National University, Seoul, Korea.
b) Department of Food and Nutrition, Dankook University, Kyunggi-Do, Korea.
c) Correspondent Author

<ABSTRACT>

Purpose :
This study examined what causes the differences between subjective probabilities of survival and 
probabilities of survival on life table among the elderly in Korea. We investigated whether the
factors such as SES, health behaviors and health status, family relationship and social activities
affect these differences.

Method :
Using the representative sample of 3,231 Koreans (aged 65-94) in 2006 Korean Longitudinal Study
of Ageing(KLoSA) data, we estimated linear regression of the differences between subjective
probabilities of survival and objective ones.

Results :
Gender and age affected the differences in probabilities of survival. And SES, assets was
significantly associated with differences but income was not. For education, only over college
degree had significant coefficient compared to lower education. The self-employed assessed his
subjective probability of survival higher than the employed or the unemployed. In health
behaviors and health status, private insurance, regular exercise, IADL and perceived health had
significant estimators. Only former smoker had negative significant coefficients, while drinking had
not. In family relationships and social activities, a number of social activities and marital status
had no significant coefficients, but living with adult children and having parent with same sex
had significant coefficients. The R-square value was 0.3594.

Conclusion :
SES, health and quality of life was a useful measure to assess one's subjective probabilities of
surviving especially in elderly. Given its ease of collection, it would be informative to future
studies in healthy life expectancy.