DESCRIPTION (provided by applicant): This proposal is one of two parallel submissions from separate institutions constituting a single project. The Principal Investigators are: Zachary Zimmer, Population Council, and Mary Beth Ofstedal, University of Michigan. The overall goal of the project is to describe and analyze health transitions and health care utilization patterns in four Asian countries undergoing rapid population aging and social and economic change. The study involves comparative analyses of recent panel surveys of older populations in Taiwan (1989-1999), Indonesia (1993, 1997, 1998), the Philippines (1996, 2000-2001), and Singapore (1995, 1999). The Asian panel surveys represent the first nationally-representative, longitudinal data available on the older population for these countries, which span a continuum of socioeconomic development. The comparative approach allows for estimates of how the potential demand for care and support varies between countries (i.e., prevalence and transitions in health outcomes) and how generalizable the links are between socioeconomic status, social support and health that have been observed in many developed nations in the West. The project has three specific aims: (1) to estimate population prevalence levels of health outcomes (functional limitation, chronic conditions, and self-assessed health status), examine recent trends in prevalence levels within countries, determine individual-level transition rates in health status, and estimate active life expectancy; (2) to investigate the effects of socioeconomic status and family network characteristics and social support on health status transitions among older adults; and (3) to assess the roles of access and need as determinants of health care utilization by examining individual-level and community resource characteristics. A variety of analytic techniques will be employed depending on the form of the dependent variable and the goal of the particular analysis. Logistic regression models will be utilized in analyses of prevalence of health outcomes and changes in prevalence levels, and a combination of logistic and linear regression techniques will be used in analyses of health care utilization. Multinomial logit models will be used for analyses of health transitions; active life expectancy will be examined using multistate life tables for different population subgroups, and structural equation models will be used to test for cross-lagged effects of socioeconomic status and health.