The Social Environment and Biomarkers of Aging Study (SEBAS) was a joint undertaking between the Taiwan Health Promotion Administration, Ministry of Health and Welfare (formerly the Taiwan Provincial Institute of Family Planning), and Princeton and Georgetown Universities, with notable contributions from others over the life of the project.

The very extensive data that were collected as part of the ongoing Taiwan Longitudinal Study of Aging (TLSA) provided a strong foundation for the underlying research questions: (1) what are the reciprocal relationships among health, the social environment, and exposure to challenge and (2) what can be learned from biomarkers about the pathways and mechanisms through which those relationships operate. The TLSA data – the first round was done in 1989 – comprised self-reported information on demographic characteristics, health and health-related behaviors, occupational and residential histories, participation in social activities, economic and educational status, and emotional and instrumental support (Chang et al. 2008; Taiwan Provincial Institute of Family Planning and Population Studies Center and Institute of Gerontology, University of Michigan 1989). The goal with the SEBAS was to update information from the most recent round of TLSA, to obtain information on exposure to stressors, and to collect measurements and specimens from the TLSA participants that would then be used to obtain biomarkers.

At the time the project began, few psychosocial surveys included the collection of biomarkers; SEBAS might have been the first to do so on a countrywide representative sample (albeit one limited to an older set of participants). As described below, the choice of biomarkers was modeled on the MacArthur Study of Successful Aging (Chang et al. 2008), which used the idea of allostatic load (McEwen and Stellar 1993) to understand how life’s challenges play out at the physiological level.

The initial pretest for SEBAS was performed in 1997–1998 (Weinstein et al. 2003); the first round was fielded in 2000 (Goldman et al. 2003, 2006) and the final round, which as described below collected performance assessments and self-reported information about health status, was completed in 2016.