3103.0: Monday, November 13, 2000 - 2:50 PM

Abstract #11355

Education differentials in mortality: Disentangling age, period and cohort effects

Diane S. Lauderdale, PhD, Department of Health Studies, University of Chicago, 5841 S. Maryland Ave. MC2007, Chicago, IL 60637, 773/834-0913, lauderdale@health.bsd.uchicago.edu

Kitagawa and Hauser’s 1960 study matching death certificates to census records revealed mortality differentials by educational attainment which decreased with age for men. Subsequent studies found increased age stratified mortality differentials by education in the 1980s. Age and period comparisons in these studies, because of their relatively short duration, are potentially confounded by cohort effects. To distinguish, age, period and cohort effects, we followed five decade-long U.S. birth cohorts (b.1890-99 ... b.1930-39) of native-born whites through four IPUMS census samples (1960-1990) to track how each cohort’s educational distribution changed as they aged, assumed to reflect differential survival. We assume there is no significant in-migration, out-migration, or change in educational attainment after age 30, and that persons report educational attainment reliably throughout life. For each sex cohort and sequential pair of censuses, we calculate risk ratios of 10-year survival by educational attainment using logit models. Educational attainment varies markedly by sex and cohort. Education effects on survival were found across all age and sex groups for each 10-year period. Period and cohort effects for these education differentials on survival are much stronger than age and sex effects. Within each cohort, educational differentials increase with age. However, this is explained by period effects. Educational differentials also widen in successive birth cohorts; education past grade 12 confers survival advantages beginning with the 1910-19 birth cohort for men and the 1920-29 cohort for women.

Learning Objectives: The participant will be able to 1. describe the association between education and adult mortality in the U.S., and 2. recognize the difficulties in distinguishing age and cohort effects in most study designs

Keywords: Mortality, Education

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA