Education and Age, Period, and Cohort Patterns of U.S. Adult Mortality: Heart Disease, Lung Cancer, and Respiratory Diseases

Ryan K. Masters, University of Texas at Austin
Robert A. Hummer, University of Texas at Austin
Thomas W. Pullum, University of Texas at Austin

Adult mortality risk in the United States has decreased significantly over the past twenty years. While the continued decreases are to be celebrated, the forces behind these decreases are not entirely understood. Scant research has attempted to disentangle cohort effects from period effects, and most explanations of the mortality reductions simply allude to lifestyle changes, medical and science breakthroughs, and/or increased education. Further still, despite the overall reduction in mortality risk, educational differences in U.S. mortality have persisted, or even widened. The persistence and/or widening of these gaps suggest that the recent mortality decreases have varied considerably by educational attainment. In this paper, we use the 1986-2002 NHIS-LMF data to analyze the variations in the age, period, and cohort patterns across both sex and educational level over the past 20 years.