Using Biomarkers as Indicators of Frailty to Understand Deviation from the Gompertz Mortality Curve at the Oldest Ages

Eileen Crimmins, University of Southern California
Dawn Alley, University of Southern California
Jung Ki Kim, University of Southern California
Arun Karlamangla, University of California, Los Angeles
Teresa E. Seeman, University of California, Los Angeles

Deviation from the Gompertz mortality curve at older ages has been theorized to be the result of population heterogeneity in mortality risk, but little research has examined this issue empirically. Using mortality follow-up data from the National Health and Nutrition Examination Surveys (N=16,697), we examine the population distribution of a set of biomarkers that represent traditional clinical risk factors for poor health outcomes; the summary measure of these biomarkers acts as an index of frailty or biological risk. Results demonstrate that mean levels of biological risk display a leveling at ages over 80 similar to observed declines in mortality. These biomarkers are associated with mortality so that individuals with the highest risk are systematically eliminated at all adult ages, leaving a more robust surviving population in the next age group. Thus, biomarkers of clinical risk provide a good indicator of the heterogeneity of the population for mortality risk.