Multistate Distributions, Lifespan Inequality and Morbidity Compression: Advancing the Debate on Aging and Health

Iñaki Permanyer , Centre d'Estudis Demogràfics
Timothy Riffe, Universidad del País Vasco & Ikerbasque (Basque Foundation for Science)

Background: Longevity increases have been accompanied by a compression in the distribution of ages at which individuals die – thus reducing lifespan inequality and making age-at-death increasingly predictable. Less clear, though, is whether the retreat of mortality towards older ages has been accompanied by commensurate morbidity declines. Assessments of countries’ longevity performance should take into consideration the fact that a non-negligible fraction of individuals’ lives is spent in less-than-good health. Aim: While countless studies have investigated the influence of morbidity on average longevity, virtually nothing is known about the effect of morbidity on lifespan inequality. In this paper, we study (i) what fraction of lifespan inequality is attributable to the years individuals spend in different health states, and (ii) what implications does this have for the formulation and testing of the compression vs expansion of morbidity debate. Data: We use self-reported health measures from the US Health and Retirement Study by sex and in 5-year periods. Methods: We model health and mortality transition probabilities and apply recently proposed multistate life table methods to estimate the number of years individuals have spent in different health states at time at death. Standard decomposition techniques are applied to break down lifespan inequality in components with clear demographic interpretations. Results: The results from this research shed new light to the longstanding compression vs expansion of morbidity debate initiated in the late 1970s. Our novel approach offers complementary insights that could not be gained through traditional approaches relying on average morbidity and mortality measures.

See paper

 Presented in Session 22. Longevity and Health