Deriving and Comparing Healthy Longevity Distributions by Gender and Health Prevalence Measures: A Statistical Moments Approach

Rami Cosulich, University of Padova
Virginia Zarulli, Department of Statistical Sciences, University of Padova, Italy

This study explores the distributions of healthy longevity, focusing on differences between men and women and across various measures of health prevalence. Using a Markov chain model with rewards and data from SHARE and Human Mortality Database, we estimate distribution moments—mean, variance, and skewness—based on prevalence of limitations in activities of daily living (ADL), self-rated health (SRH), and cognitive health (COGN). We then derive the empirical density function of health loss using the maximum entropy method. Preliminary results reveal that healthy life expectancy is increasing across all health measures, but self-rated health remains shorter in duration compared to ADL and COGN. Standard deviations show minor changes, with slight reductions at age 50, except for female cognitive health. Skewness is becoming more negative, particularly in cognitive health, where the most likely age of health loss has shifted to over 90 years. Women have narrower distributions than men and generally higher risks than loosing their health than men. This is particularly true when health is defined with self reported indicators. To further compare these distributions, we will employ visual and statistical methods. Our findings shed light on gender-specific differences and the evolving shape of healthy longevity distributions across health metrics.

Keywords: Health and Morbidity, Gender Dynamics, Inequality, Disadvantage and Discrimination

See extended abstract.