The indicator is calculated following the Sullivan method (125KB) which is widely used by experts across the world since the 1970’s. It is based on prevalence measures of the age specific proportion of population with and without disabilities and on mortality data. Its interest lies in its simplicity, the availability of its basic data and its independence of the size and age structure of the population. The health status of a population is inherently difficult to measure because it is often defined differently among individuals, populations, cultures, and even across time periods. The demographic measure of life expectancy has often been used as a measure of a nation's health status because it is defined by a single vital characteristic of individuals and populations -- death. However, the measure of life expectancy has limited utility as a gauge of a population's health status because it does not provide an estimate of how healthy people are during their lifespan.

The idea for the development of a measure of "health expectancy" (the partitioning of the demographic measure of life expectancy into healthy and non-healthy years of life) originated with a report published in 1969 by the U.S. Department of Health, Education and Welfare. The report noted that both good health and long life are fundamental objectives of human activity, but that despite the substantial rise in life expectancy in the 20th century the overall health status of the population was unknown. In fact, it was recognized that accompanying the rise in life expectancy was the emergence of chronic diseases -- thus raising concerns about the future health status of the population if death rates continued to decline.

The two components of the calculation of the HLY in the EU are the mortality tables and the self-perceived disability assessed by health surveys. Life tables which give mortality data for calculating life expectancy are fully available as a demographic long-term series based in the standard procedures of causes of death registration harmonised at EU level. As regards self-perceived disability, from 1995 to 2001, data from the Eurostat European Community Household Panel (ECHP) survey have been used for the EU-15 Member States. The successor of the ECHP, the Eurostat EU-Statistics on Income and Living Conditions survey (EU-SILC) is being launched in various countries at different times in 2004 and 2005. During the transition between end-ECHP and start EU-SILC, for the EU-15 Member States, data will be calculated by extrapolating the data on the prevalence of disability from 1995 to 2001. As disability is a phenomenon which changes slowly in time, in the calculations for 2002 and 2003 an assumption was made that the evolution of the prevalence is linear. For the new Member States national sources are used when comparable (CZ, HU, CY, MT, and PL). From 2004/2005 onwards, data from the EU-SILC survey will be used for EU-25.