Changes in disease life expectancy over time and differences between the sexes in England: an explanation through the contribution of the underlying causes

Domenica Rasulo, Office for National Statistics
Les Mayhew, City University London
Ben Rickayzen, City University London

At present, evaluation of health programs is based principally on their impact on life expectancy since mortality is more easily measured than morbidity. As morbidity and mortality are complementary aspects of a population’s health, a good measure should comprise both. Disease life expectancy, which is the average number of years an individual is expected to live with diseases, is computed by taking into account both mortality and morbidity rates. As a result, it is a good measure to use in investigating changes in a population’s health. Using this idea, the study aims to analyse the health in England in the period 1991-2005 by reference to changes in the disease life expectancy. Information about the occurrence of disease is drawn from the Health Survey for England, which has run on a yearly basis since 1991 and includes people 16 years-old and older. Death counts and populations at risk are supplied by the Office for National Statistics. The disease life expectancy is used to assess changes over time within a sex as well as snapshot differences between the sexes. The investigation of the factors triggering a variation in disease life expectancy is performed through the application of a decomposition method. The method of decomposition is based on Nusselder and Looman (2004) which enables both the causes of death and the causes of morbidity which account for a change in disease life expectancy to be separately identified. The research investigates whether the changes in disease life expectancy over time for each sex are explained by morbidity compression or expansion, and whether the factors accounting for differences between the genders vary between 1991 and 2005.