Three ways of comparing variations and inequalities in life expectancy
Comparing socioeconomic inequality in life expectancy between countries is challenging. It requires studying socioeconomic groups that can be put into harmonized categories across countries. It also requires openly available life expectancy estimates from the same time period for each country being compared.
We compare variation and socioeconomic inequality in life expectancy using three measures that can be calculated with open data (Box 1).
1. Life disparity
We use life disparity (how much lifespan differs between individuals) as a measure of variation in age at death. The measure is easily calculable using openly available country-level data. From life tables, we calculated life disparities at ages 10 years and older for each of 36 countries. A life disparity of 0 years indicates no variation in age at death, with everyone dying at the same age. A life disparity of 5 years means that the average years of life expectancy lost due to death is 5 years. Life disparity estimates are based on current mortality rates and may be poor predictors of future trends.
Although it is not a direct measure of inequality, countries with low levels of life disparity by definition have low inequality in life expectancy, while high life disparity reflects likely high inequality in life expectancy within a country.
2. Gaps in mortality between geographical areas
The second measure we use is variation in the average life expectancies of people living in different geographical areas within a country. This captures socioeconomic inequality better than simple variation across the whole population, since differences in life expectancy between places tend to reflect their characteristics (including available amenities and jobs) and populations (such as education and income levels).
We calculated the absolute and relative gaps in mortality between geographical areas with the highest and lowest mortality (10th and 90th percentile, respectively) in each country. We used population and death data from standardized areas (called NUTS3 (Nomenclature of Territorial Units for Statistics) areas) in 31 European countries. For the most part, these areas have populations between 150,000 and 800,000 people, similar to Warwick and Leeds, respectively. However, some capital cities, including London, contain several NUTS3 areas; others are classified as a single NUTS3 area. Where a city is classified as a single NUTS3 area, variations within it may be masked.
3. Inequalities in life expectancy by education
Our third measure is the gap in average life expectancy between people with different levels of education. This directly measures systematic and potentially avoidable inequalities between socioeconomic groups. We used data from the Organization for Economic Co-operation and Development (OECD) on the absolute gap in life expectancy by education level in 25 European countries, separately for men and women. Education level was categorized using the International Standard Classification of Education 2011. Available data combines these levels into low (no school attended beyond Key Stage 3/Year 9), middle (completed upper secondary school or college, including vocational or technical training) and high (completed a university degree or equivalent). Few other direct measures are available for cross-country comparisons of socioeconomic inequalities. Gaps by area-level deprivation are often reported in the UK as a proxy for individual-level deprivation but cannot be used for international comparisons of inequality because few other countries calculate similar area-level deprivation measures. Average life expectancy according to income or ethnicity would be a useful measure of inequality, but few countries report this.