Health and life expectancies: data introduction
NRS publish life expectancy (LE) and healthy life expectancy (HLE) estimates annually for Scottish areas.
The Office for National Statistics (ONS) also publish life expectancy (LE) and healthy life expectancy (HLE) estimates annually for the UK nations and their local authorities.
NRS, ONS and ScotPHO have collaborated in recent years to align methodologies and to reduce publication duplication.
Both LE and HLE can be estimated for people at various ages, not just at birth. These show that, for example, an average man aged 65 could expect to live a further 17.5 years, 9.9 of them in ‘healthy’ condition. An average 65-year-old woman could expect to live a further 19.8 years, 10.7 of them in ‘healthy’ condition.
When comparing LE or HLE from different sources or that have been produced at different points in time it is helpful to consider the following:
- What data aggregations are used (do estimates use a single year of data or do they cover multiple years).
- Have statistical modelling techniques been applied (estimates for smaller geographies often use statistical modelling techniques to remove effects of random variation, particularly within younger and older age groups).
- For HLE estimates only, what is the source of information on health condition (commonly used sources include census data, Annual Population Survey (APS) or Scottish Health/Household Surveys (SHoS/SHeS)).
In addition to the annual HLE statistics produced by NRS and ONS, every 10 years survey data from the Scotland Census may be used to calculate HLE estimates. These estimates are based on 5 years of data on deaths and populations, and a single year of data for self-assessed health (SAH) from the Scotland Census. As a result, NRS' census-based HLE estimates may differ slightly from their annual HLE estimates once the 2022 census data is published. Census-based estimates have two advantages. Firstly, the survey sample is the whole population and secondly, they can be used to produce estimates for smaller geographical areas. However, this detailed information is only collected once every 10 years, therefore it is not possible to regularly update figures.
LE and HLE are often presented alongside 95% confidence intervals, these indicate the precision in terms of the likely range of random year-to-year statistical variability in the data used to produce the estimates. It should be emphasised that it is not possible to quantify effects such as any inconsistencies in how people interpret and respond to questions about their health, or any unusual circumstances which affect the figures for a particular area or year (e.g. impact of COVID-19 pandemic on survey delivery methodology).
