1. Introduction

The Office for National Statistics (ONS) produce estimates of the age 90 and over population by single year of age and sex up to age 105 and over for England and for Wales. Corresponding estimates for Scotland are produced by the National Records of Scotland (NRS) and for Northern Ireland by the Northern Ireland Statistics and Research Agency (NISRA).

ONS carry out additional quality assurance on the estimates produced by NRS and NISRA. ONS also compile the 90 and over estimates produced by the three statistical offices to produce a set of estimates for publication at UK level.

Back to table of contents

2. Background

The mid-year population estimates published by the Office for National Statistics (ONS) include estimates by single year of age to 89, with a final category for ages 90 and over.

Prior to 2007, population estimates for single years of age beyond 90 were calculated for England and Wales (as a whole), Scotland and Northern Ireland by the Government Actuary’s Department for use in compiling national interim life tables and in producing the national population projections. The estimates were made available for research purposes but were not officially published. ONS took over the production of these estimates in 2007.

In recognition of increased interest in population estimates at the oldest ages ONS began to publish 90 and over estimates by single year of age in 2007 as experimental statistics. NRS took over the production of the estimates for Scotland in 2008 and NISRA took over the production of these estimates for Northern Ireland in 2010. The estimates have been published as National Statistics since 2011.

Back to table of contents

3. Release titles

The ONS 90 and over population estimates release is titled 'Estimates of the Very Old (including Centenarians)’.The NRS equivalent release is titled 'Centenarians in Scotland, 2003 to 2013, including mid-year population estimates for those aged 90 and over’. In Northern Ireland the release is titled 'Estimates of the Population aged 85 and over' and includes estimates for those aged 85-89 taken from the annual mid-year population estimates.

Back to table of contents

4. Methodology

ONS, NRS and NISRA all use a modified form of the survivor-ratio methodology proposed by Kannisto-Thatcher1 (the KT method) to produce single year of age population estimates for ages 90 and over. In this method, population estimates are produced from death registration data. ONS, NRA and NISRA all use death occurrence data2. The population at a given age is estimated by looking at the ratio of the number of survivors of a cohort still alive to the number of that cohort who died in the last few years. By making an assumption about the highest age at which everyone in a given cohort will have died, it is then possible to produce an algorithm using these survival ratios that will give estimates of the numbers of people alive at earlier ages for each cohort. That is, the KT method uses “age-at-death” data to build up distribution profiles of the numbers of elderly people in previous years. For example, if someone dies in 2012 aged 105, then this means that they were alive and aged 104 in 2011 and 103 in 2010, etc. By collating “age-at-death” data for a series of years, it becomes possible to make an estimate of the number of people of a given age alive in any particular year and so create age distribution profiles, assuming that migration at these oldest ages is minimal. To make estimates for the current year, it is not possible to use death data, as we are interested in the population who are currently or very recently alive. So the KT method uses an average of the last 5 years of age-at-death information to produce an estimate of the number of survivors for the most current year. The estimates for the current year and the recalculated back years are constrained to sum to the 90 and over totals in the Mid-Year Estimates (MYE) for males and females separately for the current year and the previous years. This provides users with a set of estimates by single year of age up to age 105 and over consistent with the published aggregate 90 and over population estimates.

Notes:

  1. Thatcher, R. (Summer 1999). ‘The demography of centenarians in England and Wales’, Population Trends, No. 96 pp 5-12.

  2. Number of deaths that occurred in the reference period.

Back to table of contents

5. Differences

In Scotland and Northern Ireland, the input deaths data is on a mid-year to mid-year basis by age at the start of the mid-year to mid-year period. Applying the KT method to data in this format allows population estimates by age at a mid-year to be derived directly from the input data. In England and Wales the input deaths data are on a calendar year basis by age at death. This is because historically this is the way in which deaths data for England and Wales were produced. Also, the estimates were initially calculated as input data for life tables which traditionally use calendar year deaths.

ONS make the assumption that deaths are evenly distributed over the calendar year and that birth dates of those dying at age x are also evenly spread over a year. This means that for each age the deaths in year T will be assigned to two separate years of birth cohorts. For example, if there were 8042 deaths at age 90 in 2013, it is assumed that 4021 deaths were amongst the cohort born in 1922 and 4021 were amongst the cohort born in 1923. This gives estimated deaths in a calendar year by age at the start of the calendar year. Applying the KT method to calendar year deaths in this format produces population estimates by age at the start of a calendar year. To obtain mid-year estimates, ONS interpolate between the estimates derived for 1 January year T and 1 January year T+1 to produce mid-year estimates for year T.

ONS and NRS round their estimates of the population aged 90 and over by single year of age to the nearest 10 (except where the number is greater than zero but less than or equal to 5, which is shown as ≤5), reflecting the quality of data and methodology. Given the small population of Northern Ireland, NISRA publishes estimates to the nearest person.

Back to table of contents

6. Geographical Breakdown

Estimates of the population aged 90 and over by single year of age are only available using this methodology at the country level. Below country level, migration can become more significant, for example, for older people moving from a domestic property to a residential care home.

ONS proposed an approach to producing population estimates for the 90‐94 and the 95‐plus age groups, by sex, for Lower Layer Super Output Areas (LSOAs) in England and Wales using a different methodology and invited user comment. Details can be found in the LSOA Population Estimates of the Very Old Research paper 17. Experimental Mid-2002 to Mid-2012 LSOA Population Estimates of the Very Old, England and Wales were been published as supporting information to the paper.

NRS published estimates for the population aged 90 and over for council areas in Scotland for the first time on 30 September 2015. Community Health Index (CHI) data for single year of ages 90+ was used as an initial proxy for the distribution of these ages within each Council area. This method is explained in more detail in the publication which is available in the Sub-national Population Estimates for ages 90 and over section of the NRS website. They are looking for feedback on the method and results of these developmental statistics.

Back to table of contents

7. Timing of Release

ONS, NRS and NISRA coordinate a release date that is generally at the end of September, to be available prior to the United Nations International Day of Older Persons (1 October).

Back to table of contents

8. Developments

It would be possible for ONS to reallocate deaths to a mid-year by mid-year basis for years after 1993 but prior to this date data is not available meaning there would be some small discontinuities in historical data sets.

ONS are currently conducting a review of the methodology to produce 90 and over estimates by single year of age. The review will consider the impact of moving from using deaths on a calendar year basis to deaths on a mid-year basis.

Back to table of contents