Dementia statistics for people aged 65 or over by health board
Next update expected:November 2027
Updates
- 12th November 2026
2026 data added.
- 4th December 2025
- Dataset first published.
Main information
- Designation
- Official statistics
- Data provider 1
- Office for National Statistics (ONS)
- Data source 1
- Mid-year population estimates
- Data provider 2
- Digital Health and Care Wales (DHCW)
- Data source 2
- Audit+
- Time period covered
- January 2020 to December 2026
- Dataset ID
- f2f8f44b-9b22-49aa-865e-38c8fe3e4377
Overview
- Summary of dataset and variables
This StatsWales cube provides data for people aged 65 or over: a) estimated to have dementia, b) on the dementia register (recorded as being diagnosed with dementia at GP practices), c) estimated as being undiagnosed with dementia and d) % diagnosed with dementia.
- Data collection or calculation
The number of patients diagnosed with dementia is recorded as a disease register at general practices; however clinical research suggests that a number of people with dementia are not formally diagnosed with it. In light of this, data is also presented for the estimated total number of people with dementia in Wales, based on prevalence creates published in a Lancet paper from the CFAS (Cognitive Function and Ageing Study) II study and general population data. The difference between the estimated number and the disease register number is an estimate of the number of people with dementia but have yet to be diagnosed with it.
Data is presented for patients aged 65 or older, as the large majority of dementia patients are in this age group.
From 2023 onwards, this analysis uses the age group data directly from GP systems. Prior to this, the number was estimated using QAIF extracts (which were not available by age group) apportioned by the percentage of dementia patients aged over 65 through the alternative source (GP systems).
- Statistical quality
Any insights drawn from year-to-year changes are limited because the same mid year estimate has been used for some adjacent years. For example, the mid year estimate for 2021 was used in both 2021 and 2022 and therefore the estimated number of people with dementia would be the same in both years.
The dementia analysis uses the latest available ONS mid-year population estimates (MYEs) at the time of publication. For each disease register reference date, the MYE reference year used is as follows:
- 2020: 2019 MYE
- 2021: 2021 MYE
- 2022: 2021 MYE
- 2023: 2022 MYE
- 2024: 2022 MYE
- 2025: 2024 MYE
- 2026: 2025 MYE
Coverage information of dementia disease register data
2020: There was 100% coverage of GP practices in the QAIF data for 2019-20; data was based on all 399 active practices as at 30 September 2020. However, the Audit+ data was sourced from 395 out of 397 active practices as at 14 December, as two practices which provided QAIF disease register data were unable to provide Audit+ data.
2021: The QAIF data was based on 391 out of 392 active practices as at 30 September 2021. Data was not available for W97036. The Audit+ data was based on 390 practices out of 392 active practices as of October 2021.
2022: There was 100% coverage of GP practices in the QAIF data for 2021-22; data was based on all 386 active practices as at 30 September 2022. The Audit+ data is also based on all 386 practices.
2023: In 2023, all bar one practice (W97048) submitted data.
2024: In 2024, all practices submitted data.
2025: In 2025, all practices submitted data.
2026: In 2026, all practices submitted data.
- Related reports
Published by
- Organisation
- Welsh Government
- Contact email
- stats.healthinfo@gov.wales