SHARE-HCAP

Please cite the dataset in your references as follows:

SHARE-ERIC (2025). Survey of Health, Ageing and Retirement in Europe (SHARE). SHARE-HCAP1. Release version: 1.0.0. SHARE-ERIC. Data set. DOI: 10.6103/SHARE.HCAP1.100

SHARE-HCAP provides in-depth, internationally validated cognitive assessment data for adults aged 65 and older in Europe. It is the European arm of the global Harmonised Cognitive Assessment Protocol (HCAP) network – a battery of tests designed to be delivered around the world in a standardised way.

The general aim of SHARE-HCAP is to exploit the international variation of policy, health and life circumstances in Europe to identify how biomedical and socioeconomic conditions over the life course interact to affect cognition in later life. 

This makes SHARE-HCAP a key resource for researchers studying:

• Dementia and Alzheimer's disease epidemiology
• Quality of life of people living with mild cognitive impairment (MCI)
• Social and economic determinants of cognitive ageing
• Aetiology and consequences of domain-specific impairments  
• Cross-national comparisons of cognitive health and longevity
• Brain health in countries in Asia, Africa and the Americas (see below)

What is HCAP?

HCAP stands for Harmonised Cognitive Assessment Protocol. It is an in-depth battery of cognitive tests. It also includes a report from a family member or friend on the respondent's cognitive functioning.

HCAP was developed by the US Health and Retirement Study (HRS). It is used by a global network of ageing studies, funded by the US National Institute on Aging. This allows cognitive health to be measured the same way in different countries.

As of 2026, population-based studies using the HCAP protocol have been completed in the United States, Mexico, Chile, England, Northern Ireland, the Republic of Ireland, several European countries (Czech Republic, Denmark, Germany, France, Italy), South Africa, Kenya, China, and India.

What is SHARE-HCAP and how were respondents sampled?

SHARE-HCAP is the European arm of the HCAP network. It sits within SHARE, the Survey of Health, Ageing and Retirement in Europe.

SHARE-HCAP collected data from 2,687 respondents. All were aged 65 or older. Data came from five countries: Czech Republic, Denmark, France, Germany, and Italy. These countries were chosen to represent Eastern, Northern, Western, and Southern Europe.

Respondents were selected from the main SHARE sample. Those with low scores on a word recall test were oversampled. This ensured enough people with cognitive impairment to allow valid estimates. Data were collected between May and November 2022, around five months after the main SHARE Wave 9 interview. The response rate was 75.8%.

SHARE-HCAP tests memory, executive functioning, visuospatial skills, language and orientation across Europe

Full information on the tests and the rationale for their selection is provided in Release Note 1.1 (Börsch-Supan et al., 2025).

Table 1: Cognitive tests administered to respondents

Table 2: SHARE-HCAP informant report items

Establishing prevalence estimates in cross-national cognitive research

Most multi-country dementia prevalence studies are not well harmonised. They differ in methodology, diagnostic criteria, age groups, and study year. This makes cross-country comparisons unreliable.

SHARE-HCAP solves this problem. It uses the same cognitive tests and the same diagnostic criteria across all five countries. Results are also used to validate cognition measures in the much larger SHARE parent sample (over 47,000 individuals across 27 European countries and Israel). This allows MCI and SCI prevalence to be estimated consistently for the whole of Europe, not just the five HCAP countries.

Respondents are classified as normal, MCI, or possible dementia. This classification follows standard diagnostic criteria from the US National Institute on Aging and Alzheimer's Association. The same criteria are used in other HCAP studies worldwide, so results are comparable internationally.

Initial findings using SHARE-HCAP data

Combining SHARE-HCAP with the SHARE parent sample has uncovered large variation in cognitive impairment across Europe:

• SCI prevalence ranged from around 5% (Switzerland, Sweden, Denmark, Germany) to over 20% (Spain, Portugal)
• MCI prevalence ranged from about 17% (Austria, Germany, Sweden, Denmark, Switzerland) to almost a third (Bulgaria, Cyprus, Greece, Portugal)
• Every 5-year increase in age was linked to a higher risk of progressing  to possible dementia
• Higher educational attainment in early life was linked to a large decrease in risk of cognitive impairment

Differences in education, when respondents were young, explained most of the international variation in cognitive impairment across Europe.

Methodology and further reading

Full details of the data collection, classification approach, and statistical methods are available in:

Börsch-Supan, A., S. Douhou., & B.B Tawiah (2025): Cognitive Impairment and the Long Arm of Childhood Education: Evidence from Europe. MEA-SHARE Discussion Paper 01-2025.

Börsch-Supan, A., S. Douhou et al. (2025): Release note 1.1 to SHARE-HCAP data. MEA Discussion Paper 02/2025.

How to access SHARE-HCAP data

Data is free to access for the scientific community, subject to SHARE registration.

1. Register as a SHARE user — registration information
2. Download the data via the SHARE Research Data Center
3. Read the full release documentation (see above)

Contact:

hcap(at)mea-share.eu