German Ageing Survey

Summary

The DEAS is one of the best data sources to answer questions about health in old age within Germany. Health is assessed as a multi-dimensional construct (e.g. functional, physical, subjective, and objective health measures are available). The DEAS allows investigation of individual changes over time (a period of 15 years), as well as health differences between cohorts. Moreover, health can be linked to life transitions such as retirement and widowhood. As there is a huge variety of indicators of different life domains available, underlying mechanisms of health changes over time and within age groups can be addressed. The DEAS stands out because of its interdisciplinary approach and cohort-sequential design. There are also some weaknesses. First, the majority of health indicators are self-assessed, objective health indicators are limited, bio markers are not available. Although the DEAS assesses participants over the age of 85 and participants living in care facilities, the sample is not representative for very old age (85 years and older) and persons living in retirement or care homes, therefore, health status and health changes within these important sub-populations cannot be investigated without loss of data quality. The first wave of the DEAS was limited to participants with German citizenship, since 2002 the inclusion criteria is ability to speak and understand German. Despite the effort to include foreigners and people with migration background, the DEAS still is not representative for this sub-population.

Type of data

Data Source
Survey

Type of Study
Survey same
Crosssection regular
Other: Cohort-sequential design

Data gathering method
Face-to-face
Self administered questionnaire

Access to data

Conditions of access
Available for scientific, non-profit use

Type of available data (e.g. anonymised microdata, aggregated tables, etc.)
anonymised microdata

Formats available
SPSS, STATA

Coverage

Coverage Years of collection, reference years, and sample sizes
Wave 1: Data collected in 1996 (DOI 10.5156/DEAS.1996.M.001) with a sample size of 4, 838 individuals. Wave 2: Data collected in 2002 (DOI 10.5156/DEAS.2002.M.001) with a base sample of 3,084 individuals, a migrant sample of 586 individuals, and a panel sample of 1,524 individuals. Wave 3: Data collected in 2008 (DOI 10.5156/DEAS.2008.M.001)with a base sample of 6,205 individuals and a panel sample of 1,995 individuals. Wave 4: Data collected in 2011 with a panel sample of 4, 855 individuals. Wave 5: Data will be collected in 2014. A new base sample will be drawn and the panel sample will be reassessed.

First year of collection
1996

Stratification if applicable
age (40-54, 55-69, 70-85 years), sex, region (East/West)

Base used for sampling

Geographical coverage and breakdowns
national, NUTS3-level (Kreise)

Age range
baseline samples: 40-85 years; Panel sample: 40-90 years

Statistical representativeness
Other, please specify

Coverage of main and cross-cutting topics
Within the health domain, approximately 150 comparable variables are available within each wave. More health information (up to 200 variables) is assessed at each wave, but may not be reassessed. The variables cover: physical health, functional health, subjective health, mental health, health behaviour, need of care, need of assistance, health care utilisation, sentinal health events, pain, sleep, health test and information on mortality.

Linkage

Standardisation
The data set contains various internationally harmonised standards (e.g. ISCED-97 (International Standard Classification of Education), ISCO-88 (International Standard Classification of Occupation)). Most health instruments are standardised questions and scales. Within the health domain, two-thirds of measures are comparable with international research (e.g. the European health module is assessed in parts).

Possibility of linkage among databases
Only regional linkage on district level via NUTS3-level (Kreise) is possible.

Data quality

Entry errors if applicable
Raw data is cleaned by project organisers and checked for inconsistencies; data is further checked by the Research Data Centre, then the scientific use file (SUF) is created.

Breaks
Research group changed between first and second wave (1996: Freie Universität Berlin, since 2002: German Centre of Gerontology). In 1996, interviews were conducted via PAPI, since 2002, however, they have been conducted via CAPI.

Consistency of terminology or coding used during collection
Due to change in research group, the documentation for 1996 is incomplete.

Governance

Contact information

Research Data Centre of the German Ageing Survey, German Centre of Gerontology (DZA Berlin)
Manfred-von-Richthofen-Straße 2
12101 Berlin Germany Phone: +49 (0)30 - 260740-0
Email: fdz(at)dza.de
Url: http://www.dza.de/en/fdz/research-data-centre-of-the-german-ageing-survey-fdz-deas/access-to-deas-data.html

Timeliness, transparency
The scientific use file is available about 2 years after data collection.