Programme no. 361-P
Public Health
Validation of permanent nursing home residency in a nation-wide Danish database
Anna Bebe*1, Jens Soendergaard2, Tora Willadsen3, Anni Brit Sternhagen Nielsen4, Volkert Siersma5, Jakob Kragstrup6, Astrid Raahede7, Frans Boch Waldorff8
1Research Unit for General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark, 2Research Unit of General Practice,University of southern Denmark,Odense,Denmark, 3Research Unit of General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark, 4Research Unit for General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark, 5Research Unit of General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark, 6Research Unit for General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark, 7Research Unit of General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark, 8Research Unit for General Practice,University of Copenhagen, Deparment of Public Health,Copenhagen,Denmark
* = Presenting author
Objectives: The aim of the present study is to validate the nursing home algorithm of SD.
Background: Background

There is a growing demand for long term care as the world population keeps ageing and by the year of 2050 it is expected that a third of the population will be over the age of 60. Permanent nursing home placement predicts health outcomes in the elderly, and therefore may represent an important health related proxy.

Currently, there is no nation-wide validated method to identify Danish individuals in a permanent nursing home facility.

Results: At present, nine out of ten municipalities have agreed to participate in the project and data has been received from eight municipalities. The final municipality has notified us that they currently have higher-ranking priorities than providing us with data. Therefore, a new municipality will be randomly selected. Each municipality consists of approximately 500 individuals. Thus, this project will involve approximately 5000 individuals.

Once the data is received from all participating municipalities it will be compared with SD’s data. In addition to comparing social security number, we will look at sex, marital and socioeconomic status and age. Furthermore, we will include size, location and the socioeconomic status of the participating municipalities.

Material/Methods: Ten municipalities were randomly selected stratified by region and reporting status to. Data included individuals living in a nursing home facility on 1st January 2013 (Gold Standard). We were able to retrieve all individuals living in nursing homes on 1st January 2013. Sensitivity and positive predictive value will be calculated based on these two data sets.

Conclusion: The main hypothesis of this study is that SD’s algorithm will be most accurate in rural municipalities. Mainly, since there are less addresses housing several people, compared to urban living, which contain a higher percentage of high-rise buildings.

Points for discussion:

Potential pit falls in using nursing home as a health related outcome?

What could we do to prevent that non-validated variables are used in research?