13 Feb Docent Reijo Sund: Can we use Finnish registers in scientific research? Facts and fictions about the quality of secondary data. Seminar series "Registers as a data source for effectiveness and patient safety research", Kuopio

Reijo Sund (National Institute of Health and Welfare): Can we use Finnish registers in scientific research? Facts and fictions about the quality of secondary data

The Seminar series" Registers as a data source for effectiveness and patient safety research" continues on 13.2.2014 from 14.15 to 15.45. Place: MS 1+2 in Medistudia.

Everyone is warmly welcome! Please note the change in date+place!

Extra material for the seminar series: http://www.uef.fi/en/receps/seminaarisarja

Short bio:

Docent Reijo Sund is a senior statistician working currently at the Centre for Research Methodology in Social Sciences, Department of Social Research, University of Helsinki. His main research interests have been in the general methodology of register-based data-analysis, and he has worked with register-based data at the National Institute for Health and Welfare (and its predecessor) since 1997.

Can we use Finnish registers in scientific research? Facts and fictions about the quality of secondary data

Administrative register data has been recognized to be a feasible and cost-effective way to study large, possibly nationwide, populations. The Finnish health registers offer a very attractive and flexible environment for register-based research, because the universal personal identification numbers are used in all registers since the 1960s. The main disadvantage of such register data is the secondary nature, i.e. that the data have not been collected primarily for the purposes of specific research question, and therefore that the quality of the data for the intended research purpose is typically unknown.

In this talk I postulate that to make any scientific statements concerning the quality of (secondary) data, it is essential to first outline the properties of required measures carefully under a fixed conceptual model (determine how data result from the theory), and then evaluate the observed data against these requirements (examine how well the theory can be reconstructed using the properties of actual observed data). I show how such approach works in practice and discuss the results of recent systematic review of the Finnish Hospital Discharge Register from this perspective.

Publishing year: 2014

Back to this years article listing