gms | German Medical Science

10. Deutscher Kongress für Versorgungsforschung, 18. GAA-Jahrestagung

Deutsches Netzwerk Versorgungsforschung e. V.
Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie e. V.

20.-22.10.2011, Köln

Do different matching approaches result in equivalent findings?

Meeting Abstract

  • author presenting/speaker Robert Krohn - Aqua Institut, Göttingen, Germany
  • corresponding author Boris Pöhlmann - Aqua Institut, Göttingen, Germany
  • Birgit Fullerton - Institut für Allgemeinmedizin, Johann Wolfgang Goethe-Universität, Frankfurt, Germany
  • Thomas P. Zahn - DCC Risikoanalytik GmbH, Berlin, Germany
  • Petra Kaufmann-Kolle - AQUA-Institut GmbH, Göttingen, Germany
  • Günther Heller - AQUA-Institut GmbH, Göttingen, Germany
  • Björn Broge - AQUA-Institut GmbH, Göttingen, Germany
  • Erik Bauer - AQUA-Institut GmbH, Göttingen, Germany

10. Deutscher Kongress für Versorgungsforschung. 18. GAA-Jahrestagung. Köln, 20.-22.10.2011. Düsseldorf: German Medical Science GMS Publishing House; 2011. Doc11dkvf061

doi: 10.3205/11dkvf061, urn:nbn:de:0183-11dkvf0612

Veröffentlicht: 12. Oktober 2011

© 2011 Krohn et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.de). Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.


Gliederung

Text

Background: Statutory health care funds in Germany are obliged to offer their insured a general practitioner (GP) centered health care contract to strengthen the GP in his role as a gatekeeper and focusing on better quality in medical care and cost effectiveness. To evaluate such programs properly it is important to gain a control group for a self selected study population. In this context it is of great interest whether or not different matching methods could result in different results.

Materials and methods: The statutory health insurance funds provided us with routine data for the years from 2004 to 2008. Because of the data characteristics the researcher has to cope with heterogeneity and self selection. As randomization is unfeasible due to the voluntary design of enrolment models it is essential to build up a control group via matching approach. Within two different evaluation projects we applied different approaches to construct a control group, inter alia propensity score matching.

To compare the results we applied different quality measures (T-Test, Kruskal Wallis Test and standardized mean difference (Cohen’s D)) to explore disparities between matching methods for different variables.

Results: Each approach has its own advantages and disadvantages the researcher has to choose between. The application of different quality measures showed us that there is no gold standard for matching approaches. However, direct covariate matching is of limited use in practical studies where many variables must be accounted for. The use of propensity score matching is encouraged for such studies with a broader spectrum of use.

Conclusions: Matching approaches are the method of choice to find a control for a self selected study population. The right decision in choosing an adequate approach depends on the data set, the mathematical background of the researcher and the complexity of the algorithm.


References

1.
Ariel Linden JL. Using Propensity Scores to Construct Comparable Control Groups for Disease Management Program Evaluation. Disease Management Health Outcomes. 2005;12(2):107-115.
2.
Austin PC. A critical appraisal of propensity-score matching in the medical. Statistics in medicine. 2008;27:2037-249.
3.
Riens B, Broge B, Kaufmann-Kolle P, Pöhlmann B, Grün B, Ose D, Szecsenyi J. Creation of a control group by matched pairs with GKV routine data for the evaluation of enrollment models. Gesundheitswesen. 2010;72(6):363-70.