Email updates

Keep up to date with the latest news and content from BMC Complementary and Alternative Medicine and BioMed Central.

This article is part of the supplement: Scientific Abstracts Presented at the International Research Congress on Integrative Medicine and Health 2012

Open Access Oral presentation

OA01.03. Sample size and regions of principally achievable significance in cost effectiveness studies: an example of complementary medicine

T Ostermann1*, K Boehm1 and F Krummenauer2

  • * Corresponding author: T Ostermann

Author affiliations

1 University of Witten/Herdecke, Herdecke, Germany

2 Institute for Medical Biometrics, University of Witten/Herdecke, Witten, Germany

For all author emails, please log on.

Citation and License

BMC Complementary and Alternative Medicine 2012, 12(Suppl 1):O3  doi:10.1186/1472-6882-12-S1-O3

The electronic version of this article is the complete one and can be found online at:

Published:12 June 2012

© 2012 Ostermann et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Health economic studies gain more and more importance in both conventional and complementary medicine. In most cases such studies are conducted as a combination of a RCT and a health economic evaluation (“Piggy-Back-Studies”). Therefore study planning parameters like sample size still are calculated on the basis of the outcome of the clinical parameters. This might lead to situations of underpowering.


Based on the sample size estimates given in Glick (2011), we aimed at finding health economic studies of CAM providing mean and standard deviation data on cost and outcome differences. Based on this data and on assumptions on willingness to pay and cost-outcome correlation, we constructed a model of principally achievable significance.


Based on an existing review on health economic studies in CAM, 8 of 143 studies mentioned sample size calculation parameters. However, only one study (Lin et al. 2008) provided enough data for modelling. We found a significant discrepancy between the sample size of n=46 based on conventional sample size calculation and the hypothetically needed patients of more than 350 based on our modelling approach.


Planning health economic studies should be done with great caution not to end in a situation of small power. Our approach might retrospectively give estimates of regions of significance and thus might help to interpret health economic studies not only in the field of CAM.