gms | German Medical Science

64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

08. - 11.09.2019, Dortmund

Adaptive Designs in Randomized Diagnostic Studies with Patient-Relevant Endpoints

Meeting Abstract

Search Medline for

  • Amra Hot - Institut für Medizinische Biometrie und Epidemiologie, Universitätsklinikum Hamburg-Eppendorf, Hamburg, Germany
  • Antonia Zapf - Department of Medical Biometry, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Dortmund, 08.-11.09.2019. Düsseldorf: German Medical Science GMS Publishing House; 2019. DocAbstr. 122

doi: 10.3205/19gmds090, urn:nbn:de:0183-19gmds0908

Published: September 6, 2019

© 2019 Hot et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Diagnostic accuracy studies are performed to assess how well a diagnostic test can distinguish diseased and non-diseased individuals. However, the resulting diagnostic information is only beneficial if it’s appropriately used in subsequent patient management decisions and thus clinically relevant outcomes, such as morbidity, mortality or health related quality of life, are improved in the long run. A joint evaluation of test and therapy regarding patient-relevant outcomes is the aim of randomized diagnostic trials.

Here, an important aspect is the sample size calculation, which is based on assumptions concerning the prevalence and the diagnostic accuracy, i.e. sensitivity and/or specificity, of the test as well as the suggested therapy effect from preliminary studies. Due to incorrect assumptions there is always a risk of incorrect estimation of the sample size and leading to an over- or underpowered trial.

Thus, adaptive trial designs for randomized diagnostic studies would be valuable, which allow modifications of the sample size or aspects of the study design by means of predetermined interim analyses. Until now, there is little research regarding adaptive designs for randomized diagnostic trials. As part of this work, the focus is to differentiate to what extent blinded interim evaluations concerning patient-relevant outcomes can be performed and when blinding is no longer fulfilled, without undermining the statistical validity and integrity of the trial. In the case of an unblinded or blinded interim analysis the type I error should be adjusted or not, particularly. The considering of the extent, to which statistical power and/or the type I error may be affected by a re-estimation of the sample size based on the prevalence, therapy effect or diagnostic accuracy of the test and whether a desired power is achieved at all, is a key question of this work.

In addition, another challenge is the variety of study designs proposed in the literature for randomized diagnostic studies that differ at time point of the randomization and their basic design. However, the nomenclature is not consistent. A clear distinction of these designs will contribute to the development of a sample size re-estimation formula for randomized diagnostic trials.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.