Cross-over study

from Wikipedia, the free encyclopedia

A cross-over study , and cross-over study is a special research design of evidence-based medicine to compare two treatment methods.

The effectiveness of two forms of treatment can be compared in cross-over studies. Two agents, or one agent and a placebo , are administered to the same test subjects at different times. There are at least two study phases for this. In the first phase, the subjects in the first group first receive therapy A, the subjects in the second group receive therapy B. Allocation to the groups is randomized (at random). After a defined duration of treatment, a switch is made to the second study phase: Group 1 now receives therapy B and group 2 receives therapy A. The effectiveness of the therapy is then compared both within a group (intra-group comparison) and between the two groups (intergroup comparison). One advantage of the study form is that none of the participants have to be withheld from the intervention to be examined for research reasons.

Compared to conventional studies with two parallel comparison groups, cross-over studies have the advantage that smaller differences in the effectiveness of the treatment forms become statistically significant or that fewer participants are required to demonstrate a significant difference. An important step in planning a crossover study is determining the sample required to obtain significant results. Calculations of simulated studies show the sample size from which reliable results can be expected and when an increased error quotient can be expected due to a sample that is too small.

However, this study design can only be used for a few medical questions. The carry-over effect must be taken into account , that is, the effect of the application effects of the first agent in the second application phase. Therefore, a break is required between the treatment phases, during which no application takes place, so that the effect of the first agent is no longer given. The method can therefore not be used when it comes to curing diseases, but only when it comes to alleviating symptoms in incurable chronic diseases. Special statistical methods are available for the evaluation .

This test plan will also cross-over experimental design or crossover experimental design called and in the statistical design of experiments dealt with. If the statistical properties are optimized, one speaks of an optimal cross-over test plan .

Individual evidence

  1. Glossary of Evidence-Based Medicine , accessed October 20, 2019
  2. Keyword: cross-over study . In: Markus Antonius Wirtz: Dorsch: Lexikon der Psychologie Hogrefe Verlag, Göttingen 2019. ISBN 978-3456859149
  3. NM Siyasinghe and MR Sooriyarachchi: Guidelines for calculating sample size in 2x2 crossover trials: a simulation study. Journal of the National Science Foundation of Sri Lanka. 2011, 39 (1): pp. 77–89 pdf available online . Retrieved October 20, 2019
  4. Gabi Schulgen, Martin Schumacher: Methodology of clinical studies: methodological principles of planning, implementation and evaluation (statistics and their applications) . Springer-Verlag, Berlin-Heidelberg-New York 2008, 3rd edition, pp. 305-308, ISBN 978-3-5408-5135-6
  5. ^ Bose, M. and Dey, A .: Optimal Crossover Design , World Scientific 2009, ISBN 978-981-281-842-3