The following pharmacological definition has been taken from the Pharmacology and Experimental Therapeutics Department Glossary at Boston University School of Medicine.
A form of experiment in which each subject receives the test preparation at least once, and every test preparation is administered to every subject. At successive experimental sessions each preparation is “crossed-over” from one subject to another. The purpose of the cross-over experiment is to permit the effects of every preparation to be studied in every subject, and to permit the data for each preparation to be similarly and equally affected by the peculiarities of each subject. In a well-designed cross-over experiment, if it is at all possible, the sequence in which the test preparations are administered is not the same for all subjects, in order to avoid bias in the experiment as a result of changes in the behavior of the subjects that are a function of time rather than of drug administration, or a function of drug interactions. At least, the cross-over design permits detecting such biases when they occur. The preparations under test in a cross-over experiment may – ideally, should – include one or more doses, of an experimental or “unknown” drug, one or more doses of a dummy or placebo medication (”negative control drugs”), and one or more doses of a standard drug, the actions of which are expected to be similar to those of the “unknown” (”positive control drug”). Even for the investigator with the best knowledge and intentions, the economics and logistics of experimentation may prevent carrying out a complete and perfect cross-over experiment.
The copyright of the text is hold by Trustees of Boston University. Permission has been granted for its use in this blog.