"Reduction" is one of the "3R's" alternatives to animals in research as described in The Principles of Humane Experimental Technique. Authors Russell and Burch define reduction in this context as "reduction in the numbers of animals used to obtain information of given amount and precision." In other words, scientists should seek to reduce the number of animals in experiments to the minimum required to obtain scientifically valid data. Reduction also can be achieved by obtaining more information from a given number of animals - so as to reduce the need for additional animal subjects.
Researchers are continuously increasing their understanding of effective experimental design and statistical methods. An increased awareness of the ethical reasons for reducing animal use is also on the rise. From a practical standpoint, the high cost of animal subjects provides further incentive to minimize the number of animals in research.
There are a variety of ways that scientists can reduce the numbers of animals in research. Russell and Burch note that, "One general way in which great reduction may occur is by the right choice of strategies in the planning and performance of whole lines of research." For example, a small pilot study, using a small number of animals, may indicate to researchers whether a larger study is appropriate.
In some cases, in vitro (test tube) experiments can be used to screen appropriate experimental design, and may indicate the feasibility of a larger study or ways a study could be modified to use fewer animals or less invasive procedures. Maximization of the data from each procedure, reduction of data skewing extraneous variables, data sharing to avoid repetition, harmonization of regulatory requirements for testing, and the use of computer-assisted teaching of experimental design and statistics may also streamline experiments and alleviate the need for more animals.
The field of bioinformatics can also help reduce the need for animal subjects. Bioinformatics is an emerging, interdisciplinary field that draws on computer science, mathematics, and information theory to model and analyze biological systems, especially systems involving genetic material. Bioinformatics can be used to analyze complex experimental results from multiple sources, patient statistics, and scientific literature. This amalgamation of biomedicine and computer technology can permit the gleaning of pertinent information from both past and ongoing experiments - allowing scientists better access to, as well as analytical tools to interpret, data already gathered.