According to the Guide for the Care and Use of Laboratory Animals, Eighth Edition, animal care and use protocols must contain "Justification of the species and number of animals requested. Whenever possible, the number of animals requested should be justified statistically." This is somewhat in contrast to PHS policy which states "identification of the species and the approximate number of animals to be used."
This difference seems to be a significant point of confusion for people submitting a protocol. Now that this institution is AAALAC accredited and AAALAC follows the principles of the "Guide", that's what we follow. The committee often gets the argument, "This is research.........we don't know how many animals we're going to need until we do the experiments." Unfortunately, this is not really an acceptable answer. There are several intentions underlying this requirement which I will try to outline:
This is a sticky point for everyone involved. The IACUC does not want to have to conduct scientific review, nor do they feel particularly qualified to conduct such a review. Unfortunately, it is inherently necessary when determining that the appropriate number of animals are being requested. AAALAC generally takes the view that if there isn't a clear explanation of the numbers requested, then there has not been enough thought or planning put into the project to justify the animal use. Numbers are really the area that the CHUA focuses most closely on when doing its "scientific review" because this is the one area that the committee does feel fairly comfortable evaluating.
It is always more desirable if the numbers selected are based on statistical estimations based on the data to be collected or based on the numbers required to achieve statistical significance for similar experiments conducted in the past or published in the literature. If statistics cannot be applied (or are not appropriate for the type of experiment), there is a much stronger need for a clear, logical explanation for why the numbers in each group were selected.
This is the one aspect that tends to hold up the most protocols. The number of animals requested usually appears in 3 different sections of the protocol; section B (total number), section D(v) (numbers justification) and section E (procedures). If the numbers requested are not consistent in these three areas, it raises a red flag and puts the committee in the position of not being able to clearly say they understand the experimental design. If the committee cannot understand the design, they cannot say the animal use is justified and therefore cannot approve the protocol.
The committee is not blind to the fact that animal research, by nature, does not always fit into specific, neat, justifiable groups. There is not a specific formula that needs to be followed to justify the numbers, just some thought and good explanation. Any combination of group sizes or even pooling animal tissues for in vitro work can be justified using logical explanations. It is also understood that sometimes things don't work and more animals are needed do to unforeseen circumstances. This can easily be handled with an addendum after the fact. Or, if there is strong scientific reason to believe that a certain percentage of experiments will fail, the extra animals can be justified within the original protocol. It should be realized, however, that a large percentage of experimental failure obligates the committee to conduct an even more critical review of the techniques to ensure that there are not methods available which would result in a higher success rate with less animal cost.