Extremely small sample size in some toxicity studies: an example from the rabbit eye irritation test

Area of Science:

• Toxicology
• Statistical Methods
• Animal Testing

Background:

• Conventional sample-size equations focus on estimation precision or hypothesis testing power.
• These methods may be unsuitable for diagnostic testing scenarios.
• The Draize test for eye irritancy serves as a relevant example.

Purpose of the Study:

• To evaluate the appropriateness of conventional sample-size calculations for diagnostic tests.
• To determine adequate sample sizes for classifying chemical compounds based on animal responses.
• To explore alternative sample-size determination for specific toxicological assessments.

Main Methods:

• The study discusses the limitations of standard statistical approaches for diagnostic tests.
• It considers scenarios with uniform and extreme animal responses.
• The focus is on classifying compounds as irritant or nonirritant.

Main Results:

• Conventional sample-size equations may not be appropriate for diagnostic tests.
• In cases of uniform and extreme animal responses, a small sample size (2-3 animals) can be adequate.
• This is particularly relevant for classifying compounds as irritant or nonirritant.

Conclusions:

• Sample size determination for diagnostic tests requires specific considerations beyond traditional statistical power or precision.
• Minimal animal usage may be sufficient for certain toxicological classifications.
• The findings suggest a potential for refining animal testing protocols in toxicology.