The Hazardous Substances (minimum Degree of Hazard) Regulations 2001 requires a mean Draize score and defines that as:
"mean Draize score,-- (a) in relation to acute skin irritation tests, means the mean value in at least 2 of 3 tested animals-- (i) from Draize grades measured at intervals of 24 hours, 48 hours, and 72 hours after the patch is removed; or (ii) where reactions are delayed, from Draize grades on 3 consecutive days after the onset of dermal reactions; and (b) in relation to acute eye irritation tests, means the mean value of at least 2 of 3 tested animals from Draize grades measured at intervals of 24 hours, 48 hours, and 72 hours after instillation of the substance"These regulations cover all chemicals used in NZ and include anything from cleaning products, agricultural compounds, solvents, water treatment chemicals to tattoo inks, fuel additives and fertilizers. This test is done despite significant structural and physiological differences between rabbit and human eyes:
Rabbits have a nictating membrane (third eyelid)
Rabbits produce tears less effectively than people
The pH and buffering capacity of the aqueous humour is different - human aqueous humour has a pH of 7.1 - 7.3; rabbits have a pH of 8.2
The cornea is 25% of a rabbit's eye surface area, for a human it's 7%
"the widely used Draize/FHSA rabbit eye irritation test has never been validated against any reported human database."
As well as the intrinsic difference between rabbits and humans the subjective nature of how the test animal's response is assessed is another a source of innacuracy. The assessment relies on observation by the vivisector. One person's severe irritant may be anothers mild irritant... Non-animal tests to predict eye irritancy are available and some are already validated by the OECD and are approved for use by New Zealand's Environmental Protection Authority.
The best data to base regulatory decisions on comes from a battery of at least three non-animal tests. These tests provide data based on human physiology and not on trying to approximate it with an animal's. They include computer models as well as tests that use human cells. We need to change the law as NZ is currently helping drive demand for the outdated Draize test to be done.
Note: We did have a petition for this campaign. It is now closed as the campaign was won.
 Griffith JF, Freeberg FE: Empirical and experimental bases for selecting the low volume eye irritation test as the validation standard for in vitro methods. In: Goldber AM (Ed): In Vitro Toxicology: Approaches to Validation. New York, Mary Ann Libert, 1987, pp. 303-311.