The assay requires a small blood sample (25 microlitres) which is then processed and stained with propidium iodide before the net amount of nuclear DNA is determined for each cell using a flow cytometre. The 10,000 cells analyzed for each sample is summarised as a coefficient of variation statistic (CV). All the CV values are then analyzed using a unique statistical protocol that respects the CV's status as a statistical value. The resulting analysis is highly sensitive to any changes in net DNA content of the nucleus which is reflective of genetic damage in the parent cells (see flow cytometry section).

We also consider the contaminant loading of individuals within populations to determine the relationship between DNA damage and specific contaminants. This approach was successfully used in investigating white sturgeon populations in Idaho where it was discovered that the concentration of metallic iron was significantly associated with DNA damage.

In other circumstances, such as beak deformities in chickadees in Alaska, the only consistent correlate with the birds with deformed beaks was the level of observed DNA damage.

DNA damage is a very sensitive indicator of environmental impact, much more so than physiologically-based indicators of effect. This phenomenon was clearly shown in two studies where many parameters were investigated simultaneously. In the study where Washington State chinook salmon from the lower reaches of the Columbia River were exposed to trace levels of hexavalent chromium, the DNA damage analysis clearly showed a dose response effect at 24 and 54 micrograms/litre whereas all the physiological parameters showed no effect until the dose was increased to 120 and 266 micrograms/litre. A similar phenomenon occurred when juvenile chinook salmon near Prince George were exposed to 2%, 4%, 8% and 16% pulp mill effluent over 30 days. The dose response effect for DNA damage started at the 2% exposure level while the physiological parameters did not start to show an effect until the 8% exposure level.

Sometimes the assay can give apparently anomalous results such as during the investigation of wild caught juvenile chinook salmon in the vicinity of the aluminum smelter at Kitimat. The tested wild controls and wild (presumed) exposed populations indicated results as expected, but the hatchery populations which were initially considered as controls showed by far the greatest amount of DNA damage. A subsequent analysis of the feed indicated an unusually high level of polycyclic aromatic hydrocarbons which would have accounted for the unusual spike in DNA damage.