Toxic substances statistical analysis

Prior to 9/11, an assessment of chemical plant site security by the Agency for Toxic Substances and Disease Registry (ATSDR) was considered by many to be the most comprehensive analysis that was publicly available. ATSDR researchers reviewed national statistics on domestic terrorism compiled by the FBI in 1995 and interviewed security staff from facilities and potential targets in one community with numerous chemical plants. Reviewers found the following concerns ... 

On-line system. Provides support for exposure assessments of toxic substances. Includes chemical property estimation techniques, statistical analysis, multimedia modeling, and graphics display (including models)... 

Teratogenic effects of substances on test animals are evaluated from resorption (death of the conceptus), fetal toxicity (reduced body weight), aberrations (malformations), and minor anomalies. Statistical analysis is performed by surveying the four most important parameters namely, the number of litters with malformed fetuses, increase in the average number of fetuses with defects per litter, the number of resorptions, and dead fetuses. Finally, the incidence of malformation (response) is plotted against doses administered. Any dose-response relationship should indicate the teratogenicity of the chemical under experimental conditions. 

Measures of toxicity are from two routes. Firstly, data from laboratory testing on animals by direct injection, oral or chemical administration and is usually expressed as an LD50 value. This is the dose that causes death to 50% of the population after 14 days expressed as mg or fig of substance per kg body weight. Tests are performed under standard conditions and subject to rigorous statistical analysis. The lethal concentration (LC50) unit is used for hazardous airborne concentrations and is defined as the 4-hour inhalation of which causes death of 50% of the test group within the 14 day observation period. 

---