Safety assessment is the process that results in an "acceptable daily intake" (ADI) of specific chemicals. Safety assessment has been carried out in the United States and much of the Western world since the late nineteenth century. Since the 1950s it has become much more formalized in its use by regulatory agencies to incorporate animal toxicology studies and potential exposures. Safety assessment starts with the underlying premise that exposure to a chemical, be it a drug, food additive, cosmetics ingredients, or consumer product, will occur. It also relies on animal toxicology studies. The assumption made is that humans are more sensitive to these chemicals than the most sensitive animal species tested.
Acute toxicity studies are used to determine the potential for poisoning, as well as possible antidotes. The acute LD50 (that dose that is lethal to 50% of the animals tested) is seldom directly extrapolated from animals to man, but many occupational exposure standards and categorizations of household chemicals are based on the LD50. Safety assessment for direct food additives, such as colorants or flavors, is based on the no observable effect level (NOEL) in laboratory animals, or on an approximation of that level. The ADI is determined from the NOEL (with appropriate safety factors) and is based on the percentage of any dietary component that contains the chemical, so that if a person consumes a kilogram of food per day and the compound of interest is only found in 10 percent of the food products generally consumed, that 10 percent becomes the exposure maximum that is used in the calculations.
Safety assessment for therapeutic agents is also based on animal toxicity, but it is also a function of the doses to be used, the diseases to be treated, and the conditions of the treated populations. As an example, the safety assessment of an over-the-counter (OTC) drug is more stringent than for a drug to be used for the treatment of a life-threatening situation. A therapeutic index (TI) is developed for almost all drugs and is based on the ratio of the toxic dose to the efficacious dose: The greater the ratio, the greater the margin of safety in use of the drug. A drug being used for cancer chemotherapy can have a greater toxicity and a much smaller TI than a drug for the common cold. A cancer patient may be in a life-threatening situation, but a doctor or nurse is monitoring his or her signs and symptoms of toxicity. In contrast, an individual may be taking uncontrolled amounts of a cold preparation and is probably not being monitored by a health professional. The other issue in this example is that there is a much larger consumer population for OTC products than for chemotherapeutic products, and the risks of OTC drugs may not be as well appreciated. Hence, the "involuntary" risk taker must be more protected.
Consumer products present a different challenge for safety assessment. The assumption here is that the consumer of a household product will ingest or touch the product, or inhale vapors from the product. The product is manufactured for something other than consumption, and the packaging has to be part of the safety assessment. For example, a household cleanser may be caustic and highly oxidizing. This product will cause extreme damage to tissues, and the risk is known from animal studies or by analogy to other chemicals of the same class. The only way to assure less risk to the user is through strong, vivid, unequivocal labeling, and childproof packaging.
Pesticides present a special case in safety assessment. First, most pesticides are designed to control and/or kill pests. By definition, pesticides, especially insecticides, are moderately or highly toxic when compared to most other consumer products. The safety assessment and toxicology package for a pesticide depends on whether it is going to be registered as a crop chemical, a chemical for ornamental flowers, a home-use pesticide, or some other use. The crop chemicals, be they insecticides, herbicides, fungicides, or growth regulators, complete a battery of tests from in vitro through carcinogenicity. The logic being that consumers could be exposed to minute amounts of residues from application to the crops, and there must be a large margin of safety between potential exposure and toxicity. Again, an ADI is established based on the toxicity of the chemical, the NOEL, the shape of the dose-response curve, the amount of a particular crop in a food product, how much may be eaten or drunk, and the body weight or surface area of the consumer (children are special cases). This information allows the regulators to establish an ADI expressed in milligrams (or micrograms) of a chemical that can be consumed safely in a day. ADIs are also set for different durations of exposure, but usually for a greater part of the lifetime.
Safety assessment differs from cancer risk assessment in several ways. Safety assessments are for multiple endpoints, not just cancer, and can be for multiple time points. Threshold limit values (TLVs) are examples of safety assessments that can be set for acute toxicity, eye irritation, or systemic toxicity. Safety assessments arrive at an exposure or dose limit and are not expressed as a probability, which cancer risk assessments are. Safety assessments are similar throughout the world.