Risk Analysis and Assessment

Risk Assessment

Personal, indoor, and outdoor environments contain obvious hazards such as hurricanes, earthquakes, tsunamis, oil tankers, landfills, and industrial facilities and less apparent ones such as tobacco products, alcohol, drugs, contaminated food, stairs, and slippery sidewalks. Risk assessment is a set of formal processes to estimate the likelihood of danger associated with hazards. Typically, the process is divided into four steps: (1) hazard identification, (2) dose-response assessment, (3) exposure assessment, and (4) risk characterization.

Hazard Identification

Is a substance or activity a hazard? Information comes from laboratory experiments that use bacteria, mice, rats, and other specially bred species to test biological, chemical, or physical hazards and activities for their toxicity, mutagenicity, car-cinogenicity, and other undesirable effects. Data also come from observing the consequences of human exposures. Both laboratory tests and human studies are valuable, but there are instances where they have not worked. For example, the chemical diethylstilbestrol, developed to treat morning sickness in pregnant women, had been tested in laboratory animals, yet it caused birth defects in humans. Likewise, drawing inference from human observations is hindered by confounding factors, which make it difficult to separate the effects of different hazards. For example, diesel fumes are potentially carcinogenic, but human studies have found it difficult to separate the effects of diesel fumes from tobacco smoke and other occupation exposures.

Another complication is that some studies show that a substance is a hazard, but others do not. A recent widely publicized example is electromagnetic fields created around lines that carry electrical power across the landscape. Some studies show that electromagnetic fields can cause cancer. Others, including nearly all the recent ones, show no effect. Many hazard studies are inconclusive, especially if the substance or activity is a weak hazard, or a hazard for one animal but not for another.

Dose-Response Assessment

This second stage evaluates the amount of exposure that produces a harmful effect. Dose-response assessment is essential because harmful effects are detected at high doses. Most human carcinogens, for example, were detected in a workplace where employees were exposed for many years to high doses. Animal tests in laboratories are expensive, and hence, the number of animals tested may be limited; and the doses are high and enhanced by solvents that increase the chance of producing a cancer. If a chemical, biological, or physical agent is dangerous at high exposure levels, is the same agent dangerous at a low exposure? Risk assessors look for exposure levels where a substance is not a danger or for the so-called no observed adverse effect level, or NOAEL.

Thresholds are different for every chemical, biological, and physical agent. Many heated debates have occurred over which mathematical curves statisticians should use to estimate dose and [Page 2467]response. For example, the human body is able to repair cell damage caused by low-level exposure to radiation, which suggests that there is a threshold that produces no effect. In fact, some scientists believe that a small dose is beneficial (hormesis), triggering the immune system into an accelerated protective mode. But hormesis, which was shown to exist for food nutrients such as selenium and some other metals, is highly debatable in the case of radiation (also, e.g., cigarette smoke). Regulatory agencies assume that there is no threshold for radiation with regard to cancer; in other words, any exposure increases the probability of cancer. This is not the case for other outcomes (e.g., birth defects), which use threshold models.

...