Home Page

Online Casebook

Officers

How to Join

Return to RiskWorld

SRA Risk Science & Law Specialty Group

Salmon River Concerned Citizens v. Robertson: Justifying Safety Factors in Risk Analysis

Can a government agency justify herbicide spraying based on a health risk analysis that provides a safety factor of 100 when extrapolating low-level human risks from laboratory animal studies?

In Salmon River Concerned Citizens v. Robertson, 32 F.3d 1346 (9th Cir. 1994), a public interest group challenged a Forest Service Environmental Impact Statement (EIS) that evaluated the health risks of herbicide spraying. The Ninth Circuit Court of Appeals found the risk assessment was adequate and contained standard risk assessment methodology.

The court approved the safety margins the agency built into the risk analysis to account for uncertainty in predicting the toxic effects of herbicides: (a) a safety factor of one order of magnitude (10 times) applied to the No-Observable-Adverse-Effects Level (NOAEL) measured in laboratory animals exposed at high levels to adjust for extrapolating human response at low levels of exposure; and (b) another safety factor of an order of magnitude applied to the NOAEL to allow for specially sensitive individuals.

The court did not accept plaintiffs' argument that the risk analysis failed to adequately discuss risks of herbicide exposure for persons with multiple chemical sensitivity syndrome (MCSS). The agency's safety margin of 100 (or two orders of magnitude) was considered sufficient to protect those individuals in view of a consensus in the scientific community that there is no definable, discrete MCS syndrome and it cannot be determined what causes a reaction in a chemically sensitive person.

Further, the Forest Service intentionally overstated the risks of harm posed by active ingredients to account for the uncertainty regarding the toxicity of ingredients, particularly inert substances, not previously studied by the EPA. Plaintiffs argued that two orders of magnitude were insufficient, and the EIS concedes that "unusually sensitive individuals may experience effects even when the margin of safety is equal to or greater than 100."

The position taken by the District Court, 798 F.Supp. 1434, 1442 (E.Dist.Cal. 1992) was:

An EIS need not quantify every risk, particularly less likely risks. Moreover, the risk analysis used by the government, [using] a margin of safety of 100, is the scientifically accepted method.

The Court of Appeals' standard of review was Oregon Envtl. Counsel v. Kunzman, 817 F.2d, 484, 491 (9th Cir. 1987):

[T]he court cannot substitute its judgment for that of the agency in question, as the agency has a measure of expertise in the area which the court lacks.

Thus, the District Court decision to deny a motion for summary judgment favoring the plaintiffs was affirmed.

This decision may affect how a regulatory risk assessment is performed to account for protecting sensitive individuals or subpopulations in a normal population, or how threshold limits of exposure are interpreted where plaintiffs in a civil lawsuit allege special sensitivity.

[Contributed by Wayne Roth-Nelson]

CLASSIFICATION: JUDICIAL REVIEW; REGULATORY AGENCY CHEMICAL RISK ASSESSMENT; DEFERENCE TO REGULATORS

Copyright © 1999 by Tec-Com, Inc.
Last modified July 29, 1999.