Home Page

Online Casebook

Officers

How to Join

Return to RiskWorld

SRA Risk Science & Law Specialty Group

Berry v. CSX Transportation, Inc.: Admitting Epidemiological Evidence to the Jury

What is the reliability standard of human (epidemiological) evidence of disease causation or health risk necessary to keep "junk science" from being admitted as evidence to a jury?

In Berry, 709 So.2d 552 (Fla. App.1 Dist. 1998), a Florida District Court of Appeals reversed the trial court decision to deny admission of evidence that long exposure to elevated levels of organic solvents—TCA, TCE, PCE, and mineral spirits—caused toxic encephalopathy. This disease results in brain and central nervous system alteration and dysfunction. A pivotal issue was the standard for reliability of epidemiological studies that infer causation or risk.

The appellant review summarized the reliability issues and related standards of proof:

§ Epidemiological studies did not have to establish a causal relationship between a toxic agent and a disease to be admissible in toxic tort litigation, but merely had to demonstrate an association from which the probability of causation could be inferred.

§ The court was not required to engage in highly detailed critical analysis of each relevant study, including the relative risk, confidence interval, statistical significance, selection and information biases, and other criteria of causality appropriate to considering the sufficiency, not the admissibility of evidence.

§ A strong dose-response relationship, which is the cornerstone of animal (toxicological) studies, is not required in epidemiological studies, which are performed on a retrospective basis without specific or consistent data on levels of toxic exposure and dose.

§ Neither are other attributes of controlled animal experiments, such as measured exposure levels, nor carefully controlled double-blind clinical studies, used to test the safety of new drugs, required to reach an inference of causation.

Other indications from this decision suggest that (1) statistical significance is not a prerequisite to admitting an epidemiological study; (2) the relative risk, whether greater or less than 2.0, goes to the issue of evidentiary sufficiency, not admissibility, and must be presented to a jury to weigh; and (3) the presence of one or more negative epidemiological studies does not necessarily fault the admissibility of a group of studies.

Appellants argued successfully that:

Neither the trial court nor this court can assume the role of an amateur scientist, examine the materials upon which the expert scientists rely, draw its own scientific conclusion as to whether or not the materials support the opinions of the plaintiffs' experts or not, and then declare one set of opinions the victor by excluding the other set of opinions from evidence. See Joiner v. General Elec. Co, 78 F. 3d 524, 530-33 (11th Cir.1996).

This decision offers to help straighten out the confused standard of proof for admission of scientific evidence on disease causation and health risk, particularly regarding standard human population (epidemiological) studies which, together with clinical opinion, often are the mainstay of scientific evidence brought in toxic tort lawsuits.

[Contributed by Wayne Roth-Nelson]

CLASSIFICATION: CIVIL LITIGATION; TOXIC CHEMICAL INJURY; ADMISSIBILITY OF EVIDENCE FOR CAUSATION

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