In converting the article to HTML the endnotes were lost please let me know if you wish me to mail you the complete article. 520-297-9443, danaise@uswest.net
In 1993, Robert Joiner brought his suit against General Electric Company and Monsanto Corporation ("defendants"), seeking damages for personal injuries from lung cancer allegedly caused by his exposure to electric transformers contaminated by polychlorinated biphenyls (PCBs). There is an overwhelming medical data linking PCBs exposure to cancer. Several regulatory and advisory agencies including the Environmental Protection Agency (EPA) and the National Institute of Environmental Health Sciences declared PCBs carcinogenic. In 1978, Congress banned the production and sale of PCBs because they "present an unreasonable risk of injury to health or the environment."
The United States District Court for the Northern District of Georgia (the "District Court") granted summary judgement for defendants. The District Court did not conduct a hearing or independently review the scientific studies relied on by the parties. . It acknowledged that Joiner's experts were prominent experts with international reputations, but based on the limited scientific information proffered, the court was not persuaded by the Plaintiff's experts' conclusions. In doing so the Court departed from the liberal admissibility guidelines established in Daubert. It implicitly held that the "gate-keeping" function of the Court gives the judge, not the jury, the power to dismiss a plaintiff 's case if the judge does not agree with plaintiff's experts’ conclusions rather than the expert's scientific reputation or methodology.
The United States Supreme Court (the "Court"), has affirmed the Joiner court ruling. In doing so it rejected the traditional test used to defeat a motion for summary judgement. The Court acknowledged that in a summary judgement motion, disputed issues of fact are resolved against the moving party, but reasoned that the question of admissibility of expert testimony is not an issue of fact and is thus reviewable under the abuse of discretion standard. After Joiner, the access of a Plaintiff to a jury trial has thus been severely curtailed.
What motivated the Court's radical departure from the established judicial history of liberal admissibility of evidence and staunch protection of the right to a jury trial? Some members of the Supreme Court were obviously concerned about runaway jury awards in toxic tort cases. Justice Breyer in his concurrence argued that manufacturers need protection from large jury awards and thus judges must exercise their gate-keeping function to curb tort liability.
In preparing this review I studied the legal briefs, expert depositions and the scientific studies proffered to the Joiner court. As a lawyer I was concerned about the Court's decision to rule on the merits of the case without an opportunity to hear the experts or read the scientific articles. As a scientist I was concerned that several key scientific data about cancer biology essential to understanding toxic tort claims were unavailable to the court.
Judges are not scientists. Unaided by experts and confronted with complex scientific questions, they must rely on whatever personal scientific background they may have. To defeat a summary judgement motion after Joiner the lawyer must therefor thoroughly brief the court about the scientific methodology and basic science, which underlie the parties' positions. This article attempts to provide, as an example, background information which may be helpful to the court in a toxic tort case like Joiner.
THE WEIGHT OF EVIDENCE METHODOLOGY
The Joiner Court reviewed only four studies proffered by
the Plaintiff and rejected each one for some flaws. In evaluating and rejecting
each study proffered, rather than the testimony as a whole, the Court departed
from its admissibility criteria established in Daubert which required
that the admissibility of evidence should be established by preponderance of
proof.
Under the preponderance of the evidence test, the Joiner court needed to look at all the evidence proffered as a whole and decide if such evidence could lead to the conclusion that the defendant was more likely than not liable for the plaintiff's injury. Instead, the Joiner court has looked at each study as if it was the only study proffered and demanded that each study meet the test of statistical significance. Statistically significant scientific results reflect 95-99% certainty that the results are not coincidental. By holding that only statistically significant studies are admissible, the court implicitly rejected the " weight of the evidence methods" used by most environmental scientists including the EPA and the World Health Organization. In doing so it created an almost insurmountable barrier to the admissibility of scientific studies involving toxic tort.
The weight of the evidence methodology is a systematic analysis of a given phenomenon looking at all available data including epidemiological studies, animal and human experiments and cell and tissue studies. Analysis is particularly difficult in Toxic tort cases. Toxic tort cases require scientific evidence to unveil the hidden link between injury and exposure to particular hazards. Cancer, birth defects, or other injuries begin insidiously, often becoming obvious many months or years after exposure. Many such ailments may have multiple causes, not all of which are known to science, and not all exposures to health hazards lead to injury. Although some exposures leave telltale biological markers in the body, plaintiffs normally rely on indirect evidence to verify that exposure caused harm.
Ethical and scientific limitations on human and animal research make epidemiological studies the preferred way to prove or disprove the harmfulness of a substance. Epidemiological studies, however, can rarely unequivocally point to a substance as the only causative harmful agent. To improve the sensitivity and specificity of a study, scientists rely on statistical analysis, most often the use of the relative risk index. Relative risk index is the ratio of incidence of disease (or death) among exposed individuals compared with the incidence among the control group. Thus, if 40 out of 1,000 members of the exposed group contract a particular disease, compared with 20 out of 1,000 members of the control group, the relative risk is 40 divided by 20, or 2.0. The Daubert Court held that for an epidemiological study to demonstrate causation under the preponderance of evidence standard, the relative risk must be greater than 2.0. The relative risk factor must also be statistically significant: that is, the possibility that the difference between the tested and the control group might be due to chance must be small. For example the odds that two tosses of a coin will produce all heads or all tails are 50/50. With three tosses, the odds of all heads or all tails are only one of four, and a fairly tossed coin is likely to land on the same side 25 times in a row only once out of 16,777,216 tosses. The larger the sample group, the smaller the risk of random error. To show statistical significance and allow for confounding causes not associated with the toxic agent in question, a sample size of several thousand subjects is needed. Studies on the effect of cholesterol on heart disease or the impact of breast screening on breast cancer mortality involved 50,000-100,000 subjects.
Analytic problems are often encountered where a study involves a small group of subjects at risk, such as exposure to a toxic agent in a manufacturing plant. A researcher can show that the mortality among the workers was twice that of the control (SMR=2) and thus clearly probative but due to sample size this may not meet the level of statistical significance associated with scientific certainty: 95-99%. Rather than reject the study, the scientific community will analyze the probative results of the study in the context of the applicable science as a whole. They will review similar studies by other scientists supporting the same conclusions and the paucity of published data negating the results of the study. The study results are then further bolstered by animal studies that show that in a controlled environment the agent in question does cause cancer. Most scientists, including scientists at the EPA and the World Health Organization when conducting a large-scale epidemiological study is not feasible have adopted this weight of the evidence methodology.
As an illustration of the difficulties posed in statistical analysis of toxic tort study note the following: Joiner proffered a study of Italian capacitor manufacturing workers who, like Joiner, were exposed to high concentration of PCBs in electric transformers. The study showed that cancer death rates among workers exposed to PCBs were significantly increased when compared to the local or national population. When the authors then analyzed death rates from lung cancer, the sample size of the male workers ( 540) was too small to establish statistical significance. The authors therefor concluded " Even the three cases of death from lung cancer were more than expected but not significantly so" . The district court seized on the wording " not significant " to reject the study. The court, admittedly not having read the Italian study, has misinterpreted the conclusions of the study. The study clearly showed that the Italian workers, both males and females, exposed to PCBs had a significantly higher mortality rate from cancer than the national or local population. The mortality rate from lung cancer in the larger women worker group ( 1540) also reached statistical significance. As to the smaller group of male workers, the difference between the observed mortality from lung cancer in the exposed group and the control did not reach statistical significance but demonstrated relative risk of 2.5. The Supreme Court has already determined in Daubert that relative risk of 2.0 is sufficient to meet the plaintiff's burden of proof in a civil case. Thus the study has shown beyond reasonable doubt ( 95% certainty) that exposure to PCBs is associated with increased mortality from cancer, and lung cancer in women. The study has also shown by the preponderance of evidence higher risk of mortality from lung cancer in male workers exposed to PCBs. In his dissent, Justice Stevens pointed to the unreasonable requirement placed on the plaintiff by the district court to admit only results that are statistically significant: "Moreover, it should be noted that under Georgia law, which applies in this diversity suit, Joiner need only show that his exposure to PCBs "promoted" his lung cancer, not that it was the sole cause of his cancer. "
CANCER BIOLOGY IN TOXIC TORT CASES
Defendants demanded that the Joiner prove that his lung cancer was directly caused by PCBs. In granting Defendant's motion, the court failed to recognize the distinction between agents which directly induce cancer to those which amplify its growth. Cancer arises when a single cell accumulates a number of mutations, usually over many years, and finally escapes from most restraints on proliferation. The mutations allow the cell and its descendants to develop additional alterations and to accumulate in increasingly large numbers, forming a tumor that consists mostly of these abnormal cells.
It is widely accepted by the scientific community that in chemically induced cancer, three phases evolve. They are initiation, promotion and progression. Initiation involves an irreversible genetic alteration. The transformed cell, however, may not replicate fast enough to be of clinical significance and thus initiation by itself is believed to be incapable of causing a tumor. Promotion, on the other hand, involves the expansion (increase in numbers) of the initiated cells into a benign tumor or pre-neoplastic cancer. Promoting agents do not damage genes but instead selectively enhance the growth of tumor cells or their precursors. Thus most cancers develop in response to a combination of both carcinogenic initiators and promoters. Progression relates to the events that take the pre-neoplastic initiated and promoted tumor into a malignant, carcinogenic state.
The distinction between carcinogenic initiators and promoters is important when one looks at the impact of environmental carcinogens (cancer promoting agents). Although the risk of a few types of cancer has declined dramatically in developed countries in this century, the incidence of the most significant forms of the disease has increased. Cancers of the lung, breast, prostate and colon have all become more frequent in developed countries as exposure to dangerous chemicals in the environment became more common. It is estimated that only a quarter of all cancers are considered "hard core cancers" These cancers would develop even in a world free of external influences, simply because of the production of initiating carcinogens within the body and the occurrence of unrepaired genetic mistakes. Epidemiologists have shown, however, that in most cases promoting carcinogens in the environment play a profound role. For example, one would expect that smoking will cause cancer of the lung, but it is less understood why smoking would cause cancer of the esophagus, bladder and pancreas and probably of the stomach, liver and kidney. Furthermore, it appears that smoking alone does not fully explain the high rate of cancer in smokers. Ecological studies indicate that lung cancer rates in polluted cities exceed those in rural areas. Such studies suggest that long-term exposure to high levels of air pollution could increase lung cancer risk by about 50 percent, especially among smokers. These studies thus lend support to Joiner’s position that the combination of smoking and PCBs, rather than smoking alone, caused his cancer.
To develop cancer, it is necessary for an external growth signal, a carcinogen, to reach the cell nucleus and effect uncontrolled replication. When a growth factor binds to a cell receptor, the receptor becomes transiently activated. A cascade of events is then generated signaling to the nucleus that a growth factor has been encountered and that cell proliferation should begin. Under normal situation the activation of the receptor is transient and the resulting cell proliferation is self-limiting. When Oncogenes are present, however, the receptor activation is prolonged and cell proliferation is continuous.
Oncogenes are cancer-promoting genes that are altered versions of the normal genes that control cell growth. These genes influence the production of growth factors that enable malignant cells to proliferate continuously. In most cases, two Oncogenes must cooperate to induce a fully malignant phenotype. Oncogenes are thus associated with different clinical stages of cancer, from initiation to progression to metastasis. A good example is the association of the RAS Oncogenes with carcinoma of the colon. Recent studies have demonstrated a relatively high incidence of RAS-Oncogenes mutations in carcinomas of the colon. Villous adenoma, a pre-malignant precursor of colon cancer, contains a significant but lower concentration of the RAS Oncogenes. These studies suggest that high concentration of the RAS Oncogene may be responsible first for the development of benign tumors and with increased concentration to their progression to malignant tumors.
These studies are also of great importance in explaining the carcinogenic effects of PCBs. Dr. Lucy Anderson a scientist at the prestigious National Cancer Institute has observed for some time that the environmental Nitrosamine causes cancer in mice. When the mice were additionally exposed to PCBs the incidence of liver and lung tumors doubled. The occurrence of multiple lung tumors was enhanced four folds by PCBs. The majority of liver tumors encountered when the mice were sacrificed at 58 weeks were pre-malignant adenomas. At 72 weeks, however, the majority of tumors were clearly cancerous suggesting a gradual progression from benign adenomas to malignant cancer in 26 weeks. Dr. Anderson's data clearly suggest that PCBs are powerful promoters of cancer once the cell is induced to transform by another agent (Nitrosamine). Cancer initiated by Nitrosamine alone is slow growing and many of the lesions appear to be precursors rather than full-blown cancer (adenomas). Upon exposure to PCBs, cancer, both lung and liver, become far more virulent and benign pre-malignant lesions become malignant.
The district court dismissed Dr. Anderson’s study because it showed PCBs cancer promotion in mice rather than man. It is a well-accepted scientific methodology to rely on mice for cancer studies. The mouse short life span allows for evaluation of cancer growth in a way that in human study may take 60 years or more. Dr. Anderson has shown that in 58 weeks 39% of mice exposed to PCBs developed cancer as compared to only 4% in the control group. 58 weeks in a mouse life are equivalent to 9-year latency period of developing cancer in humans after exposure to PCBs.
V. CONCLUSION
The Joiner court did not explicitly overrule Daubert; instead it enunciated a rule that appellate courts should be deferential to the district courts and use "abuse of discretion" as the sole standard of review. In dicta, the court encouraged the lower courts to tighten admissibility criteria for expert testimony with the hope of reducing the financial burden of large jury awards on industry. It is thus likely that increasing number of courts will grant defendant's motion for summary judgement after applying intense scrutiny to the plaintiff's experts' testimony. Most disturbingly, they will do so without scientist's input and solely on the basis of the legal briefs proffered by counsels.
Before Joiner the Plaintiff could defeat a summary judgement motion by showing that the experts have credentials, their methodology is sound and that " [r]easonable minds could differ as to the import of the evidence". After Joiner these arguments are no longer sufficient to defeat summary judgement motion in complex scientific litigation. The lesson from Joiner's defeat is that a motion in opposition to defendant's motion for summary judgement may be the only opportunity to present detailed scientific data to the court. Therefor such motion must include a detailed scientific presentation explaining in a language that a non-scientist can clearly understand why the expert testimony is sound and should be heard by the jury.
In converting the article to HTML the endnotes were lost. I will be glad to send you the entire article by mail
Pleases e -mail to danaise@uswest.net