There are many nits that a reader could pick with the third edition of the Reference Manual, but one non-trivial issue is raised by the epidemiology chapter’s pronouncement that:
“Epidemiology is concerned with the incidence of disease in populations, and epidemiologic studies do not address the question of the cause of an individual’s disease178.”[1]
According to the Manual’s authors, the so-called specific-causation question is “beyond the domain of the science of epidemiology.” Epidemiologists do not investigate whether “an agent caused a specific plaintiff’s disease.”[2] The chapter insists that “[t]his question is not a question that is addressed by epidemiology. Rather, it is a legal question with which numerous courts have grappled.”[3]
Later on in the chapter, the authors repeat their opinion when they insist that the use of a threshold relative risk is “not epidemiology or an inquiry that an epidemiologist would undertake.”[4]
Strictly speaking the authors are correct that an epidemiologic study itself typically does not address the question individual causation. The authors of an epidemiologic study have no one person in mind, as does the factfinder in a civil action. The notion that epidemiology as a scientific discipline does not address the question of individual causation, however, seems just wrong. Tellingly, the authors’ footnote pointed to case law and a law review article, and not a single scientific source. The chapter does reference another legal source, the Third Restatement, which repeats the gist of the epidemiology chapter’s categorical statement:
“Scientists who conduct group studies do not examine specific causation in their research. No scientific methodology exists for assessing specific causation for an individual based on group studies. Nevertheless, courts have reasoned from the preponderance-of-the-evidence standard to determine the sufficiency of scientific evidence on specific causation when group-based studies are involved.”[5]
The chapter’s broad, sweeping characterization fails to consider:
- genome-wide association studies that may identify genes or mutations that are highly penetrant and which identify a causal association in the carriers of the genes; and
- epidemiologic studies that identify associations, shown to be causal, with risk ratios sufficiently great (say over 20) such that virtually all cases of the studied outcome would be avoided by removing the exposure that gave rise to such a large risk ratio; and
- epidemiologic studies that allow causal associations to be inferred in limited sub-populations, with sufficiently high relative risks that the studies support specific causation opinions.
Perhaps even more important than the above counter-examples, the chapter authors ignore the myriad instances in which epidemiologic studies and clinical trials, and analyses of both, directly inform clinical judgments about individual patients or subjects. Clinicians use studies of groups, such as clinical trials, to identify therapeutic benefits from medications and other interventions. These data directly inform their prescription decisions for individual patients, or their decisions to recommend surgical or other medical interventions to individuals.[6] The classic text on medical decision making describes how group-level data provide the basis for individual clinical decisions on therapy:
“1.5 How do I choose among several risky treatment alternatives?
Choosing among risky treatment alternatives is difficult because the outcome of most treatments is uncertain: some people respond to treatment but others do not. If the outcome of a treatment is governed by chance, a clinician cannot know in advance which outcome of the treatment will result. Under these circumstances, the best way to achieve a good outcome is to choose the treatment alternative whose average outcome is best. This concept is called expected value decision making.”[7]
Medical practitioners and scientists frequently use epidemiologic data, based upon “group-based data” to make individual diagnostic judgments. The inferences from group data to individual range abound in the diagnostic process itself, where the specificity and sensitivity of disease signs and symptoms are measured by group data. Physicians must rely upon group data to make prognoses for individual patients, and they rely upon group data to predict future disease risks for individual patients. Future disease risks, as in the Framingham risk score for hard coronary heart disease, or the Gale model for breast cancer risk, are, of course, based upon “group-based data.”[8] A search on the phrase “prediction models” yielded 662,297 results in the National Library of Medicine PubMed database.
The epidemiology chapter has taken its position on the irrelevance of epidemiology to specific causation through all three editions of the Reference Manual. Perhaps its authors will rethink their dogma in the fourth.
[1] RMSE3d at 608. The internal footnote, 178, pointed to: “178. See DeLuca v. Merrell Dow Pharms., Inc., 911 F.2d 941, 945 & n.6 (3d Cir. 1990) (“Epidemiological studies do not provide direct evidence that a particular plaintiff was injured by exposure to a substance.”) (emphasis added in this post). The emphasis in the quote is mine because the DeLuca court suggested by implication that epidemiologic studies provided the basis for inferences about specific causation. The footnote also cited another case that simply cited an earlier edition of the Manual, which hardly advances the inquiry into whether the Manual was correct in the first place. See In re Viagra Prods. Liab. Litig., 572 F. Supp. 2d 1071, 1078 (D. Minn. 2008) (“Epidemiology focuses on the question of general causation (i.e., is the agent capable of causing disease?) rather than that of specific causation (i.e., did it cause a disease in a particular individual?)” (quoting the second edition of this reference guide)). Finally, the Manual cited a state court case, In re Asbestos Litig,, 900 A.2d 120, 133 (Del. Super. Ct. 2006), and a law review article, Michael Dore, “A Commentary on the Use of Epidemiological Evidence in Demonstrating Cause-in-Fact,” 7 Harv. Envtl. L. Rev. 429, 436 (1983).” Admittedly, the chapter’s position can be found in writings of other legal commentators, although repetition hardly makes it any less wrong. See, e.g., Andrew See, “Use of Human Epidemiology Studies in Proving Causation,” 67 Def. Couns. J. 478, 478 (2000) (“Epidemiology studies are relevant only to the issue of general causation and cannot establish whether an exposure or factor caused disease or injury in a specific individual.”); Melissa Moore Thomson, Causal Inference in Epidemiology: Implications for Toxic Tort Litigation, 71 N.C. L. Rev. 247, 255 (1992) (“statistic-based epidemiological study results should not be applied directly to establish the likelihood of causation in an individual plaintiff”);
[2] RMSE3d at 609 & n.179. Inconsistently, the epidemiology chapter reports, without criticism, that some courts have allowed expert witnesses to opine about specific causation, without detailing how epidemiologic science was helpful to the specific causation issues. See RMSE3d at 609 n. 181 (citing Ambrosini v. Labarraque, 101 F.3d 129, 137–41 (D.C. Cir. 1996); Zuchowicz v. United States, 870 F. Supp. 15 (D. Conn. 1994); Landrigan v. Celotex Corp., 605 A.2d 1079, 1088–89 (N.J. 1992)).
[3] RMSE3d at 609.
[4] RMSE3d at 611 n.186.
[5] See Restatement (Third) of Torts: Liability for Physical and Emotional Harm § 28 cmt.
c(3) (2010) (cited at RMSE3d at 610 n.182).
[6] See, e.g., Robert H. Fletcher, Suzanne W. Fletcher, and Grant S. Fletcher, Clinical Epidemiology: The Essentials (5th ed. 2015) (treating prognosis, diagnosis, treatment, and prevention as topics of consideration, all involving individual patient decisions, in clinical epidemiology); Grobbee & Arno W. Hoes, Clinical Epidemiology: Principles, Methods, and Applications for Clinical Research (2d ed. 2015).
[7] Harold C. Sox, Michael C. Higgins & Douglas K. Owens, Medical Decision Making 6 (2d ed. 2013).
[8] See, e.g., Ewout W. Steyerberg. Clinical Prediction Models: A Practical Approach to Development, Validation, and Updating (2d ed. 2019).
