Abstract
Statistical data are frequently used, in the law and elsewhere, to establish or measure causal relations based on associations. The gold standard for doing this is the randomized, double-blind experiment. Here are the key features of such an experiment.
The original version of the book was revised. An erratum can be found at DOI 10.1007/978-1-4419-5985-0_15
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The term is something of a misnomer because what is computed is not a risk but the fraction of a risk attributable to a cause. Sometimes the term etiologic fraction is used, which is more accurate but less common. For convenience of reference we use the more common term.
- 2.
The incidence rate of a disease is a velocity, like the speed of a car, except that it is measured, not in miles per hour, but in the proportion of the population that newly contracts the disease in a relatively short time period. For example, the incidence of heart disease among male smokers 65–70 years old is about 0.1% per month, or 1.2% per year, assuming the incidence of heart disease is fairly constant over the year. If incidence varies over time, calculations become more accurate as the unit time interval shrinks to zero and the rate approaches the instantaneous incidence, a limit concept equivalent to the hazard concept discussed in Sect. 11.1, and analogous to the instantaneous velocity of a car.
The concept of incidence should be distinguished from that of prevalence, which is the proportion of the population that has the disease at a given time and thus includes the effect of current and past incidence and disease duration. Attributable risk calculations can be made in terms of incidence or prevalence, although incidence rates are more appropriate to disease prevention.
- 3.
See, e.g., Daubert v. Merrell Dow Pharmaceuticals, 43 F.3d 1311 (9th Cir. 1995) (Bendectin and birth defects); Hall v. Baxter Healthcare Corp., 1996 U.S. Dist. Lexis 18960 at p. 15 (D. Ore. 1996) (silicone breast implants and connective tissue disease).
- 4.
See Landrigan v. Celotex Corp, 127 N.J. 404, 605 A.2d 1079 (1992) (1.55 relative risk of colon cancer from exposure to asbestos plus an absence of other causal factors is sufficient for plaintiff’s proof of causation).
- 5.
In re Joint Eastern and Southern District Asbestos Litig. (John Maiorana), 964 F.2d 92, 97 (2d Cir. 1992) (“[P]laintiff did not need to provide epidemiological evidence of a certain magnitude [presumably referring to an RR ≥2.0] in order to defeat a summary judgment motion because she did not rely on epidemiological studies alone.”)
- 6.
In comparing two populations, adjustments may sometimes be needed for differences in age distribution. In the direct method of adjustment, a weighted average is computed from age-specific rates, where the weights are the proportions of people in each age bracket in the reference population.
- 7.
See, e.g., cases cited in note 3 at p. 302; see also In re Agent Orange Liability Lit., 597 F. Supp. 740, 787–795 (E.D.N.Y. 1984) and 611 F. Supp. 1223, 1231–1234 (1985), aff’d in rel. part, 818 F.2d 145, 171–173 (2d Cir. 1987) (Agent Orange and a variety of illnesses).
- 8.
Some of the hypothetical examples given below are from Mitchell, Boyce & Wilcox, Biomedical Bestiary: Epidemiological Guide to Flaws and Fallacies in the Medical Literature (1984).
- 9.
After Bradford-Hill, The environment and disease: association or causation?, 58 Proc. Roy. Soc. Med. 295–300 (1965).
- 10.
A unit of absorbed radiation in which 100 ergs of energy are absorbed per gram of tissue. By way of comparison, background radiation delivers about 0.3–0.4 rads per year to humans, most of it from radon.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
Finkelstein, M.O., Levin, B. (2015). Epidemiology. In: Statistics for Lawyers. Statistics for Social and Behavioral Sciences. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-5985-0_10
Download citation
DOI: https://doi.org/10.1007/978-1-4419-5985-0_10
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4419-5984-3
Online ISBN: 978-1-4419-5985-0
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)