By Dr. Dade W. Moeller
Posted: Wednesday, January 1, 1997
ARTICLES
Publication Date: January 1, 1997
Some "experts" are blaming an array of environmental substances for numerous diseases and planetwide phenomena. In The Enemy Within: The High Cost of Living Near Nuclear Reactors (1996), for example, Jay M. Gould, Ph.D., claims that releases of radioactive materials from nuclear power plants have had serious, wide-ranging adverse effects.
An effective approach to challenging such claims one used more and more in the field of law is to use the principles of "good science" to evaluate them.
The Principles of "Good Science"
All true science is based on the scientific method, which encompasses certain error-preventing principles. One can effectively evaluate claims by asking questions that reflect these principles:
1. Have the investigators or claim-makers (I refer to both herein as "CMs") detailed the research methods they used? The classical approach to fact-finding begins with the formulation of a testable hypothesis an educated guess framed in precise language. A hypothesis that is untestable "Angels can fly," for example is useless in a search for facts. Good hypotheses stem from repeated, objective observations. The quality of the hypothesis determines the feasibility of experimentation.
2. Have the CMs proposed, in scientific terms, a plausible explanation for the alleged cause-and-effect relationship? "The devil [or toxic waste] made me do it" has little or no explanatory value. Buzzwords, generalities, and circular statements for example, "Acupuncture treatment influences the flow of vital energy" may have promotional utility, but they only muddle attempts to explain.
3. Did the CMs factor into their conclusions all the relevant data? Conclusions or claims that stem only or preferentially from data consistent with preconceived notions are not credible. Any credible theory must be consistent with all relevant data. Scientists must adapt theories to conflicting data; they must not pick and choose data to enhance a pet theory.
4. Did the CMs base their claims on data (facts) from scientific studies or on subjective reports (unfounded testimonials, for example), misinformation, or disinformation (propaganda)? Anecdotes (subjective reports) and opinions, even expert opinions, have limited value in a search for facts. Often, opinions are acceptable expedients to filling informational gaps. But the consensus opinions of organizations within the scientific community are almost always more reliable than the views of fringe researchers even those who write bestsellers.
5. Did the investigators determine whether the "data" they used were reliable? Often, the conclusions of investigators stem from "data" that is, what at least appear to be data generated by other investigators. Whether scientific "data" are actualities facts depends on the currentness of the data, on experimental design, and on the degree of compliance with that design. The careless execution of defective study designs cannot generate useful data; rather, it generates nondata that are mistakable for facts. To discover the number of animals in a locality, for example, a researcher must define "animals." If the study proceeds without a clear-cut definition, the resulting count could include creatures the researcher had not intended to be tallied, such as humans or protozoans.
6. Did the CMs consider all rational attempts at interpreting the data and rule out attempts that were obsolete or intrinsically deficient (because of miscalculations, for example)? Selective "interpretation" of data may promote an agenda, a belief, or a standpoint, but it adds nothing to knowledge.
7. Did the CMs factor into their conclusions other analyses credible analyses of the issue? In other words, do their conclusions jibe with analyses published in respectable scientific journals? True scientists compare theories. Science is not an accumulation of conclusions but a process that demands resolving discrepant findings and conclusions.
8. Did outside experts in the field of the CMs evaluate their work before the CMs publicized their conclusions or claims? Such evaluation is termed peer review. True scientists do not bypass peer review, because it can pinpoint blind spots in their research.
An Example of Nonscience
According to The Enemy Within, low-level radiation:
* has made certain bacteria, including those that cause tuberculosis, resistant to antibiotics;
* has "mutated the spirochete" responsible for Lyme disease;
* is responsible for AIDS, which he terms "a newly mutated sexually transmitted disease";
* has increased the incidence of breast cancer, prostate cancer, pneumonia, and septicemia (blood poisoning); and
* is responsible for a decrease in infant birth weights, a decrease in SAT scores, and "an alarming rate of decline" in the number of species in tropical rain forests..
Let's consider Dr. Gould's claims in the light of "good science":
* In his book, Gould neglects to put forward a testable hypothesis for any of his claims (see question 1). He thus complicates any attempt to disprove them. For example, he states that low-level nuclear radiation from man-made sources, unlike radiation from natural sources, weakens human immune responses. This claim is analogous to attributing more harmfulness to water that has turned to ice in a refrigerator than to identical water frozen (under otherwise equal conditions) in a pan outdoors by winter weather.
* He also neglects to clarify the cause-and-effect relationships he proposes (see question 2). Generally, he simply states that radioactive materials were released into the environment and that the incidence of a particular disease increased. He leaves to his readers' imaginations such factors as radiation doses.
* Gould supports some of his claims anecdotally or nebulously (see question 4). For example, he writes that "many" workers at facilities operated by the U.S. Department of Energy are "dying prematurely of cancer"; and he ascribes other effects to "unreported releases" of extremely short-lived radioactive materials that "may have occurred."
* He does not discuss conclusions discrepant with his own (see question 6). For example, he blames nuclear radiation for an increase in the resistance of certain pathogens to antibiotics without referring to any of the many studies of antibiotic resistance that have uncovered factors unrelated to nuclear radiation.
* In his discussion of radiation and breast cancer, Gould does not account for all the relevant data (see question 3), does not address other possible explanations (see question 6), and does not deal with whether other analyses support his conclusions (see question 7). Similarly, he notes that there were more AIDS-related deaths in May 1986 than in May 1985 and ascribes this increase to the accident at the Chernobyl nuclear plant; he does not discuss whether the increase might have been due to a difference in the number of diagnosed AIDS cases.
In support of his claim that releases of radioactive material from commercial nuclear power plants have led to an increase in breast cancer in the United States, Gould cites an interoffice National Cancer Institute memo from Dr. Charles E. Land dated January 6, 1995. Although Dr. Land, a longtime member of the NCI's Radiation Epidemiology Branch, did state in the memo that Gould's calculations were correct, he also:
* conveyed the NCI's conclusion that Gould's study "has little epidemiological credibility"; and
* said that the institute's reanalyses of Gould's data "in no way support" his conclusions.
If Gould had submitted any chapter of his book for publication in a scientific journal, the journal editors would have required peer review (see question 8) of the submission. In my opinion, the outcome for any of Gould's chapters would have been outright rejection.
Science or Science Fiction?
All truly scientific endeavors involve certain error-preventing principles. The results of activities that do not conform to all of these principles are not necessarily scientifically useless, but they are at best unreliable.
Dade Moeller, Ph.D., is Professor Emeritus, Harvard University, and the founder of Dade Moeller & Associates, Inc., a group of specialists in environmental and occupational sciences.
C. E. Land, "Comments on J. M. Gould and E. S. [sic] Sternglass Manuscripts." Bethesda, Md.: National Cancer Institute; memorandum to Devra L. Davis, Office of the Director (January 6, 1995).
Scientific Peer Review and Conflict of Interest
A scientific peer review is an attempt by scientists to review, critically and objectively, research that is, or at least purports to be, scientific. Usually, the purpose of peer review is to determine whether work merits publication, but the procedure is also used to evaluate published research relevant to the establishment of public policies.
In an editor's note in the January 1997 issue of The Networker, the Internet newsletter of the Science and Environmental Health Network, Carolyn Raffensperger argued, "Allowing scientists with conflicts of interest to serve as peer reviewers undermines the scientific process: dollars speak louder than data." She thus implied that scientists who receive funds from an industry have biases that disqualify them as peer reviewers. Raffensperger did not, however, define "conflicts of interest."
Alleging conflict of interest is a double-edged sword. Many grassroots environmentalist organizations rely on purported scientific experts whose court testimonies on behalf of such groups account for much of their income. Does such remuneration sully these witnesses any less than industrial salaries tarnish scientists? Should we try to bar both "career" expert witnesses and industry scientists from participating in scientific peer review?
The same issue of The Networker included an article by Jim Schermbeck, a member of the grassroots environmentalist group Downwinders at Risk. Schermbeck recommended that grassroots organizations seek "sympathetic scientists for collaboration with grassroots activists" and create their "own pool of scientific experts." Is this not a recommendation for conflict of interest to wit, interest in advancing a cause versus interest in objectivity?
Cindy F. Kleiman, M.P.H.
(From Priorities, Vol. 9, No. 1)
