Scientific Fraud and Bad Reporting

Cancer risk assessment

According to a 2022 study in the Netherlands, over the last three years, one in two researchers had engaged frequently in at least one 'questionable research practice', with 'not submitting or resubmitting valid negative studies for publication' being the most common practice. The fields of life and medical science had the highest prevalence (55.3%) of engaging questionable practices compared to other disciplines.

One example is cancer risk assessment:

Cancer risk assessment is a critical feature of environmental health and regulatory policy used by governments to determine the environmental risks of chemicals and radiation. It answers the question as to what the risk is, at the population level, of developing cancer at different exposure levels.

Professor Edward Calabrese of the University of Massachusetts, Amherst, has spent the last 20 years researching the historical foundations of cancer risk assessment, showing the fraudulent science behind the linear non-threshold (LNT) dose-response model adapted by practically all US agencies for chemical and radiation risk assessment.

He recently presented a comprehensive summary of the development of the linear non-threshold model of cancer risk assessment contained in about 100 papers in the peer reviewed industrial hygiene, toxicology, health physics, and radiation health scientific literature. Included are examples of scientific misrepresentations and misconduct that were offered to support the conclusions and perspectives offered. ⁽²⁾

The foundations of cancer risk assessment represents a century of significant uncorrected mistakes and scientific misconduct

He states: “The foundations of cancer risk assessment represents a century of significant uncorrected mistakes and scientific misconduct, dominated by powerful self-interests and politicized ideological actions involving the US National Academy of Sciences, Science journal, multiple Nobel Prize winners, and elite leaders of the field of radiation genetics, from the 1920s to the 1990s.”

• The linear non-threshold (LNT) model assumes that every increment of a chemical or radiation dose, no matter how small, constitutes an increased cancer risk for humans.
• The contrasting threshold dose-response model assumes an exposure level below which there is no increased cancer risk.

The impact of the LNT model guiding federal agencies is that every questionable molecule of a chemical or radiation must be eliminated, or public health is at risk. This has resulted in billions of dollars spent to eliminate chemicals to levels that have no impact on human health and the unnecessary removal of products, many of which are critical today.

The linear-non-threshold assumption exists with 'assumption' as the all-important word that needs to be taken literally. While no one disputes that high doses of radiation cause harm, no one has proof that low levels cause harm. Surprisingly, the scientists and government bodies that adhere to the LNT assumption will tell you that no proof of harm at low levels is even possible because the risk is too low to measure statistically. In the absence of proof, they say, the only prudent course is to play it safe by assuming that low levels or radiation cause harm, reports Lawrence Solomon. ⁽³⁾

Enhanced health in animals and humans from low dose ionizing radiation

Yet, here are some clearly contrary examples:

• T. D. Luckey estimates that there are more than 2,000 published research papers showing enhanced health in animals and humans from low dose ionizing radiation. ⁽⁴⁾
• A book by Charles L. Sanders titled Radiation Hormesis and the Linear-No-Threshold Assumption reports on a case of 'an almost perfect study in a human population that demonstrates the highly significant protective effects of near-continuous exposure to gamma radiation.' This case involved more than 180 apartment buildings that had been constructed in Taiwan in the early 1980s using recycled steel that was subsequently discovered to have been contaminated with radioactive cobalt-60. The 10,000 people who were housed there received large doses of radiation over a period of 9 to 20 years that, according to LNT theory, should have led to a total of 302 cancer deaths over the 1983-2003 period studied, 232 of which would have been ordinarily expected had no radiation exposure occurred, with the additional 70 stemming from the exposure. To the researchers' surprise, however, only seven cancer deaths were found, 225 fewer than would have occurred had the buildings been free of radiation. Instead of radiation increasing the death toll by 30%, it may have reduced the death toll by a staggering 97%. The number of birth defects among children born in this radioactive environment also confounded LNT theory. Instead of 48 defects expected, just three occurred.
• Researchers Maurice Tubiana and colleagues report that the LNT has resulted in medical, economic and other societal harm.. Advances in radiation biology in the last two decades, and the discovery of defenses against carcinogenesis render the LNT obsolete. (5)

Literally hundred of thousands of medical workers exposed to frequent low-level radiation have experienced similar positive health benefits

In industrialized countries, we now treat more than one million patients with radiotherapy, including 50 percent of all US cancer patients, with obvious positive results. Literally hundred of thousands of medical workers exposed to frequent low-level radiation have experienced similar positive health benefits. ⁽⁶⁾

In spite of the science, governments continue to use the LNT model. Misuse of this model has produced spending in excess of $1 trillion in the United States alone for negligible health benefits just for governmental environmental cleanup programs, while truly significant measures that would protect public health remain unfunded. It's long overdue that serious reconsideration be given to the LNT model.

Another issue: various studies over the last decade or so have shown that journalists, while all to eager to participate, are often not the source of misreporting of science news. Much of it can be traced back to the press release, and even to study authors themselves. ⁽⁷⁾

One study looked at this body of research, looking specifically at the science titles and their effect on journalistic coverage. Specifically the authors reviewed scientific studies about Alzheimer's disease (AD). Much of the AD research is conducted in mouse models. There is no no-human animal that gets AD. Yet a lot of the basic science research, therefore, depends on animal models, which reflects some marker or aspect of the disease that may tell us something about the biology of AD itself. ⁽⁸⁾

The authors report: "To this end, we analyzed a sample of 623 open access scientific papers indexed in PubMed in 2018 and 2019 that used mice either as models or as the biological source for experimental studies in AD research. We found a significant association between articles' titles and news stories' headlines, revealing that when author's omit the species in the paper's title, the writers of news stories tend to follow suit."

Exaggerated claims were most often traced to the press releases at the scientist's institutions

Specifically, if the article title declared that the study involved mice then 46.2% of news headlines did also. If the title did not mention mice, then only 10.4% did. This is a very large difference that traces directly back to the published science article itself.

While this is only one factor affecting the accuracy of science news reporting, it highlights that scientists themselves and journal editors, take significant responsibility for the quality of reporting about their own research. This goes beyond crafting the title of the article. Prior research has also found that when researchers exaggerate their own research in the abstract or discussion of their paper this is likely to translate to exaggerated reporting.

In another study, exaggerated claims were most often traced to the press releases at the scientist's institutions: "40% of the press releases contained exaggerated advice, 33% contained exaggerated causal claims, and 36% contained exaggerated inference to humans from animal research. When press releases contained such exaggeration, 58%, 81%, and 86% of news stories, respectively, contained similar exaggeration, compared with exaggeration rates of 17%, 18%, i and 10% in news when the press releases were not exaggerated." This is a huge difference. ⁽⁹⁾

All of this does not mean that medical studies are of no value or that health reports are always wrong. It simple serves as a warning that science is fluid, not static or absolute. It does suggest that every time you see a headline claiming that X causes cancer of that Y prevents it, some skepticism might be in order.

References

1. Susan Goldhaber, "The lasting impacts of scientific fraud," acsh.org/news, February 13, 2024

2. Edward J. Calabrese, “Cancer risk assessment: it's wretched history and what it means for public health,” Journal of Occupational and Environmental Hygiene,” March 7, 2024

3. Lawrence Solomon, “Radiation's benefits,” Financial Post, September 2010

4. T. D. Luckey, Radiation Hormesis, Boca Raton, CRC Press, 1991

5. Maurice Tubiana et al., “The linear-no-threshold relationship is inconsistent with radiation biologic and experimental data,” Radiology, 251, 6, 2009

6. Jay Lehr, “Low-level radiation benefits health,” Environment & Climate News, August 2011

7. Steven Novella, "The causes of bad science reporting," sciencebasedmedicine.org, June 16, 2021

8. Marcia Triunfol and Fabio C. Gouveia, "What's not in the news headlines or titles of Alzheimer disease articles?, " PLOS Biology, June 15, 2021

9. Petroc Sumner et al, "The association between exaggeration in health related science news and academic press releases: retrospective observational study," British Medical Journal, 2014;349; g7015