Monday, July 2, 2012

How to Bias Studies on the Biological Effects of Radiation While Avoiding Internal Validity Errors

Follow Up on "Debating Radiation Safety"

The debate on the biological effects of radiation is incredibly political since findings have regulatory implications. The discussion below will demonstrate that laboratory and fieldwork findings are contingent upon a range of variables, especially time frame.

Findings on the biological effects of radiation dose from laboratory studies on animals vary considerably depending upon:

a. the type of radiation the animals are exposed to

b. the length of time of exposure

c. the time frame after which the animals are killed in order to analyze their cells and/or DNA. 

Studies that expose the animals only to gamma and then kill them immediately after exposure find far less biological damage than studies that follow the animals across greater lengths of time after exposure.

Furthermore, studies that look at internal ingestion of radionuclides find far greater health effects, including death.

Recent laboratory research studies on the effects of radiation explain some of these divergences in findings.

The “bystander effect” and “delayed effect” describe the phenomena whereby cells not directly targeted by the radiation can exhibit damage and produce damaged progeny over time.

Furthermore, the research on alpha particles demonstrates conclusively that transversal of a single alpha particle can break DNA bonds. 

Let us now look briefly at studies of internal ingestion of plutonium during World War II.


The Manhattan Project was not restricted to development of the nuclear bomb. It also entailed research on the biological effects of radiation given military recognition that fallout would accompany the nuclear blast. Research studies on the effects of injected plutonium on human health are particularly troubling.

The United States Atomic Energy Commission, established by the Atomic Energy Act of 1946, sponsored experiments between 1945 and 1947 that involved injecting eighteen human subjects with plutonium to calibrate body burden (Walker Permissible Dose, p. 16).

The researchers did not anticipate the subjects would benefit from the studies and deliberately withheld information about the nature of the studies from the patients.(Walker p. 16). 

Paul Langley studied the details of Joe Hamilton’s work on the health effects of plutonium studied as part of the Manhattan Project.

The study ”The Metabolism of the Fission Products”  was later presented at the Thirty-second Annual Meeting of the Radiological Society of North America, Chicago, Ill., Dec. 1–6, 1946.  Hamilton’s concluded from his research that ingestion of plutonium is significantly more damaging than external exposure to beta or gamma irradiation:

[excerpt] The most important hazard that arises from the release of nuclear energy are radiations produced directly from fission and subsequently emitted by the resultant fission products and plutonium. The fission products can produce injury either as an external source of radiation or, if they gain entry into the body, by acting as an internal radioactive poison, quite analogous to radium poisoning. This latter consideration is a major concern, since the amounts required within the body to produce injurious effects are minute compared to the quantities necessary to induce damage by external beta and gamma irradiation.

The findings were clear: ingestion of alpha emitters is a clear and demonstrable threat to health in quantities far lower than what is required to produce the same biological damage from external radiation.

The U.S. Atomic Energy Commission knew without doubt that ionizing radiation posed significant health risks. It also knew the public would be alarmed if the agency’s research findings were disclosed.

In The Name of Science, Andrew Goliszek describes a classified AEC document dated April 17, 1947 titled “Medical Experiments in Humans” that reads: “It is desired that no document be released which refers to experiments with humans that might have an adverse effect on public opinion or result in legal suits. Documents covering such field work should be classified secret” (p. 126).

see also


Since I've posted on this subject I'm not going to re-write my previous analysis. In the following blog posts you can read about how researchers can introduce bias by inappropriately generalizing their findings.

The study criticized in these posts exposed animals to gamma only and killed them immediately after exposure. The researchers then argued (absurdly) that their findings of "no effects" should be generalized to fallout from a nuclear plant accident.

April 2, 2012 Environmental Health Perspectives Editorial on Evaluating Low Dose Effects of Chemicals (with Implications for Ionizing Radiation)

April 30: A Plant in Environmental Health Perspectives

May 17 The Plant Erupts into a Poison Fruit

June 12: Poisoning the Well


 As explained here by Huang, Snyder and Morgan in “Radiation-induced genomic instability and its implications for radiation carcinogenesis”[i]:
The biological effects of ionizing radiation in mammalian cells include gene mutation, chromosomal rearrangement, cellular transformation, cell death and carcinogenesis. For many years, the central dogma in radiobiology has been that the nucleus, specifically the DNA, is the principal target for the biological effects of radiation. Following irradiation, the initial radiation-induced DNA damage is converted into a mutation or chromosomal aberration during subsequent DNA repair and is expressed by the irradiated cell and its progeny. Over the last 10 years, evidence has accumulated demonstrating that those same deleterious effects can occur in the progeny of irradiated cells at delayed times after radiation exposure
Two of the most important effects of exposure recently explicated are “bystander effects” and “delayed effects.”[ii]  Both of these effects can produce genomic instability. These effects were discovered as laboratory analysis examined the effects of ionizing radiation on both DNA repair and cell death. 
           It was discovered that “bystander” cells not directly targeted or irradiated by radiation were often impacted nonetheless.  In particular, bystander effects have been detected as occurring at low doses of exposure: “At lower levels, some or all of the effects are likely to have been initiated not by direct radiation effects on the cell, but by the bystander effect, in which radiation damage to one cell can lead to biological changes in surrounding cells. . .”[iii]
Effects on bystander cells often take time to develop, they are therefore “delayed effects.” Genomic instability results from these bystander and delayed effects:
“Genomic instability is an all-embracing term to describe the increased rate of acquisition of alternations in the genome. Radiation-induced instability is observed in cells at delayed times after irradiation and manifests in the progeny of exposed cells multiple generations after the initial insult. Instability is measured as chromosomal alterations, changes in ploidy, micronucleus formations, gene mutations and amplifications, microsatellite instabilities, and or decreased plating efficiency. . . . there are multiple pathways for initiating and perpetuating induced instability  (Morgan, Non-Targeted and Delayed Effects, p. 567).
In effect, the researchers discovered that cells that initially survive irradiation are often instable. New cells produced by the instable cells are also affected. Thus, radiation-induced genomic instability can result in a “cascade of genomic events that increase the rate of mutation and chromosomal change in the progeny of that irradiated cell”; however, the precise signaling events between cells that initiate and perpetuate the instability remain undisclosed.

[i]           Lei Huang, Andrew R Snyder, and William F Morga.  Radiation-induced genomic instability and its implications for radiation carcinogenesis. Oncogene (2003) 22, 5848–5854. P., 5858.

[ii]           William F. Morgan. Non-targeted and Delayed Effects of Exposure to  Ionizing Radiation: I. Radiation Induced Genomic Instability and Bystander Effects In Vitro. Radiation Research, 159, 567-580.

[iii]              Antony M. Hooker, Madhava Bhat, Tanya K. Day, Joanne M. Lane, Sarah J.Swinburne, Alexander A. Morley and Pamela J. Sykes. The Linear No-Threshold Model Does Not Hold for Low-Dose Ionizing Radiation. Radiation Research, 162, No. 4 (Oct., 2004), pp. 447-452.


  1. Fascinating post. Have you ever heard of Irwin Bross? He strongly criticized animal "safety" testing.

    'In addition to hiding the hazards of dangerous pharmaceutical drugs and vaccines, animal tests are also used to hide environmental hazards from the public. Irwin D. Bross, Ph.D., discovered that the Hyde Park chemical dumpsite in western New York was an environmental danger to the nearby population, despite animal tests that "proved" that the toxins dumped there were not a hazard.

    'In a 1983 article entitled "How Animal Research Can Kill You", Dr. Bross explained: "At Hyde Park and many other dumpsites (including Love Canal and the West Valley nuclear dumpsites) the residents and workers have been repeatedly assured by State and federal agencies that their low-level exposures were 'harmless'. My research showed that these assurances by government agencies, assurances based largely on animal studies, were false, and that there were excess risks of cancer and other adverse health effects among the persons exposed to these 'safe' levels. By science and common sense, when epidemiological studies of humans are positive and laboratory studies of animals are negative, it is prudent public health practice to accept the human evidence as a guide. My employer, the New York State Health Department, not only disregarded the human evidence of the dumpsite hazards, it harassed me into early retirement from the state service."

    'Dr. Bross went on to explain how fraudulent animal tests are convenient for government bureaucracies wishing to hide environmental hazards from the public: "From the bureaucrat's standpoint, the beautiful aspect of animal research is that whatever you want to claim can be 'proved' in this way. Among experienced public health scientists it is well-known that you can 'prove' anything with animal studies. This is because there are so many different animal models and each system gives different results. By selecting whatever results happen to support a particular position (and ignoring the results to the contrary), one can come out with the desired 'conclusion'. This obviously is not the way that genuine science works but this is how animal studies are commonly used. Whenever government agencies or polluting corporations want to cover up an environmental hazard, they can always find an animal study to "prove" their claim. They can even do a new animal study which will turn out the way that they want by choosing the 'right' animal model system. If you happen to be one of the millions of Americans who has been or will be exposed to dangerous mutagens that are officially called 'safe', the games that are played with animal research can kill you."

    'Dr. Bross was Director of Biostatistics at the Roswell Park Memorial Institute for Cancer Research for 24 years.'

    (source -

  2. Very interesting history of Dr. Boss

    I'll follow up on his analyses.

    It really is depressing how compromised science has become - and perhaps always was....