ARE NUCLEAR REACTORS
SIGNIFICANT THREATS
TO HEALTH?
*We are in the process of updating these web pages on energy.  The accident at Fukushima has not been covered, nor
more recent data from the Life Span Study from the Radiation Effects Research Foundation.  

1: INTRODUCTION

Fear of radioactivity is the primary source of opposition to the use of commercial nuclear power reactors; the fear is
not only of the reactors themselves, but also of the transportation and storage of used reactor fuel.  Opposition to
commercial nuclear power developed before health data were available from what are now two of the most relevant
sources. This web page is focused on the lessons to be learned from peer-reviewed medical data, and includes papers on
the effects of the accident at Three Mile Island.

The major health concern is increased incidence of cancer, or death from it.  The link between radioactivity and
diseases other than cancer is more difficult to establish because much higher levels of exposure are required to cause
them (
1).

The three sections that follow present some important information about cancer, information that is neutral in its
impact on the nuclear power controversy.  We suggest that you to read these sections because the information will be
helpful no matter what your opinion is about nuclear reactors.  The news media frequently omit information that is
needed to understand and interpret claims and assertions about the causality of cancer.
CONTENTS OF THIS PAGE:
1: INTRODUCTION
2: THE ORIGINS OF CANCER (Some basics-1)
3: TWO IMPORTANT ATTRIBUTES OF CANCER (Some basics-2)
4: DRAWING VALID CONCLUSIONS ABOUT CAUSALITY (Some basics-3)
5: THE CONSEQUENCES OF REACTOR ACCIDENTS: TREATED           
OBJECTIVELY OR EXAGGERATED?
6: TWO EVENTS THAT ARE PROVIDING HIGH QUALITY  DATA
7: THREE MILE ISLAND
8: COMPARING THE THREE EVENTS
9: CONCLUSIONS
10:
ENDNOTES PAGE (The endnote numbers in the text also link to this
page.)
11:
TABLES - HEALTH (Here, the tables in the text are placed on a
separate page for printing.  The links in the text also lead to this page.)
Please note that our web pages are correctly formatted for printing by Internet
Explorer
but not by Mozilla Firefox (we have not tested other browsers).  We will try
to address this problem.  Thank you for your patience.
Ionizing radiation, the energetic
emissions from radioactive
atoms occurs in several different
particle forms (alpha rays, beta
rays, gamma-rays, X-rays, and
neutrons).  These forms of
radiation are able to remove
electrons from atoms in
molecules, thus creating ions, a
process called ionization.  The
production of the ions alters the
chemical properties of the
affected molecule, especially
DNA (the genetic material), and
this is the cause of the biological
damage.

The international standard (SI)
prefixes and their
(abbreviations): kilo(k) =
thousand; mega(M) = million;
giga(G) = billion; terra (T) =
trillion; peta (P) = quadrillion
TYPES OF
RADIOACTIVITY
The amount of any radioactive
isotope is usually measured as the
number of nuclei that decay in a
given time interval.  The standard
unit is the Becquerel (Bq) which is
1 decay per second.  (An older
unit, the Curie (Ci), is 37 billion
decays per second.)

The Becquerel measures only the
number of decays, and not their
strength (energy).  The strength of
a radioactive dose is the energy
(measured in Joules) deposited in
a kilogram of mass (J/k) by the rays
from a source of radioactivity.  
There are two international
standard (SI) units in use, the Gray
(Gy) and the Sievert (Sv).  The
Gray is the  unit of absorbed dose
(1 J/kg). The Sievert measures the
"dose equivalent" which is equal
to the absorbed dose in Gray
multiplied by a quality factor (Q)
which depends on the type of
radiation absorbed.  Doses
produced by gamma rays, X-rays
and beta rays are assigned a Q of
1. The Q can be as high as 20 for
alpha radiation and ranges from 6
to 20 for neutrons depending on
their energy.
MEASURES OF
RADIOACTIVITY
2: THE ORIGINS OF CANCER (Basic Things It Helps to Know - 1)

To understand the role of radioactivity in causing cancer, it is useful to understand
contemporary scientific ideas about the origins of the disease.  A cancer or tumor,
either benign or malignant, originates when a single cell loses control over its rate of
division; the new cancer cell produces daughter cells faster than it should; if these
daughter cells break away from the original clump and grow in other parts of the body,
the cancer is called metastatic.  

Cell division is the biological process that produces new cells to replace those that
have died normally in the course of life (
1).  The cellular machinery that regulates the
rate of division is exceptionally complex and is still being unraveled by scientists.  
Control over the rate of cell division is so complex that it requires the proper
functioning of several hundred genes in each cell in the body.  These genes do the job of
making sure that each cell divides at the proper time, never faster or slower than
needed to maintain the human body over the decades of life.  Some cells stop dividing
fairly early in life, and it is significant that these non-dividing cells never become
cancerous.  Nerve cells and muscle cells are among the non-dividing cells in adults
(there are brain cancers, but they occur in cells that help nerve cells to function but
are not actually nerve cells).

Genes are made of a substance called DNA.  Alterations of the chemical structure of
DNA are called mutations, and mutations are the major cause of the malfunction of
genes.  Radioactivity is not the only agent that can cause mutations; chemical agents
are another, perhaps more common cause.  In addition, natural, unavoidable
mutations occur when the DNA is duplicated in the normal process of cell division.  
Human DNA is comprised of about 3.2 billion “letters” (known technically as the “four
DNA bases”), and a change in any one of these letters to different one of the four is a
mutation.  The duplication of the DNA is among the most accurate processes known,
but it is not perfect.  Its error rate is about 1 letter in 10 million to 100 million, so that
every time a cell divides, a number of mutations (roughly 30 to 300) are produced.  
The significance of these natural mutations as causes of cancer is supported by the
facts that 1), cells which do not divide do not become cancerous, and 2), that repeated
trauma can induce cancer because cell division is stimulated during the repair of the
injury.  Finally, we are all born with a burden of mutations inherited from our parents,
and these mutations are the causes of inherited, or familial, dispositions toward
cancer.

3: TWO IMPORTANT ATTRIBUTES OF CANCER (Basic Things It Helps to Know - 2)

Linking an increase in cancer with an exposure to a carcinogen of any sort (chemical
or physical) is difficult for two reasons.  One is the fact that cancer is a common
disease.  In the US 23% of all deaths are caused by cancer (
2:Table B, p.5; 3:Table C, p.
8
).  Furthermore, about twice as many people develop a cancer in their lifetime than
die from one, so about 40% of the population will develop a cancer.  Thus, in a
population of 1 million, over 400,000 people will develop one or more of various kinds
of cancer and about (“about” is important) 230,000 will die from it.

The second, greater difficulty in attributing causation results from the fact that the
incidence of cancer is variable (as alluded to immediately above).  It varies
significantly among different racial and ethnic groups.  For example, according to the
Centers for Disease Control the incidence of all cancers in both sexes in the year 2004
ranged (in number of cases per 100,000 people) from 470 cases for blacks, to 455 for
whites, to 357 for Hispanics (
4); and the percentage of all deaths that were caused by
cancer in 2004 were 21.8% for Blacks, 23.5% for White non-Hispanics, and 20.0% for
Hispanics (
3:Table F, p.14).   This means that if a comparison of cancer deaths were
made between a million non-Hispanic whites and a million Hispanics there could well
be 35,000 fewer deaths among the Hispanics (3.5% of 1 million).  Specific types of
cancer also vary widely in different racial groups (
5); and its occurrence is also
dependent on age, sex, and aspects of lifestyle such as smoking, drinking, or diet, and
also varies with geographic region (
4).

The incidents that we have analyzed below in our attempt to evaluate the damage
from exposure to radioactivity involved groups of from tens to hundreds of thousands
of people.  The data on variability cited above and in (
3:Tables D, p.9; E, p.12; and F, p.
14
; 4; and 5) point out that among groups of hundreds of thousands people of differing
composition of race, ethnicity, age, sex, and life style there might well be differences in
the incidence of cancer on the order of thousands of people.  If groups of tens of
thousands are examined, the variation could well amount to differences of hundreds
of cases.  The critical and inescapable conclusion is that it is technically difficult to
distinguish the effects of exposure to a carcinogen from the intrinsic variability in the
occurrence of cancer.  This fact alone is the source of the controversy over the impact
of events like reactor accidents.

4: DRAWING VALID CONCLUSIONS ABOUT CAUSALITY (Basic Things It Helps to
Know - 3)

There is no way to distinguish between cancer caused by radioactivity and that caused
by other mechanisms.  Two important pieces of information are almost always needed
to conclude that an exposure to radioactivity caused health damage: 1) information
on the exposure received by individuals, not just by groups; and 2) a suitable control
population, needed to factor the intrinsic variability of cancer out of the effects of the
exposure.
Second, about 80% of the harm occurred among the 11,000 or 12,000 people who were subjected to high amounts of
radioactivity ranging from 200 to 4000 times a single year’s natural background radioactivity, while only 157 excess
solid tumors and 83 excess deaths occurred among the 33,000 or 37,000 people subjected to moderate exposure–80 to
200 times background.  As we shall see, the workers who were sent in to clean up the mess at Chernobyl–the worst
reactor accident in history– were (except for the few present during the first hours) subjected to less than 200 times
background, and the accident at TMI released very much less radioactivity when compared to Chernobyl (see
Among other important findings from the LSS is the apparently complete absence of inherited genetic effects,
surprising because they were originally predicted to be the most serious consequences of the bombs (
6).  Exposure in
utero caused no greater damage than it did among the newborn (
6).  Except for those people who received over 250
times background, the radioactivity is estimated to have shortened life expectancy by 2 months (
6).

The story has not yet ended.  Just under half of the people irradiated by the bombs were still alive in 1998, and there is
reason to expect that elevated incidence of disease and death will continue.  It is possible that at the time the study
ends, the excess death and disease attributed to the bombs will have about doubled from the current numbers.  There is
as yet not sufficient understanding of the effects of radioactivity on health to predict whether the proportion of excess
cancers will increase, remain the same, or decrease as the study continues (
6, 11).

                                                                                      Chernobyl

The accident at Chernobyl has been reported in detail in many places (
14). The reactor was of a design not used in
western commercial reactors: the “moderator’ was graphite (see below); the core was not contained within a pressure
vessel; and the building was not a “containment” building because its roof was of ordinary industrial construction–not
built to contain an explosion.  The accident occurred on April 26, 1986 during a poorly managed training exercise.  
During the exercise core cooling was lost, following which the core exploded.  Because there was no containment vessel,
and also because of the relatively light construction of the roof, a significant portion of the burning core blew up
through the roof, and also set the roof on fire.  The core was intensely radioactive and the firefighters who responded in
the first few hours were exposed to extreme amounts of radiation.  Two reactor operators were killed directly by the
explosion, and 134 persons who worked at the site in the hours immediately following were diagnosed with acute
radiation sickness, and in subsequent months between 28 and 47 people died of acute radiation sickness (the cause of
death was not clear with some of these people) (
15).  The core continued to burn and much more radioactivity was
released by the fire than the amount from the initial explosion (
Figure 1).  Heavy exposure to the local population
caused by the fire continued for ten days.  

Several hundred thousand workers, almost all men, were engaged in cleaning up the contamination at the reactor itself
and in the surrounding highly contaminated region.  (These people are known variously as clean-up workers,
liquidators, or emergency workers.)  Information on individual dose (although of varying quality) was obtained for a
large number of these workers, many of whom were Russian military personnel and professional radiation physicists.

Thus, the affected population in the Soviet Union can be looked on as two groups, the general population and the clean-
up workers.  Since better data on exposure were obtained from the clean-up workers, health effects among them are
easier to interpret, with the one exception of thyroid cancer among young people living in the vicinity of the reactor
whose incidence was very high, as will be mentioned later.  Several peer-reviewed scientific papers on the health effects
observed in both the clean-up workers and the general population were published in 2007 in the journal
Health Physics
(
15, 16, 17, 18, 19).  A consortium of United Nations agencies, the World Bank, and the governments of Belarus, the
Russian Federation, and Ukraine published a report in 2005 (
20) that summarizes the accident and reaches
conclusions about health effects very similar to those in the Health Physics papers; we use the more recent data.

Several groups (cohorts) of clean-up workers for whom at least some individual dosimetry was available were enrolled
in studies to monitor their health status.  These men came from all over the former Soviet Union, so the control group
possible is the general male Russian population with proper age-specific incidence rates (
16).  

One cohort of 55,718 men who received an average dose of 130 mSv (=54 times one year’s background) which places
them within the range of the moderately exposed A-bomb survivors (5 to 200 mSv), was monitored for the incidence of
solid tumors;.  The data are complete through 2001 (15 years after the accident) and they show no trend upward in the
incidence of solid tumors (
Figure 2) (16), suggesting that the longer-term prospects are for an impact similar to that
shown by the A-bomb survivors after 53 years (see below).

A second cohort of 71,870 workers who received an average dose of 107 mSv (= 45 times background) was monitored for
leukemia, which is among the most sensitive to radioactivity and fastest appearing of all cancers (
21). The data as of
the year 2003 show that of 71 cases that developed, 6 could be attributed to the exposures received at Chernobyl (
18);
the excess cancers appeared only in those men who received more than 150 mSv (some received up to 215 mSv).  Also all
the excess cancers appeared by 1996 and none from 1997 to 2003, expected from the characteristics of the disease (
21).  
It’s significant to note that there were fewer cancers  than the number expected among the men who received less than
150 mSv. This illustrates the uncertainties of ascribing cancer to exposures unless the exposures are very high,
producing a lot of cancer, the number of subjects is large and/or the controls are highly appropriate.

A third cohort of 61,017 men who received an average dose of 160 mSv (= 67 times background) were monitored for
various cerebrovascular diseases; the data as of the year 2000 show as many seemingly “protective” effects of the
radiation (i.e. less disease than expected) as harmful effects (
16).  This again illustrates the uncertainties of ascribing
cancer to exposures.

A fourth cohort of 65,905 men who received an average dose of 110 mSv was monitored from mortality from any cause or
from malignant cancers.  For death from any cause, there seemed to be a trend upward with time until the year 1997
after which it fell, but the data are difficult to interpret because for all years, the mortality among the clean-up workers
was less than the expected number.  Also there was no upward trend with time for death from malignant cancer, whose
incidence was also less than expected (
16).

The general population living in the contaminated areas comprises 5.5 million people who have received an average
total whole body dose of 10 mSv (and a range in dose from a few mSv to about 100mSv) cumulative over the 17 years
following the accident (
17, 18).   There were 116,000 people living close to the reactor who were evacuated, and they
received an average dose of 33 mSv, while those continuing to live close to the reactor (270,000 people) have received an
average dose of 50 mSv.  Background radioactivity is about 2.4 mSv per year (
6).

The accident released about 1760 petaBq of radioactive iodine (I-131) (
19, 22).  This isotope is concentrated by the
thyroid gland and is known to cause thyroid cancer.  A significant incidence of thyroid cancer has occurred among
residents of the area, especially children (
Figure 3) (19).  The numbers are so large that controls are not needed to show
causation for younger people, but the incidence decreases as the age at exposure increases.  Increased incidence is
expected to continue (
19).

Because leukemia is so sensitive to radiation several studies were undertaken to monitor it in the general population
(the studies of leukemia in the clean-up workers are described above), and these are summarized in (
18).  The author
states “...
it is concluded that, possibly apart from Russian cleanup workers, no meaningful evidence of any statistical
association between exposure and leukemia risk as yet exists.


7: THREE MILE ISLAND

The accident occurred on March 28, 1979; its causes were a combination of faulty equipment, poor design, and operator
error.  It has been written about extensively, and an excellent recent account is found in (
7).  In this section the primary
concern is the consequences to the health of the population living near the reactor, although the probability of a similar
accident will be considered below.

One of the sources of misinterpretation of the TMI accident is the way in which it has been compared to Chernobyl by
the news media.  Although a vastly larger amount of radioactivity was released from Chernobyl, popular accounts
almost uniformly equate the two events and don’t provide any detail.  In this section we present medical data on health
effects with numerical comparisons of the amounts of radioactivity released by the two accidents.  But even a visual
comparison of the two reactors makes it clear that the accidents were of a very different order.  
Figure 4 shows both
reactors after their respective accidents.  The building housing the reactor at Chernobyl, which was not a containment
building because of its weak roof, was severely damaged, only the shell of the building and rubble remain, whereas there
is no external evidence of damage to the containment building at TMI.  (Note that the very tall curved structures are
cooling towers which are used at coal plants as well as nuclear plants; they are not reactor containment buildings.)  It
was not understood until 1985 that about half the core at TMI  had melted (
7), but had not breached even the pressure
vessel.  The next barrier against escape of radioactivity to the environment is the containment building which, at TMI,
was able to retain escaped radioactive krypton gas for sixteen months until the decision was made to vent it
deliberately (
7).

Three peer reviewed papers on the health effects of the TMI accident have been published by epidemiologists (
24, 25,
26).  The first two reports state that there was no measurable cancer resulting from the accident while the third
disagrees.  Interpretation of these conflicting conclusions depends to a significant degree, as shall become clear, on the
amount of radioactivity released.  Measurements and estimates were made during and after the accident by the
Environmental Protection Agency, the Food and Drug Administration, the State of Pennsylvania, the Department of
Energy, the Nuclear Regulatory Commission, and the company that owned the reactor (
7) and these values were
accepted by a number of investigating commissions and panels in the months following the accident; even the Union of
Concerned Scientists agreed that the amount of radioactive krypton (the most abundant radioactive element emitted
during the accident) was not a serious health concern (
7).

The “official” values for exposure to nearby residents were estimated to be 0.1 mSv average, or 0.4 mSv maximum (
24,
25).  These range from 1/6 to 1/10 of background radiation for a single year, and were not expected to have resulted in
any harm to health.  

The first study, on the incidence of cancer, (
24) was based on a 6 year  follow-up of a population of 159,680 people
(including both exposed and unexposed); the expected values for cancer were derived from the relevant age and sex
data for the entire U.S., the same method used in the studies of the Chernobyl accident; this is one of the major
methodogical differences between this paper and paper (
26).  Most oncologists believe that six years of follow-up is
somewhat too short for many cancers to manifest themselves following their initiation; ten years or more is considered
most likely (
21).  Therefore, the Hatch study directed some focus on leukemia and childhood cancers because these do
tend to become symptomatic sooner after their initiation (
21).  The study was probably begun so soon after the accident
because of the anxiety that the accident produced in the local population.  In their conclusions, the authors state: “
No
associations were seen for leukemia in adults or for childhood cancers as a group” and “Overall, the pattern of results
does not provide convincing evidence that radiation releases from the Three Mile Island nuclear facility influenced
cancer risk during the limited period of follow-up.


The paper by Talbott, et al. (
25) reported mortality statistics, including mortality from cancer, with 19 years of follow-
up, in a cohort of 32,135 people, and using the population of three surrounding counties as controls.  The authors state:
Overall cancer mortality in this cohort was similar to the local population....” and “In conclusion, the mortality
surveillance of this cohort, with a total of almost 20 years of follow-up, provides no consistent evidence that
radioactivity released during the TMI accident (estimated maximum and likely gamma exposure) has had a significant
impact on the mortality experience of this cohort through 1998.
”  Dr. Talbott is the director of the ongoing study of the
TMI cohort, which is being supported by the Pennsylvania Department of Health.

The paper by Wing, et al., (
26) takes issue with the findings–by criticizing the methods–of the paper by Hatch, et al.
(
24), and the Wing paper’s conclusions are that the emissions from TMI did cause a measurable amount of cancer.  The
Wing group’s claim is that 197 cases of cancer were in excess out of a total of 2831 cases (from all exposure groups).  
They also claim a large increase in incidence of leukemia, but this conclusion is based on a calculated excess of 8 cases
out of a total of only 15 cases; these are very small numbers.  

They used exactly the same data set of cancer incidence that the Hatch group used, so the disparity between their
conclusions results from differences in the methods of analysis.  The Wing and Hatch groups used different statistical
models.  And, unlike the Hatch group, the Wing group calculated the expected number of cancers using three years of
pre-accident cancer statistics from the same communities where the exposures occurred.  This is a more closely
matched control set in terms of the characteristics of the people, but three years might not provide sufficient data for a
valid baseline.  Also, these results are surprising for a six year follow-up period.  

The final technical comment on the Wing group’s conclusions concerns the amount of radioactivity that was released.  
As mentioned above, the official estimates were so small that radiation scientists expected no health effects.  In the
Introduction to their paper, the Wing group acknowledges that their results make sense only with a higher dose: “
If the
premise that maximum doses were no higher than average annual background levels is not open to question, then no
positive association could be interpreted as evidence in support of the hypothesis that radiation from the accident led
to increased cancer rates.
”  The Wing group presents evidence that some residents of the affected area were exposed to
as much as 600 to 900 mSv, a level that approaches the highest that the A-bomb survivors were subjected to, and which
exceeds exposures of the clean-up workers at Chernobyl.  The evidence for this, however, is of questionable quality.  It
was obtained from analyses of blood samples taken 15 years after the accident (
27, 28).  The analytical method used
(cytological analysis for chromosome anomalies) is not yet generally accepted as a method for determining exposure,
especially not from blood samples taken this long after an exposure, and where the control samples, which are
absolutely essential for this assay, were taken from a population of Russian people, and are
a priori inappropriate (29).  
Thus, the conclusion that there were high exposures at TMI is not supported by credible scientific evidence.

8: COMPARING THE THREE EVENTS

The reactors at Chernobyl and TMI had about the same power capacity, 1000MW at Chernobyl and 850 MW at TMI.  But
the accident at Chernobyl was far more violent than that at TMI, as evidenced both by the physical damage to the
buildings that housed them (
Figure 4) and by the relative amounts of radioactivity released (Figure 1). The reactors
were of very different design.  An essential component of all power reactors is called the “moderator,” and the
moderator used at Chernobyl was graphite, while at TMI it was the cooling water itself (which therefore served two
purposes).  When a water moderated reactor loses its water supply, the fission reaction slows down, whereas just the
opposite happens when a graphite moderated reactor loses its cooling water.  It was probably the acceleration of the
fission reaction that caused the explosion and initial release of radiation at Chernobyl, but then the graphite caught
fire and burned intensely for 10 days leading to the release of much additional radiation.  Also, there was no pressure
vessel surrounding the core of the Chernobyl reactor.  Finally, the roof of the building housing the Chernobyl reactor
was of ordinary industrial construction, not strong enough to contain the explosion (
14).  The containment building at
TMI was able to retain a radioactive gas (85-xenon) for sixteen months, until the gas was vented deliberately (
7).  These
differences of design and construction between the two reactors explain the great difference in the amount of
radioactivity emitted from the accidents.

The phrase “Chernobyl on the Chesapeake” has been used to stimulate opposition to building the new reactor at Calvert
Cliffs, but it would be more accurate to describe Chernobyl as a “Titanic” among reactor designs.  An accident like that
at Chernobyl simply cannot happen in a water moderated reactor.

For the induction of leukemia, three of the most potent isotopes (because they are deposited in bone) are strontium-90
(Sr-90) and cesium-134 (Cs-134) and cesium-137 (Cs-137).  The accident at Chernobyl released about 8 petaBq of Sr-90
(peta = a million times a billion) and a total of 57 petaBq of the two cesium isotopes (
14, 22), whereas at TMI none of
these isotopes was released.  Radioactive iodine can cause thyroid cancer and the epidemic of this cancer among the
young living near Chernobyl is clearly the result of emissions of radioactive iodine.  The accident at Chernobyl released
1760 petaBq of iodine-131 (
22) whereas at TMI 0.00056 petaBq were released (23) about 3 million times less than at
Chernobyl.

Table 3 compares the health effects from the three events; unlike Tables 1 & 2, Table3 uses data on the A-bomb survivors
taken from a paper in which the incidence of cancer for an
eight year follow-up period was reported (30), matching
more closely the
six year follow-up period used by Wing, et al.  The exposures received by the A-bomb survivors or the
clean-up workers range from 100-fold to 25,000-fold higher than those experienced by the residents near TMI (
Table
3).  It is almost universally accepted that the risk of cancer increases with the dose (6).  Because the data from the A-
bomb survivors is of much higher reliability than those from any other study and reports a much lower incidence of
cancer, we conclude that the cancer reported by Wing et al. is in error, possibly because of the statistical methods that
the group employed.  The differences in the severity of the health effects among the three groups might have been
caused by their ethnic and/or racial differences  (
6) if, and only if, the people living near TMI were orders of magnitude
more sensitive to radioactivity than are Russian men or the Japanese residents of Hiroshima and Nagasaki.  This is very
unlikely.
It’s intuitive that harm from radioactivity is related to the amount of exposure; but when an accident or other event
spreads radioactivity over a large area it is erroneous to assume that everyone in that area received exactly the same
dose.  Individual exposures could vary dramatically, and those who, unfortunately, received a higher than average
dose are more likely to be affected.  Therefore, strong conclusions demand knowledge of what happened to each person.

The need for a control group is imposed by the variability in the incidence of cancer.  If everyone had exactly the same
chance of developing cancer, then an increase caused by exposure to a carcinogen could be detected and measured
easily.  But this is unfortunately not the case, and a control group that resembles the exposed group in every respect
(race, ethnicity, age and sex distribution, lifestyle, etc.) except for the exposure is essential.  In some of the figures
shown below, the incidence of disease in the exposed population bounces above and below the incidence in the control
population; this is a manifestation of the intrinsic variability in the appearance of cancer. We will mention one
instance where the increase in cancer caused by radioactivity was so large that a control population is not needed to
attribute cause (
6).

5: THE CONSEQUENCES OF REACTOR ACCIDENTS: TREATED OBJECTIVELY OR EXAGGERATED?

The accidents at Three Mile Island (TMI) in 1979 and at Chernobyl in 1986 are the worst in the history of commercial
nuclear power generation.  Both events hardened the fear of radioactivity that had originally developed in response to
the atomic bombings of Japan and to the global radioactive fallout produced by testing nuclear weapons (
7).  Nuclear
reactors were equated to atomic bombs, so there was fairly widespread belief that reactors could undergo nuclear
bomb-like explosions.  The movie
The China Syndrome is an iconic example of the speculation about catastrophic
damage resulting from a serious reactor accident.  But the accident at Chernobyl far exceeded the severity of the
accident scenarios talked about in the movie, and we shall see that the damage to health caused by Chernobyl seems far
less severe than the movie implied it would have been.  At the time of the two reactor accidents, quantitative evidence
about harm to health was not as extensive or as robust as it is today, although there was already some solid data from
Japan that was not reported in the news media.

Unfounded exaggerations and warnings of catastrophe occur throughout the history of the anti-nuclear movement.  
The following quotations are found in the first chapter of (
7):
    a) “We are irreversibly committed to one million deaths from radiation.”
    b) “... render an area the size of Pennsylvania permanently uninhabitable.”
    c) A prominent anti-nuclear lawyer stated that he was so “... seriously afraid of nuclear accidents [that he wanted
    to] put people in jail.”
    d) A well known consumer advocate stated that nuclear power was “... a terrible hazard [that could cause] the
    greatest destruction that this country has ever known.”

As recently as this year, warnings of dramatic damage were put forward as reasons to oppose the new reactor:
    a) “The Department of Energy in 1982 estimated that for each reactor at Calvert Cliffs, a meltdown would likely
    result in 5,600 fatalities, 15,000 peak injuries, and 23,000 cancer deaths.  It stands to reason that those figures
    would be greater now, given the growth in population and development in Southern Maryland.” (8).
    b) “... the industry is vying to build the first nuclear plants since the deadly [emphasis added] accident at Three
    Mile Island in 1979.” (9).
We will see that the accident at Chernobyl, which did not occur until 1986, as yet has caused few fatalities, certainly
not the 5,600 or 23,000 alluded to above–even among the most highly exposed workers (but see below about thyroid
cancer in children at Chernobyl).  Furthermore, it was not yet understood in 1982 that the core at TMI-2 had partially
(about 50%) melted, without significant release of radioactivity (
7).  Thus, opinion common in 1982, even among
experts, needs to be seriously revised in light of experience from the ensuing 26 years.

6: TWO EVENTS THAT ARE PROVIDING HIGH QUALITY  DATA

    The atomic bombings

Two of history’s most tragic events, the atomic bombings of Hiroshima and Nagasaki, happen to be the source of the
best data on the effects of exposure to radioactivity on health.  No one knows how many people died instantly in the
blasts, but the number was probably tens of thousands. More died in the following months from acute radiation
poisoning, but about 150,000 people survived and some of these people were enrolled in what is known as the Life Span
Study, whose purpose is to improve understanding of the effects of the exposure on health.  The study has been a joint
effort of Japanese and American scientists and physicians.

The importance of the data from the LSS is widely acknowledged, both by medical researchers and committed anti-
nuclear activists.  The following is a quotation from the BEIR Committee (
6).
    “The Life Span Study (LSS) cohort of the survivors of the atomic bombings in Hiroshima and Nagasaki continues
    to serve as a major source of information for evaluating health risks from exposure to ionizing radiation and
    particularly for developing quantitative estimates of risk.  The advantages of this population include its large
    size...; the inclusion of both sexes and all ages; a wide range of doses that have been estimated for individual
    subjects; and high-quality mortality and cancer incidence data.”
Dr. John Gofman, a physician and a prominent opponent of nuclear power, commented in a similar vein (10):
    “One of the enormous scientific merits of this study is the plan to follow-up these individuals for their complete
    lifespans.... The study includes a large unexposed group and a very great range of doses.... It is the only careful
    long-term study which includes persons at all ages at the time of exposure.... In addition to the inclusion of all
    doses and all ages, the A-bomb Study includes both sexes.... Moreover, the A-bomb Study is a study following
    exposure of the entire body to ionizing radiation.  Thus, the study can address the problem of radiation induced
    cancer in general.... Unlike studies which must rely on Vital Statistics for cancer-rates among unexposed
    population-samples, the A-bomb Study provides its own control or reference group, internal to the study.”

The three most recent major scientific papers (
11, 12, 13) from the Life Span Study include data up to 1997 or 1998 (52
or 53 years after the bombings) and are summarized in
Tables 1 and 2; and several  important conclusions can be made
from the data.  First, the number of survivors suffering from the effects of the blasts is probably much smaller than
many people would guess if asked to estimate those numbers.  The data show that there were only 925 cases of excess
cancer and 765 deaths (from all diseases, including cancer) among the 45,000 or 49,000 exposed residents of the two
cities who were enrolled in the LSS after a period of more than 50 years.  
Table 1: Cancer incidence* among the atomic bomb survivors and their controls from 1958 to 1998 (11).
 
Number of
people
Exposure
(mSv)**
Exposure relative to
background***
Total
Cases
Excess
Cases
Percent in
Excess
Controls ****
60,792
0 to 5
0 to 2 times
9,597
NA
NA
Moderately
exposed
33,316
5 to 200
2 to 80 times
5,374
157
2.9%
Heavily exposed
11,319
200 to 4,000
80 to 1,600 times
2,477
693
28.0%
Exposed (total)
44,635
5 to 4,000
2 to 1,600 times
7,851
850*****
10.8%
* Solid tumors only.
** The radioactivity from the bombs.  Exposure exceeding 8,000 mSv is almost 100% lethal within months.
*** Background is 2.4 mSv per year.
**** Controls include people who received from zero to low exposures.  In this study a group of 26,580 residents of the two cities who were not in the cities
at the time of the bombings
were included in the controls.
***** There were 75 excess cases of blood cancers in the Life Span Study between 1950 and 1987, bringing the
total excess cases of cancer to 925.
Table 2 : Cancer mortality* among the atomic bomb survivors and their controls from 1950 to 1997 (12).
 
Number of
people
Exposure
(mSv)**
Exposure relative to
background***
Total Deaths
Excess
Deaths
Controls****
37,458
less than 5
less than 2 times
3,833
NA
Moderately exposed
37,382
5 to 200
2 to 80 times
3,945
83
Heavily exposed
11,732
200 to 4,000
80 to 1,600 times
1,557
357
Exposed (all)
49,114
5 to 4000
2 to 1600 times
5,502
440*****
* From solid tumors only.
** The radioactivity from the bombs.  Exposure exceeding 8,000 mSv is almost 100% lethal within months.
*** Background is 2.4 mSv per year.
**** Controls include people who received from zero to low exposures.  In this study a group of 26,580 residents of the two cities who were not in the cities
at the time of the bombings were
not included in the controls.
***** There were 75 excess cases of various blood cancers (e.g. leukemia) in the Life Span Study participants between 1950 and 1987 (
12) and 250
noncancer deaths between 1950 and 1997 (
11); assuming that all of the blood cancer patients died, the total excess deaths from all causes
becomes 765
.  As with the solid tumor deaths, the bulk of the noncancer deaths (216 out of 250) occurred in the heavily exposed group (11).

9: CONCLUSIONS

Warnings of severe health damage from reactor accidents were quoted in section 5.  But data in peer reviewed scientific
papers on the consequences of both the atomic bombings of Japan and the severe reactor accident at Chernobyl show
that these two events, which produced relatively high exposures, harmed a surprising small number of people.  There is
also no credible scientific evidence that the accident at TMI caused measurable harm to health.  Therefore, our
conclusion is that there is no credible evidence that water moderated nuclear power reactors have resulted in any
measurable, let alone significant, harm to health.  We are not implying that reactors are perfectly safe, ionizing
radiation is potentially very harmful.  But, there have been about 400 reactors in service for the past 25 years (
31), and
the only accident of note was at TMI 30 years ago.  Many reactors are decades old, but they are functioning without
serious mishaps, and there is every reason to believe that the newer design being proposed for Calvert Cliffs will be even
safer.  An obviously enhanced safety feature is the double walled containment building, which Constellation has
designed to withstand the impact of a large commercial jet airplane.  

A firm conclusion from the analysis presented above is that any accident that might occur would have to be far more
severe than that at TMI, possibly approximating the Chernobyl accident, before even measurable health effects would
be expected.  The benefits of avoiding of these small risks must be measured against the certainty of massive
environmental degradation and the possibility of an unreliable electricity supply (
32) if renewable energy from wind,
sunlight, or biomass were to be implemented on a scale adequate to provide a significant portion of our energy needs.

Last edited: 7/7/09 7:30 PM
Table 3: Comparison of A-bomb survivors, Chernobyl clean-up workers, and residents near TMI.
 
Excess:
Exposure relative to
background*
Follow-up
(years)
Number of
subjects
Lung
Cancer
Solid
Tumors
Leukemia
TMI: Hatch et al.
(
24)
**
0
0
0.04 to 0.16 times
5
78,500
TMI: Wing et al.
(
26)
44
134
10
0.04 to 0.16 times
5
78,500
A-bomb
(1950-1965) (
13)
**
**
57
4 to 4,000 times
16
49,114
A-bomb
(1958-1965) (
30)
38
77
**
5 to 4,000 times
8
49,114
Clean-up
workers (
16)
0
0
**
54 times
16
55,718
Clean-up
workers (
17)
**
**
6
45 times
16
71,870
* Background is 2.4 mSv per year.
** Not studied
NAVIGATION BAR:
SIDEBAR