Neurobehavioral and Cognitive Performances of Children Exposed to Low-Dose Radiation in the Chernobyl Accident The Israeli Chernobyl Health Effects Study

Abstract

Exposure to low levels of ionizing radiation after the Chernobyl accident in the Ukraine could potentially have influenced the neurobehavioral and cognitive performances of exposed children. A cohort study of adolescents who were children at the time of the accident and who subsequently emigrated to Israel was conducted in 1998-2001. A total of 1,629 children (59% of all 2,769 invited) were included in the study (41% from higher contamination areas, 25% from lower contamination areas, 34% from noncontaminated areas). Mean scores of the Raven Standard Progressive Matrices Test were highest in children in all exposure groups whose parents had a high level of education. No overall relation was found between the cognitive function scores of the child and his/her putative radiation exposure level. Conners' test T scores did not differ significantly by level of exposure. Mothers of all exposure groups who were pregnant at the time of the accident gave their children significantly higher Conners' test scores than did those who were not pregnant. Scores for hyperactivity and attention-deficit/hyperactivity disorder were significantly higher among those who were in utero at the time of the accident. These results do not show differences of neurobehavioral or cognitive performance in exposed versus nonexposed children. There is a possible behavioral effect among offspring of pregnant mothers or mothers of very young children in all exposure levels.

Full text

453
Am J Epidemiol
2004;160:453–459
American Journal of Epidemiology
Copyright © 2004 by the Johns Hopkins Bloomberg School of Public Health
All rights reserved
Vol. 160, No. 5
Printed in U.S.A.
DOI: 10.1093/aje/kwh231
Neurobehavioral and Cognitive Performances of Children Exposed to Low-Dose
Radiation in the Chernobyl Accident
The Israeli Chernobyl Health Effects Study
N. Bar Joseph
1
, D. Reisfeld
1
, E. Tirosh
2,3
, Z. Silman
1
, and G. Rennert
1,3
1
CHS National Cancer Control Center and Department of Community Medicine and Epidemiology, Carmel Medical Center and
Technion S. Neaman Institute, Haifa, Israel.
2
The Hannah Khoushy Child Development Center, Bnai Zion Medical Center, Haifa, Israel.
3
Technion Faculty of Medicine, Haifa, Israel.
Received for publication May 7, 2003; accepted for publication March 19, 2004.
Exposure to low levels of ionizing radiation after the Chernobyl accident in the Ukraine could potentially have
influenced the neurobehavioral and cognitive performances of exposed children. A cohort study of adolescents
who were children at the time of the accident and who subsequently emigrated to Israel was conducted in 1998–
2001. A total of 1,629 children (59% of all 2,769 invited) were included in the study (41% from higher
contamination areas, 25% from lower contamination areas, 34% from noncontaminated areas). Mean scores of
the Raven Standard Progressive Matrices Test were highest in children in all exposure groups whose parents
had a high level of education. No overall relation was found between the cognitive function scores of the child
and his/her putative radiation exposure level. Conners’ test T scores did not differ significantly by level of
exposure. Mothers of all exposure groups who were pregnant at the time of the accident gave their children
significantly higher Conners’ test scores than did those who were not pregnant. Scores for hyperactivity and
attention-deficit/hyperactivity disorder were significantly higher among those who were in utero at the time of the
accident. These results do not show differences of neurobehavioral or cognitive performance in exposed versus
nonexposed children. There is a possible behavioral effect among offspring of pregnant mothers or mothers of
very young children in all exposure levels.
accidents, radiation; child; cognition; health; neurobehavioral manifestations; radiation
Abbreviation: ADHD, attention-deficit/hyperactivity disorder.
The accident at the Chernobyl Nuclear Power Plant in the
Ukraine in 1986 was the largest nuclear reactor accident ever
reported. The accident caused the dispersion of radioactive
fallout over large areas, which exposed millions of people to
different amounts of radiation (1).
Although a lot of attention was first paid to the physical
health consequences of the radiation-exposed groups,
concern was also raised regarding the neurodevelopmental
sequelae in those who were exposed to radiation in utero or
at a young age. It could be hypothesized that prenatal and
early childhood irradiation of the brain could lead to
neurobehavioral dysfunction (2).
Studies of the survivors of the atomic bombs in Hiroshima
and Nagasaki, Japan, indicated that the fetal brain is most
susceptible to high doses of ionizing irradiation for 8–15
weeks after conception. Fetal exposure to high doses of radi-
ation increased the risk of mental retardation, small head
size, subsequent seizures, and poor performance on conven-
tional tests of intelligence (3, 4).
After the Chernobyl accident, an excess of thyroid cancers
among exposed children was reported and, more recently,
some genetic changes of unclear importance have also been
described in children possibly influenced by the Chernobyl
radiation exposure (5, 6). Psychological stress, anxiety, and
fear, especially in adults, were reported in several studies
Correspondence to Dr. Gad Rennert, CHS National Cancer Control Center, Carmel Medical Center, Haifa 34362, Israel (e-mail:
rennert@tx.technion.ac.il).
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from

454 Joseph et al.
Am J Epidemiol
2004;160:453–459
after the Chernobyl accident (7–12). Studies of the impact of
the Three Mile Island accident showed in one study greater
psychological distress among women who were pregnant or
had young children (13) but showed in another study no such
effect (14). Further, young children living near the Three
Mile Island reactor did not differ from children living next to
another reactor in their prevalence of behavioral problems
(15).
In addition to several hundred thousand individuals who
were relocated within the former Soviet Union after the
disaster, hundreds of thousands more left the country. Many
of them emigrated to Israel. Israeli immigrants from the
Chernobyl region manifested high levels of anxiety and
concern about radiation. Additionally, they showed inaccu-
rate preconceptions about the health risk of chronic radiation
exposure and sought extensive health care for what they
feared would be long-term illnesses caused by the Chernobyl
Nuclear Power Plant disaster (16, 17). No differences were
found in the cognitive and neuropsychological functioning
of evacuated children who were in utero or up to 15 months
of age at the time of the accident when they were compared
with their classmates in Kiev (18).
The present study, among others, was designed to evaluate
possible influences of the exposure to ionizing radiation after
the Chernobyl accident on the neurobehavioral and cognitive
performances of Israeli immigrants who were young chil-
dren at the time of the accident.
MATERIALS AND METHODS
Study population
All children aged up to 4 years (including in utero) at the
time of the accident who emigrated to Israel from the Gomel
(highly exposed) region and a convenience sample of immi-
grant children from the Mogilev and Kiev (mildly exposed)
regions and from the (nonexposed) cities of Moscow and St.
Petersburg, as well as other unexposed areas of Belarus,
were invited to participate in the study. Exposure levels of
the various areas were determined according to geographic
cesium contamination maps of the International Atomic
Energy Agency. Altogether, 1,629 children (58.8 percent of
2,769 children sampled and invited, whose mailing
addresses were confirmed from a selected cohort of 4,319
children) who emigrated to Israel between 1989 and 2000
were included. Of these, 667 (41 percent) children came
from areas with high contamination, 408 (25 percent) came
from areas of low contamination, and 554 (34 percent) came
from noncontaminated areas. Compliance rates of samples
of children with an identified address are as follows: 65
percent for children from Moscow-St. Petersburg, 62.2
percent for children from other Belarusian cities, 60.2
percent for children from Mogilev, 57.0 percent for children
from Gomel, and 54.9 percent for children from Kiev.
The area of origin in the former Soviet Union was defined
as the settlement of origin in their home countries recorded
by the Ministry of Immigrant Absorption in Israel upon
registration of their families for immigration. This indicator
was validated by information collected during the partici-
pant’s interview. Final assignment to area of exposure was
based on the true origin of the child and not on the informa-
tion appearing in the official documents.
All the children were invited, with their mothers, to the
CHS National Cancer Control Center at the Carmel Medical
Center in Haifa, Israel. The Israeli Chernobyl Health Effects
Study took place during the years 1998–2001. Participating
children were thus aged 12–18 years at the time of the study.
Data collection
After signing an informed consent form, the child’s
mother was interviewed by trained Russian-speaking inter-
viewers who used a standardized questionnaire. The ques-
tionnaire sought demographic data, residential and personal
radiation exposure (for dose estimation) data, and informa-
tion on exposure to various risk factors during the relevant
pregnancy, disease status, and health behavior patterns of the
relevant child. All children undertook a cognitive test and a
questionnaire evaluating their behavior. A complete physical
examination, self-reported puberty staging according to the
Tanner scale, and venous blood tests including hematology
and thyroid panels were also conducted but are not discussed
in this report. Assessment of attention-deficit/hyperactivity
disorder (ADHD) was carried out using the short Conners’
Rating Scales-Revised for children and their mothers (19).
Nonverbal intelligence was examined by the Raven Standard
Progressive Matrices Test (20). This test is not language
dependent and, therefore, functions well in a new immigrant
population.
Measurement of cognitive ability.
Cognitive ability of the
children was measured by using the Raven Standard
Progressive Matrices Test (1996 edition) (20). This
nonverbal intelligence test was selected because the study
population of immigrant children had diverse levels of
knowledge of Hebrew. The test consists of 60 diagrammatic
puzzles. Each puzzle has one part missing, which the child
has to fill in with one of the options provided after proper
explanation by the test administrator. No time limit was set
for completion of the test.
The child’s score was calculated as the total number of
problems solved correctly. A table of percentile norms (20,
table SPM 14) was used to interpret the scores. These norms
enabled us to compare the performances of children in the
various exposure groups.
Assessment of attention.
The assessment of ADHD was
carried out by using Conners’ Rating Scales-Revised. These
scales are a comprehensive assessment of psychopathology
and problem behaviors of children and adolescents.
The Adolescent Self-report Scales and the Parent Rating
Scales (which relate to the child’s behavior) (19) were
chosen for this study. The short forms (27-item question-
naire) yielded scores on four parameters (appearing in both
the adolescent and parent questionnaires): oppositional prob-
lems, cognitive problems, hyperactivity, and risk of ADHD.
Children were provided with a Hebrew version of the ques-
tionnaire, and their mothers were provided with a Russian
version. Conners’ Rating Scales-Revised are available in
only English, Spanish, and French. Because we needed a
Hebrew version for the children and a Russian version for
their mothers, the questionnaires were translated into
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from

Neurobehavior and Cognition in Radiation-exposed Children 455
Am J Epidemiol
2004;160:453–459
Hebrew and Russian and then translated back into English to
validate the translation. The crude scores of the Conners’ test
were translated into T scores, which are age and gender
specific. T scores of 65 or more usually indicate a significant
clinical problem, while T scores between 56 and 64 warn of
a possible problem.
Statistical analyses
Statistical analyses using SPSS/PC, version 10.07 (SPSS,
Inc., Chicago, Illinois), software included comparison of
different exposure groups (all groups, or highest exposure
group vs. others) by means of analysis of variance. After
studying individual scores, we used models that adjusted the
test scores for the following potential effect modifiers, iden-
tified in the univariate analysis: child’s gender, birth order,
number of siblings, family status (two parents or single
parent), timing of exposure (in utero vs. not in utero), and
parent’s level of education.
RESULTS
Of the 2,769 children approached, 1,629 (58.8 percent)
participated. Of these, 270 (16.6 percent) were in utero at the
time of the accident (table 1). The mean age of the study
participants at the time of the accident in 1986 was 1.48
(standard deviation, 1.31) years. Of the participating chil-
dren, 48 percent were females, and 52 percent were males.
No major differences were found in the ages at the time of
the accident (p = 0.62) and in the gender distribution (p =
0.88) of the children as related to the different exposure
groups. Most children (69 percent) had one sibling, 21
percent had no siblings, and 10 percent had two or more
siblings. Single-parent families comprised 20 percent of all
study families. None of the participating children suffered
from significant mental or physical handicaps at the time of
the study.
The age of the participating children at the time of the
study ranged from 12 to 18 years. About a fourth of the chil-
dren were aged 12–13 years at the time of the test, 45 percent
were aged 14–15 years, and 31 percent were aged 16–18
years.
The exposure groups differed significantly in the level of
education of their parents. Higher levels of education for
both parents were noted in residents of the Moscow-St.
Petersburg group, followed by residents of Kiev and Minsk
(low- or no-exposure group), and were lowest in members of
the high-exposure groups (table 2).
Cognitive assessment by the Raven Standard
Progressive Matrices Test
Mean scores (percentiles) of the Raven Standard Progres-
sive Matrices Test were highest in children in all exposure
groups whose parents had a high level of education. The
mean scores of the matrices test, adjusted for parent’s level
of education, differed significantly (p < 0.0001), although
the difference did not correspond to the level of radiation
exposure (table 3). No other variable was found to signifi-
cantly influence the results. Children who were in utero at
the time of the accident did not score differently from older
children in the same exposure area.
Attention assessment by the Conners’ test
The children’s test results for the Conners’ test did not
show any significant difference by radiation exposure level,
after controlling for gender, birth order, number of siblings,
family status, and parent’s level of education. Rates of all
exposure groups were within a normal range (score of less
than 50) (table 4).
Significant differences in the oppositional and hyperac-
tivity subscale scores of the mother’s test, but not in the
subscale scores for cognitive function and ADHD, were
noticed between the different exposure groups. However,
TABLE 1. Child population, by residency at time of accident, Israeli Chernobyl Health Effects Study, 1998–
2001
* According to an International Atomic Energy Agency map, 1989.
† A total of 26 children (3.9%) lived in very high contaminated areas (15–40 Ci/km
2
).
Residential area
Level of
cesium 137
(Ci/km
2
)*
No. of
children
approached
Participating children
Children in utero
at time of accident
No. % No. %
Gomel region, Belarus, and other
exposed areas in Belarus and the
Ukraine
1–40 1,169 667† 57.0 98 15
Mogilev region, Belarus <1 216 130 60.2 14 11
Kiev region, the Ukraine <1 506 278 54.9 48 17
Other regions in Belarus ≤0.1 598 372 62.0 72 19
Cities of Moscow and St. Petersburg,
Russia ≤0.1 280 182 65.0 38 21
Total 2,769 1,629 270 16.6
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from

456 Joseph et al.
Am J Epidemiol
2004;160:453–459
these differences were unrelated to the degree of radiation
exposure (highly exposed areas compared with all other
exposure areas or a comparison of all five exposure levels
separately). Most scores were within the normal range (score
of less than 56), while the other scores were within the
borderline range (score of 56–60) (table 4). Mothers with a
high level of education rated their children significantly
lower (on all scales except for the oppositional subscale)
than did less educated mothers.
Mothers’ scores were significantly higher (p ≤ 0.001 for
each of the subscales) than their children’s scores. This was
true for each of the five residence groups. The subjective
assessment of the children’s health status by their mothers
was found to correlate negatively with the mother’s score on
the Conners’ test. In all radiation exposure groups, mothers
who rated their children highest (good to excellent health
status) scored lowest on the Conners’ test (p < 0.001).
Mothers from all radiation exposure groups who were
pregnant at the time of the accident rated their children with
significantly higher Conners’ test scores in three of four
categories than did those who were not pregnant. The scores,
however, were mostly in the normal range (table 5).
A negative correlation was found between the scores of the
Raven matrices and the scores of Conners’ test (mother’s or
child’s score). All correlation coefficients were very low but
statistically significant. The highest correlation coefficients
were found between the Raven score and Conners’ test score
for the cognitive subscale (r = 0.2 for mothers, r = 0.1 for
children; p < 0.0001 for both).
DISCUSSION
No relation was found between exposure to radiation
(proxy estimated by location of residence at the time of the
accident) and performance in tests measuring neurobehav-
ioral and cognitive performance. This result may either
reflect a true lack of relation between radiation and behav-
ioral performance or may be the result of one of numerous
biases. This study was a large, population-based study in
which a whole cohort of children emigrating from exposed
areas in the former Soviet Union was recruited and included
a convenience sample of children from nonexposed areas as
well. The sampling frame came from governmental registra-
tion files and is believed to have a high validity and lack of
TABLE 2. Level of parent’s education, by exposure group, Israeli Chernobyl Health Effects Study, 1998–2001
Exposure group
Mother’s level of education Father’s level of education
9–11 years Technical High 9–11 years Technical High
No. % No. % No. % No. % No. % No. %
Gomel and other exposed areas 91 14 295 44 281 42 174 26 232 35 256 39
Mogilev 8 6 57 44 65 50 27 21 55 42 48 37
Kiev 24 9 99 36 154 56 46 17 92 33 138 50
Other Belarus 26 7 132 36 214 58 51 14 112 30 205 56
Moscow and St. Petersburg 1 0.5 33 18 148 81 7 4 28 16 142 80
Total 150 9 616 38 862 53 305 19 519 32 789 49
TABLE 3. Matrices’ mean scores, by parent’s level of education, Israeli Chernobyl Health Effects Study,
1998–2001
* Numbers in parentheses, standard deviation.
† Adjusted for mother’s education.
Gomel Mogilev Kiev Other Belarus
Moscow and
St. Petersburg
Mother’s education
9–11 years 50 (29)* 52 (23) 61 (30) 46 (25)
Technical school 54 (29) 49 (29) 57 (31) 58 (27) 60 (33)
University 67 (27) 61 (29) 76 (21) 69 (27) 78 (24)
Father’s education
9–11 years 52 (27) 46 (28) 63 (29) 54 (29) 52 (30)
Technical school 56 (30) 57 (29) 63 (28) 58 (27) 58 (31)
University 66 (28) 57 (30) 74 (24) 69 (26) 78 (24)
Total† 57 (29) 52 (28) 64 (29) 58 (30) 64 (30)
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from

Neurobehavior and Cognition in Radiation-exposed Children 457
Am J Epidemiol
2004;160:453–459
selection bias. Compliance in this study was about 60
percent and was nondifferentially distributed among the
various exposure groups. This, most probably, reflects a
nonattendance that is not disease status related. If sick chil-
dren were more apt to participate in this study, and if disease
in these children was related to radiation exposure, then the
data would be biased against the null. Because we did not
find a relation between residence in areas affected by the
Chernobyl accident and the degree of neurobehavioral or
cognitive performance in our study population, we assume
that no such bias occurred. In addition, the health status of
the participating children was not found to differ between the
cases and the controls in our series. Another possible bias is
a misclassification bias in which children from seemingly
exposed areas would be counted as nonexposed and vice
versa. We tested the residence registration in the sampling
frame and found it to correlate fully with the actual place of
residence in 1986. A more serious question is whether resi-
dence in Gomel truly corresponds with high radiation expo-
sure. Data from the former Soviet Union and from maps
produced by the International Atomic Energy Agency
several years after the accident have pointed at the Gomel
region (or oblast) as the area with the highest contamination
level outside the 30-km zone. Nevertheless, it is believed
that urban dwellers as a group were exposed to lower radia-
tion levels than were rural dwellers. Jews in the former
Soviet Union were mostly urban dwellers. This fact, coupled
with evidence that the Chernobyl accident as a whole was a
low-radiation-exposure event, could lead to studying and
comparing groups that do not differ enough between them-
selves to show an increased risk, if such risk actually exists.
No good measure of individual exposure to radiation is
available.
Other findings in the study may indirectly validate the
study results. The educational level of the parents was found,
as expected, to be a strong positive predictor of performance.
Good correlations between cognitive parameters in both
tests and higher Conners’ test scores of mothers versus their
children correspond to the current knowledge in this field.
Interestingly, the neurobehavioral assessment of ADHD
showed that mothers who were pregnant at the time of the
accident scored their children with significantly higher
scores than did nonpregnant women for their children. This
was shown without any relation to exposure. Children’s
assessment scores, in the hyperactivity and the ADHD index,
were significantly higher among those who were in utero at
the time of the accident. This too was unrelated to exposure
level.
Our results may suggest that mothers who were pregnant
or had very young children at the time of the accident,
TABLE 4. Conners’ mean scores adjusted for mother’s education, Israeli Chernobyl Health Effects Study,
1998–2001
* Numbers in parentheses, standard deviation.
† ADHD, attention-deficit/hyperactivity disorder.
Gomel Mogilev Kiev Other Belarus
Moscow and
St. Petersburg
Child’s scores
Oppositional 48 (6)* 48 (7) 48 (5) 48 (7) 49 (6)
Cognitive 47 (7) 48 (8) 47 (7) 46 (6) 47 (7)
Hyperactivity 46 (7) 46 (7) 45 (7) 45 (7) 45 (7)
ADHD† 47 (8) 48 (9) 46 (7) 47 (7) 47 (7)
Mother’s scores
Oppositional 53 (10) 55 (12) 54 (10) 55 (10) 58 (11)
Cognitive 51 (8) 52 (8) 52 (7) 52 (8) 53 (8)
Hyperactivity 57 (13) 61 (16) 58 (14) 58 (13) 61 (13)
ADHD 51 (8) 52 (9) 52 (8) 52 (7) 54 (8)
TABLE 5. Mean item score in Conners’ test, by pregnancy status at time of accident, Israeli Chernobyl
Health Effects Study, 1998–2001
* Numbers in parentheses, standard deviation.
† ADHD, attention-deficit/hyperactivity disorder.
Conners’ test item
Pregnant
(
n
= 270)
Not pregnant
(
n
= 1,346)
p
value
A (oppositional) 55.23 (10.28)* 53.77 (10.00) 0.029
B (cognitive problems) 51.61 (7.60) 50.81 (7.65) 0.119
C (hyperactivity) 56.92 (13.72) 56.82 (13.44) 0.002
D (risk of ADHD†) 52.44 (7.80) 51.36 (7.74) 0.036
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from

458 Joseph et al.
Am J Epidemiol
2004;160:453–459
regardless of their location in the former Soviet Union, had
higher levels of anxiety and fear of the radiation. This might
stem from lack of information and inaccurate preconceptions
about the exact magnitude and spatial distribution of radia-
tion. Such a sense of fear might either distort their perception
or, alternatively, have been transmitted to their children who
showed higher hyperactivity and ADHD index scores among
the younger ages.
Similar results were found by Ginzburg (7), who reported
high levels of anxiety and stress among adults from contam-
inated and noncontaminated villages. Nearly half of this
study population, regardless of the true level of exposure,
were unsure if they had a radiation-related illness. Two other
studies from the Chernobyl area reported a high risk of
psychiatric disorders among women with children under the
age of 18 years residing in the exposed areas (21, 22). A
study on the psychological effects of the Three Mile Island
accident showed that women who were pregnant or had
young children experienced the greatest psychological
distress after the accident (13). The trauma brought on by the
Chernobyl accident, experienced by mothers, was reflected
in another study in their perceptions of their children’s well
being, particularly somatic symptoms. This anxiety,
however, was not transmitted to the children themselves (8).
Two previous studies of the Chernobyl accident suggested
lower cognitive performance among children exposed prena-
tally when compared with nonexposed children (2, 23). One
of these studies (2) included evacuated children who were
described as having been significantly more exposed than
our study population. The second study (23) related to prena-
tally exposed children only, and it was based on very small
numbers of children. Furthermore, these studies did not
control for possible confounders (mainly, level of parent’s
education) as did our study, which did not demonstrate a
similar lower performance. It is possible that emigration to a
new place, remote from the irradiated area, played a role in
lowering the anxiety level in the family.
We could not evaluate what, if any, influence the site of
residence in Israel had. Moreover, we could also not deter-
mine if the level of education received by these children in
Israel had an effect. However, because of Israel’s manda-
tory-education law, it highly likely that all these young chil-
dren received formal education in Israeli schools. Given the
small size of Israel and given that most of our participants
came from one of nine cities in the country, it is unlikely that
residence in Israel biased our results.
In a comparison of the cognitive and neuropsychological
functioning of children who were between in utero and up to
15 months of age at the time of the Chernobyl disaster and
who were evacuated to Kiev and their Kiev school class-
mates, no difference in performance was found (18).
Self-assessment of the child’s health by the mothers was,
as expected, negatively correlated with the attention assess-
ment scores of the child. Health self-assessment has been
found in many studies to correlate positively with physical
health status (24).
In conclusion, the results of our large, population-based
study of children who had been exposed to low-dose
ionizing radiation do not show differences in neurobehav-
ioral or cognitive performance compared with results of
nonexposed children. Thus, low-radiation events probably
do not result in deviations in neurobehavioral and cognitive
performance. Lower neurobehavioral performance of chil-
dren who were in utero or young at the time of the accident,
in all exposure groups, may possibly hint at a broader,
country-wide, Chernobyl-related effect resulting in
increased levels of anxiety in the pregnant mothers or
mothers of very young children in the former Soviet Union.
Attention should be paid by agencies caring for relevant
former Soviet Union populations to the possible effects of
the anxiety of their subjects on the behavioral and cognitive
functioning of the children. Providing validated information
with regard to the true health effects of the Chernobyl acci-
dent may help to lower the level of anxiety.
ACKNOWLEDGMENTS
This study was part of the Israeli Chernobyl Effects Study
supported by a grant from the US Agency for International
Development, US State Department. The study was operated
through the S. Neaman Institute at the Technion-Israel Insti-
tute of Technology.
The authors would like to thank the pediatricians Drs.
Irena Chistyakov, Marina Umansky, and Leonid Ganelis and
interviewers Tanya Agranovsky and Yael Aerov for their
help in carrying out the study.
REFERENCES
1. Ilyin LA. Experience from Chernobyl in the context of current
problems of radiation. In: Radiation protection in nuclear
energy: proceedings of an international conference on radiation
protection in nuclear energy. Vienna, Austria: International
Atomic Energy Agency, 1988.
2. Nyagu AI, Loganovsky KN, Loganovskaja TK. Psychophysio-
logic aftereffects of prenatal irradiation. Int J Psychophysiol
1998;30:303–11.
3. Yamazaki JN, Schull WJ. Perinatal loss and neurological
abnormalities among children of the atomic bomb. Nagasaki
and Hiroshima revisited, 1949 to 1989. JAMA 1990;264:605–
9.
4. Otake M, Shull WJ. In utero exposure to A-bomb radiation and
mental retardation; a reassessment. Br J Radiol 1984;57:409–
14.
5. Weinberg HS, Nevo E, Korol A, et al. Molecular changes in the
offspring of liquidators who emigrated to Israel from the Cher-
nobyl disaster area. Environ Health Perspect 1997;105(suppl
6):1479–81.
6. Weinberg HS, Korol AB, Kirzhner VM, et al. Very high muta-
tion rate in offspring of Chernobyl accident liquidators. Proc R
Soc Lond B Biol Sci 2001;268:1001–5.
7. Ginzburg HM. The psychological consequences of the Cherno-
byl accident: findings from the International Atomic Agency
Study. Public Health Rep 1993;108:184–92.
8. Bromet EJ, Goldgaber D, Carlson G, et al. Children’s well-
being 11 years after the Chernobyl catastrophe. Arch Gen Psy-
chiatry 2000;57:563–71.
9. Cwikel J, Rozovski U. Coping with the stress of immigration
among new immigrants to Israel from Commonwealth of Inde-
pendent States (CIS) who were exposed to Chernobyl: the
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from

Neurobehavior and Cognition in Radiation-exposed Children 459
Am J Epidemiol
2004;160:453–459
effect of age. Int J Aging Hum Dev 1998;46:305–18.
10. Yevelson II, Abdelgani A, Cwikel J, et al. Bridging the gap in
mental health approaches between East and West: the psycho-
social consequences of radiation exposure. Environ Health Per-
spect 1997;105(suppl 6):1551–6.
11. Cwikel J, Abdelgani A, Goldsmith JR, et al. Two-year follow
up study of stress-related disorders among immigrants to Israel
from the Chernobyl area. Environ Health Perspect 1997;
105(suppl 6):1545–50.
12. Zilber N, Lerner Y. Psychological distress among recent immi-
grants from the former Soviet Union to Israel. I. Correlates of
level of distress. Psychol Med 1996;26:493–501.
13. Houts PS, Cleary PD, Hu TW. The Three Mile Island crisis:
psychological, social, and economic impacts on the surround-
ing population. University Park, PA: Pennsylvania State Uni-
versity Press, 1988.
14. Houts PS, Tokuhata GK, Bratz J, et al. Effect of pregnancy dur-
ing TMI crisis on mothers’ mental health and their child’s
development. Am J Public Health 1991;81:384–6.
15. Cornely P, Bromet E. Prevalence of behavior problems in three
year old children living near Three Mile Island: a comparative
analysis. J Child Psychol Psychiatry 1986;27:489–98.
16. Weinberg AD, Kripalani S, McCarthy PL, et al. Caring for sur-
vivors of the Chernobyl disaster: what the clinician should
know. JAMA 1995;274:408–12.
17. Remennick LI. Immigrants from Chernobyl-affected areas in
Israel: the link between health and social adjustment. Soc Sci
Med 2002;54:309–17.
18. Litcher L, Bromet EJ, Carlson G, et al. School and neuropsy-
chological performance of evacuated children in Kiev 11 years
after the Chernobyl disaster. J Child Psychol Psychiatry 2000;
41:291–9.
19. Conners CK. Conners’ Rating Scales-Revised: technical man-
ual. New York, NY: Multi-Health Systems, 1997.
20. Raven JC, Court JH, Raven J. Standard progressive matrices.
San Antonio, TX: The Psychological Corporation, 1996:50–66.
21. Havenaar JM, Van den Brink W, Van den Bout J, et al. Mental
health problems in the Gomel region (Belarus): an analysis of
risk factors in an area affected by the Chernobyl disaster.
Psychol Med 1996;26:845–55.
22. Havenaar JM, Rumyantzeva GM, Van den Brink W, et al. Long
term mental health effects of the Chernobyl disaster: an epide-
miologic survey in two former Soviet regions. Am J Psychiatry
1997;154:1605–7.
23. Kolominsky Y, Igumnov S, Drozdovitch V. The psychological
development of children from Belarus exposed in the prenatal
period to radiation from the Chernobyl atomic power plant. J
Child Psychol Psychiatry 1999;40:299–305.
24. Mossey JM, Shapiro E. Self rated health: a predictor of mortal-
ity among the elderly. Am J Public Health 1982;72:800–6.
by guest on June 7, 2013http://aje.oxfordjournals.org/Downloaded from