SEARCH

SEARCH BY CITATION

Keywords:

  • atomic bomb survivors;
  • General Health Questionnaire;
  • lifestyle;
  • mental health;
  • self-rated health

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

To elucidate the effects of the bombing on the atomic bomb survivors’ mental health, a mental health survey was conducted using a 12-item version of the General Health Questionnaire (GHQ-12) and a mail survey on atomic bomb exposure conditions and lifestyle using a self-administered questionnaire. A total of 3526 atomic bomb survivors in Nagasaki responded and a high GHQ-12 score, as defined when the responses to four or more items were positive, was observed in 296 (8.4%) subjects. It was indicated that the risk of a high GHQ-12 score will decrease 0.98-fold with every 1-year increase in age, and will increase 1.45-fold and 1.70-fold in those who lost family members due to the bombing and those who had acute symptoms, respectively, compared with those who did not. It was indicated that the atomic bomb exposure has affected survivors’ mental health and that the care of their mental health is important.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Nagasaki City was severely damaged by the blast, heat and radiation after the atomic bomb was detonated on 9 August 1945. Approximately 18 000 houses and buildings were destroyed by the blast or burned out by fires which broke out in many places soon after the explosion, and approximately 74 000 people died by the end of 1945 while approximately 75 000 people were injured.

Although more than 50 years have elapsed, many people are still suffering from late effects of radiation exposure. The late somatic effects of atomic bomb radiation exposure have been extensively investigated: for example, a significant increase with radiation dose has been demonstrated for the incidence or mortality rate of leukemia and many kinds of malignant tumors. However, studies on mental or psychological effects of atomic bomb exposure have been limited. A psychiatric study in November 1945, which reviewed past histories of 50 subjects randomly selected from 192 inpatients exposed to the Nagasaki atomic bomb, revealed only four patients with mental disorders.1 A larger study conducted in 1956 revealed 533 neurosis cases among 7297 atomic bomb survivors in Nagasaki and reported that the proportion of neurosis cases was higher by twofold or more in subjects with acute symptoms due to radiation exposure (9.7%) than in subjects without such symptoms (3.9%).2 A study on the somatic and mental health conditions of 10 522 atomic bomb survivors from Hiroshima and Nagasaki reported on the basis of the Cornell Medical Index that the frequency of subjects having easy fatiguability, feeling sick, anxiety, uneasiness and depression was higher in the proximally exposed than in the distally exposed.3

Recently, several epidemiological studies have been conducted on the psychological effects of large-scale natural or artificial disasters such as earthquakes, floods, volcanic eruptions, war and nuclear power plant accidents. A mental health survey of residents in the Gomel region of Belarus affected by the Chernobyl nuclear power plant accident revealed a higher prevalence of mental health problems in evacuees and mothers with children under 18 years of age.4 A cohort study of children in Kiev indicated a higher frequency of neuroticism in evacuees from the Chernobyl zone to Kiev than in controls who had been in Kiev prior to the Chernobyl accident,5 while another study in Kiev found no difference in mental health problems between the evacuated children and the control group of their classmates.6 In addition to the Chernobyl accident, some studies of the Three Mile Island nuclear power plant accident reported mental health problems among people living around the site. Psychiatric symptom levels assessed using the Symptom Checklist-90 were elevated after the restart of the nuclear reactor,7 and were consistently elevated at the 10th anniversary8 among mothers of young children living within 16 km of Three Mile Island nuclear power plant. A study of the psychiatric effects of the eruption of Mt Unzen in Kyushu, Japan suggested, on the basis of the 30-item version of the General Health Questionnaire (GHQ-30), a higher prevalence of psychiatric problems in evacuees than in the general population.9 Some studies on US veterans who participated in the Gulf War revealed a higher prevalence of post-traumatic stress disorder10–12 and major depression.12

The objective of the present study was to elucidate the mental and psychological effects of the atomic bomb on the basis of a mental health survey using 12-item version of the GHQ13 (GHQ-12) and a mail survey about the conditions of atomic bomb exposure and current lifestyle in Nagasaki atomic bomb survivors.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Subjects

The Law Concerning Medical Services for the Atomic Bomb Exposed became effective in 1957. People fulfilling one of the following four conditions were officially recognized as atomic bomb survivors and were issued (by their respective local municipalities) the Atomic Bomb Health Handbook which ensures free medical service.

  • 1
    Directly exposed: those who were in Nagasaki or Hiroshima city, or in the designated neighboring areas at the time of the bombing.
  • 2
    Early entrants: those who entered designated areas, which are approximately within 2 km from the hypocenter, within 2 weeks after the bombing.
  • 3
    Other affected victims: those who were in the affected area after the bombing for purposes such as relief activities, burying corpses, and others.
  • 4
    Prenatally exposed: those who were exposed while in the uterus of mothers who fulfilled one of the preceding three conditions.

The aforementioned law was abrogated and revised as the Law Concerning Support for the Atomic Bomb Exposed, which became effective in 1994.

As of 31 March 1997, there were 58 431 atomic bomb survivors who were issued the Atomic Bomb Health Handbook from Nagasaki City Government (43 801 directly exposed; 9843 early entrants; 3661 other affected victims; and 1126 prenatally exposed). Information on Nagasaki atomic bomb survivors including their demographic data (date of birth and gender), exposure conditions (distance between exposed place and the hypocenter, and shielding conditions), health examination findings and mortality (cause and date of death) were stored on a computer database at the Atomic Bomb Disease Institute, Nagasaki University School of Medicine.

The subjects of the present study were 3526 directly exposed Nagasaki atomic bomb survivors who underwent mental health examinations and who responded to the mail survey described below. The present study was reviewed and approved in July 1997 by the institutional ethical committee of Nagasaki University School of Medicine.

Mental health survey

A total of 8510 atomic bomb survivors underwent a mental health survey at Nagasaki Atomic Bomb Casualty Council Health Management Center during the periods October 1994–January 1995; November 1995–February 1996; and April 1996–August 1996, on the occasion of their health examination at the center. Some, however, underwent the survey two or more times and 7669 people underwent the first mental health survey.

We assessed the mental health status of atomic bomb survivors using the GHQ-12, which was devised as a short version of the original 60-item version. The GHQ-12 is a reliable and convenient self-rating questionnaire, and is frequently used for screening minor psychiatric disorders.14,15 The Japanese version of the GHQ-12 completed by 7669 people in the present study was prepared by psychiatrists from the Department of Neuropsychiatry, Nagasaki University School of Medicine.

Mail survey

A mail survey was carried out among the 7669 people who completed the GHQ-12. The mail survey was carried out on 1 August 1997 for 7597 people living in Nagasaki City as of 1 July 1997 by sending them a self-administered questionnaire with a cover letter explaining the purpose and methods of the study, and a stamped return envelope. By 10 September 1997, 4890 (64.4%) subjects had responded, 87 (1.1%) refused, 51 (0.7%) letters were returned because of unknown address, and others (33.8%) did not reply. Among the respondents, only the directly exposed atomic bomb survivors were included in the analysis, and the number of subjects remaining for analysis was 3526 (1261 men and 2265 women). The sample quartiles of the age at the time of the mail survey (called age, hereafter) and the distance of the place of exposure from the hypocenter (called distance, hereafter) were (61, 66, 71) years and (2.5, 3.3, 4.1) km, respectively.

The questionnaire included three types of questions: questions on demographic and lifestyle factors, questions about late effects of the bombing and those on conditions of atomic bomb exposure. The first type of questions were about (i) drinking habits; (ii) smoking habits; (iii) regular mealtime; (iv) moderate exercise; (v) marital status; (vi) solitary life; and (vii) feeling of satisfaction in daily life. The second type of questions were about (viii) self-rated current health status; (ix) concern about persistent late effects by the bombing; and (x) concern about future appearance of late effects due to the bombing. The last type of questions were about (xi) loss of family members by the bombing; (xii) loss of friends or acquaintances by the bombing; (xiii) number of acute symptoms (burn, trauma, epilation, diarrhea, stomatitis, fever, anemia, hemorrhage, vomiting, stomachache, headache, dizziness and disturbance of consciousness); and (xiv) damage to housing.

Data analysis

We assigned a score of 0 or 1 to the first two and the last two answers, respectively, and defined the GHQ-12 score as the sum of the 12 scores: the GHQ-12 score thus takes an integer from 0 to 12. On the basis of our previous study, which indicated a higher prevalence of psychiatric disorders in people with a GHQ-12 score >3 (62.7%) than in those with a GHQ-12 score of ≤3 (18.6%),15 we classified the subjects into two groups (i.e. high-score and low-score groups) depending on whether their GHQ-12 score exceeded 3.

Distribution of the distance was compared between men and women, and among age groups by the Wilcoxon rank–sum test and the Kruscal–Wallis test, respectively. Association between the frequency of subjects in the high-GHQ-12-score group and the factors including the distance and items in the mail survey were analyzed. The chi-square test was used for nominal scale data such as loss of family members due to the bombing while the Cochran–Armitage trend test was used for ordinal scale data such as the number of acute symptoms.

Furthermore, simultaneous effects of the atomic bomb exposure conditions on the frequency of subjects in the high-score group were analyzed using linear logistic models with gender, age, distance, loss of family members due to the bombing, loss of friends or acquaintances due to the bombing, appearance of acute symptoms and damage to housing as covariates. Except for age and distance (the original data of which were used), the covariates were dichotomized and coded as follows. Gender: 1 for female and 0 for male; loss of family members due to the bombing: 1 for yes and 0 for no; loss of friends or acquaintances due to the bombing: 1 for yes and 0 for no; appearance of acute symptoms: 1 for one or more and 0 for none; and damage to housing: 1 for complete or moderate damage and 0 for little damage. Starting from a logistic model including all of these covariates and their two-factor interaction terms, we selected the most appropriate model on the basis of Akaike's information cirterion (AIC),16 by restricting the model in the class to make them hierarchical in the sense that an interaction term is included in the model only if the corresponding main factors are all included in the model. Once the most appropriate model was selected, the maximum likelihood estimation of the model parameters was carried out and then the odds ratio and its 95% confidence interval were calculated for each covariate in the model. Likelihood ratio statistics were used for the calculation of the confidence intervals. FREQ, LOGISTIC and NPAR1WAY in the the SAS® system17 were used for the calculations.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Demographic and lifestyle factors, and mental health conditions

A high GHQ-12 score (i.e. score ≥ 4) was observed in 296 (8.4%) subjects and the frequency of subjects in the high-GHQ-12-score group by category for each of the demographic and lifestyle factors is shown in Table 1.

Table 1.  Classification of subjects by demographic and lifestyle factors, and GHQ-12 score
 GHQ-12 score  
Demographic and lifestyle factors0–34–12TotalP
  1.    Simple logistic regression analysis;  chi-square test excluding unknown; §  Cochran–Armitage test excluding unknown.

  2.  GHQ, General Health Questionnaire.

Age (years)    
 50–59644 (89.9)72 (10.1)716 
 60–691574 (92.0)137 (8.0)1711 
 70–79892 (92.0)78 (8.0)970 
 80–120 (93.0)  9 (7.0)1290.27
Gender    
 Male1154 (91.5)107 (8.5)1261 
 Female2076 (91.7)189 (8.3)22650.89
Drinking habits    
 Drink currently1344 (93.0)101 (7.0)1445 
 Drank but stopped176 (88.4)23 (11.6)199 
 Do not drink1554 (90.7)159 (9.3)17130.02
 Unknown156 (92.3)13 (7.7)169 
Smoking habits    
 Smoke currently512 (90.5)54 (9.5)566 
 Smoked but stopped506 (90.8)51 (9.2)557 
 Do not smoke2050 (91.8)184 (8.2)22340.54
 Unknown162 (95.9)  7 (4.1)169 
Having regular mealtime    
 Yes2949 (92.1)254 (7.9)3203 
 No220 (84.9)39 (15.1)259< 0.01
 Unknown  61 (95.3)  3 (4.7)  64 
Taking moderate exercise    
 Yes1648 (92.8)128 (7.2)1776 
 No1495 (90.4)158 (9.6)16530.01
 Unknown  87 (89.7)10 (10.3)  97 
Marital status    
 Married2283 (92.5)186 (7.5)2469 
 Widowed, divorced or unmarried878 (89.6)102 (10.4)980< 0.01
 Unknown  69 (89.6)  8 (10.4)  77 
Solitary life    
 Yes487 (88.5)63 (11.5)550 
 No (living with family members)2239 (92.3)196 (7.7)2535< 0.01
 Unknown404 (91.6)37 (8.4)441 
Feeling of satisfaction in daily life    
 Very satisfied243 (98.0)  5 (2.0)248 
 Fairly satisfied2212 (94.9)118 (5.1)2330 
 Fairly unsatisfied606 (83.8)117 (16.2)723 
 Very unsatisfied101 (68.2)47 (31.8)148< 0.01§
 Unknown  68 (88.3)  9 (11.7)  77 

Although not significant (= 0.27, simple logistic regression), a decrease with age was observed while no difference by gender was observed in the frequency of subjects with a high GHQ-12 score. All lifestyle factors except smoking had a significant association with the frequency of subjects in the high-score group. The frequency of high-GHQ-12-score subjects was approximately 1.7-fold higher in those who stopped drinking than in those who currently drank (= 0.02), while no significant (= 0.30) difference was observed in the frequency of high-GHQ-12-score subjects between those who stopped drinking and those who do not drink. Subjects with a high GHQ-12 score were observed approximately 2.1-fold more frequently in those who had irregular mealtimes than in those who had regular mealtimes (< 0.01). Similarly, the ratio of the frequency of high-GHQ-12-score subjects in the two categories was significantly different from 1 for moderate exercise, marital status and solitary life: the ratio was approximately 1.4 (= 0.01), 1.4 (< 0.01) and 1.5 (< 0.01), respectively. A feeling of satisfaction in daily life was most strongly associated with the frequency of high-GHQ-12-score subjects. The daily life was (very or fairly) satisfactory for 2578 (73.1%) subjects while it was (fairly or very) unsatisfactory for 871 (24.7%), and the frequency of the high-GHQ-12-score subjects was approximately 4.6-fold higher in those who were dissatisfied (18.8%, 164 of 871 subjects) with their daily life than in those who were satisfied (4.8%, 123 of 2578 subjects).

Concern about late effects from atomic bomb exposure and mental health conditions

A significant association was observed between concern about late effects of the atomic bomb exposure and mental health conditions (Table 2). A total of 1329 (37.7%) subjects rated their health as bad and 188 (14.1%) of the 1329 subjects had a high GHQ-12 score. The frequency of the high-GHQ-12 subjects was approximately 3.3-fold higher in those who rated their health as bad than in those who rated their health as good or medium (< 0.01). Concern about persistent late effects by the bombing was observed in 1068 (30.3%) subjects, among whom 131 (12.3%) had a high GHQ-12 score. The frequency of the high-GHQ-12-score subjects was approximately 1.9-fold higher in those concerned about persistent late effects due to the bombing than in those who were not (< 0.01). Among the 2031 (57.6%) subjects who were not concerned about persistent late effects due to the bombing, 1298 (63.9%) were concerned about future appearance of late effects by the bombing. Subjects with a high GHQ-12 score were observed approximately 1.9-fold more frequently in those who were concerned about future appearance of late effects by the bombing than in those who were not (< 0.01).

Table 2.  Classification of subjects by concern about late effects after the bombing and GHQ-12 score
 GHQ-12 score  
Concern about late effects after the bombings0–34–12TotalP
Self-rated current health status    
 Good136 (98.6)  2 (1.4)138 
 Medium1843 (95.0)97 (5.0)1940 
 Bad1141 (85.9)188 (14.1)1329< 0.01
 Unknown110 (92.4)  9 (7.6)119 
Concern about persistent late effects after the bombing    
 Yes937 (87.7)131 (12.3)1068 
 No1894 (93.3)137 (6.7)2031< 0.01§
 Unknown399 (93.4)28 (6.6)427 
Concern about future appearance of late effects after the bombing    
 Yes1195 (92.1)103 (7.9)1298 
 No574 (95.7)26 (4.3)600< 0.01§
 Unknown125 (94.0)  8 (6.0)133 
   Asked of those who responded ‘no’ to the previous question;  Cochran–Armitage test excluding unknown; §  chi-square test excluding unknown.
 GHQ, General Health Questionnaire.

Atomic bomb exposure conditions and mental health conditions

Each of the atomic bomb exposure conditions had a significant association with mental health conditions except for distance and loss of friends or acquaintances due to the bombing (Table 3). The frequency of subjects with a high GHQ-12 score showed a tendency to decrease with the distance at which the subjects were exposed to the atomic bomb, but the decrease was not significant (= 0.17, simple logistic regression). A total of 1221 (34.6%) people lost a family member due to the bombing and the frequency of the high-GHQ-12 subjects in such people was approximately 1.4-fold higher than in those who did not lose a family member due to the bombing (< 0.01). The number of subjects was approximately 2.0-fold higher in those with acute symptoms than in those without an acute symptom (< 0.01). The houses of 1672 (47.4%) people had been completely or partially destroyed or burned by the bombing and the frequency of the high GHQ-12 score subjects in such people was approximately 1.6-fold higher than in those whose houses had been little affected by the bombing (< 0.01).

Table 3.  Classification of subjects by atomic bomb exposure and GHQ-12 score
 GHQ-12 score  
Atomic bomb exposure0–34–12TotalP
  1.    Simple logistic regression analysis;  chi-square test excluding unknown; §  Cochran–Armitage test excluding unknown.

  2.  GHQ, General Health Questionnaire.

Distance of place of exposure from the hypocenter (km)    
  –1.0  68 (84.0)13 (16.0)  81
 1.1–2.0529 (90.1)58 (9.9)587
 2.1–3.0739 (90.6)77 (9.4)816
 3.1–1894 (92.8)148 (7.2)20420.17
Loss of family members due to the bombing    
 Yes1097 (89.8)124 (10.2)1221
 No1881 (92.8)147 (7.2)2028< 0.01
 Unknown252 (91.0)25 (9.0)277
Loss of friends or acquaintances due to the bombing    
 Yes1619 (90.8)164 (9.2)1783
 No986 (92.4)81 (7.6)10670.14
 Unknown625 (92.5)51 (7.5)676
No. acute symptoms    
 None1610 (94.3)98 (5.7)1708
 One629 (91.3)60 (8.7)689
 Two or more991 (87.8)138 (12.2)1129< 0.01§
Damage to housing    
 Complete602 (88.9)75 (11.1)677
 Moderate899 (90.4)96 (9.6)995
 Little1251 (93.2)91 (6.8)1342< 0.01§
 Unknown478 (93.4)34 (6.6)512

Simultaneous effects of the atomic bomb exposure conditions on mental health conditions

The linear logistic model selected as the most appropriate for describing the frequency of subjects in the high score group included the following factors as covariates: gender, age, loss of family members, appearance of acute symptoms and damage to housing. No two-factor interaction terms were included in the selected model.

Table 4 shows the estimate and the 95% confidence interval of the odds ratio for each factor included in the model. The findings indicate the following.

Table 4.  Estimates and CI of odds ratios for gender, age and exposure
FactorComparisonOdds ratio Estimate95%CI
  1.  CI, confidence interval.

GenderFemale vs male1.110.84–1.47
AgeBy 1-year increment0.980.96–0.99
Loss of family membersAny loss vs no loss1.451.10–1.90
Acute symptomPresent vs absent1.701.27–2.31
Damage to housingModerate or complete damage vs little damage1.240.93–1.66
  • 1
    Although not significant, the risk of a high GHQ-12 score will be higher by approximately 1.11-fold for women than men if they are similar with respect to other factors.
  • 2
    The risk of a high GHQ-12 score will significantly (< 0.05) decrease by approximately 0.98-fold with every 1-year increase in age after adjustment for other factors.
  • 3
    The risk of a high GHQ-12 score will be significantly (< 0.05) higher by approximately 1.44-fold in those who lost a family member than in those who did not if they are similar with respect to other factors.
  • 4
    The risk of a high GHQ-12 score will significantly (< 0.05) increase by approximately 1.70-fold in those who had at least one acute symptom than in those who did not have any acute symptoms if they are similar with respect to other factors.
  • 5
    Although not significant, the risk of a high GHQ-12 score in those whose house was completely damaged or burned will be approximately 1.23-fold higher than in those whose house was little affected.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The results of the present study were based on 3526 directly exposed atomic bomb survivors who underwent the mental health survey and responded to the mail survey as well. Because 1906 directly exposed atomic bomb survivors did not respond to the mail survey although they had undergone the mental health survey, we compared the two groups with respect to the available factors to address the issue of non-response bias. The quartiles of the age distribution were (61, 66, 71) years and (59, 65, 72)  years in the respondents and the non-respondents, respectively, and hence the difference in the age distribution of the two groups was negligible in effect. Furthermore, no significant difference was observed in the distance distribution between the two groups. The gender ratio (female/male) was 1.8 and 1.4 in the directly exposed respondents and the directly exposed non-respondents, respectively, and the difference was significant (< 0.01, chi-square test). The proportion of subjects with a high GHQ- 12 score was 8.4% and 10.4% in the directly exposed respondents and the directly exposed non-respondents, respectively, and the difference was significant (< 0.02, chi-square test). This may partly be explained by Table 4 and the aforementioned fact that the former group was older by 1 year than the latter group on the average.

The questionnaire used in the present study was designed on the basis that the atomic bomb operated in two different ways: as an enormously powerful bomb and as a strong radioactive source. The questionnaire, thus, included questions about conditions and late effects of atomic bomb exposure as well as enquiries about lifestyle and demographic factors.

No significant association between the distance with the frequency of a high GHQ-12 score was indicated in the present study either by simple logistic regression analysis (Table 3) or by multiple logistic regression analysis (Table 4). The distance, however, had a significant association with the frequency of a high GHQ-12 score when it was used as a means of classification into groups as in Table 3 (< 0.01, Cochran–Armitage trend test). However, statistical analysis of continuous data based on such a classification is not appropriate because no rationale exists in the definition of cut-off points for classifying the continuous data.

A larger-scale study carried out on 10 522 atomic bomb survivors in Hiroshima and Nagasaki3 reported that the frequency of easy fatiguability, feeling sick, anxiety, uneasiness and depression was significantly higher in those exposed within 2 km of the hypocenter than in those exposed beyond 3 km from the hypocenter. The study used some items of the Cornell Medical Index in assessing the mental health conditions, but the Cornell Medical Index was designed for primarily assessing somatic health conditions rather than mental health conditions. Instead, we applied the mental health scale of GHQ for assessing the mental health conditions.

Another similar study with GHQ-12 was carried out on 2047 atomic bomb survivors from Hiroshima.18 The study compared the frequency of positive response to respective items of GHQ-12 between those exposed within 2 km of the hypocenter and those exposed at 2 km or further, and observed a significant (< 0.05) difference between the two groups only for ability to concentrate: the frequency was significantly higher in the first group than in the second group. A similar analysis was conducted in the present study: the frequency of positive responses to respective items of GHQ-12 was higher in those exposed within 2 km of the hypocenter than in those exposed at 2 km or beyond for all 12 items. A significant (< 0.05) difference between the two groups was observed for two items about feelings of unhappiness and depression, and about thinking of oneself as worthless.

Conditions of atomic bomb exposure other than the distance of the exposed place from the hypocenter were not considered in the aforementioned two studies, while the present study indicated that atomic bomb exposure conditions such as the loss of family members and appearance of acute symptoms were significantly associated with a high GHQ-12 score rather than the distance.

A significant association between concern about late effects of atomic bomb exposure and mental health condition was indicated (Table 2). However, it does not necessarily imply that the former caused the latter: to the contrary, the poor mental health status may have caused the concern about late effects of radiation exposure. Further studies are necessary to clarify the causality.

Appearance of acute symptoms was also strongly associated with mental health conditions (Table 4). Among the 13 acute symptoms, diarrhea was most prevalent (668 subjects or 18.9%), followed by trauma (635 subjects or 18.0%) and anemia (569 subjects or 16.1%), and for each symptom except burns, the frequency of high-GHQ-12-score subjects was significantly higher in those who had the symptoms than in those who did not. The present findings are similar to those of a previous study, which reported the higher prevalence of neurosis in atomic bomb survivors with acute symptoms compared with those who were free of such symptoms.2

The present study has some limitations in that the GHQ-12 is primarily used for screening psychiatric disorders, and in that the conditions of atomic bomb exposure were based on subjects’ memory recall, more than half a century after the experience. Nonetheless, the findings of the present study indicate that atomic bomb exposure has affected survivors’ mental health and that the care of their mental health is important.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

This study was supported in part by a Grant-in-Aid for Scientific Research (07457209) from the Ministry of Education, Science, Sports, and Culture from 1995 to 1996, and a Grant-in-Aid for Research on Atomic Bomb Disease from the Ministry of Welfare from 1994 to 1997. The authors thank Drs Kazuyasu Yoshitake, Koichi Takada, Toshihiro Otsuka, Keiko Hatada, Hiroyuki Sugasaki, Yuka Ishizaki at the Department of Neuropsychiatry, Nagasaki University School of Medicine, for valuable comments in developing the questionnaire, Dr Tomoko Hata at the Nagasaki Atomic Bomb Casualty Council Health Management Center for executive support in the mental health survey and Mr Ken-ichi Yokota at the Biostatistics Section, Atomic Bomb Disease Institute, Nagasaki University School of Medicine for technical support in data management.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  • 1
    Okumura N, Hikita H. Results of psycho-neurological study on atomic bomb survivors. Kyushu Neuro-Psychiatry 1949; 1: 5052 (in Japanese).
  • 2
    Nishikawa T, Tuiki S. Psychiatric investigation of atomic bomb survivors. Nagasaki Med. J. 1961; 36: 717722 (in Japanese).
  • 3
    Yamada M, Kodama K, Wong FL. The long-term psychological sequelae of atomic bomb survivors in Hiroshima and Nagasaki. In: Hubner KF, Fry SA (eds). The Medical Basis for Radiation Accident Preparedness. Elsevier, New York, 1991; 155163.
  • 4
    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: 845855.
  • 5
    Korol N, Shibata Y, Nakane Y. Psychosomatic health status of children exposed to the Chernobyl accident. Acta Med. Nagasaki 1998; 43: 5761.
  • 6
    Bromet EJ, Goldgaber D, Carlson G, et al. Children's well-being 11 years after the Chornobyl catastrophe. Arch. Gen. Psychiatry 2000; 57: 563571.
  • 7
    Dew MA, Bromet EJ, Schulberg HC, Dunn LO, Parkinson DK. Mental health effects of the Three Mile Island nuclear reactor restart. Am. J. Psychiatry 1987; 144: 10741077.
  • 8
    Dew MA, Bromet EJ. Predictors of temporal patterns of psychiatric distress during 10 years following the nuclear accident at Three Mile Island. Soc. Psychiatry Psychiatr. Epidemiol. 1993; 28: 4955.
  • 9
    Ohta Y, Araki K, Kawasaki N, Nakane Y, Mine M, Honda S. Study on psychiatric problems of evacuees of the volcanic eruption of Mt Unzen: Factor analysis of General Health Questionnaire (GHQ-30). Jpn Bull. Soc. Psychiatry 1995; 3: 109129 (in Japanese).
  • 10
    Southwick SM, Morgan III CA, Darnell A et al. Trauma-related symptoms in veterans of Operation Desert Storm: A 2-year follow-up. Am. J. Psychiatry 1995; 152: 11501155.
  • 11
    Proctor SP, Heeren T, White RF et al. Health status of Persian Gulf War veterans: Self-reported symptoms, environmental exposures and the effect of stress. Int. J. Epidemiol. 1998; 27: 10001010.
  • 12
    Wolfe J, Proctor SP, Erickson DJ et al. Relationship of psychiatric status to Gulf War veterans’ health problems. Psychosom. Med. 1999; 61: 532540.
  • 13
    Goldberg DP. The Detection of Psychiatric Illness by Questionnaire. Oxford University Press, London, 1972.
  • 14
    Nakane Y, Michitsuji S. Results from the Nagasaki Center. In: Ustun TB, Sartorius N (eds). Mental Illness in General Health Care. John Wiley and Sons, Singapore, 1994; 139209.
  • 15
    Nakane Y, Honda S, Mine M, Tomonaga M, Tagawa M, Imamura Y. The mental health of atomic bomb survivors. In: Nagataki S, Yamashita S (eds). Nagasaki Symposium Radiation and Human Health: Proposal from Nagasaki. Elsevier, Amsterdam, 1996; 239249.
  • 16
    Akaike H. Statistical predictor identification. Ann. Inst. Stat. Math. 1970; 22: 203217.
  • 17
    SAS Institute. SAS/Stat User's Guide, Version 6, 4th edn. SAS Institute, Cary, NC, 1990.
  • 18
    Sasaki H, Maeda R, Ito C. Psychological study of A-bomb survivors. Hiroshima Med. J. 1996; 49: 400403 (in Japanese).