Next | ToC | Prev



CHAPTER 4

Overview of a Uniquely Valuable Database




          Since 1950, more than 90,000 of the survivors of the atomic bombings at Hiroshima and Nagasaki have been enrolled in a lifetime study of their health. (Details are in Chapter 5.) The study, first conducted by the Atomic Bomb Casualty Commission (ABCC), was transferred in 1975 to control by the Radiation Effects Research Foundation (RERF), which has its headquarters at 5-2 Hijiyama Park, Minami-Ku, Hiroshima 732, Japan. and true magnitude of the radiation risk.

          RERF is equally funded by the government of Japan through its Ministry of Health and Welfare and by the government of the United States via the National Academy of Sciences (NAS) under contract with the Department of Energy (DOE).

          Cancer-observations among the survivors are reported by RERF in a series of Technical Reports (TR). Generally, these reports also appear in the journal Radiation Research, but usually in abbreviated form and after considerable delay. The two versions do not necessarily report identical conclusions (TR-9-87, p.35 versus Pr88, p.458).

          The major follow-up reports on cancer-mortality among the survivors have been made, for the past sixteen years, in four-year increments as indicated below:

Lifespan Follow-Up :


          One of the enormous scientific merits of this study is the plan to follow-up these individuals for their complete lifespans. With the A-Bomb Study's follow-up already extended for 37 years after the exposure to radiation, the study is telling us far more about cancer-induction by ionizing radiation than several other studies whose follow-up extends only 15 or 20 years beyond an exposure.

          Nonetheless, the A-bomb follow-up is far from complete. Readers can see from Tables 4-A and 4-B that almost two-thirds of the participating A-bomb survivors were still alive in 1982. Those who were young, during the 1945 bombings, account for an ever-growing share of the cancer-observations, of course. In the coming decades of the study, they will contribute very important information on the duration and true magnitude of the radiation risk.

All Dose-Levels :


          A unique scientific merit of this study can be seen from the entries in Table 4-A, Column D: The study includes a large unexposed group and a very great range of doses. This permits examination of dose versus cancer-response within a single study. If cancer-rate rises with dose in successively higher Dose-Groups, the argument for causality is stronger than if there were only one exposed group to compare with an unexposed Reference Group.

          In the T65DR dosimetry, the average internal organ-dose in Dose-Group 3 is estimated at only 11 centi-sieverts or rems. This is an organ-dose commonly received, medically, in special diagnostic procedures such as fluoroscopy, angiography, and in some isotope procedures such as radio-iodine uptake studies. At the other extreme is an average organ-dose of about 264 cSv (rems), a very high dose indeed when it is received by the entire body.

          Because the study also includes exposures between these extremes, it is one of the very few studies capable of revealing whether the cancer-risk per dose-unit is the same at all dose-levels or not. If it is the same, then the dose-response is linear. But if the cancer-risk per dose-unit is more severe at low total doses than at high doses, humanity needs to know it badly, because most human exposures occur at low total doses.

All Ages :


          The A-Bomb Survivor Study is uniquely valuable in additional ways. It is the only careful long-term study which includes persons at all ages at the time of exposure (ATE);   see Table 4-B, Column A. Thus it has already been capable of revealing what other studies could only suggest: Relative to the spontaneous cancer-rate, the cancer-risk per dose-unit is higher for persons who are young ATE than for persons who are old ATE. As the follow-up period extends beyond its current limit, the A-Bomb Study will enable analysts to quantify age-sensitivity with far more certitude than is possible now.

Both Sexes :


          In addition to the inclusion of all doses and all ages, the A-Bomb Study includes both sexes. This is not the case in many studies. Studies which are based on occupational or medical exposures are often limited or nearly limited to one sex.

All Cancer-Sites :


          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. By contrast, studies of equal population-size which involve the exposure of only a few organs will necessarily have many fewer cancer-cases to evaluate, and thus findings from such studies will be inherently less reliable due to random fluctuations in small numbers.

          Even in the A-Bomb Study, analysts can create a "small-numbers problem" if they practice excessive subdivision of the data by cancer-type, dose, age, sex. It is in the nature of numbers that any database can be rendered inconclusive by excessive subdivision. However, when analysts refrain from such questionable practices, the A-bomb database is a uniquely powerful resource for conclusive human evidence.

Definite Exposure-Date :


          Another enormous scientific merit of the A-Bomb Study is the relative absence of confounding variables, which can muddy and even invalidate so many other studies.

          For instance, in the A-Bomb Study, all the exposed groups received a single acute dose at the same time. In spite of some doses from residual radioactivity, no one doubts that the dominant exposure was from the explosion and fireball (Chapter 8). The definite exposure-date eliminates the problems which plague studies based on occupational exposures, where there is no clearly defined time-interval between exposure (which takes place over years) and death from cancer. Such studies can provide no reliable indication of the latency period or the duration of radiation's carcinogenic effect. (Even the dosimetry becomes muddier than usual, if cancer occurs during the period of exposure.) By contrast, the A-Bomb Study is capable of revealing, over its course, some valuable information about the duration of the carcinogenic effect, and about minimum, maximum and average latency periods.

Its Own Internal Control-Group :


          The most important confounding variable which is absent from the A-Bomb Study is the need to use an outside control-group (a group not closely related to the exposed group).

          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. Thus, if there is something peculiar and special about the 1945 populations of Hiroshima and Nagasaki -- something which would give them spontaneous cancer-rates unlike the rates elsewhere in Japan -- we can expect such a factor to be randomly distributed at the same rate among those who happened to be exposed by the bombs at one dose-level or any other dose-level, or not to have been exposed at all. (Radial distance, from where the bombs happened to drop, is the prime determinant of dose in the overall study.)

Summary :


          The A-bomb database is so valuable that it would be a real blow against human welfare if its scientific worth were undermined by irregular handling -- a danger discussed in detail by Chapters 5 and 6.







Table 4-A

Overview by Dose-Groups of the A-Bomb Study, 1950-1982 (T65DR Dosimetry). Raw Data.

================================================================================
| Col.A  Col.B   Col.C   Col.D    Col.E    Col.F    Col.G     Col.H      Col.I |
|                        RBE=2                                                 |
|RERF'S           TRUE   MEAN      TRUE  INITIAL    TOTAL   PERSONS (RAW VALUE)|
|EIGHT   KERMA    MEAN   ORGAN-    MEAN  PERSONS   DEATHS     STILL      FATAL |
|DOSE-   RANGE   KERMA   DOSE   AGE ATB       IN    1950-     ALIVE    CANCERS |
|GROUPS (RADS)  (RADS)   (cSv)  (YEARS)     1950     1982      1982  1950-1982 |
|                                                                              |
|  1       0       0.0     0.0    28.7     37173    12798     24375       2376 |
|  2      1-9      3.0     1.5    27.8     28855     9563     19292       1779 |
|  3     10-49    21.8    11.0    28.9     14943     5170      9773       1055 |
|  4     50-99    70.6    35.4    29.1      4225     1476      2749        314 |
|  5    100-199  142.5    71.4    27.9      3128     1029      2099        253 |
|  6    200-299  243.6   122.1    26.1      1381      455       926        131 |
|  7    300-399  345.2   173.2    26.6       639      215       424         51 |
|  8      400+   526.4   264.1    27.1       887      337       550         91 |
|                                       ====================================== |
|                                          91231    31043     60188       6050 |
================================================================================
NOTES ----- ATB means "at time of bombing" in 1945. Cancers exclude leukemia.
Columns B, C, E, and F come from Table 26-E.
Column D comes from Table 9-C, Row 11.
Columns G and I are sums coming from RERF's diskette "R10ALL" (see Tables 26-A,B,C,D).
Column H is Column F minus Column G.
          Reminder:   Fatal cancers are absolute numbers, not rates per 10,000 initial persons.





Table 4-B

Overview by Age-Bands of the A-Bomb Study, 1950-1982 (T65DR Dosimetry). Raw Data.

==================================================================================
|  Col.A     Col.B   Col.C  Col.D    Col.E    Col.F    Col.G   Col.H      Col.I  |
| RERF'S                     RBE=2                                               |
|  FIVE               TRUE    MEAN     TRUE  INITIAL   TOTAL  PERSONS (RAW VALUE)|
|AGE-BANDS   KERMA    MEAN  ORGAN-     MEAN  PERSONS  DEATHS    STILL      FATAL |
|IN YEARS,   RANGE   KERMA   DOSE   AGE ATB      IN    1950-    ALIVE    CANCERS |
|  ATB      (RADS)  (RADS)   (cSv)  (YEARS)    1950     1982     1982  1950-1982 |
|                                                                                |
|  0-9 YR    0-400+   18.4    9.23      4.1   18402      728    17674         93 |
| 10-19 YR   0-400+   29.8   14.95     14.6   19224     1715    17509        349 |
| 20-34 YR   0-400+   27.1   13.58     27.0   17691     3075    14616        949 |
| 35-49 YR   0-400+   24.1   12.07     42.0   20903    11234     9669       2788 |
|  50+ YR    0-400+   19.1    9.58     58.5   15011    14291      720       1871 |
|                                            =================================== |
|                                             91231    31043    60188       6050 |
==================================================================================
Columns A, B, C, and E come from Table 26-E.
Columns D, F, G, and I are calculated from the Master File 26 (A,B,C,D).
Column H is Column F minus Column G.
          Reminder:   Entries for mean doses and for fatal cancers include the entire study-population -- the unexposed groups as well as the exposed. Column I excludes leukemia. The radiation-sensitivity of the young, not apparent here, becomes clear in Chapter 15.









Next | ToC | Prev
back to RIC | CNR | radiation | rat haus | Index | Search