contents.txt                 10/11/94

These files include detailed tabulations and documentation of
the data used in the analyses of LSS cancer incidence for
the period from 1958-1987 for solid cancer and 1950-1987 for
leukemia, lymphoma, and myeloma. The papersby Thompson et al's
Cancer Incidence Part II: Solid tumors (Radiation Research
137 Supplement:11-67, 1994) and Preston et al's Cancer Incidence
Part III: Leukemia, lymphoma and multiple myeloma (Radiation
Research 137 Supplement:68-97, 1994) present the results of
RERF's analyses of these data.

In the analyses in Thompson et al's paper (Radiation Research
137 Supplement:11-67, 1994) separate person-year tabulations
were made for each organ dose. The solid cancer table included
here was defined using colon dose (TR87DATA.DAT). In order to
allow you to carry out analyses based on doses to the other
organs, we have included a table of city and age-at-exposure
specific organ dose adjustment factors (DS86ADJF.DAT).
The leukemia, lymphoma, and myeloma cancer table included was
defined using bone marrow dose (HEMA87.DAT).



Data file       Description

TR87DATA.DAT    LSS solid cancer incidence
                file
DS86ADJF.DAT    Organ dose adjustment
                factor file
HEMA87.DAT      Leukemia incidence file

Documentation
DOCUMENT.TXT    Documentation for the above
                three data files
CONTENTS.TXT    Descriptions for all of the
                files

Sample analysis
script file
TR87DATA.SCR    Script file for reading
                TR87DATA.DAT
TR87MOD.SCR     Script file to fit the
                basic models
DS86ADJF.SCR    Script file for computation
                of other organ doses using
                factors from DS86ADJF.DAT
HEMA87.SCR      Script file for reading
                HEMA87.DAT
HEMAMOD.SCR     Script file to fit the
                basic models of leukemia,
                lymphoma, and myeloma

Sample analysis
log file
TR87DATA.LOG    Log file of TR87DATA.SCR
                and TR87MOD.SCR
HEMA87.LOG      Log file of HEMA87.SCR and
                HEMAMOD.SCR


The solid cancer and leukemia data sets are detailed
tabulations of person years, case counts, and summary data
constructed from data on individual survivors. The
population for the solid tumor data set includes data on
80,206 survivors while the leukemia data set includes
information on 86,594 survivors. These totals include
survivors with DS86 shielded kerma estimates greater than 4
Gy. The data sets are structured to make it easy to exclude
survivors with total shielded kerma estimates above 4 Gy as
was done in the published reports.

The solid tumor data set is based on data obtained from the
Hiroshima and Nagasaki tumor registries together with LSS
mortality follow-up data. Since the tumor registries did not
start operation until 1958, follow-up is limited to the
period from 1958 to 1987. The ABCC/RERF leukemia registry
was the primary source of data on the incidence of leukemia,
lymphoma, and myeloma incidence however, data were also
obtained from the Hiroshima and Nagasaki tumor registries.
Because of the availability of the Leukemia Registry data,
follow-up for the leukemia data set begins on October 1,
1950. The difference in the number of survivors on the two
data sets is a consequence of the different starting dates
for the follow-up periods. As in the principal analyses of
the papers noted above, case counts are limited to first
primary cancers diagnosed in the registry catchment area.

Data on individual survivors were stratified on city, sex,
age at exposure (five year intervals), calendar time, and
dose to produce these data sets. Dose categories in the
leukemia data set are defined in terms of total bone marrow
dose while dose categories in the solid cancer table are
defined using total dose to the colon. The leukemia data
set is identical to that used in the analyses of Preston et
al. The solid cancer data set is identical to that used in
the pooled analysis of all solid tumors and in the analyses
of colon cancer risks. However, other site-specific
analyses in Thompson et al were based on data sets of the
same basic form in which the data were grouped by the
appropriate organ dose. It is impractical to distribute all
of these data sets. However, as noted above, the files here
contain city and age at exposure specific conversion factors
that can be used to compute estimates of doses to other
organs along with a command script that provides an explicit
illustration of how one can use these factors to compute doses
for other organs.

Each record in the main data files includes indicators of
sex, city, organ dose category, age-at-exposure category,
calendar time period, and other factors. The basic data for
each record in these tables include: person years, migration
adjusted person years, the number of people entering the
study, mean values of attained age, age at exposure, time
since exposure, and year. Doses are summarized by the mean
values of the gamma and neutron organ doses and the mean RBE
10 weighted total organ dose (i.e. gamma dose plus ten times
the neutron dose). The solid tumor table includes case
counts for all solid tumors as a group and for 31 specific
tumor types. The leukemia table includes counts for all
lymphomas, non-Hodgkin's lymphomas, multiple myeloma, all
leukemias, acute myelogenous leukemia, chronic myelogenous
leukemia, acute lymphocytic leukemia, and adult T-cell
leukemia.

The solid tumor data set has 3249 records while the leukemia
data set has 4894 records. The files are ASCII text files
in which the records have a fixed format and fields are
separated by blanks so that they can easily be read into
spreadsheets or other data analysis programs.

In addition to detailed documentation of the content of each
of the data files, EPICURE command script and log files for
fitting some of the models that were used in the analyses
for the published cancer incidence reports are also included
on the disk. The models that can be fit using these scripts
include: those used for the standard excess relative risk
estimates for solid cancers and excess absolute risk models
for all leukemias, non-Hodgkin's lymphoma, and multiple
myeloma. These scripts also illustrate how the data can be
read and how to convert colon dose estimates to approximate
values for other organs.

If these data are used as the basis for analyses in any
publication including working papers or technical reports, a
statement of acknowledgment must be included in the
manuscript. This statement should read:

      This report makes use of data obtained from the Radiation Effects
      Research Foundation (RERF), Hiroshima and Nagasaki, Japan. RERF
      is a private, non-profit foundation funded by the Japanese Ministry
      of Health, Labour and Welfare and the U.S. Department of Energy
      the latter through the National Academy of Sciences. The conclusions
      in this report are those of the authors and do not necessarily
      reflect the scientific judgment of RERF or its funding agencies.


Please send a copy of any manuscripts which make use of
these data to:

          Archives Unit, Library and Archives Section
          Information Technology Department
          Radiation Effects Research Foundation
          5-2 Hijiyama Koen
          Minami-ku
          Hiroshima Shi 732-0815  JAPAN


Copyright 2003.
Radiation Effects Research Foundation