Radioactive Aerosols - Aerosols: Measurement, Dosimetry, and Health Effects - page 395

were present. Personal dosimeters for radon and its progeny have not been developed until recently
and their usage has been limited; hence, detailed information on concentrations within mines and
time spent in various locations has not been available to most epidemiological studies.
In the epidemiological studies, these ideal approaches have been replaced by various pragmati-
cally determined strategies for exposure estimates that draw on measurements made for regulatory
and research purposes and extend the measurements using interpolation and extrapolation to com-
plete gaps for miners in particular years. Additionally, missing information for mines in the earliest
years of some of the studies was completed by either expert judgment or by recreation of operating
conditions. It should be noted that very little has been done to quantitatively assess uncertainties
related to the replacement of this ideal approach with the pragmatic approach.
This is a very important issue because, in reality, uncertainties in the assessment of the dose
and even exposure are many times greater than statistical errors in the assessment of lung cancer
mortality, which make risk assessment very uncertain. In Alterman (1974), Urusov (1972), Ruzer
et al. (1995), and measurements of radon decay product concentrations in different working
sites, the direct measurement of the activity in the lungs of individual miners and lung cancer
mortality and lung sickness for different groups of miners was studied in nonuranium mines of
Tadjikistan. It was shown that variations in the radon and radon decay product concentrations
vary by a factor of 2-10 and that calculated radon progeny intake varies by a factor of 3-14. A
similar nonuniformity was found in lung cancer mortality and lung sickness. The uniformity
approach can also be the source of uncertainties much greater than the statistical errors in the
assessment of mortality.
From the point of view of dose (alpha particle energy deposited in the lung) assessment of all the
factors, that is, concentration of radon decay products measured by standard procedure, the ratio
between this measurement and the concentration in the breathing zone, volume breathing rate, and
deposition coeficient are equally important, including eficiency in using respirators, because all
these factors affect the amount of radioactivity deposited in the lung.
In this section, based on the quantitative analysis of accuracy of dosimetric factors, we will show
that the conception of uniformity is incorrect in terms of both dosimetry and epidemiology.
As a result, we suggest that intake, which takes into account not only radon progeny concen-
tration in the working site but also the physical load of work for different groups of miners and
the scenario of exposure, is a better measure of radon and its decay products dosimetry than
exposure.
15.10.1.1 Discussion on Miner Radiation Dosimetry: Quantitative Approach
In every miner epidemiological study, two important parameters should be determined together
with their errors (uncertainties):
1. Lung cancer mortality
2. Radon decay products dosimetric characteristics: concentration, exposure, activity in the
lung, intake, dose of alpha-radiation
In a majority of published studies, only the assessment of concentration and exposure takes place,
despite the fact that both are characteristics of air and not of tissue irradiation. It is not even a char-
acteristic of air really breathed in by miners, as was shown in studies (Domanski et al., 1989).
It should also be mentioned that to achieve reliable data on risk assessment, uncertainties (errors)
for both mortality and dosimetry must be comparable.
One of the most comprehensive studies of this kind is BEIR VI, which summarizes dosimetric
and epidemiological data from 11 local studies in uranium and nonuranium mines in different
countries. Similar data from the former Soviet uranium industry were not included in this report,
because at the time of preparation data were yet to be classiied. In the report BEIR VI, two objec-
tive criteria were established for inclusion in the study: