United States
Environmental Protection
Agency
Office of
Radiation Programs
Washington, D.C. 20460
EPA 520/6-86-006
April 1986
Radiation
v>EPA
Comparison of Working
Level Ratios in Houses
Occupied by Smokers and
Non-Smokers
-------
COMPARISON OF WORKING LEVEL RATIOS IN HOUSES OCCUPIED
BY SMOKERS AND NON-SMOKERS
by
R. J. LYON
F. AU
J. M. HANS, JR .
--',
~'-
Cx..
;\:..>'
December 1985
I'
<-,I
"
'_\
d
~
u.s. Environmental Protection Agency
Office of Radiation Programs-Las Vegas Facility
P.O. Box 18416, Las Vegas, Nevada 89114
-------
DISCLAIMER
This report has been reviewed by the Office of Radiation Programs-
Las Vegas Facility, U.S. Environmental Protection Agency. Mention of trade
names or commercial products does not constitute endorsement or
recommendation for their use.
i i
-------
PREFACE
The Office of Radiation Programs of the U.S. Environmental Protection
Agency carries out a national program designed to evaluate population
exposure to ionizing and nonionizing radiation and to promote development of
controls necessary to protect the public health and safety.
Within the Office of Radiation Programs, the Las Vegas Facility conducts
in-depth fi~ld studies of various radiation sources (e.g., nuclear facili-
ties, uranium mill tailings, and phosphate mills) to provide technical data
for environmental impact assessments as well as needed information on source
characteristics, environmental transport, critical pathways for population
exposure, and dose model validation. The Office of Radiation Programs-
Las Vegas Facility also provides, upon request, technical assistance to
Western States and to other Federal agencies. The Las Vegas Facility parti-
cipated in a radiation survey in Butte, Montana. The primary purpose of the
survey was to intercompare various commercially-available methods of
measuring radon and radon decay products; and in doing so also expand
previous data relative to the sources which might elevate radon concentration
in the homes. Measurements of radon concentrations and working levels are
used in this report to characterize a lifestyle influence.
Readers of this
Programs-Las Vegas
requests for further
report are encouraged to inform the Office
Facility of any omissions or errors.
information are also invited.
of Radiation
Comments or
~~~
Sheldon Meyers, Acting Director
Office of Radiation Programs
i i i
-------
ABSTRACT
The equilibrium ratios of measured radon concentrations and working
levels in sixty-eight homes in Butte, Montana, were cataloged according to
smokers and non-smokers in each home. The ratios were statistically
evaluated to determine if there were differences in the average of the ratios
depending on the number of smokers in the homes. The equil ibrium ratio was
hi gher in homes with smokers compared to homes with no smokers. Th is rat i 0
was independent of the number of people smoking within the home.
iv
-------
Disclaimer
Preface
Abstract
List of Tables
List of
Figures
and Conclusions
Summary
Acknowledgements
Introduction
The Butte Study.
Design
Quality Assurance
Data Base
Data Selection.
Resu lts
and
Discussion
References
CONTENTS
i i
i i i
iv
vi
vi
vii
viii
1
3
3
4
4
4
6
15
v
-------
Number
Number
LIST OF TABLES
1.
List of the Yearly Average of Radon Concentrations,
Working Levels, and Working Levels Ratios for the
Butte Study Houses. . . . . . . . . . . . . . . .
.....
2.
Working Level Ratio of Continuous Measurements. .
.....
3.
Working Level Ratio of RPISU and PERM Measurements
From SILOM, ILOM, and NLOM Houses. . . . . . . .
.....
4.
Working Level Ratio of Grab Sampling Measurements
.....
LI ST OF FIGURES
1.
Distributions of the Yearly Averages of Radon
Concentrations, Working Levels, and the Working
Level Ratios for 68 Butte, Montana, Homes. . . . . . . . .
2.
Distributions of the Yearly Averages of the Working
Level Ratios for Homes of Smokers and Non-Smokers
.....
vi
Page
7
11
12
13
Page
8
9
-------
SUMMARY AND CONCLUSIONS
Measurements of radon and radon decay products were made in Butte,
Montana, from October 1981 through March 1983. Of the sixty-eight homes
sampled, thirty-six were occupied by persons who did not smoke. Twenty of
the houses had one smoker and eleven of the houses had two smokers. No
smoker information was collected for one house. Working level ratios
(working level x 100 divided by pCi Rn/liter) were computed for the three
types of sampling: continuous, weekly, and instantaneous samples.
Statistically, all three types of measurements show significant difference
between the working level ratios of residences of non-smokers and residences
of smokers. The working level ratio in houses with non-smoking residents
are significantly lower than the working level ratio in houses that are
occupied by people who smoke. These ratios were 0.29 to 0.31 for the homes
of non-smokers and 0.37 to 0.51 for homes occupied by smokers. Thus a
lifestyle, such as smoking, can significantly affect the ratio of indoor
radon decay products to radon.
vii
-------
ACKNOWLEDGEMENTS
We are indebted to Mr. Larry Lloyd and the staff of the Montana
Department of Health and Environmental Sciences for their commitment to
excellence in conducting the field measurements portion of the study. We
also recognize the tireless efforts of Mr. Allen Sparks, Computer Sciences
Corporation, who mastered the data base and processed the data required to
produce this report.
viii
-------
INTRODUCTION
The Butte Study arose from the investigation of the use of phosphate
slag in Butte, Montana. During the fall of 1977 and the spring of 1978, the
Montana Department of Health and Environmental Sciences (DHES) staff
di scovered that phosphate slag, produced at a nearby e 1 ementa 1 phosphorus
plant, was being used for fill, construction, and paving. Construction
products were pre-stressed concrete beams and concrete blocks. Structures
containing these products were identified and measured for indoor working
levels (WL) during the spring of 1978. Working level measurements were made
in structures which contained no phosphate slag for comparison purposes.
Many of these structures exhibited working levels well above those which
contained phosphate slag. Expansion of the sampling program indicated that
elevated work i ng 1 eve 1 s were common, and other potent i a 1 sources of indoor
radon became suspect.
The Montana DHES requested a grant in July 1974 from the U.S.
Environmental Protection Agency's Office of Radiation Programs (ORP) to
expand the measurement work. The ORP awarded DHES a contract to expand the
measurements and to investigate and identify the sources of radon
responsible for the elevated working levels in Butte homes (L183).
The measurement of radon decay products (RDP) is both labor intensive
and expens i ve. New radon and RDP measurement dev ices were appear i ng in the
market place and offered hope of reducing the RDP measurement costs. The
contract with DHES was amended and expanded to evaluate the performance of
these new products. The primary purpose of the expansion of the Butte Study
was to compare the ability of a variety of radon and RDP measurement devices
to measure radon concentrations and RDP levels in homes. The evaluation is
in progress.
1
-------
Although the study was designed to compare the performance of a variety
of instruments used to measure radon and radon decay products, other
interest i ng data evo 1 ved from the study. Measurements of radon and radon
progeny in houses in Butte, Montana, during the 18 months beginning October
1981 gave abundant data for statistical comparisons. The houses studied
contained different numbers of smokers and the question arose as to the
effect of smoking on RDP concentrations.
The following pages describe the study and show the results of smoking
on the working level ratios within the houses.
2
-------
THE BUTTE STUDY
DESIGN
Radon and working level measurements were conducted in sixty-eight
houses. These houses were divided into three groups. The first group
contained four houses and was designated Super Intensive Level of
Measurements (SILOM). The second group contained 16 houses and was
designated Intensive Level of Measurement (ILOM). The third group of homes
was designated Normal Level of Measurement (NLOM) (Ha85) (Ny83).
Houses selected for the study were not randomly selected. The SILOM
homes were selected to provide a broad range of radon and RDP concentrations.
The ILOM and NLOM houses were selected from those, where possible, that had
prior RDP measurements. Some houses were selected because of the willingness
of the occupants to participate in the study. Even though the house
selection is somewhat biased, it is believed that the selections represented
a reasonable cross section of the Butte house population and radon and RDP
concentrations.
SILOM houses were monitored using continuous radon gas monitors (Mo80),
passive environmental radon monitors (PERM's) (Ge77), alpha track detectors
(A181), grab samples (Th71, Th72), continuous working level monitors (Mo80),
and radon progeny integrating sampling units (RPISU's) (Sc74). ILOM houses
received all of the above less the continuous monitors' measurements. NLOM
houses were monitored using only alpha track detectors, and intermittent
measurements with PERM's, RPISU's and grab samples.
3
-------
QUALITY ASSURANCE
Quality assurance and control were an integral part of the study.
Procedures for operating, calibrating, handling, packaging, and storing all
of the measurement devices were prepared before the study began. This
program has been described by Nyberg (Ny85).
DATA BASE
There are 100,875 valid measurements in the data base. Files were
created for time intervals and the data were sorted into the files. Files
were generated for replicate measurements, averages, and normalized
averages. Several programs were written to display and to perform
statistical analyses of the data. In addition, a meteorological file of
hourly measurements and a file of outdoor radon measurements have been added
to the data base. Information was gathered concerning the houses and
occupants by questionnaire.
DATA SELECTION
Four houses had measuring devices which continuously measured radon and
RDP. These four houses had PERM's and RPISU IS wh ich measured week ly radon
and RDP throughout the 18 month study. Grab samples (instantaneous
samp 1 i ng) of radon and radon progeny were also taken week ly duri ng the
study. Sixty-four other houses were studied but continuous .measurements
were not included.
Initial survey questionnaires were filled out for each house studied.
Data concerning the number of smokers residing in each house were collected.
The data show that 36 of the houses were occupied by persons who did not
smoke. Twenty of the houses had one smoker and eleven of the houses had two
4
-------
smokers. The number of smokers residing in each house remained the same
throughout the study. No smoking information is available for one house.
Since the presence of cigarette smoke is reported (Ar73) (Co82) to affect
the ratio of radon decay product and radon concentrations, correlation
statistics were calculated for the number of smokers and working level ratio.
5
-------
RESULTS AND DISCUSSION
An overview of the Butte data base is presented in table 1 and figures
1 and 2. Table 1 1 ists the annual average of radon concentrations, working
levels (WL's) and working level ratios (WLRBs) in Butte houses. There is a
small difference in definition between the terms equilibrium ratio and
working level ratio; however, for the purpose of this report, these terms
shall be considered as having the same meaning and will be used
interchangeably. Likewise, all radon decay product measurements cited here
resulted in calculated working levels therefore these terms are also used
i nterchangeab ly. The work i ng 1 eve 1 rat i os were ca 1 cu 1 ated by d i v i ding the
working level measurement for a given time period by the radon measurement
for the same time period. Both measurements were made in the same room of
each house at the same time.
The annual average radon concentrations are the average of two week ly
PERM measurements for 52 weeks for the first twenty houses. The radon
concentrations for houses 21 through 68 are the average of two simultaneous
one-week PERM measurements made quarterly for one year. The annual average
work i ng 1 eve 1 va 1 ues for the first twenty houses are the average of 52
one-week RPISU measurements. Houses 21 through 68 were measured one week
per quarter with RPISU's. The working level ratio is the average RPISU
measurement in working levels times 100 divided by the average PERM
measurement in pCi/l. Figure 1 contains frequency distributions of the
annual average values as presented in table 1. The mean radon concentration
for the 68 homes is approximately 6 pCi/l ranging from 2.5 to 16 pCi/l at
the 68 percent confidence level. The mean working level is approximately
0.02 ranging from 0.007 to 0.06 at the 68 percent confidence level. Figure 2
6
-------
TABLE 1. LIST OF THE YEARLY AVERAGES OF RADON CONCENTRATIONS, WORKING
LEVELS, AND WORKING LEVEL RATIOS FOR THE BUTTE STUDY HOUSES
House WL+ Radon WLR+++ House WL+ Radon WLR+++
(pCi/1)++ (pCi/1)++
BS001 0.1810 71 .00 0.25 BS035 0.0073 2.55 0.29
BS002 0.0730 19.60 0.37 BS036 0.0425 7.38 0.58
BS003 0.0202 11.60 0.17 BS037 0.0079 1.70 0.46
BS004 0.0100 2.83 0.35 BS038 0.0076 5.78 0.13
BS005 0.0972 31.90 0.30 BS039 0.0079 4.52 O. 18
BS006 0.0814 21.60 0.38 BS040 0.0098 3.64 0.27
BS007 0.0398 9.54 0.42 BS041 0.0796 27.00 0.29
BS008 0.0416 14.70 0.28 BS042 0.0090 3.08 0.29
BS009 0.0291 11 .80 0.25 BS043 0.0043 1.55 0.27
BS010 0.0564 14.30 0.39 B5044 0.0040 0.86 0.47
B5011 0.0262 8.13 0.32 B5045 0.0257 9.58 0.27
B5012 0.0388 11 .40 0.34 B5046 0.0054 1.10 0.49
B5013 0.0410 20.50 0.20 B5047 0.0067 3.50 0.19
B5014 0.0853 39.90 0.21 B5048 0.0085 2.02 0.42
B5015 0.0332 11 .50 0.29 B5049 0.0042 1. 15 0.37
B5016 0.0688 20.90 0.33 B5050 0.0448 8.55 0.52
B5017 0.0257 8.23 0.31 B5051 0.0223 12.60 0.18
B5018 0.0297 5.67 0.52 B5052 0.0145 6.35 0.23
B5019 0.1010 31.90 0.32 B5053 0.0107 3.29 0.33
B5020 0.0792 24. 10 0.33 B5054 0.0097 3.55 0.27
B5021 0.0573 19.60 0.29 B5055 0.0038 1.89 0.20
B5022 0.0112 4.55 0.25 B5056 0.0182 9.25 0.20
B5023 O. 0100 5.38 0.19 B5057 0.0290 11.10 0.26
B5024 0.0199 4.63 0.43 B5058 0.0030* 0.43 0.70
B5025 0.0186 4.03 0.46 B5059 0.0109 2.55 0.43
B5026 0.0267 1 3.40 0.20 B5060 0.0098 2.04 0.48
B5027 0.0104 3.04 0.3' B5061 0.0287 6.94 0.41
B5028 0.0025 1.00 0.30 B5062 0.0076 1.53 0.50
B5029 0.0561 9. 14 0.61 B5063 0.0051 1.53 0.33
B5030 0.0557 21. 10 0.26 B5064 0.0020* 0.54 0.37
B5031 0.0024 0.63 0.38 B5065 0.0122 3.37 0.36
B5032 0.0195 9.84 0.20 B5066 0.0171 11.70 0.15
B5033 0.0303 8.62 0.35 B5067 0.0154 5.49 0.28
B5034 0.0234 4.29 0.55 B5068 0.0115 2.45 0.47
+WL = Working Level - average of 52 week ly RP I5U measurements for houses
B5001-B5020, average of quarter ly one-week RP I S U s amp 1 e s for houses
B5021-B5068.
++Radon (pCi/l) = average of two PERM's per house for 52 weeks for houses
B5001 - B5020; average of two one-week PERM measurements
made quarterly for one year for houses B5021 - B5068.
+++WLR = Working Level Ratio = Average WL x 100/Average Radon (pCi/l)
* = Only grab sample average available
7
-------
40
35
30
>, 25
lJ
<::
~
~ 20
a-
~
L
LL 15
10
o
o
40
35
30
>, 25
u
<::
II)
~ 20
a-
t)
L
LL 15
10
o
40 r
35
,
30 j
j
>, 25
u
<::
II)
~ 20
a-
t)
L
LL 15
':L
o
-------~----- ~---_._---~--
//
-~l
40
45
50
10
15 20 25 30 35
Radon Concentration (pCi/l)
0.12
I
I
--J
0.14
o
0.02
0.04
0. 06 0. 08
Workins Level
0.1
8
0.9
FiQure 1. Distributions of the Yearly Average~ of Radon Concentrations,
Working Levels, and the Working Level Ratios for 68 Butte, Montana, homes.
0.3 0.4 0.5
Working Level
0.8
-------
>-.
u 6
s:::
ill
:::5 5
0-
ill
L 4
LL
>-.
u 6
s:::
ill
:::5 5
0-
ill
L 4
LL
10
9
8
7
3
r:
/'
//'
/~
2
j
0.1
0.5
0.66
0.74
0. 18
0.26
0.34
0.58
0.42
Norklng Leve1
(smokers)
Rat 1 0
10
I
I
I
I
I
I
I
I
i
I
I
I
9
8
""/1
~
)
3
2
o
0.1
-G---,---~
0.5 0.58 0.66 0.74
(non-smokers)
0.26
0.34
0.42
Ratio
0.18
Norklng Level
Figure 2. Distributions of the Yearly Averages of the Working Level Ratios
for Homes of Smokers and Non-Smokers.
9
-------
The working level ratio of continuous measuring devices was calculated
by dividing weekly average continuous working level monitor measurements by
the week ly average radon gas mon i tor measurements and mu 1 tip lyi ng by 100.
These working level ratios were averaged over all weeks of the study and all
houses that had no smokers in residence. The average of the weekly
continuous working level ratios was also calculated for the one house that
had a smoker.
In calculating the various t-values given in this report, the following
equation was used
Xl ;t X2
t =
S2 S2
L+L
nl n2
Eq. 1 (Za84)
where Xl ~ X2
S2
p
nl & n2
= working level ratio means
= pooled variances of means
= samples sizes
The continuous radon measurements were taken during the second and
third half-years of the study from the SILOM houses. Using all the
cont i nuous measurements the average work i ng leve 1 rat i os are presented in
table 2.
10
-------
TABLE 2. WORKING LEVEL RATIO OF CONTINUOUS MEASUREMENTS (ENTIRE STUDY)
Mean Standard Error No. of Weekly Averages
Non-smokers 0.3031 0.00930 168
Smokers (1) 0.3683 0.00561 51
Using all the continuous measurement data,
indeed highly significant since t = 6.00 is
t-value of 1.98 at a = 0.05.
the presence of a smoker is
greater than the tabulated
A similar comparison of the working level ratio of the PERM and RPISU
data can be made. Tahl~_3~ives the working level ratio statistics of the
weekly RPISU measurements and the weekly PERM measurements. The data cover
the measurements collected in the fi rst year as well as the ent i re study
period (one and one half years).
11
-------
TABLE 3. WORKING LEVEL RATIO OF RPISU AND PERM MEASUREMENTS FROM SILOM,
ILOM, AND NLOM HOUSES
Mean Standard Error No. of Measurements
ONE YEAR
Non-smokers 0.3037 0.00614 521
Smokers (1) 0.4216 0.02026 244
Smokers (2) 0.3950 0 . 01160 214
Smokers (lor 2) 0.4092 0.01208 458
ENTIRE STUDY
Non-smokers 0.2936 0.00556 641
Smokers (1) 0.4005 0.01710 297
Smokers (2) 0.3918 0.01125 225
Smokers (lor 2) 0.3968 0.01087 522
12
-------
A comparison of the means shows no significant difference between one
and two smokers in a home but a significant difference between the homes of
smokers and non-smokers. The t-values, in the case of one smoker versus two
smokers, are t (1 year) = 1.14 and t (entire study) = 0.42; the t for
smokers versus non-smokers are t (1 year) = 7.79 and t (entire study) = 8.45,
respectively. The tabulated t-value at a = 0.05 is 1.96.
Table 4 shows the mean and error of the working level ratio based on
grab sampling for the first year of study and the entire study period from
the SILOM, ILOM, and NLOM houses. Grab sampling and counting of radon and
radon decay product measurements were completed weekly in 20 houses and
quarterly in 48 houses during the study.
TABLE 4. WORKING LEVEL RATIO OF GRAB SAMPLING MEASUREMENTS
Mean
Standard Error
No. of Measurements
ONE YEAR
Non-smokers
Smokers
Smokers
Smokers
(1 )
(2 )
(1 or 2)
0.3043
0.4813
0.5373
0.5081
0.01264
0.03354
0.06297
0.03484
554
282
259
541
ENTIRE STUDY
Smokers
Smokers
Smokers
(1)
(2 )
(1 or 2)
0.3031
0.4572
0.5261
0.4881
0.01164
0.02806
0.05918
0.03071
691
340
276
616
Non-smokers
13
-------
Grab sampling confirmed the mean comparisons of other measurements. The
t-value comparing smokers are t (1 year) = 0.78 and t (entire study) = 1.05.
The t-values for smoker versus non-smoker means are t (1 year) = 5.50 and t
(entire study) = 8.45. The tabulated t-value is 1.96 at a = 0.05, indicating
a significant difference in the means of the working level ratio's from
houses with smokers and from houses with no smokers.
An ana lys is of vari ance was made on the PERM and RP ISU data for
different seasons. There was a significant difference between the means of
the working level ratios for each time period: winter months, full year,
and entire study. The Student-Newman-Keuls multiple range test showed that
the mean working level ratio was significantly lower in homes with no
smokers. The test showed no difference between one or two smokers in the
homes. The data shows, therefore, that the presence of a smoker or smokers
may significantly affect the ratio of indoor radon decay products to radon.
Such an effect has been reported previously (Co 78, Is 86, and NCRP No. 77).
The mechani sm suggested by these authors is that tobacco smoke provi des
condensation nuclei to which radon decay products may attach. As the smoke
particles are small enough to remain suspended in air, the attached radon
decay products are available for collection whereas if there were no smoke
the radon decay products would rapidly migrate to walls, furniture, curtains,
and other large objects in the indoor space.
14
-------
REFERENCES
A181 Alter, H.W. and R.L. Fleischer, 1981, "Passive Integrating Radon
Monitor for Environmental Monitoring," Health Physics, 40, 693-702.
Ar73 Archer, V.E., J.K. Wasner, and F.E. Lundin, 1973, "Uranium Mining and
Cigarette Smoking Effects on Man," J. Occup. Med., 15(3), 204-211.
C082 Cohen, B.L., 1982, "Failure and Critique of the BEIR III Lung Cancer
Risk Estimates," Health Physics 42, 267-284.
C078 Cooper, G. and G. Langer, 1978, "Limitations of Commercial Condensation
Nucleus Counters as Absolute Aerosol Counters," J. Aerosol Sci. 9, 65.
Ge77 George, A.C. and A.J. Breslin, 1977, "Measurements of Environmental
Radon with Integrating Instruments," Workshop on Methods for Measuring
Radiation in and Around Uranium Mills (Edited by E.D. Harward)
sponsored by Atomic Industrial Forum, Inc., 7101 Wisconsin Ave.,
Washington, D.C. 20014.
Ha85 Hans, J.M., Jr., R.J. Lyon, and M.
of Indoor Radon and Radon Decay
Fami 1y Homes," Proceedi ngs of the
Physics Society, January 6-10, 1985,
Israeli, 1985, "Temporal Variation
Product Concentrations in Single
Midyear Symposium of- the Health
Colorado Springs, CO.
Is86
Israeli, M., 1986, "Deposition Rates of Radon Progeny in Houses," To
be published in Health Physics in 1986.
15
-------
M080 Momeni, M. H., A.J. Zielen,
Kisieleski, 1980, "Systems
Radon-Ra222 - Concentrat i on
(ANL/ES-88).
J.E. Miranda, Jr., N.D. Kretz, and W.E.
for Continuous Measurements of Ai rborne
and Work i ng Leve 1, II NUREG Report CR-1412
Ny83 Nyberg, P.C., 1983, "Method for the Calculation of Radon Response
Characteristics of Integrating Detectors," Health Physics, Vo. 45, No.
2 (August), pp. 544-550.
Ny85 Nyberg, P.C., 1985, "Calibration and Quality Assurance Techniques for
a Major Radon Measurement Compari son Study, II Proceedi ngs of the
Midyear Symposium of the Health Physics Society, January 6-10, 1985,
Colorado Springs, CO.
Sc74 Schiager, K.J., 1974, IIIntegrating Radon Progeny Air Sampler, II Am.
Indust. Hygiene Assoc. J., March 1974, pp. 165-174.
Th71 Thomas, J.W., 1971, IIDetermination of the Working Level of Radon
Daughters by the Modified Tsivoglou Method, II Department of Energy
Environmental Measurements Laboratory, NY (unpublished ~ote).
Th72 Thomas, J.W., 1972, IIMeasurement of Radon Daughters in
Counting of Air Filters, II Department of Energy
Measurements Laboratory, Rep. HASL-256, New York, NY.
Air by Alpha
Environmental
Za74 Zar, J.H., 1974, Biostatistical Analysis, (Reprinted by Prentice-Hall,
Inc., 1984), p. 128.
tr u.s. Government Printing Office: 1986-491.191/52927
16
------- |