g    UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
          v            WASHINGTON, D.C. 20460
                                    EPA-SAB -RAC-S 9 -01 7


      25 » 1989                                          Tur AQ
Honorable William K, Reilly
Administrator
U. S. Environmental Protection Agency
401 M Street, SW
Washington, DC  20460

SUBJECTS  Science Advisory Board Review of the Radon
          Measurement Proficiency Program

Dear Mr. Reilly:

     On September 23, 1987 the Director of the Office of
Radiation Programs revested that the Science Advisory Board
review the Radon Measurement Proficiency Program*  The request was
referred to the Radiation Advisory Committee which formed the
Radon Measurement Subcommittee to conduct the review.  The
Subcommittee held two open meetings.  The first was held at the
Eastern Environmental Radiation Facility in Montgomery, Alabama
where the Radon Measurement Proficiency Program is housed on
January 26-27, 1988 and the second was held February 16-17, 1988
at Region I Headquarters in Boston,  The Subcommittee also held a
writing session July 7, 1988 in Boston.

     The Subcommittee addressed performance standards,
statistical methods, standard measurement protocols,
participant's procedures, blind tests of passive devices,
consensus standards and voluntary accreditation, and user fees,

     The Subcommittee's recommendations include:

     o    development of separate objectives for devices or
          methods used for screening, diagnostic measurements,
          and exposure evaluations,

     o    consideration of different testing protocols for
          passive devices and active measurements,

     o    design of the testing program to obtain independent
          measurements from each device of method to be tested,

-------
r—
                 o
testing should be done blind, wherever practicable as
in the instance of passive measmfeaes.t.devices».and

strongly urgess the assignment of a full-tine
statistician to the program at least until it is well
established.
                 These  and other recommendations are described in greater
           detail  in the attached report which was approved by the Executive
           committee January 31,  1989,

                 The  Science Advisory Board was pleased to participate in
           this  review and appreciates  the opportunity to be briefed on the
           activities  of the Radon Measurement Proficiency Program.  We
           request that the Agency consider the advice contained here and
           respond in  writing to our recommendations.

                                          Sincerely,
                                          Raym
                                          Executive Committee
                                          William J.  Setat|lr  Chairman
                                          wadi a
-------
J     *
                             SCIENCE ADVISORY BOAMD
                          RADIATION ADVISORY COMMITTEE
                         RADON MEASUREMENT SUBCOMMITTEE
                                     REVIEW
                                     OF THE
                          OFFICE OF RADIATION PROGRAM'S
                      RADON MEASUREMENT PROFICIENCY PROGRAM
                                  JANUARY 1989

-------
                             NOTICE
           *
     This report, has been written as part of the activities of
the  Science  Advisory  Board,  a public  advisory group  providing
extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection  Agency,   The
Board is structured to provide a balanced expert assessment of the
scientific issues  related to problems  facing the Agency,   This
report has  not  been  reviewed  for approval  by  the agency  and,
hence, the  contents of the  report do not  necessarily  represent
the views  and policies of  the Environmental  Protection  Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.

-------
              U, S, ENVIRONMENTAL PROTECTION AGEKCY
                     SCIENCE ADVISORY BOARD
                  RADIATION ADVISORY COMMITTEE
                 RADON MEASUREMENT SUBCOMMITTEE

                             ROSTER

                            CBAIRMAM

Dr, Keith J, Schiager, 100 Orson-Spencer Hall, University of
     Utah, Salt Lake City, Utah 843,12

                       MEMBERS/CONSULTANTS

Dr. Andreas George, Aerosols studies Division, Environmental
     Measurements Laboratory, U,S, Department of Energy,
     376 Hudson Street, New York, NY 10014

Dr, John Barley, Box M268, Hoboken, Hew Jersey  07030

Dr, Nan Laird, Department of Biostatistics, Harvard School of
     Public Health, 677 Huntington Avenue, Boston, MA  02115
                       EXECUTIVE SECRETARY

Mrs, Kathleen W. Conway, Deputy Directorf Science Advisory Board/
     A-101F, U.S. Environmental Protection Agency, 401 M street,
     S.W., Washington, DC  20460

                         STAFF SECRETARY

Mrs. Dorothy M* Clark, Secretary, Science! Advisory Board, A-101F,
     U.S. Environmental Protection Agency, 401 M Street, S.W.,
     Washington, DC  20460

-------
                        TABLE OP COm|ai.TS

1.0  INTRODUCTION                                      1

2.0  SOMMABY OF RECOMMENDATIONS                        2

     2,1  Performance Standards                        2
     2,3  Statistical Methods                          2
     2,3  Standard Measurement Protocols               3
     2.4  Participants1 Procedures                     3
     2.5  Blind Tests of Passive Devices               3
     2.6  Testing of Active Devices and
          Operators                                    3
     2,7  consensus Standards and Voluntary
          Accreditation                                3
     2,8  Adequate Program Support                     4

3,0  DEVELOPMENT OF THE RADON MEASUREMENTS
     PROFICIENCY PROGRAM (RMPP)                        5

4.0  PROFICIENCY TESTING OBJECTIVES                    7

     4.1  Performance Standards                        7
     4.2  Statistical Methods                          9

5,0  PROFICIENCY TESTING PROCEDURES                   10

     5.1  Standard Measurement Protocols              10
     5.2  Participants' Procedures                    11
     5.3  Blind Tests of Passive Devices              11
     5,4  Testing of Active Devices and
          Operators                                   12
     5,5  Radon Concentration standard                13

6.0  LONG-TERM PLANNING FOR THE RMPP                  14

     6.1  Consensus standards and Voluntary
          Accreditation                               14
     6,2  Financial Support Needed for Adequate
          Program                                     14

APPENDIX A - QUESTIONS SUBMITTED TO THE SAB
             BY THE EPA

-------
                           ABSTRACT

     The Radon  Measurement  Proficiency Program  (RHPP)  of  the
Environmental Protection Agency's  Office  of Radiation  Programs
was reviewed by the  Science  Advisory  Board's Radiation  Advisory
Committee (SAB/RAC).   The SMPP was established in 1986 to provide
states and homeowners with  some assurance that individuals  and
organizations making  radon measurements were, in fact,  competent
to do so.

     The SAB/RAC"s  recommendations  include:    definition of
separate objectives  for  screening,  diagnostic,  and exposure
measurementsi the  need  for  independent  exposures  when  testing
devices  submitted  by participants!  improved statistical  support
for  the program?  blind  testing  for  passive  devices,   and
consideration of establishing  consensus standards, voluntary
accreditation,  and user  fees.

-------
1* 0  INTRODUCTION

     Upon receiving a request from the Director of the  Office of
Radiation Programs (ORP)  the Science Advisory Board's Radiation
Advisory Committee formed a Radon Measurements Subcommittee to
review the Radon Measurements Proficiency Program (RMPP)»   The
Subcommittee held a public meeting at the Eastern Environmental
Radiation Facility (EERF) of the EPA in Montgomery,  Alabama on
January 26-27, 1988,  The ESRP staff briefed  the  Subcommittee on
the RMPP and provided a  detailed tour of the  facilities used for
the RMPP and  other  programs conducted at th& EERF.  On February
16-17, 1988 the Subcommittee met at EPA*s Region I Headquarters in
Boston,  Massachusetts,   Staff  from  the -National   Bureau  of
Standards (NBS) described the structure and functions of the NBS-
sponsored National  Voluntary Laboratory  Accreditation Program.
The  Subcommittee met  again in  Boston  on  July 7,   1988,  to
complete the writing of the report,

      The Subcommittee considered the eight questions initially
posed by the EPA, as well as oral and written comments  from
members of the public.  The Subcommittee was pleased with  the
open nature of the discussions with the EPA staff, and  encouraged
by the fact that several of its  recommendations have been  adopted
and put into effect during round 5 of the proficiency testing,
even before this report was completed.

     This report includes a brief summary Of the Subcommittee^
recommendations, a short history of the development of  the RMPP,
followed by a discussion of the  major topics considered by the
Subcommittee«

-------
2.0  SUMMARY OF RECQMMEMPaTIOHS

2.1 Performance Standards

     The EPA may need to define separate objectives for devices
or methods used only for screening measurement and those that are
to be used for diagnostic purposes or exposure evaluations.
Different testing protocols should also be considered for
different categories of measurements, e.g. passive devices versus
active measurements, since a single measurement by one
measurement technique may carry greater significance than a
single measurement by another technique.

     For passive devices used only  for  screening  measurements, a
total uncertainty of a factor of 2 from the Jcnown test value at
the 90% confidence level is recommended.  (See Sec, 4.1)

     For passive devices used for doslmetry purposes, i.e. for
determining long-term average concentrations of radon or radon
progeny, a total uncertainty of 25% at the 90% confidence level
is recommended, (see Section 4.1)

     For active devices used for screening, diagnostic or
dosimetric purposes, a total uncertainty of 25% at the 30%
confidence level is recommended, (see Section 4.1}

     The EPA should design the testing program to obtain several
in
-------
2.3, Standard Measurement Protocols
           f
     The EPA should require that the information provided to the
client by the measurement company before and after the
measurement include clear statements of the purpose and proper
interpretation of the measurement. (See Section 5.1)

2 , 4  Participants f Procedures

     Detailed procedures for exposing,  evaluating (counting) and
calibrating each passive measurement device, as veil as the
methods used for calculating the radon concentration, should be
submitted with the application for admission to the proficiency
testing program,  (See Section S.2)

     For active measuring devices, the application should include
performance characteristics, calibration data, qualifications of
existing operators and details of the training program for future
or unia0jitj.fj.e5i operators, (See Section 5.2)
2,5  Blind Tests of passive Devices

     The technical advantages of blind testing are sufficient
that the EPA is encouraged to adopt this system as soon as
feasible. (See Section 5.3)

2,6  Testing of Active Devices and Operators

     The accepted use of grab sampling should be made conditional
on the use of good procedures by a qualified operator,  (see
Section 5,4)

     Active devices admitted to the RHP Program should be
certified as to their original calibration and continued
maintenance of that calibration.  (See Section 5.4)

     The qualification and/or training of the operators of active
devices must be retested or reviewed periodically*  (See Section
5,4)

2.7  Consensus standards and voluntary Accreditation

     The Subcommittee recommends that serious consideration be
given to establishing consensus standards and voluntary
accreditation procedures »

     EPA should solicit the advice  of various organizations that
establish  consensus  standards  and  voluntary  accreditation
programs and  should work  toward converting  the  RMPP to  such a
voluntary program as expeditiouely as possible.

-------
2*8  Adequate Program Support
           r
     The Subcommittee notes that a successful program will
require ade
-------
3.0  BEVBIOPMENT OF THE B&DQN MEASUREMENTS  PROFICIENCY PROGRAM
     EPA established the STOP in response to specific needs and
urgent requests far help from private laboratories and state
agencies*  Concern, and even fear,  among residents in areas where
high indoor radon concentrations had been found, and widespread
media coverage, led to numerous requests for radon measurements,
One of the immediate effects of the rapidly increasing demand for
radon  measurements  was the  sudden appearance  of  numerous
individuals or companies offering  radon measurements  for a fee.
Since many  of these  individuals  and companies had  no previous
experience and no apparent  qualifications related to indoor radon
measurements  or  evaluations,  state  agencies  became  rightly
concerned.  The HIPP  was established by the EPA as  a prompt and
effective response to the growing problem,

     The  EPA  contacted companies  already  involved  in  radon
measurements and  encouraged  them  to participate in voluntary
proficiency testing.  During these initial phases of the program,
it was  necessary  and normal to  assist  inexperienced companies
with measurement  equipment  and methods in  order to  assure the
development and availability of radon measurement capabilities to
meet the needs  of  the public.    Companies  who  were  able  to
demonstrate adequate  proficiency  were listed  in  a "proficiency
report" distributed by the EPA to state agencies.  For companies
providing radon  measurement services,  listing  in the  EPA's
proficiency report has become a de facto license to do business.

     When the 1MPP was first initiated in early 1986 , the EPA
facilities were not adequate to conduct the program*  The EPA
received  assistance  from  the   Environmental  Measurements
Laboratory  (EML)  of the Department of Energy  (DOE)  in New ¥ork
City, where  the  first three  rounds  of proficiency  testing were
conducted.  The program was originally designed as a single test
process that would be repeated every three _ months with results
reported  after each round  of testing.  Round  one,  conducted at
EML  in  April  1986,  result  in  35 participants  listed,"  85 were
listed after round three in November 1986,  Recognizing the need
for  larger  and more versatile facilities  for  radon measurements
calibration and testing, the EPA built a new radon chamber at its
Eastern  Environmental Radiation Facility  (EEHFJ  in Montgomery,
Alabama.  Round 4 of the FMPP was conducted in  the new chamber.
Round 5 was conducted during the summer of 1988.

     Until the Subcommittee's review, EPA conducted all of the
proficiency testing in a batch mode.  For logistical reasons,
all participants were required to submit devices  or equipment for
testing within the same limited time interval,  and all devices
were exposed to the same test atmosphere.  T?he  scheduling of
groups of similar devices for simultaneous testing has been
necessary to  accommodate  all participants  with the  existing
facilities and personnel.

-------
     Calibration and testing of radon and radon progeny
measurement*devices and instruments requires a large-volume test
chamber   containing  very  well-known  radon  and   progeny
concentrations,  chambers in which appropriate concentrations of
radon and other  atmospheric constituents can be established and
maintained  for extended  periods  of  time exist  in  only  a few
locations in the  country,  Even the  few existing chambers were
not  designed  for testing  of  large numbers  and varieties of
instruments and devices.   In addition to  the  limited number of
test chambers,  the very nature of the measurements restricts the
number of tests that can be accommodated  in a single  chamber.  To
minimize uncertainties in the  determination  of the true average
radon concentration to which a device is exposed., it is desirable
to  maintain a  nearly constant  radon concentration  during the
testing period.  The most commonly  used radon measurement devices
are passive devices  such as charcoal canisters  and alpha track
detectors.  Since most of the passive measurement devices require
exposure times  of  several days,  it is preferable for most
applications to  change the test atmosphere  no  more often than
weeKly,

     The older of the two radon chambers  at the EERF  is quite
small (3,6 cubic meters)  and access to the test atmosphere  is by
means of only a few ports and a small panel*  Air flows through
the chamber in a single pass,  with  radon  being  injected into the
incoming air and the outlet being exhausted to the outdoor
atmosphere.   The new chamber has a  volume of  42 cubic meters and
the air is recirculatad.   Operators may walk  into the new chamber
to install or exchange devices, an
-------
accumulated activity.   The temperature  and humidity of the
atmosphere in each chamber can. also  be  controlled, as can the
concentration of small airborne aerosols, i.e. condensation
nuclei.

4.0  PROFICIENCY TESTING OBJECTIVES

4,1  Performance standards

     In order to demonstrate proficiency  in making radon
measurements, participants and measurement devices must be tested
against an appropriate performance standard.   The test procedure
must be appropriate to  ascertain  that  the  measurement device or
method does,  in fact,  meet  the performance  standard.   The
following two questions posed  by  the EPA relate directly to the
performance standard and the means  for demonstrating measurement
proficiency,

     1PA Question li "The sample design was selected to insure
that a measurement company could,  within  80 percent confidence,
identify a radon level as being greater than 4 pCi/L when in fact
the true level is 8 pCi/L» with a  single  measurement in a house.
Is the sample design now in use sufficient to meet SMPP program
objectives?

     EPA Question 2i "EPA presently evaluates  all devices by the
same measurement criteria, i.e. four measurement results must be
within 25  percent of  the  known  chamber  levels,    EPA  is
considering changing  this criteria  (sic)  to  obtain  better
confidence in companies* abilities.  Are  the criteria outlined in
the BMPP Background Document scientifically sound?**

     The Subcommittee agrees  that,  at  concentrations  of the same
order  of magnitude as the  EPA's present  radon concentration
guideline  (Rcc),  a  screening measurenent  made with a  passive
device  solely  to  determine whether follow-up  measurements are
warranted should  produce  a result within a factor of 2  of the
known  test value almost all (e.g. 90%) of the time.  Although a
possible error  of  a  factor of 2 may  appear to be excessively
large,  the  Subcommittee believes  this degree  of uncertainty is
acceptable for  screeninet measurements because?   (1)  the
measurement protocol inherently introduces  bias on the high side
of the true annual average concentration  by more than a factor of
2, and  (2) greater  precision  is unnecessary and  could result
in increased costs to homeowners.   Since  many decisions regarding
radon  mitigation  may  be  based  on  a  very  small  number  of
measurements, the  total  uncertainty accepted  in the measurement
should include the uncertainties of  both  bias and precision.

     Since it is possible, under the revised procedures adopted
for round five, to deliver independent  exposures to four devices
within a short time, this should be  adopted as the minimum number
of  measurements used  for  determining the  acceptability  of
performance.  As discussed in Sec. 4.2f the acceptance criteria
originally proposed by the EPA, as well as those recommended by

-------
the  Subcommittee,  are based  on statistical assumptions that
include  independence  of  test  measurements.   As  additional
facilities become available,  this minimum number of  samples could
be increased.

     The Subcommittee strongly  recommends against  stating the
1MPP objectives in terms of the RCG for  lifetime exposure,
i.e. 4 pCi/L,  since the value  becomes  easily misinterpreted in
the context of  the EPA's screening measurement protocol because
conditions in the house are selected deliberately  to maximize the
radon concentration measured.   Too  much emphasis  has been given
to this numerical  value  already and it  is, unfortunately, being
interpreted as a hard and fast  line between "safe"  and "unsafe".
The  1FA should make  a deliberate effort to discourage  this
regulatory attitude toward radon measurements  and  mitigation,

     To remain listed as proficient in measurements, a company
should be required to repeat its performance testing
periodically since performance  may change over time  due to factors
such as  aging  or  new  equipment,  changes   in  operators  and
analysts,  etc.   The  exact interval  for  retesting should be
developed by the EPA, staff on the basis  of  cumulative performance
data and availability of facilities.  The Subcommittee recommends
that participants that are new to the program be  retested within
a year after  first being  listed as proficient.   The retesting
interval could be  increased after a participant has demonstrated
competence two  or  three  times.   In fact,  the retesting interval
could easily be determined by the cumulative performance over the
past two  or three tests on  the basis  of  total  error (accuracy
plus precision).

     The Subcommittee considers that separate  objectives, and
testing protocols may be appropriate for different types and
conditions of measurements.   Because of the  much greater cost,
sampling  devices and  methods  that  require an operator  should
produce measurements of much better quality than  can be obtained
by  passive  devices.    Also,  measurements  made to determine the
need for, and/or extent of,  mitigation activities  should, because
of  the  potential  cost to the  homeowner,  be of  higher quality
than screening  measurements  made simply to determine if further
assessment is  necessary.   For passive  devices  that may be
distributed in  various ways and  returned  by  mail,  the emphasis
should be placed on the  consistency of  instructions to the user,
handling,  processing and  reporting procedures.   For  active
devices that require an operator, both the  adequacy  of the device
and the competence of the operator should be evaluated.

     Passive devices used for dosimetry, i.e.  for  determining
long-term average concentrations, should produce results that are
within 25 percent of the known  test value at least 90 percent of
the time.   This level of uncertainty is achievable with devices
currently available  and should be  required to assure homeowners
that mitigation costs are truly justified.
                           8

-------
     The EPA should separate education and calibration assistance
from  proficiency  testing because  the  support  activities
appropriate to the first two activities are not appropriate to e
objective testing program*   A company, or individual operator,
should be allowed to participate  in the RMPP only after
demonstrating understanding of  the devices or methods intended
for use. and after some type of  calibration has been performed.
Because of the lack of commercial calibration facilities EPA may
wish to continue to allow its chambers to be used for
calibrations, but this function should not be conducted
simultaneously with that of proficiency testing,

4*2  statistical Methods

     The current EPA testing procedure,  as  stated in the program
reports ana as presented to the  Subcommittee,  requires  that the
average of four measurements fall within 251 of the test chamber
value.    The Subcommittee  learned,  however,  that  all  four
measurements were normally made simultaneously  in  the  same
atmosphere.  Beyond the faot that the test measurements could not
be  considered  to  be  independent,   thus  invalidating  the
statistical assumptions on which  proficiency was based,  the
Subcommittee was  concerned that  the procedure  allowed  too many
opportunities for unscrupulous  participants to obtain the results
by means other than actual  measurement.

     The current performance criteria do not talce into account
uncertainties in both precision and accuracy in  determining
proficiency.  Changes to the procedures  initially proposed by the
EPA would require a larger number of  measurements and would
consider both precision and accuracy, but the measurements would
still not be independent  and, consequently,  the  evaluation of
accuracy  and precision would  be incorrect.   The  dilemma  lies
primarily in the batch mode of operation and in the small number
of truly  independent measurements  that can  be made at  any one
time.

     Until additional testing facilities are available, the
Subcommittee recommends that the  1PA test  fewer devices,  if
necessary, but  assure  that tests  are independent,  i.e.,  at
different  concentrations.   Analysis  of the results  in previous
rounds  which  examine  the   within-round  and  between-round
variability within  companies may help  to  answer questions about
sample sizes.

     The statistical methods initially proposed  for the EPA's
testing program are inappropriate in  view of the physical and
economic constraints on the SERF  testing facility.  Developing a
set of statistical performance  criteria  and methods for running a
proficiency program requires commitment  from professional
personnel with statistical expertise  who can interact closely
with the EEAfs laboratory staff.  Close  interaction is necessary
to ensure that the statistician clearly  understands the
objectives of the program and the limitations of the testing

-------
environment.   Likewise, the  laboratory staff  needs input  from
statisticians as to the appropriateness of the procedures  front a
statistical point of view,  The  Subcommittee  strongly recommends
that a  statistician be assigned on a  full-time,  on-site  basis
until the program is well  established.

     All of the performance criteria discussed above are based on
independent test measurements.  To ensure this,  the measurements
should be made of different chamber atmospheres,  either at
different times  in the same  chamber or  in different  chambers.
The range of radon concentrations should represent values of most
interest but varied unpredictably, an
-------
that indicates  only  whether further measurements are  desirable,
but gives no indication of actual annual exposure.
           *
5.2  Participants'  Procedures

     EPA Question 6:   "Within the parameters of the  EPA
measurement protocols the RMP strives to follow each company's
standard operating procedures for testing,   EPA does not  set  the
exposure period, the sampling period or  the  counting procedures.
Accommodating the large variety  of  standard  operating  procedures
causes  logistical  and  operational  problems during performance
test.   Is  this a reasonable  and meritorious approach or  should
EPA set standard RMP test procedures?"

     The EPA should assure that a participating company cannot
claim that it failed a proficiency test because the  test  was  not
conducted in accordance with the company's  normal operating
procedure.   To do this, the EPA should determine the minimum  cumulative
exposure claimed to be detectable by the company, or the  maximum
exposure time required for correct response,  and assure that
tests are within these  criteria.   As long as the total  exposure
of the device, i.e. picocurie-days per liter, exceeds the minimum
claimed to be detectable by the participant,  and the decay  or loss
of  radon   from a  charcoal  canister does  not   exceed   the
normal amount, the participant's criteria will be satisfied.

     The procedures for exposing, evaluating (counting) and
calibrating each passive measurement device,  as well as the
methods used for calculating the concentration,  should  be
submitted with the application for admission to the  proficiency
program.  The EPA should evaluate and approved the procedures
before devices may be submitted for testing.

     For active measurement devices, the application should include
performance characteristics and calibration data. The
qualifications of existing operators, and the training  program
for future or unidentified operators, should be included  with the
applications for active devices.

5.3. Blind Tests of Passive Devices

     EPA Question 5:   "Participants in the  RMP  include  analysis
laboratories and distributors,  i.e. middlemen operating between
homeowners and analysis companies.   EPA intends to evaluate
distributors' performance in  conjunction with, but  different
from,  the  current chamber  test  process.    For  example, we  are
proposing distributors  undergo a yearly double  blind test  once
they have  established themselves  with a  participating  analysis
laboratory.  We would not require these  companies to participate
in further laboratory performance tests.  This would minimize the
impact  on  these smaller  companies and  allow  for  more  rigorous
testing of true laboratories.  Is this reasonable?"

     The blind test would be extremely valuable for  monitoring
the performance of companies that distribute passive devices  by


                           11

-------
mail  or  over the counter at retail outlets.   It  should be noted
that  these blind tests are also valuable in assessing performance
of  the  primary  measurement  laboratories.    However,   some
situations  cannot be  monitored effectively  by  this  technique,
e.g.  companies that deliver passive devices directly to the home,
Because  radon  decays  over time, there is also  a  time  constraint
for receiving  a passive  device  at  a distant location,  sending it
to the EERF  for exposure, returning it to the surrogate homeowner
and then mailing it back to the company who may also have to mail
it to an analytical laboratory*

      The advantages of blind testing are sufficient that  EPA is
encouraged  to adopt  this  system  as  soon  as  feasible.   The
advantages  of  blind testing  includet   (1) the participant does
not know that  the device is being tested,  nor  the  exposure,  (2)
the   handling  and reporting procedures are tested,   as  well  as
the measurement itself,  and  (3) the procedures of the distributor
are tested as well as those of the processing laboratory,  The
exposures of the passive devices could be carried out throughout
the year, allowing for a better distribution of effort with time.

5.4   Testing of Active Devices and Operators

      SPA Question Ss  "Currently, EPA supports the usage of grab
sampling as  a screening measurement method by the inclusion of
the method in EPA measurement protocols and in the RHP*  Given
the controversy in the scientific community surrounding its
reliability  as a screening measurement, should the RHP continue
to categorize grab sampling as a screening method or should we,
for example, establish a new category for diagnostic methods?11

      While grab  sampling is more appropriate  for diagnostic and
remediation  purposes,  it can be used as a  valid  screening method
particularly when prompt results are  required.  In  spite of sane
controversy  over  its reliability,  grab  sampling  is just  one of
several   short-term   measurement  techniques,  which include
screening measurements of only  a  week  or  two  with passive
devices.   The  advantages of on-site grab samples by  a qualified
operator are that results  and  interrelations can be available
immediately  without waiting  for mail  deliveries.  The  accepted use
of grab  sampling should be  Made  conditional  on the use of good
procedures for sampling, counting and calculation by a qualified
operator, since the method can be very easily abused.

      Active  monitoring devices  for grab, continuous and
integrating  sampling are used widely for both screening and
diagnostic purposes.  Because these instruments are larger in
siie  and in  many cases require a trained operator for deployment
in the field, they require test criteria different from those
required by  passive devices.  The outcome of the measurement
depends  largely on the individual operator and on the
original calibration of  the  instrument with sources procured from
the NBS  or from a laboratory which, in turn, was cross calibrated
.with  an  NB5  radon and radon progeny reference atmosphere,
Submission of one instrument of one type to the EPA HMPP does

-------
not guarantee that all others of the same type will perform
satisfactorily.  Several things can go wrong in  the field
requiring specialized training for problem recognition.  The
qualifications of the operators of active devices must be
reviewed periodically by EPA to ascertain the reliability of the
measurement results.  Commercial devices admitted to the program
should be certified as to their original calibration and
procedural steps that will be taken by the operator to maintain
their original calibration.

     The criteria should require the operator to maintain
documented quality control and measurement procedures for each
type of instrument.  Air leakage around a filter and improper
counter efficiency are the most common areas where problems may
occur.  The  operator should be able  to  recognize  them  and take
proper action  to correct  them.   The  participant  should submit
sampling procedures  with  a check list of operational procedures
to  be followed  before and  after exposure.   In  summary,  good
procedures for  calibration, sample  counting,  calculation and
the  qualifications of  the  operator  must be the  basis  for
accepting a company into the RMPP.

     EPA Question 3:  "Some measurement devices  can make Multiple
measurements in a 24-hour period while others can make a maximum
of two a day.  Requiring the same number of samples for all types
of devices creates logistical and financial problems for EPA and
companies.  Can any alternatives be recommended?"

     Active devices the logistical problems of requiring multiple
measurements  for  devices  that  require  manual  reading  and
resetting may  be alleviated by  applying  testing  procedures and
criteria  different from those  used  for  passive detectors.
Instruments and procedures for diagnostic  purposes and exposure
measurements may require  more  restrictive  criteria than those
required for screening measurements.

5.5  Radon Concentration Standard

     EPA Question 7s "in the absence of a NBS radon standard, EPA
bases their radon chamber calibration procedure  on radon values
established through national and international intercomparison
tests.  Does this approach sufficiently substantiate the accuracy
of our chamber levels?"

     The EERP uses both intercomparison tests and its own
laboratory standard for determining radon concentrations and
should continue  to  do  so.  It  is  generally accepted  within the
scientific community that a valid laboratory standard for radon can
be  prepared  from  an  NBS  radium  solution primary standard.
Careful handling,, including opening the NBS vial,  dilution and
preparing aliguots can give results  that are accurate  to within
one  percent.    Subsequent  transfer of  the   radon  gas to  a
scintillation  cell  or  ion  chamber  can  be  done   within
approximately the  same uncertainty.   The accuracy requirements
for radon exposure  chamber  calibration  are  of the  order of five
                           13

-------
percent, so a competent laboratory can calibrate its chamber in a
manner acceptable to users.
           f
     A potential source of error in the standard method of
calibrating scintillation cells from a standard radium solution
is in the transfer of moisture to the cells as a result of bubbling
air through the solution.  Condensation in the cellst  even when
so slight as to go unnoticed, can lead to substantial  loss of
detection efficiency.  One technique for preventing condensation
is simply to assure that the radon extraction operation is
performed in a room that is substantially cooler than  the room in
which the cells will be stored for radon progeny ingrowth and
subsequently counted.  Cooling of the bubbler flask in a
refrigerator prior to use may serve the same purpose.   Such
precautions should be explicitly included in the calibration
procedure.

6.0  LONG-TERM PLANNING FOR THE RMPP

6.1  Consensus Standards and Voluntary Accreditation

     Because of the recognized voluntary nature of the RMPP, the
Subcommittee recommends that serious consideration be  given to
establishing consensus standards and voluntary accreditation
procedures.    This  approach  would utilize a  broad  range  of
existing expertise  in  measurement technology and  in  statistical
analysis through the participation of other agencies and industry
representatives.   This approach has  been used  successfully  for
many years  by ANSI, ASME,  ASTM,  etc.   It might be  possible  to
establish a  program  under,  or  comparable  to,  the National
Voluntary Laboratory Accreditation Program sponsored by the NBS.

6.2  Financial Support Needed for Adequate Program

     The Subcommittee anticipates a continuing and increasing
demand for indoor radon measurements and the concomitant
requirements for calibration and testing, and encourages the EPA
to act now to assure the availability of the necessary facilities
and programs.  It is essential that adequate funding be provided
for the RMPP without depleting support for other important EPA
radon programs.  The Subcommittee strongly urges the SPA to
investigate all avenues to support the Radon Measurements
Proficiency Programs.  One possibility is through user fees* It
should be noted the latter regard that industry presentatives
present at the meeting urged the use of such fees particularly
if it would help maintain the announced testing schedule.  They
noted that the delay of several months in starting round 5 has
been damaging to companies who have invested in equipment but are
not able to conduct business until they are included on the KMPP
listing.
                           14

-------