Improvements Needed in the Environmental Protection Agency's Testing Program for Radon Measurement Companies

Published by the Government Accountability Office on 1990-05-23.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

-                              United States General    Accounting Office
                                                 -...- .._____

    For Release                   Improvements       Needed   in   the         Environmental          Protection
    on Delivery                   Agency's    Testing    Program         for     Radon    Measurement
    Expected    at
    9:30 A.M. EDT                 Companies
    May 23, 1990

                                  Statement     for the Record by Richard                      L.   Hembra,    Director         ,
                                  Environmental      Protection    Issues
                                  Resources,      Community,    and Economic              Development

                                  Before     the
                                  Subcommittee      on Super-fund,   Ocean and Water
                                  Committee      on Environment    and Public Works
                                  U.S. Senate


                                 <~8SLC~ ~jwQc0                                                                           --
                                                 /                                                       GAO Form 160 (12/87)
Mr. Chairman     and Members of      the   Subcommittee:

        We are pleased    to present    a statement       for the record     that
discusses     (1) our report1     on the Environmental        Protection     Agency
(EPA) program that assesses radon measurement companies'
competency,      and (2) the results      of our current      review on changes
needed in this program to provide            homeowners with greater         assurance
that radon measurements        are accurate,      and on state      efforts    to
control    radon measurement companies.           This work was done at the
request of the House Committee on Science,                Space and Technology      and
the results      were presented    in testimony      before    the Committee's
Subcommittee      on Natural    Resources,    Agriculture      Research,    and
Environment      on May 16, 1990.

       Radon, a naturally     occurring,     colorless,     odorless      gas has been
shown to cause lung cancer.          EPA estimates      that 20,000 lung cancer
deaths each year can be attributed           to indoor     radon.       As a result,
EPA and the Public      Health Service      have advised      residents        to test
their    homes for radon and take action         when elevated        levels     are
found.2     In 1986, to help ensure that homeowners obtain                   accurate
radon measurements,       EPA published     procedures     for taking        radon
measurements and established         the voluntary      Radon Measurement
Proficiency    (RMP) program.       More recently,      Congress passed Public
Law 100-551,    commonly referred        to as the Indoor Radon Abatement Act
of 1988, which directed       EPA to undertake        a variety     of activities      to

1Air Pollution:      Uncertainty     Exists    in Radon Measurements        (GAO/RCED-
90-25, Oct. 16,      1989).
2While EPA maintains       that no safe level    exists,   the agency
recommends that homeowners take action          whenever annual average
radon levels    are believed     to be greater   than about 4 picocuries     per
liter  of air.     EPA estimates    that the risk of being exposed to
annual radon levels      of 4 picocuries     is comparable    to smoking half a
pack of cigarettes     each day.
address the radon problem.     A number of these activities                         including
the RMP program,  were already   underway as part of EPA's                        radon

       The RMP program assesses the capabilities                of companies
providing    test results      to homeowners.        The objectives        of the
program are to (1) assist           the states    and the public        in selecting
companies that have demonstrated             competence in measuring           indoor
radon, and in the long-run            (2) provide    assurance     to the public       that
companies'     test results      are accurate     through    the use of
standardized      measurement and quality         assurance     procedures.         TO
achieve these objectives,           EPA envisioned      a federal/state        approach.
EPA would be responsible           for testing    the proficiency         of firms,
whose participation       in the program would be voluntary,                 and would
encourage firms to adopt quality             assurance    procedures.         The states,
according    to EPA officials,         would determine      any additional
regulation     of firms such as mandatory participation                 in the RMP
program and mandatory        adoption     of quality     assurance      programs.


        EPA has made considerab le progress                 in ach ieving       the initial
objective       of the program.            Since 1986 EPA has assessed companies'
competency in measuring               radon on six occasions.            It has also
published       the results        in national      reports     that are distributed
throughout        the country        and in individual        state reports.         The
reports    list       the companies that have demonstrated                 proficiency      and
device or devices            they used to demonstrate             it.  The number of
firms demonstrating             proficiency      through    the RMP program has grown
dramatically.            For example,       24 firms demonstrated          proficiency      in
EPA's first         test held in 1986 while about 660 firms were listed                        in
EPA's latest         proficiency        report   (issued    in January 1990).           Of the
660 firms,        EPA lists      about 260 as national            companies.

        Even with the increased    number of firms demonstrating
proficiency     in measuring  radon, EPA and the public       still    cannot be
assured that all companies meet proficiency         standards       or that
measurement results      are accurate  on a day-to-day     basis.      As we
reported    in 1989, this assurance    does not exist    primarily      because

      --   the voluntary  nature     of the program allows   firms to market
           measurement devices      that have not been tested    or that
           failed  a test and

      --   the program does not require        measurement companies to
           implement quality      assurance  programs that ensure quality
           measurement results      on a day-to-day    basis,   and,
           consequently,     companies may be providing       homeowners with
           inaccurate    results.

Further,      only a limited     number of states   have assumed the
responsibllitles       EPA envisioned     in the federal/state    approach.    Our
current     work shows that only nine states        have developed    programs
that provide       some regulation     of radon measurement companies.


        Radon occurs naturally       almost everywhere.       Current      estimates
are that the average U.S. home contains            about 1.5 picocuries            of
radon.      Several different      devices are available      to measure radon
levels    in the home.        For example,  the popular     charcoal      canister
measures radon over 2 to 7 days.            Another popular       device,     the alpha
track detector,        measures radon for longer     periods      such as 3 months
to a year.         Both devices,   which can be purchased       from various
retail    outlets,     must be sent to laboratories       for analysis        after
being exposed to radon.           Some more costly   devices      that require
skilled     operators,     such as the continuous    radon monitor,         can
measure radon and provide          more immediate   results     without
laboratory      analysis.      Companies that provide     the laboratory
analysis    results     or   the   results     through    the    use of   instruments      used
by a skilled    operator  are         tested    through    the    RMP program     and are
called  primary   companies.9

        To pass the RMP program and be listed                 in the proficiency
report,     EPA requires        a primary     company to (1) follow          the
appropriate      measurement protocols,            (2) demonstrate        the ability       to
get test results           to the proper homeowner, and (3) demonstrate                   the
ability     to measure radon to within             25 percent      of actual     levels.       To
meet the first          requirement,      EPA generally     relies      on a Company’s
statement     in the application           that it follows       the protocols.          To
meet the second and third              requirements,      companies must pass a
proficiency       test,     which includes      correctly     analyzing      devices
exposed to known levels              of radon and reporting          the results      to EPA
for verification.


       In October 1989 we reported         that 87 percent          of the devices
companies had tested       in the RMP program in 1988 met the RMP
requirements,      thus demonstrating      a minimum level          of competency in
measuring radon.        However, we also reported           that the voluntary
nature of the program allows          firms to market devices            that fail   the
program or that have not been tested             in the program.           When
companies are allowed        to market devices        without      demonstrating   a
minimum level      of competency    in measuring       radon,      consumers have no
assurance     that they are receiving        accurate     results.       In fact,  in a
few of the cases cited        below it appeared the companies may have
been providing       homeowners with inaccurate          measurements.         The

3Primary companies either          have laboratory     capabilities      to analyze
radon measurement devices          after   they have been exposed to radon or
measure the radon levels          and analyze the results         with their  own
instrumentation      and operators.        Secondary companies provide
services    ranging    from distribution       of radon devices       to home
inspection      and consultation.        Secondary companies must use a
primary company to analyze           the radon devices.
following         are cases we reported   which   illustrate           the    lack       of
controls        in the RMP program:4

       --   One large and a few small companies were marketing                               devices
            that had not been tested  in the RMP program.

       --   One large     company was marketing      a device          that     did      not    meet

            the    RMP requirements.

       --   Several     small companies   were marketing          devices            after
            failing     the proficiency   test.

       --   A  few small companies that tested           some     of    their  devices in
            the RMP program had been marketing            other        devices  that had
            not been tested  in the program.

In addition,         a company not in our sample but identified      through
discussions        with   contracting personnel, was analyzing    devices     in
its laboratory          under another name after it failed    the proficiency


       One objective     of our current       review was to address changes in
the RMP program that could provide              homeowners with greater
assurance    that radon measurements are accurate.             To help  answer
this objective       we interviewed    officials     from a sample of the radon
industry   for their     views and we found that they generally          agreed
that all measurement         companies should be required        to demonstrate

4In our review,    we sampled all 11 large primary     companies   (having
30 or more secondary      companies) and 100 of the 347 small     primary
companies (having     fewer than 30 secondary   companies)   that
participated    in the 1988 testing.    The precise  estimates    can be
found on pages 7 and 50 of our Oct.      16, 1989, report.
Proficiency     ln testing   radon.5  Of the 32 interviewed,       27 officials
said participation       in the RMP program should be mandatory.          The
reasons given for making participation       mandatory    included    the

      --   Radon health      effects are      severe enough       to warrant obtaining
           assurance     from companies       that they are       meeting RMP

      --   The public      needs assurance that they are dealing  with
           reputable     firms,  and the RMP program is the Only  means
           available     to companies for demonstrating  competence    in
           measuring     radon.


        An effective       quality       assurance   program is EPA's best         assurance
that    radon measurement firms are performing                 quality    testing     on a
day-to-day     basis.        The primary       purpose of documented quality
control     1s to assure that the capability                demonstrated     during
performance       testing     is maintained        until  the next periodic
evaluation,       according       to the former Chief,        Office   of Radiation
Measurement,        National      Institute      of Standards    and Technology.

SWe interviewed       a judgemental        sample of 32 radon measurement
industry    representatives         from 26 commercial,          3 university,      and 2
state laboratories        and 1 federal       laboratory,        all  of which provide
radon measurement services.               We pretested      our survey instrument
with representatives          of 3 commercial        laboratories.        To obtain
viewpoints      from all types of applicants,              we divided     the universe    of
843 applicants       for participation        in EPA's 1989 proficiency             test
into four strata        and then randomly selected              and interviewed
applicants      from each strata.          We selected       30 applicants,      one of
which had already        been    interviewed      in the pretest.         Information
obtained    reflects     only the views of those interviewed                  and cannot be
considered     representative         of the entire       universe    of applicants.
        Although     EPA recommends certain           quality     assurance     procedures
for radon measurement companies,                  the agency does not require
companies to develop and implement such procedures                        as a condition
for program participation.               Recommended procedures           include:
controlled      calibrations       of measurement devices            in a known radon
environment,       such as in a calibration             chamber; background         and
duplicate      measurements;       written      procedures     for attaining       quality
assurance objectives:            a system for recording           and monitoring        the
results     of quality       assurance     measurements;       and maintenance       of
control     charts    and related       statistical      data.

       If  measurement companies do not develop and implement adequate
quality    assurance    programs,   they may be providing        homeowners with
inaccurate     results.     For example,     in our prior    work we found that
only 12 of 21 companies we interviewed             that participated    in the
1988 testing      were calibrating     their   equipment.     One of the nine
companies that did not calibrate      its equipment                 failed      the 1988 test
with a loo-percent   error  but had been marketing                    the measurement
device for a full   year before   the test.    After                calibrating      its
equipment,   the company retested    and passed.

         In addition,      not requiring      measurement companies to implement
quality     assurance      programs seems inconsistent        with EPA's agency-
wide quality       assurance     policy    for EPA-sponsored     environmental
monitoring      and measurement efforts.           This policy    requires     every
measurement project           to have a written      and approved quality
assurance plan and applies              to all EPA program offices,        regional
offices,     laboratories,       contractors,     and grantees.

        EPA's own Science Advisory   Board has also recommended that the
agency require     radon measurement companies to maintain    documented
quality    control  and measurement procedures   for measurement

devices.6       Specifically,    the     Board recommended that both detailed
descriptions      of calibration        procedures    and calibration       data for
certain     types of measurement         devices be submitted        with the
application      for admission     to    the proficiency     testing     program.

        EPA program officials         expressed    concern about enforcing       a
quality    control     requirement      and managing the costs associated          with
it.     Nevertheless,       EPA recently    established    a task force to assess
changes that are needed in the RMP program.                 In a recent    meeting,
program officials         told us that the task force was developing             a
proposal     for changes in the RMP program that would include
requiring     measurement companies to have quality             assurance   programs
as a condition        for participating       in the RMP program.       In addition,
as authorized       by the 1988 radon legislation,          EPA is developing        a
user-fee     proposal     to cover costs      of the RMP program.


       Of the 32 officials        from the radon testing      industry   we
interviewed,       31 said some quality     assurance  should be required                as a
condition      for participating      in the RMP program.       Some of the
reasons these officials          gave for requiring   quality     assurance
included     the following:

       --   At the present     time, the radon measurement area is                wide
            open to abuse     because homeowners cannot see, taste,                or
            smell radon.

       --   Quality    assurance    requirements      would     force industry      to
            develop    good quality     control    systems      and standardize
            industry    practices.

6The Science Advisory         Board is a group of            independent   scientists
who review the quality         and sufficiency     of        scientific   data
underlying  regulatory        development    of some         EPA actions.
      --   A quality control   system is needed as a link between a
           company's performance    on the proficiency test and its
           everyday work.


        Another objective       of our current       review was to determine         the
status of state radon programs that exercise                   control   over radon
measurement companies.            In designing     the RMP program using a
federal/state      approach,      EPA envisioned       that states,     through
accreditation      or certification       programs,       would exercise     some degree
of control     over the reliability,         consistency       and quality     of the
measurement data companies provide               homeowners.       According     to EPA
officials,     the agency envisioned          that states      would make
participation      in the RMP program mandatory and establish                  and
enforce  quality   control    requirements     through   state programs.      We
conducted a telephone      survey of radon coordinators          in 50 states
and the District     of Columbia to determine         the progress   states   were
making in establishing      programs.      (See app. I for a summary of the
survey informat    on on the status        of state radon programs.)

      The survey we conducted      showed that only nine states           have
programs that e ther certify,        license,     or accredit   radon
measurement companies.        All nine programs have a requirement            that
radon measurement companies participate            in EPA's RMP program or in
a similar  state-run     program.    However, only five of the programs
are mandatory.      Three of the states       with voluntary    programs are
attempting   to pass legislation       or implement regulations        that would
make their   programs mandatory.        The fourth    state has no plans to
change its voluntary      program.     Of those nine states       that have
programs,  only five have quality         assurance   requirements.
       An additional   20 states    may establish      a program in the future.
Twenty-two    state coordinators     said their     states   probably   would not
have a program.      Two reasons coordinators        gave for not having a
program were a lack     of funding     or resources      for indoor air
problems in general,      including    radon, and a lack of legislative
authority    for such programs.

       Not only do states    vary in whether or not they require
mandatory participation      in their      programs,    but they also vary in
the number of requirements       in their      programs.    The result   is that
each program gives a different          level   of assurance    to homeowners
regarding   the accuracy  of    measurements.        For example,   five of the
nine programs require           companies to calibrate           their    equipment
periodically        and follow    other quality       assurance       procedures.    Four
of nine programs provide            for on-site      inspection       of measurement
companies.         The only requirement        common     to all    nine programs is for
companies to participate            in EPA'S RMP testing           program and/or a
similar      state-run    testing     program.     (See table      1.1, app. I for a
comparison       of the various       requirements      of the nine programs.)

       Although,      we did not attempt          in the survey        to establish          why
some states        had certain       requirements      and while others          did not, one
reason may be the lack of guidance                  from EPA. For the most part,
EPA has not defined           the degree of control            it wants states           to
exercise     over measurement companies.                In a May 1988 report,               EPA
stated    such control        could include       registration,        certification           and
licensing.7         According      to the report,       registration,         certification,
and licensing        differences        center around whether the control                   process
is mandatory,        whether nonparticipating             (e.g.,    non-certified           or non-
licensed)       firms are excluded          from the market,         and whether a fee is
charged by the state            for the process.          Licensing      is viewed as the
most   restrictive        form of "quality        control,"       while registration            is

7Key Elements of a State             Radon Proqram,        EPA 520/l-88-006,          Office     of
Radiation  Programs.
the least     restrictive.     The report          noted that the RMP program
illustrates      a form of registration            with no fee--the least
restrictive      form of control.

In an April     1990 meeting,  EPA program officials     told us that the
radon task     force was proposing   that a document describing     a model
state certification     program be developed     for guidance  to states.


       Since radon has been identified              as a national      health     problem,
EPA and the Public         Health Service       have advised       homeowners to test
their    homes and take action          if elevated      levels    are found.      We
believe     that to make health-based           decisions       homeowners need more
assurance that the radon test results                 they rely on are accurate.
Homeowners would have greater              assurance     that radon measurements             are
accurate      if (1) participation         in the RMP program were made
mandatory,       and (2) radon measurement          firms were required          to meet
minimum quality        assurance     requirements      as a condition      to
participation.          In designing      the RMP program, EPA relied           a great
deal on states        to develop programs that would exercise                some    control
over the reliability,          consistency,       and quality      of the measurement
data companies provide           to homeowners on a day-to-day            basis.       While
some states       have developed       programs,    most    have not.

        We are not opposed to states          exercising     control       over
measurement companies.            However, to provide       homeowners with a
minimum level        of assurance     that radon measurements are accurate,               we
believe     actions     need to be taken at the national            level.      In our
May 16 testimony          before  the House Subcommittee        on Natural
Resources,       Agriculture     Research,   and Environment,         we recommended
that the Congress provide            EPA authority     to require      companies to
participate       in and successfully       pass the RMP program before
marketing      their    devices   to the public.       In addition,        to help ensure
that    radon measurement companies are providing               quality       measurement
results  on a day-to-day      basis,     we recommended that EPA establish
quality  assurance    requirements       for the different       measurement
devices and, as a condition          for participating      in the RMP program,
require  measurement firms to demonstrate              that they have developed
and implemented    programs that will         meet the requirements.         Finally,
to ensure the development       of state      programs that provide       a minimum
degree of control     and consistency        over radon measurement companies,
we recommended that EPA develop           and issue guidance       on the type of
state programs and level       of control       it believes    is needed at the
state level   in order to provide          homeowners with adequate assurance
that radon measurements are accurate.

APPENDIX I                                                                    APPENDIX I
                         STATUS OF STATE RADON PROGRAMS

         To obtain       a nationwide     perspective      on how states     exercise
control      over radon measurement companies,               we conducted     a telephone
survey of state radon coordinators                  in the 50 states      and the
District      of Columbia.          The following     is a summary of their
responses        concerning      (1) whether radon levels         in homes are a
problem in their           state    (2) whether their       state certifies,
licenses,        or accredits       companies that perform radon measurements             in
their     state,     and if so, the elements          included    in their    program and
 (3) their       states'    efforts     to provide    radon information       to the


       Thirty-five      of the 51 radon coordinators       said they believe
radon levels       are a problem in their    state.     Coordinators       in the
states    of Alaska,      Arizona, Arkansas,  Hawaii,    Indiana,      Louisiana,
Mississippi,       Oklahoma, North Carolina,      South Carolina,       Texas and
the District       of Columbia said radon was not a problem.             Radon
coordinators       in California,   Nevada, New Hampshire,        and South Dakota
did not know whether radon was a problem in their               state.


       Nine of the 51 state radon coordinators                    told us that their
states    have programs to certify,           license,      or accredit      companies
that perform radon measurements             in their       state.      The remaining   42
coordinators       said that their     states      do not have such a program.
Of the states       with programs,     five have mandatory             programs:
Delaware,     Florida,    Iowa, Pennsylvania,           and Virginia.        The four
states    with voluntary       programs are Kentucky,             Nebraska,   New Jersey,
and North Dakota.         All nine states         publish     a list    of proficient
radon testing       companies.
APPENDIX I                                                                      APPENDIX I

       Coordinators        from 20 of the 42 states            that do not have radon
measurement       accreditation        programs expect their         states     to have
programs in the future.               Six of the 20 coordinators           anticipate
having programs within             1 year,    4 others     foresee   having a program
within    2 years,      while the remaining           10 expect to establish
programs but do not know when.                  The coordinators       for the remaining
22 states      without     programs do not expect their             states    to establish
programs in the near future.                 Although    these coordinators         do not
foresee     their    states     establishing       a program,     12 of them believe
that radon is a problem in their                 states.

        To help defray     the cost of the state    certification,        three
states,     (Florida,   Iowa, and Pennsylvania)    are currently        charging
application       fees, and the fourth    state, New Jersey,       will  begln            to
charge a fee when its state certification           program becomes
mandatory according        to a state official.     Iowa is the only state
requiring      all radon measurement   companies to post a bond before


       We asked the coordinators      in the nine states     with programs to
describe    both their   requirements    for radon testing     companies and
the state activities       that were included    in their   programs.       The
number of requirements        in the nine programs varied.       The only
requirement     common to all nine programs is for companies to
participate     in EPA's RMP testing     program and/or a similar        state-
run testing     program.     Examples of other requirements      include      the

       --   Seven  of the programs require           measurement         companies to     meet
            minimum educational   requirements            for   critical     personnel.
APPENDIX I                                                                          APPENDIX I

      --   Seven states    have established  minimum radon experience
           requirements    for a measurement company's critical

      --   Five programs require      companies   to calibrate     their
           equipment  periodically     and follow    other quality       control
           procedures   such as (1) routine      checking    of equipment
           accuracy,   (2) procedures     to ensure that measurement
           equipment  is operating     properly,    and (3) record keeping.

      --   Four programs     call     for     on-site      inspection    of measurement

The following     table    compares         the    requirements     of   the nine    state

APPENDIX I                                                                  APPENDIX I

Table   I.1   Comparison      of Program       Requirements
                                           Mandatory program
State program requirements                 E PA E VJJ g
Companies required to participate
in WA's IFlP testing program and/
or state-rm testing program                X    X       x   x   x   X   x    x

Minimun education required
for critical company psrsonnel             X            X       X   X   x    x

Minimum radon experience
required for critical  personnel           X    X       X       X   X        X

Canpenies required    to submit
radon test results                         X    X       X       X   X        X

State program includes radon
training for measurement companies         X                    X   X   X

Companies required to calibrate
their equipment periodically               X     X              X   X        X

Other quality assurance procedures
required ((1) routine checking of
equipment accuracy, (2) procedures
to ensure that measurement
equipment is operating properly,
and (3) record keeping)                    X        X               X        X      5
State program calls    for on-site
inspection                                 X        X                        X      4
Companies and radon specialists
required to pay a fee                      X        X

State program includes     blind
testing of companies                             X

Companies required    to post a bond                            X


        We asked state coordinators    to furnish  information   on the ways
their    states  communicate with homeowners.     All states   have some
means    of communlcatlng   indoor radon problems to the public.       The
APPENDIX I                                                                                         APPENDIX I

following   list  summarizes the different                         ways states           are      providing
radon information    to homeowners:

     --    19 states      have a radon         hotline

     --    50 states      distribute         EPA's      Citizen's         Guide         for     Radon.

     --    48 states  distribute  listing                     of   proficient       radon
           measurement companies.

      --   49 states      distribute         other      radon       brochures/publications.

     --    3 states     run    television,           radio,        or print       ads.

     --    32 states use other outreach                       methods      to provide             consumers
           with radon information.


      While 32 states   provide   some funding    for radon activities,
only 18 have designated     money specifically      for radon programs in
the state budgets.     The remaining    14 states    use funds from larger
departmental  budgets.     As shown below,     the amount of funds budgeted
for radon varied   among the 18 states:

      --   Four states budget           over $1 million   for                   radon         (Florida,       New
           York, Pennsylvania,           and New Jersey).

      --   Five states        budget between $100,000 and $1 million
           (California,        Washington,  Connecticut;Alaska,   and

      --   Six states      budget less than $100,000   (Maryland,                                 Iowa,   New
           Hampshire,      Wisconsin,  Indiana, and Idaho).
APPENDIX I                                                               APPENDIX I

     --   Three states   did not    tell  us their    funding   levels
          (Kentucky,   Minnesota,     and Tennessee).