GENERAL46067

• Permit No: M1988112
• IBM Index Class Name: General Documents
• Doc Date: 7/17/1992
• Doc Name: NEWSPAPER ARTICLE ADVANTAGES PITFALLS IN CHOOSING LABS
• Media Type: D
• Archive: No

r..-
:,dminis:er:nC o test, the laboratory Should at leas:. be able t0
~scxblisb chat:
• The PyM1iculnr instrument or equipment usod was in proper
working order and properly calibrated;
• The sample wns taken and merntained properly -including
nlaiatonnnce of a proper "chain of custody" - prior W testing,
• The test was administered properly;
• The operawr was competent and qualifiad; and
• Atlequnto contemporaneous records of procedures and
results e7(let.
Although lhean criteria sound simple to satisfy, complying with
them can be exacting. Sampling end henaling protocols can
cotriuse even the ntort experienced lalioratnry and field neraua-
neL
For some tests, composite sampling ie allowed, white for others
it is not. Some enmoles are light-sensitive; some arc temperature-
sensitive. Some must bo analyzed imme.dialely wl»lo others can
wait. 1)iffcreat substances require that different amounts be
rxsted lu result in legally defensible daat. Same samples em be
held in glass containers, while others must be contninod in
olnstin. Some sampling procedures or mating procedures arc
prams mcross-contamination problems.
(3vcu these concerns, o ]nbaratary should be able m produce
doemnentation and proof that internal procedures for mnin-
cnimng sample intzgrity were followed far a particular batch of
samples. Firiafly, the nrnst defensible daw cnnres before a court
wiry a currrpletn chain•of•ctrslady. The fniluro to meet all
regaireatents necessary fur legally defensible analytical data
does not automatically mean that the data will be unacceptable
-but it ccttunly goes to rho hoart of the datay credibility.
IL is easiest fur a laboratory to estoblish drat corraet
procedures were used fur a particular act of samples if uniform,
detailed and written procedures for handling anti analysis of the
samples exist at the laboratory; if proof that these procedures
aetmlly wets followed far s particular bet of samples etials; and
if it can be shown [hat all personnel handling tha samples and
pcrnrming the analyses were trained and caoablc.
TS avoid facing a charge of professional negligence, a
cmualtant's best defense s [o vac "due diligence" when selecting
a laouratory for a client. This means that the consultant made a
comoreheasive systematic anti documented im estigatien prior to
retaining the laborxlurye services. Tha ideal audit is rigorous
enmgh to reveal unerationel flaws of rho laboratory that would
atFect its ability w produce accurate. legally defensible data.
The consultant's audit alto should be tailnrod to ferret nut
thnso laboratories likely to produce data that ore inconsiswnt
wits the client's etrstegic objectives.
Finally, the consultant should provide the contract ]aliorstory
with a precise xnd written scope of wort: that sari furor the tests
w bs rue., the subecnnees to be restart, :nechodolornes w be usod
and the dncuatentatinn and cerliflcations required. If something
goes awry with the lest results, the scope of work may assist in
treeing logal liability.
The ir,llowing scenarios illustrate potential consultant liability
awing from indefensible data. The laboratory witkr the lowest
anelyeis price 's not necessarily lire best manxgetl mre. If an
inadequntaiy trained chamist at that laboratory produces
erninsous dote, the lobarxtary anti consulting fins [hat chose it
wild expect to be lrr)d liable for dxruoges.
[n a second scanaria a laboratory prcduree xcc::ra:e result^^..
Rut when those resulW are challenged in court, t}tey arc rejected
for iusuflicent ducuwenteliun. Agoin, this might suh7ect the
cmtsul[ing fins w allegations of legal liabiiil}. If rite laboratory
followed iffi normal procedures, tshich du nu; require es:ecsi•:e
documentation, it might have n reduced deg•ee of linbilicy, or,
depending an the written scope nF work, no liability' nc all.
Another case might involve a laboratory tha,carefully followed
internal pracoduree and ducumeatntion. Nere.^theless, one ot'the
laboratory's analysts made an error that was not detected by the
laboracor•y/a quality control :yslem. If th.' consulting firm
conduced a tkrorough investigation oC the laboratory and
monitored rte protocols over time, the conaul:nnc might be less
liable if the eror results in damages to the client. The laboratory,
then, would 6e the most liltely liability target.
Also, since analytical laboratories are certified on a subatance-
epecific basis, the lab audit should determine if the substance in
question is covered by the certification. The certificafian
requirement far all the target substances shut ld be ineorpornted
into one writton scooe of work Cot• the lab.
These examplos underscore the pntontial perils that flow from
what at fast blush seems like are easy task - aelect.ing a
laboratory. Tltey further illustrate the need to demand proof from
the laboratory of its good practices and to write a precise anti
appropriate scope of work
The Contract Laboratory Program (CLPI administered by EPA
servos as a model for producing legally defensible annlvcical dale.
Date from the CLP are classified as Level IV dale, in a ]ticrnrchy
of EPA's ane'}-4a1 programs. 9: Level IV (CLP;, ana:;ti_;!
procedures used to determine cnrtcentration and [hc operational
ronnnrc•d un ~x+Xe as
THE TRIANGLE
LABORATORIES GROUP
DIOXINS/FURANS • ORGANICS •• INORGANICS
Research Trlanple Park, NC [phone]
Houston, TX [phone]
Cdumbus, OH [phone]
Atlanta, GA [phone]
ntuv,v uGl'ST Iv92 Clrcle 40 On cartl. a5
Your AnalYllCal Source for