An article on risk assessment and how corporations
and government decide to take risks with public health, courtesy of
Rachel's Environment and Health News.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RACHEL'S ENVIRONMENT & HEALTH NEWS #800
http://www.rachel.org September 16, 2004
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
THE CHEMICAL WARS, Part 3
by Peter Montague
[Continuing: We have been describing the philosophy ofenvironmental
regulation in the U.S. Basically, it is a "proveharm" system --
anything goes until someone can "line up thedead bodies" and prove that
significant harm is occurring. Whenthat happens, which is rare, then a
multi-year, ormulti-decade, battle begins in which underfunded
andunderstaffed government regulators bargain with a phalanx
ofcorporate lawyers and scientists-for-hire. Eventually theyhammer out
a compromise between public health and corporatepurposes. The
compromise becomes an enforceable regulation --until one corporation or
another decides to mount a challengeand the dance begins anew.
The "prove harm" system rests on three assumptions: (1) Humanscan
determine the "assimilative capacity" of every populationof humans and
animals and every ecosystem on Earth -- thecapacity to absorb damage
without suffering permanent, seriousharm. (2) Once the "assimilative
capacity" of a river, or apopulation of humans or birds, has been
determined, we will setregulatory controls to keep the harm within
"acceptable"limits; and (3) We already know which substances and
activitiesare harmful or, in the case of activities we never
suspectedwere harmful, we will we warned of possible dangers
bytraumatic but sublethal shocks.
Obviously the system really hinges on assumption #1 -- that wecan
determine the "assimilative capacity" of an ecosystem, orof a
population of polar bears or humans. For this purpose, aspecial
technique has been developed called "risk assessment."Risk assessment
is the linchpin of the "prove harm" regulatorysystem, and the main
intellectual armor of industrialpolluters. But this emperor is wearing
no clothes. Let's take alook.]
Of course there's nothing wrong with trying to assess risks. Weall do
it every day. But there's an important differencebetween our own
personal risk assessments andcorporate/governmental risk assessments.
When we assess risk in our own lives, (a) we examine risks thatwe
ourselves are willing to take; (b) we compare our options;and (c) we
use all available information; and (d) we weigh notonly the risks we
face but also the benefits. For example, wemight ask ourselves, "Can I
just dash across this street in themiddle of the block, or, given the
shoes I'm wearing and thearthritis in my left knee, should I walk to
the corner andcross with the light? Is saving a minute or two worth the
riskof being hit by a truck?" We compare risks and benefits, weassess
our alternatives, we consider all the availableinformation, and we
weigh the risks we ourselves are willing totake.
In contrast, corporate risk assessors almost always (a) assessthe
dangers of a single pre-determined option, and (b) assessdangers that
they intend to impose on others, usually withouttheir informed consent;
and (c) examine only thescientifically-proven evidence, ignoring other
kinds ofinformation such as historical precedents, worker knowledge,and
community preferences; and (d) ignore the benefits (or lackof them) to
those who will be enduring the dangers. Basically,the main use of
corporate/governmental risk assessment is toestablish how much damage
corporations and governments can getaway with and to label that damage
"acceptable."[1]
Typical questions that corporate/governmental risk assessmentsanswer
would include, How much dioxin can aluminum smeltersdischarge into the
Columbia River basin without thinning theBald Eagle population to
extinction? How many trout canfamilies along Lake Michigan eat each
month before theirchildren's IQs are diminished 5 points? How much
benzene can wemaintain in the air of this factory without killing more
than 1in every 10,000 workers? Will this urban trash incinerator killno
more than one in each million citizens who breathe itsfumes?
Risk assessment serves corporate purposes because it involveslarge
quantities of scientific data, all of it subject tolimitations and
uncertainties that can be disputed foreverwithout resolution. Where
data are lacking or disputed,assumptions and judgments must be
substituted for facts. TheNational Academy of Sciences put it politely
when it said,"Risk assessment techniques are highly speculative, and
almostall rely on multiple assumptions of fact -- some of which
areentirely untestable."[2] In 1983 the National Academy
identifiedat least 50 points during the course of a cancer
riskassessment where choices had to be made on the basis ofprofessional
judgment, not science.[3] Corporatescientists-for-hire can select and
manipulate the data andchoose particular assumptions (often silently),
allowing themto reach almost any conclusion they set out to reach yet
stillpackage it as "science" even though the conclusion is based
onjudgment and is not in any way reproducible.[4]
Risk assessment provides corporations other major benefits aswell.
Because risks are expressed mathematically (theprobability of x
occurring during y years of exposure tochemical z), troublesome
questions of right and wrong cannotarise, and most of the public is
left out of the process. Thusrisk assessment gives corporate goals a
patina of "soundscience," prevents ethical considerations from muddying
thedebate, and keeps the affected citizens locked out of thediscussion.
Risk assessment now guides all environmental management, notmerely the
control of chemicals. Before cutting new roads intoa national forest,
the government completes a risk assessmentto decide how many roads
would decimate the bear population.Ocean fisheries are managed by risk
assessment to determine the"maximum sustainable yield" of fish. Risk
assessment determinesallowable drug residues in beef, allowable
pesticide residuesin food, allowable withdrawals of water from rivers
andaquifers, allowable contamination of drinking water, limits onthe
discharge of particulates and toxic chemicals fromcoal-fired power
plants, auto emission limits, livestockgrazing allotments on arid
lands, allowable harvests ofendangered species, fishing and hunting
quotas, workplaceexposure limits, radiation limits in medical settings,
cleanupstandards for contaminated sites, and on and on.
Risk assessment is so fundamental to the "growth and rapidinnovation"
culture that the technique is now taught at mostlarge colleges and
universities. There are several scholarlyjournals devoted to it. Many
books have been written on thesubject, including several by the
National Academy of Sciences.The federal government sponsors research
to elaborate andrefine risk assessment techniques, and it trains risk
assessorsin places like Mexico and the Ukraine, intending to
"harmonize"the response to corporate harms world-wide. Risk
assessmentresearch institutes at places like Harvard are
generouslyfunded by important corporate risk-makers like Monsanto
andDow, and the work of these institutes is injected directly
intofederal "risk policy." Professional societies of risk assessorsmeet
each year in resort locations to swap war stories andshare their latest
techniques. Assessing risks has become amajor industry unto itself. It
is no exaggeration to say thatthe modern industrial system with its
culture of "rapidinnovation at any cost" could not maintain its present
coursewithout risk assessors to run interference.
In the last decade, however, risk assessment has come underwithering criticism from at least a dozen perspectives:
1) Because of genetic makeup, individuals differ markedly intheir
susceptibility to poisons. Some people are far moresensitive than
others. For example, some people cough andwheeze when they walk down
the detergent aisle at the grocerystore; others don't. Furthermore,
many people suffer fromchronic conditions (asthma, diabetes, etc.), so
riskassessments cannot reasonably assume, as they typically do,that
only healthy young adults are exposed.
2) Risk assessors try to account for human variability byapplying a
"safety factor" of 10 to their numerical estimate ofrisk. But such a
number has little to do with science. Safetyfactors are often little
more than guesses. Why not a factor of11 or 17 instead of 10? Even
calling it a "safety" factor ismisleading because who can say it offers
safety?
3) Risk assessments of chemicals are conducted on singlechemicals, but
in the real world we are all exposed to mixturesof chemicals day in and
day out. Furthermore, many studies havenow shown that harmless amounts
of individual chemicals, incombination, can add up to a harmful
dose.[5] The healtheffects of mixtures are far too complex for science
to sortout, yet mixtures are what we encounter in our daily lives,
sotesting single chemicals is misleading and often beside thepoint.
Corporate scientists-for-hire may pretend that, withsufficient testing,
the problem of mixtures can be mastered.But when asked where the
resources will come from to test allpossible combinations of even 1000
chemicals, they grow silent.There are 41 billion possible combinations
of 1000 chemicalstaken in groups of 4. Even if we could test a
millioncombinations a year, which we can't, it would take 41,000
yearsto complete such a battery of tests.
4) Some chemicals are only biologically active during a briefperiod of
time (a "window of vulnerability") in the developmentof an organism, so
toxicity must be tested during those exacttimes.[6,7,8,9,10] Chemicals
tested during other times willappear to be less potent or even inert.
5) In the case of some hormone-disrupting chemicals, low dosescan cause
greater endocrine disruption than high doses. Morethan 100 studies have
now confirmed that this phenomenon isreal.[11] This seems to happen
because the hormone system isactive at low doses but becomes
overwhelmed and stopsresponding at higher doses. Traditionally,
chemicals have beentested at the highest doses that laboratory animals
couldtolerate, but now we know that high-dose tests may missimportant
toxic effects that only occur at low doses. Many ofthe high-dose tests
that have been completed to date (and uponwhich federal regulations are
based) are of very limited valuefrom a public health perspective and
need to be re-done at muchlower doses.
6) We now know that cells respond differently to chemicals,depending on
their prior history of exposure.[12,13] Inaddition, whole organisms
(mice, humans) exhibit similarbehavior: response to a chemical is
strongly conditioned byprior exposure. For example, a person who smokes
a cigarettefor the first time reacts with lightheadedness and
perhapsnausea but a habituated smoker develops a craving for
cigarettesmoke and feels sick without it. Furthermore, after a
heavysmoker quits smoking, he or she will be "sensitized" tosecond-hand
smoke thereafter, reacting to it much morepowerfully than a person who
has never smoked. Thus individualhistory of exposure to a chemical can
dramatically affectresponse. This important phenomenon is not taken
into accountin the toxicity tests that underlie chemical risk
assessments.
7) It has now been established that cells respond differentlyto pulsed
exposures to some chemicals, compared to continuousexposures. Thus a
pattern of repeated exposures interrupted byregular intervals of
non-exposure elicits a different responsecompared to cells continuously
exposed.[14,15] "For example,when animals respond to
gonadotropin-releasing hormone, thepulse frequency of stimulation is
more important than theaverage level of the hormone."[14]
8) Medical understanding of the role of inflammation in diseaseis now
changing substantially. Inflammation is a sign that theimmune system
has been incited, and animals (or humans) withinflammation react
differently to chemical exposures thananimals without inflammation.[16]
9) We now know that many dose-response relationships are notlinear.
Indeed, the shape of dose-response curves is thesubject of an extensive
body of contentious literature, yetrisk assessors continue to rely most
often on the simplifyingassumption of linearity. This simplifying
assumption makes manyrisk assessments possible but it may also make
them wrong.
10) Thousands of potentially important biochemical reactionsare ignored
during risk assessments. Current federal protocolsfor examining the
tissues of experimental animals weredeveloped before the advent of
biochemistry and molecularbiology. After animals are dosed and then
killed for tissueanalysis, their organs are examined visually for gross
damage,but microscopic examination of the organs is not
typicallyrequired -- much less the sophisticated analyses made
possibleby modern biochemistry and molecular biology. Animal testing
isdecades behind current biology, and will likely remain so foreconomic
reasons. Thorough examination of dosed animals wouldbe far more
expensive than the simple examinations that arestandard today (and
which already cost in the range of $20,000to $100,000 per test).
Even when animal tissues are examined under a microscope, notall tissue
types are examined. All organs are composed ofvarious types of cells,
and each type would need to be examinedto claim that a thorough
investigation had been conducted, butthis is not done.
Thus thousands of distinct biochemical mechanisms are notexamined,
because no one requires them to be (to keep costsdown). Cognition,
behavior, fertility, disease resistance, malereproduction, chronic
neurotoxicity, immune alteration andhormone function (critical to
hundreds of biochemical systems)are all ignored in typical risk
assessments.[17]
In sum, thousands of potential injuries are missed by typicalgross
visual (and occasional microscopic) examinations inanimal toxicity
tests.
11) The vulnerable period of development is not tested. Withrare
exceptions, the period of greatest vulnerability(corresponding to the
human period of life from conceptionthrough age 18) is not tested in
laboratory animals. Adultanimals are tested. In addition, effects on
second and thirdgenerations are not typically looked for.
12) The commercial forms of chemicals tested in the laboratorymay bear
little resemblance to chemicals of the same name foundin environmental
food chains. Depending on source of exposure,pathway through the food
chain, and weathering effects,chemicals measured in humans or other
animals can havedistinctly different characteristics from "pure"
commercialforms of chemicals, meaning that many risk assessments
areconducted on chemical species that are not encountered in thereal
world.[18]
It must be obvious that these shortcomings of risk assessmentcannot be
remedied because there simply aren't enoughlaboratories and enough
money to take into account all thesources of variability listed above.
And if corporations and government agencies cannotsystematically take
these biological phenomena into account,they should acknowledge that
their risk assessments are hardlymore than window dressing, having
little to do withreproducible science, intended mainly to mollify
anapprehensive public.
[To be continued.]
============
Reprinted with permission from: Peter Montague, "The ChemicalWars," New Solutions Vol. 14, No. 1 (2003), pgs. 19-41.
[1] Mary O'Brien, Making Better Environmental Decisions; AnAlternative
to Risk Assessment (Cambridge, Mass.: MIT Press,2000; ISBN
0-262-65053-3).
[2] Quoted in Anthony B. Miller and others, EnvironmentalEpidemiology,
Volume 1: Public Health and Hazardous Wastes(Washington, DC: National
Academy of Sciences, 1991), pg. 45.
[3] United States General Accounting Office, Chemical RiskAssessment;
Selected Federal Agencies' Procedures, Assumptions,and Policies
[GAO-01-810] (Washington, D.C.: United StatesGeneral Accounting Office,
August, 2001), pg. 31.
[4] A major study of risk assessment was conducted by 11
Europeangovernments during the period 1988-1990, and published by
theCommission of the European Communities under the titleBenchmark
Exercise in Major Hazard Analysis in 1991. The 11governments
(Netherlands; Greece; Great Britain; Denmark;Italy; Germany; France;
Belgium; Spain; Finland; andLuxembourg) established teams of their best
scientists andengineers and set them to work on a single problem:
analyzingthe accident hazards of a small ammonia storage plant.
Privatecompanies like Rohm & Haas, Solvay, Battelle, and
Fiatcontributed experts as well. The results were stunning: the 11teams
varied in their assessment of the hazards by a factor of25,000.
Analyzing the hazards of a single, small plant handlingonly one
chemical, these world-class "risk experts" reachedwildly different
conclusions. For example, the individual riskat the "refrigerated
storage site" was calculated by one groupof experts to be one-in-400,
but by another group of experts tobe one-in-10-million. (Figure 3.5,
pg. 58 of the Benchmarkstudy.) See Commission of the European
Communities, BenchmarkExercise on Major Hazard Analysis. 3 Volumes.
(Luxembourg,Luxembourg: Commission of the European Communities, 1991).
[5] David O. Carpenter, Kathleen Arcaro, and David C.
Spink,"Understanding the Human Health Effects of Chemical
Mixtures,"Environmental Health Perspectives Vol. 110 Supplement
1(February, 2002) pgs. 25-42.
[6] Beverly S. Rubin, Mary K. Murray, David A. Damassa, Joan C.King,
and Ana M. Soto, "Perinatal Exposure to Low Doses ofBisphenol A Affects
Body Weight, Patterns of Estrous Cyclicity,and Plasma LH Levels,"
Environmental Health Perspectives Vol.109, No. 7 (July 2001), pgs.
675-680.
[7] K.S. Landreth, "Critical windows in development of therodent immune
system," Human and Experimental Toxicology Vol.21, Nos. 9-10 (Sep-Oct,
2002), pgs.493-498;
[8] M.C. Garofolo, F.J. Seidler, M.M. Cousins, C.A. Tate, D.Oiao,
and T.A. Slotkin, "Developmental toxicity of terbutaline:Critical
periods for sex-selective effects on macromoleculesand DNA synthesis in
rat brain, heart, and liver," BrainResearch Bulletin Vol. 59, No. 4
(Jan. 15, 2003), pgs. 319-329;
[9] T.A. Lindsley and L.J. Rising, "Morphologic and neurotoxiceffects
of ethanol vary with timing of exposure in vitro,"Alcohol Vol. 28, No.
3 (Nov., 2002), pgs. 197-203;
[10] M.R. van den Heuvel and R.J. Ellis, "Timing of exposure toa pulp
and paper effluent influences the manifestation ofreproductive effects
in rainbow trout," EnvironmentalToxicology and Chemistry Vol. 21, No.
11 (Nov., 2002), pgs.2338-2347.
[11] Erik Baatrup and Mette Junge, "Antiandrogenic PesticidesDisrupt
Sexual Characteristics in the Adult Male Guppy(Poecilia reticulata),"
Environmental Health Perspectives Vol.109, No. 10 (October 2001), pgs.
1063-1070.
[12] Nicholas T. Ingolia and Andrew W. Murray, "HistoryMatters," Science Vol. 297 (Aug. 9, 2002), pgs. 948-949.
[13] Upinder S. Bhalla, P.T. Ram, and R. Iyengar, "MAP
KinasePhosphatase As a Locus of Flexibility in a
Mitogen-ActivatedProtein Kinase Signaling Network," Science Vol. 297
(Aug. 9,2002), pgs. 1018-1023.
[14] M.S. Berrill, S. Bertram, B. Pauli, D. Coulson, M. Kolohon,and D.
Ostrander, "Comparative sensitivity of amphibiantadpoles to single and
pulsed exposures of the forest-useinsecticide fenitrothion,"
Environmental Toxicology andChemistry, Vol. 14, No. 6 (1995), pgs.
1011-1018;
[15] R.B. Naddy, K.A. Johnson, and S.J. Klaine, "Response ofDaphnia
magna to pulsed exposures of chlorpyrifos,"Environmental Toxicology and
Chemistry Vol. 19, No. 2 (2000),pgs. 423-431.
[16] P.E. Ganey and R.A. Roth, "Concurrent inflammation as adeterminant
of susceptibility to toxicity from xenobioticagents," Toxicology Vol.
169, No. 3 (Dec 28, 2001), pgs.195-208.
[17] U.S. Environmental Protection Agency, Health Effects
TestGuidelines; OPPTS 870.4100 Chronic Toxicity [EPA
712-C-98-210](Washington, D.C.: U.S. Environmental Protection Agency,
1998.)
18. S.L. Schantz, J.J. Widholm, and D.C. Rice, "Effects of PCBExposure
on Neuropsychological Function in Children,"Environmental Health
Perspectives Vol. 111, No. 3 (March 2003),pgs. 357-376.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RACHEL'S ENVIRONMENT & HEALTH NEWSEnvironmental Research
FoundationP.O. Box 160New Brunswick, N.J. 08903Fax (732) 791-4603;
E-mail:
Pulp Pollution Primer
