a

I "4

8/31/83

Mr. Frank Graham Jr.
AUDUBON Magazine

Dear Sir,

I read with considerable interest your article in the July, l983
AUDUBON concerning dioxin, and Dow Chemical. Although I am
presently a Visiting Professor at the University of Miami School of

Medicine, I was at one time the senior EPA scientist responsible for the

Alsea study referenced in your article. While the generalizations made
by the Dow people are familiar, they in no way accurately reflect the
study findings, or for that matter, the reason the study was undertaken.

The investigation of spontaneous abortions in Alsea, Oregon began
with a 1978 notice in the Federal Register which informed the public
that a Rebuttable Presumption Against Registration(RPAR)of was
in place. As EPA noted, and summarized in the RPAR, evidence was avail-
able which indicated the the herbicide, or it's trace contaminant, 2,
fetotoxic and carcino-
genic effects in laboratory animals. The notice solicited public comment
for, and against registration of the chemical. In response to the sol-
icitation, EPA received a letter from an Oregon woman who felt that she
and seven other women had experienced 13 miscarriages as a result of
exposure to during forest spraying. The miscarriages took place
between 1972 and 1977, and closely followed(within six weeks)the appli-
cation of the herbicide. Since the mid 1970's, the Office of Pesticide
Programs, EPA has employed a person to investigate any pesticide related
incident thought to be adversely effecting the public health or environ:
ment. The title of the position is Pesticide Incident Response Officer,
and 1 approved the Alsea investigation as a normal function of that pro?
gram activity, and for no other reason. And, considering that 
may be associated with cancer, birth defects, and miscarriages in ani-
mals, there was more than sufficient reason to carefully review the Alsea
experience.

I don't intend to get into a debate on the merits of the study.1t
is far too complex a proposition to discuss in this format. There are
literally reams of information, pro and con, in the court and adminis-
trative hearing testimony concerning this study. I will say however, that
at no time did EPA authorize a scientific paper to be released on the
Alsea study. At some point during the early suspension phase of the act-
ivities, a severely edited version of the findings, originally intended
for use as an insert in the RPAR Position Document, was released
to the public. As a result, many negative comments concerning the quality
of the Alsea study were forthcoming from members of the scientific comm-

1_ i3 9 ?1 E)

 in the study area betwe

 

unity. The release of this document as presented was obviously a mistake,
since it was treated by the proponents of the chemical, some members of
the scientific community, and others, as the scientific justification for
the March, 1979 suspension of I will also say that not one criti-
cism raised by the reviewers of the early report has gone unanswered dur-
ing the litigation process, and the answers are now part of the public
record. I would also say that at no time has any critic of the Alsea study,
or of the subsequent su5pension of talked directly to me, or to
my knowledge any person directly associated with the conduct and analysis
of the study. Although, I have felt constrained to not comment publicly
on the Alsea study during the litigation process, recent developments
concerning dioxin contamination around the country, and confusion as to
possible human effects frequently expressed in the media, the Ho Chi Min
City conference on dioxin related health effects, and the many misrepre-
sentation of the facts associated with the Alsea study findings, prompted
me to reSpond to your article.

The comments by the Dow people concerning the lack of exposure data
in Alsea were totally inaccurate. I don't intend to get into a long dis?
cussion of usage in the Alsea study area, however, suffice it to
say, that approximately 20,000 pounds of the herbicide were 
. As to 
tically assoc1ations between the time and level of 
spray in the study area, and a subsequent increase in spontaneous abort-
ions was clearly shown. The Hospitalized Spontaneous Abortion 
for the study area was positively correlated with spray applied
two months previously, and abortion indices based on conception cohorts
were positively correlated with spray which occurred within one month of
conception. Both relationships held true for six year totals as well as
for individual values by month and year. The positive correlations with
the spray patterns were supplemented by the findings that the median
gestational age at the time of the abortion was lower in the study area
compared to the control area. The study area also reported proportionally
more spontaneous abortions in the gestational age range of zero to eight
weeks compared to the control area. Finally, the overall HSAI in the study
area was significantly elevated over that of the control area.

Concerning exposure of the women to the herbicide, it was my feeling,
and subsequently the position of EPA that it: was through the water supply,
and not directly by spray. That is, the women were likely ingesting the
TCDD residue with the drinking water(oral ingestion)as opposed to expos?
ure by spray drift(dermal and respiratory exposure). It is well known
that TCDD residues are bound to soil particulates. I believed it
likely that residues,from the Spray, were bound to the soil particles, and
then transported to the various water sources of the Alsea population. I
felt this scenario to be possible because many of the people in the Alsea
area use a water holding pond as the source for their household water
supply. A holding pond is simply an area of a creek, or gully, constru-
cted with a dam in order that the flowing water can be contained. The -
holding pond is located on the hill where the house is located, but some-
what above the dwelling. The water runoff is stored in the holding pond,
and then by means of gravity, or mechanical pump, transported to the house

2
1? Fifth)

  

?below. During the hearing in the Federal District court in Midland, Mich-
igan, it was the Dow position this premise was not possible. However,

when EPA sampled the sediment from the holding pond of one of the original

eight women, we found TCDD residues greater than the one part per billion
limit which EPA considers to be the minimal safety threshold. Dow then

took the position that since the residue was on the sediment, it was on the
bottom of the pond and presented no risk to the drinker. However, I sub-
sequently observed water from a kitchen faucet in the home of one of the
eight women, and found it to be very turbid, with visible particles of
sediment. In fact, during my visit in December, 1979, the resident of the
house told me that the water during the spring and summer months(spraying
season)was considerably more turbid than I observed. It did not at the
time, nor does it now, seem inconceivable that TCDD residues could be trans-
. ported on such sediment directly to the waiting embryo. If such contact is
made within the first 30 days following conception, the embryo is,at that
time,at great-risk to chemical insult, possibly resulting in abortion of
the conceptus. .

As to "minute quantities", there is in fact no known "No Observed
Effect Level" for TCDD in animals or humans. Anyone talking to the issue
of very small concentrations of any chemical residue must,or at least
should be aware of the No Observed Effect Level, and the potential long?
term risk posed by a chemical not meeting this standard in a laboratory
setting. This is particularly true with a very toxic chemical, such as
TCDD. The facts are that we don't know the minimal dosage of TCDD necess-
ary to kill or damage a human embryo. That is the bottom line, and all the
discussion of "small quantities" is essentially meaningless. Therefore, the
EPA took the prudent position that since there is no known level, below
which damage will not occur, and since there was evidence in primate
studies that small doses of TEDD would result in spontaneous abortions, a
potential public health hazard existed.

Comparing the Seveso, Italy incident with Alsea is truly comparing
apples with oranges. In a country historically reluctant to acknowledge
and report abortions, the population of Seveso was possibly exposed to a
mist or vapor of TCDD, as the result of an industrial accident. We don't
know for example, how many people were actually exposed, how many women
were pregnant at the time of exposure, the period of gestation at the time
of exposure, etc. There is absolutely no information available on the
severity of exposure to the total population, or more importantly, on the
amount of residue actually absorbed by human tissue, by any adequate
quantitative measurement. It is interesting that when Dow scientists want
to condemn the Alsea study for bad science, they always refer to Seveso
to support their position. There is of course an additional characteristic
of the Seveso incident that makes it absolutely inappropriate to compare
the findings with Alsea. Exposure to the Seveso population was by dermal,
or respiratory contact. Alsea exposure was by oral ingestion. Certainly,
no competent toxicologist would attempt to compare orally induced effects
with effects developed as a result of dermal or respiratory exposure in 
any good laboratory study, without extremely rigid controls.

As to the Nitro Nest Virginia situation, the exposure mechanism is
not clear. However, one could make the assumption that it was either by

3


-.. 
A

~52 

.42.:

 . .. . I
mg?! m, 
. . -., . 

  
  
 
 
 

    

aecxiauwv . .

or respiratory contact. Given that the chemical is produted in a

rather closed system, it is extremely likely that any exposure would be
result of an industrial accident, perhaps similar to the Seveso in-

s?gif' cident. If this is the case, then the presentation of the exposure as

sufficient to create long-term effects, if they exists, is probably not

justified. If the expoSure should come from the day to day operation of

- a relatively closed production system, then one could assume virtually
no absorption into the worker's tissue. Simply to associate employment

.q with exposure, and finally to absorption is inappropriate in attempting

,Er to quantify exposure_history. However, assuming that exposure, and absorp-

tion occurred, one must then look to the presentation of the data before

conclusions may be drawn as to effects. Certainly, in-a closed system,

unless accidental exposure occurred, one would not expect to see overt

conditions, such as chloracne. In the long-term one would not expect to

observe diseases requiring extensive periods of time for development, un?

less, the worker's were exposed to very low dosage over an extensive

time period to a very hot chemical, a carcinogen. Interestingly,

three soft-tissUe sarcomas were identified in the Nitro worker populat-

ion. However, no emphasis was placed on the finding or reporting of these

very rare cancers because only three were found. Unfortunately, this is

a seven fold increase over what one would expect to find in a normal

population. Thus, one might expect these findings to merit considerable

attention, particularly so, since physicians treating Vietnam veterans

have reported seeing unexpectedly large numbers of soft?tissue sarcomas.

In point of fact, to-produce an "ironclad connection between dioxin .

residues in the environment and illness or death of any kind" one would

only have to look at the available record to locate case after case of

animal death or birth defects. As to human association,_one would have to

--dose a population with measured amounts of TCDD, control fOr confounding
variables, and measure the effects over time, before.an "ironclad" dose
response finding coUld be determined. Morally, and ethically such an
approach poses great difficulties to the researcher, and his or her spon?
soring institution, and is seldom practiced in other than the must rigidly
controlled clinical, or laboratory studies. The great majority of epidemi-
logic studies don't-meet the criteria necessary to establish cause-effett,
and in fact, it is only through statistical assoc' tions, and the pre-
ponderance of data on a given-topic that one is q?gally able to infer
such a relationship. For example, "ironclad" data has not as yet been es-
tablished between smoking and lung cancer. However, the preponderance of
data certainly suggest the relationship exists, and that smoking may be
hazardous to pne's health. Unfortunately, no matter how pervasive the
evidence to the contrary, one can usually get an argument over cause-effectdon't believe I can close without some comment on Dr Kenaga's
characterizatiOn of'academic scientists "commonly" perverting the scien-
tific process to gain favor a grantor, or other financial source. I
have had the pleasure to serve on Graduate Faculties for more.than six
years, and have worked for an additional l6 years in the government bureaua
cracy, largely in the area of grant and contract funding to the academic'
community. It has been my experience that the reasearchers with whom I've
been fortunate to work, are highly individualistic, independent minded
people with an abiding interest in the truth.

Jack Griffith,Ph.D
Visiting Professor

16948