• Home
• About ATSDR
• Press Room
• A-Z Index
• Glossary
• Employment
• Contact Us
• CDC

ATSDR en Español

Search:

Skip directly to: content | left navigation | search

---

Find a Publication

1. ATSDR > Public Health Assessments & Consultations

*    Concerns raised by community members about cancer incidence in vicinity of these areas are addressed in the Community Health Concerns section of this report.
**    The Massachusetts Department of Environmental Protection is planning fish sampling to address this public health issue.

Source: Census of Population and Housing, 1990: Summary Tape File 1A (Massachusetts) [machine-readable data files]. Prepared by the Bureau of the Census. Washington, DC: The Bureau [producer and distributor], 1991.

* A household is an occupied housing unit, but does not include group quarters such as military barracks, prisons, and college dormitories.

Source: Census of Population and Housing, 1990: Summary Tape File 1A (Massachusetts) [machine-readable data files]. Prepared by the Bureau of the Census. Washington, DC: The Bureau [producer and distributor], 1991.

The following section was not available in electronic format for conversion to HTML at the time of preparation of this document. To obtain a hard copy of the document, please contact:

Agency for Toxic Substances and Disease Registry
Division of Health Assessment and Consultation
Attn: Chief, Program Evaluation, Records, and Information Services Branch,
MS E-56
1600 Clifton Road NE, Atlanta, Georgia 30333

ATSDR received eight letters commenting on our Public Health Assessment for the Materials Technology Laboratory (U.S. Army), Watertown, Massachusetts, December 26, 1995. If the accuracy of a statement was questioned, we responded by verifying or correcting the statement in the document. We answer specific health questions in the following section from each letter. Following this section, we summarize the respondent's major concerns or questions about past radiation exposure and answer them.

FUTURE EXPOSURE

1)

A couple of respondents expressed concern about future exposure scenarios. "Table 1 presents a summary of the potential exposure pathways on site. There is no evaluation of subsurface contamination on the AMTL parcel. AMTL is included only as a potential source of contamination to the Charles River. ATSDR appears to have assumed whatever remediation takes place (as yet undecided) will address all other potential exposure pathways. This assumption seems premature since a final technology has not yet been selected and there has been no demonstration of technology effectiveness." " The report concludes that it is "unlikely that people would be exposed to contaminants at levels of health concern." However, there is no quantitative assessment of what the risks might be. ATSDR simply assumes no one will be exposed because the areas are not currently publicly accessible. For future exposures, again no risks are calculated, and they rely on the fact that "before these properties can be released for reuse, residual contamination must be cleaned up to levels considered protective of public health according to guidelines set by state and federal regulatory agencies."

ATSDR evaluated the potential for past, current and future exposure. Our role in current or future exposure scenarios is to stop, prevent, or reduce harmful exposures. For past exposures, our role is to try and provide information on the likelihood that illness or disease will occur as a result of that exposure. In the case of MTL, we looked at current exposures and recommended immediate public health actions (i.e., fish sampling lead to a fish advisory, sewer sampling lead to more sampling to better define contaminant levels). When future remediation or landuse decisions are made, it is the regulator's role to determine future cleanup levels for the protection of public health. ATSDR is available, if requested, to provide and evaluation of human exposure with regard to proposals for land use changes.

PAST EXPOSURE TO CONTAMINATED GROUNDWATER

2)

A couple of respondents had questions as to whether past use of neighborhood wells for both drinking and garden irrigation could have posed a health hazard. "It appears that testing in well bottoms, and water and sewer drains in adjacent neighborhoods should be conducted to determine migration patterns. The government should make the Watertown citizens aware of any contamination existing in these locations so the populace can be aware of possible effects on progeny and factor this into their decisions to have offspring." "We all ate fruits and vegetables from the garden at our house on Nichols Ave., from my grandfather's garden on Berkley St (not labeled on any of your maps) and especially from my other grandfather's garden on Laurel St. He grew a wide variety of vegetables and fruit, and we ate a lot. The garden was watered from a well that my grandfather had dug himself. There were other residents who had dug wells in the neighborhood. Having early memories of eating food from the gardens that were watered by that well water, I am overwhelmed that the first warning to not use them is given in this report."

Past industrial activity in the area contaminated groundwater in some parts of the community. Berkley Street (where your grandfather had a well), was near both the Watertown landfill and a rubber company - two potential sources for groundwater contamination. The Massachusetts Department of Environmental Protection (MDEP) has investigated contamination associated with a number of waste sites in Watertown. Currently, if MDEP determines that groundwater contamination at any of these sites currently poses a health hazard to community members, actions are taken to reduce or eliminate exposure. Their evaluation includes potential vapor transfer from contaminated groundwater into indoor air. (92)

Since the groundwater in the area is currently contaminated, it was likely contaminated in the past. However, even if contaminants were present, uptake in plants would have been influenced by factors such as the chemical type, its concentration, and the kind of vegetable grown. The primary groundwater contaminants in high concentrations are solvents which do not bioconcentrate in crops. Some metals are the primary contaminants that do bioconcentrate. Uranium metals produced from burning or by exposure to air (uranium dioxide (UO₂) and triuranium octaoxide (U₃O₈)) are insoluble in water. (69) Most metals are generally not very soluble and therefore tend not to travel easily or very far in groundwater from a source of contamination. Uranium uptake from the soil is restricted to the root system and is only transported to the outer root membrane, not the interior root. No significant translocation of uranium from the soil to above ground parts of the plants has been observed in research studies; however, root vegetables grown in contaminated areas may be a slight source of exposure. Fruit and vegetables grown in uranium contaminated areas may contain some uranium from airborne dust on the surfaces. However, very little of uranium ingested would be absorbed in the body. (69)

ATSDR is not aware of any sampling data for wells previously used by nearby residents so we can not evaluate individual exposure consequences. However, our experience at other hazardous waste sites shows that generally, non-subsistence consumption of crops watered with low concentrations of contaminants posed no health hazard.

RECORD SEARCHES

3)	One respondent wanted a more thorough search of all government documents referring to the Arsenal, an oral history of the people who lived in the area and especially those who worked at the Arsenal, recorded and made a part of the public record.

The Army conducted these interviews. Those documents are part of Administrative Record kept at the Watertown public library.

ATSDR also conducted numerous interviews with past employees as part of our evaluation. Those conversations are referenced in this document.

SUMMARY OF MAJOR CONCERNS ON PAST AIR EXPOSURE TO DEPLETED URANIUM

QUESTIONS ON POSSIBLE HEALTH EFFECTS

4)

One family, with a history of cancer (myelocytic leukemia, anaplastic thyroid cancer), thyroid dysfunction, and multiple sclerosis (MS) is particularly concerned that longtime residence in Watertown near the Arsenal contributed to health problems. "My sister died of anaplastic thyroid cancer in October, 1995. It had metastasized to the lungs. Doctors at Mass General Hospital have confirmed that the only known cause of this type of cancer is from radiation exposure. However, she never worked at any job involving any type of radiation exposure. However, she did live on Nichols Ave and on Putnam Street from 1935 to 1955. Could her type of cancer have been related to her exposure to any type of emissions from the Arsenal Complex?" "All members of my family who lived on Nichols Ave until 1955, mother, father, two sisters, a brother, and myself have been diagnosed with some form of thyroid dysfunction. I remember a breeze coming from the direction of the Arsenal. My aunt, who lives on Laurel St., and has Crohn's disease, often talked about how lucky we were to have the breeze, especially during the summer. That breeze at night always kept us cool." " As your report states, MS is supposed to be related to a virus contacted some time before the age of 15 years. However, it is my understanding that genetic disposition may also play a role. Is it possible that my proximity to nuclear and radiological activity, such as what occurred at the Arsenal during the 1930s through the 1950s, have influenced my risk of having MS?" " My nephew died of acute myelocytic leukemia at age 20 in July, 1994. He was my sister's youngest son (who lived on Nichols Ave and Putnam Street). According to the American Cancer Society and New England Medical Center, the only plausible causes are exposure to radiation and hydrocarbons. However, he grew up in the relatively clean environment of Kennebunk, Maine. He was a college student at St Anselm's College in New Hampshire. Is any of this related to his mother having lived so close to the Arsenal during her formative years?"

We cannot answer person-specific health effect questions because we did not perform person-specific exposure evaluations. However, we looked at the location the streets where the family had residences (Nichols Avenue (west of Coolidge Hill on our Figure 8, Putnam Street (NW of Elm), and Laurel Street (between Cypress and Dexter) in relation to the GSA property (where burning occurred between 1955 and 1967) concluded that the houses were too far from the Arsenal to plausibly be impacted by the DU burning. The closest street to the GSA property was Nichols street; ~ 2350 feet (~ 715 meters) away. Because DU is extremely dense, even small particles carried in smoke would quickly settle to the ground in the general proximity of the burn pit. The radiological surveys to determine the extent of contamination at the GSA property showed that radiologic contamination is primarily confined to surface soil in discrete, on-site areas. No contamination was found at the boundary of the site outside the new fence or on the southwest side of Greenough Blvd. (38)

For exposures after 1967, our evaluation showed that the maximum effective dose equivalent (EDE) resulting from DU incineration at Building 43 was less than 200 mrem per year (mrem/y), whole body, to an individual living within 200 meters (or ~ 650 feet). As the distance one lived from the Arsenal increased, the effective dose decreased. Since we estimated that the lifetime cancer risk was less than the risk of cancer expected from background levels of radiation at 200 meters, people potentially exposed outside of 200 meters were unlikely to have been exposed to harmful levels of DU. Laurel, Putnam, and Nichols Streets were between ~1173 and 2053 feet (~360 to 665 meters) from Building 43.

Thyroid Dysfunctions and Cancers

Even if DU exposure did occur in some other way (e.g., working or playing nearer to the Arsenal), the plausible health effects from DU exposure do not include thyroid dysfunctions, thyroid cancer, or MS. Uranium exposure has not been implicated in thyroid disease. (103) The only correlation of thyroid cancer and a radioactive source is the ingestion of radioactive iodine (the Arsenal had no iodine releases) or high dose-rate exposure to ionizing radiation such as with the medical treatment for ringworm of the scalp. (104, 105) Thyroid cancer is a relatively uncommon neoplasm accounting for about 1.5% of all cancers in the U.S. (106) The primary risk factors for thyroid dysfunction or thyroid cancer are diet, hormonal etiology, and a positive family history. (16, 107)

Thyroid cancer is more common in women than in men with a rate of 5.9 cases per 100,000 women compared to a rate of 2.2 cases per 100,000 men in 1984. The majority of cases occur between the ages of 25 and 65 years of age. (108) There are several types of thyroid cancer. In the United States, papillary cancer is the most common (58% of cases) followed by follicular (24%) and anaplastic (15%). Medullary cancer accounts for only about 3 percent of thyroid cancers in the U.S. (109)

Thyroid cancers appear to be increased in both areas with iodine deficiency (follicular cancer) and in areas with an abundance of iodine (papillary cancer). A history of nodular goiter is seen in about 80% of patients with anaplastic carcinoma. (110) Medullary thyroid cancer is the type of thyroid cancer most influenced by heredity. Nearly 20% of the cases of medullary thyroid cancer are associated with an autosomal dominant genetic defect. (107)

In 1907, infants began to be treated for thymus enlargement with external radiation (with X-rays). Subsequently, irradiation was used to treat enlarged tonsils and adenoids, mastoiditis, sinusitis, hemangiomas, lymphadenopathy, and acne. In 1950, the first study linking thyroid cancer to therapeutic irradiation of the head and neck was reported. Many studies have since confirmed this association. (111) Women appear to be at increased risk from these exposures. Risk is also increased in patients who are younger at the time of the irradiation.

A review of the cancer incidence in the census tracts near the Arsenal showed that thyroid cancer incidence in the census tract CT 3703 (near the GSA property) was significantly elevated in total incidence (Standard Incidence Ratio (SIR) of 325) as well as among females. However, the 95% confidence interval (CI) was very broad (CI =127-656) indicating an unstable statistic and leaving considerable doubt about the true SIR. (16) The true SIR could be much lower or higher than the observed SIR. It appears that Putnam, Laurel and Nichols street are in this census tract. The thyroid cancer cases in CT 3703 were not geographically concentrated (clustered) nor did any cases reside close to the MTL property (within one-quarter mile). For any exposures, but particularly radiation, one would expect to see health effects in the population impacted by clusters, starting nearer the exposure source. This is not the case for any of the cancer incidences reported in the four census tracts in Watertown. (16)

Multiple Sclerosis (MS)

MS has also not been associated with depleted uranium or any radiation exposure. Some studies have reported possible links between MS and occupational exposure to metals, but the predominance of epidemiologic data suggests that MS is associated with an infectious process. (89)

One respondent asked, "Since most people with MS tend to live with it for a long time, what is being done to determine the number of people who are living or have lived in the community and are suffering from the disease? If children were affected by contamination in the 1940-1960s, there could be many people in their 30's - 60's with the disease who have yet to show up on the mortality lists".

Since there has not been an association between the disease and depleted uranium, no further studies with respect to the Arsenal are recommended. However, MDPH has provided a copy of their Health Outcome Data consultation to the Multiple Sclerosis National Society, Massachusetts Chapter, "for their consideration of the prevalence estimate and appropriate followup". (16)

Myelocytic Leukemia

As the respondent stated, radiation and exposure to hydrocarbons (particularly benzene) are risk factors associated with myelocytic leukemia (one of the cancers experienced by the family), but the family lived on Nichols and Putnam Street, approximately 2347 and 4302 feet (or 715 and 1311 meters) from the GSA property. We believe that exposures from the contamination at the GSA property were unlikely at that distance.

QUESTIONS ON MEDICAL SURVEILLANCE, HEALTH STUDIES, AND HEALTH STATISTICS REVIEWS

5)	One respondent wanted to know when the medical surveillance program was first implemented and whether records exist for the entire time period that workers may have been exposed.

We were not able to determine the exact date of the medical surveillance program but do know that it started before 1974. (112)

Workers were provided copies of their medical records upon request. Once an employee left the Arsenal, the records were sent to the National Personnel Records Center, 111 Winnebago Street, St. Louis, MO 63118 (314-425-5761 [civilian] or 314-263-3901 [military]). (60) These records are filed by name, date of birth, social security number and agency name. (Ethel Duberry, MTL, provided the information about the records center. A records center employee indicated how the records were filed.)

6)	One respondent wanted to know why there was no discussion of the lifestyle and land use behaviors of the local population during the years of hazardous exposures.

We looked at aerial photographs to determine approximate distances to nearby residences and industrial buildings. Our exposure assumptions for DU burning were that burning continued continuously for 8 hours per day, every day of the year and occurred over a 20 year period. Investigating lifestyle and land use behaviors would likely show less exposure, but we wanted our evaluation to be very conservative. Even with very conservative assumptions, our evaluation showed the amount of depleted uranium (DU) released to air from MTL and the GSA site was unlikely to be at levels considered harmful to residents of the surrounding community.

7)

A couple of respondents had questions about how the people who moved away from Watertown were accounted for in the health statistics report. "The report discusses that MDPH is considering further evaluation of risk factor information and residential histories in relation to cancer as a result of the Arsenal. We understand that such surveys currently do not take into account prior residents based on the statistical methodologies employed by the governments. This approach would seem to make sense for a steady state problem where deceased residents leaving Watertown equals deceased residents entering Watertown. However, this steady state case is unlikely if significant community contamination occurred as a result of the Arsenal. ATSDR should recommend that future surveys solicit input from past residents. If the government does not choose to include past residents, it should justify the basis for this statistical sampling methodology in its reports." "The health emphasis seems to be on the people living (or recently deceased) of Watertown in the 1990s. That is not the same population that may have been effected by the nuclear and radiological material during the 1930s through the 1950s, when the safeguards that were eventually required were not in place. What about the people who moved away? What about their children?" "I found the report emphasized the importance of death in how seriously the medical evidence was weighed. While I don't quarrel with the seriousness of death, there is also a weightiness to the impairment on the quality of life of those who suffer from diseases that do not necessarily lead to death. A cancer study should be continued and expanded to include past residents of Watertown and their progeny. And, while the medical documentation is appreciated, I believe that more recent studies can add to the report. Also, has there been any gathering from the extensive medical community: individual doctors, Mt Auburn Hospital, Waltham Weston Hospital, and the whole Boston Medical Complex?"

Our exposure evaluation did not show that residents were exposed to harmful levels of DU, so follow up of people who moved from the area was not recommended.

The MDPH is considering further evaluation of risk factor information and residential histories for certain cancer cases which appear concentrated in Watertown. For any exposures, but particularly radiation, one would expect to see health effects, such as disease clusters, in the impacted population, starting nearer the exposure source. This is not the case for any of the cancer incidences reported in the four census tracts in Watertown. Moreover, the types of cancers that MDPH considered statistically elevated, non-Hodgkin's lymphoma and stomach cancer, are not associated with exposure to depleted uranium. MDPH is not associating the cancer cases with the Arsenal. Their follow up is part of the state's role in providing good public health practice and is designed to determine possible causes or risk factors. We will pass your recommendation onto them as they continue their follow up. However we do not have any information that shows "significant community contamination occurred as a result of the Arsenal". Moreover, we have no evidence that Watertown residents have been exposed to any environmental contamination from the Arsenal.

QUESTIONS ON EXPOSURE AND DOSE RECONSTRUCTION

8)	What were your actual dose calculations and your assumptions?

The radiological dose assessment made for the Watertown Arsenal Building 43 incineration required a series of steps. Essentially, we obtained information from the facility in which the burn rates and air flows were discussed, but did not contain specific details. We also received information from MTL on the amounts of depleted uranium a "typical burn" would comprise as well as some information on stack air sampling. However, the facility did not have specific values for some of the parameters we needed to model the dose estimates. From discussions with MTL staff and based on the information received, ATSDR estimated the maximum and minimum amounts of air flow in the crude incinerator, and the maximum and minimum amounts of DU in a burn. We next utilized a computer program, "Crystal Ball," which uses a statistical method known as Monte Carlo analysis to estimate the most probable releases. After we derived this value, the CAP88-PC dose assessment computer program was used to estimate the population dose of individuals around Building 43 on the arsenal property. For the stack height, ATSDR used the height of Building 43 plus the additional stack height above the roof, a total of 10.4 meters (~ 34 feet). The CAP-88PC default wind library for the Boston area was used for the probable wind speed and direction parameters. The radiation dose was also estimated using recommended values from the International Commission on Radiological Protection (ICRP) and related ICRP dose conversion factors.

As discussed in the Public Comment release of this document, radiation dose estimates were based on very protective (conservative) assumptions. We used the following assumptions - burning continued continuously for 8 hours per day, every day of the year and occurred over a 20 year period. Arsenal records for 1972-1977 indicate that DU burns usually lasted 2 hours and occurred 5 times or less a month (44). Comparing the likely burn rate scenario to the conservative scenario we modeled, our doses most likely are greatly overestimated and are still well below (about 60%) the current dose limit recommendations of the International Commission on Radiological Protection (ICRP) and the public exposure regulations.

9)	How do you justify your dose assumptions when you didn't consider data or information from the 1930s - 1950s - a period when more emissions were discharged?

The amount of depleted uranium (DU) released to air from MTL and the GSA site was unlikely to be at levels considered harmful to residents of the surrounding community. We base our conclusion on the fact that few, if any, residents lived close enough to sites to have been exposed. (7) Further, radiation dose estimates, based on very protective assumptions, showed a lifetime cancer risk less than that expected from background radiation levels.

Here is a summary of what we know about Uranium and depleted Uranium use between the 1930s and the 1950s. Although uranium was purified in the 1840s, its use in glass apparently has dated back as far as 79 A.D. Prior to the 1930s, uranium was used in research laboratories in relatively small quantities. In the 1930s, two European scientists discovered that uranium could undergo fission, releasing large amounts of energy. Upon the urging of Albert Einstein, President Franklin Roosevelt agreed to develop an atomic bomb. This led to large scale development of nuclear weapons under the Manhattan Project which was one of the top military research projects in the early to mid 1940s. (113) Small scale uranium and thorium research projects were conducted in building 421 at the Arsenal between 1946 and 1953. (7) During the production of weapons-grade uranium (U-235), depleted uranium (DU) is produced as a by-product of uranium enrichment. Hence, DU did not exist in sufficient amounts until the mid to late 1940s and therefore could not be present at the Arsenal prior to this time.

The Arsenal began its incineration of DU at the GSA site in 1955 and continued those operations until 1967 (7,24) at which time, incineration was shifted to Building 43. DU is very dense, in fact, if you melted it down and dropped in lead, the lead would float. Because DU is extremely dense, even small particles carried in smoke would quickly settle to the ground in the general proximity of the burn pit. The radiological surveys to determine the extent of contamination at the GSA property showed that radiologic contamination is primarily confined to surface soil in discrete, on-site areas. No contamination was found at the boundary of the site outside the new fence or on the southwest side of Greenough Blvd. (38)

Conceivably, very small particulates could have dispersed at greater distances from the source. However, deposition of these fine particles would be widely scattered, and consequently, measurable amounts of DU would not occur in localized areas distant from the source.

Based on aerial maps and photographs obtained from the Army, the population and population density surrounding the GSA property during those burns was very small (7). The aerial photographs show that the closest houses to the burn area on the GSA property in 1952 were on Berkley and Clarendon Streets approximately 1500 feet (~ 457 meters) away. (7) Across the Charles River, the closest buildings (unknown if they are houses), were approximately 1000 feet ( 305 meters) away. (7) As the distance one lived from the Arsenal increased, the DU concentrations would decrease. We believe that exposure at these distances was unlikely. Even in the unlikely event that people off site did inhale DU particulates, exposures would have been rare and the resulting dose to the lung would have been below levels of health concern.

One respondent asked if there could have been other exposures here from open pits of uranium sludge. (She heard this at a meeting held in January 1994, when a woman was reading from an EPA report that referred to pits located at the GSA property in the late 1930s). As we discussed above, uranium did not have wide-spread use in the 1930s. We could not find the report referring to these sludge pits. However, during the 1920s and 1930s, workers at the Arsenal were making steel, iron and nonferrous castings. (114) It is likely these pits could have been waste products from the casting molds and not uranium sludge. Further, if uranium was in these pits, it could not have migrated far enough to reach the residential areas.

For a more details on the history of chemical and radiological use and disposal at the Arsenal, see ABB Environmental Services, Inc. Former Watertown Arsenal Preliminary Assessment, Watertown, Massachusetts. October 1993.

10)

"The report states that very conservative assumptions were made concerning the frequency and duration of DU burns and such assumptions probably overestimate the actual duration and rate of burning. The report should also provide a discussion regarding to what extent the 1981 stack data might underestimate the emissions during years before engineering controls were installed and what impact that might have on the overall estimate of dose from DU incineration at Building 43."

The 1981 air data used in the radiation dose estimation model is from the time period after engineering controls reduced the amount of DU released to outside air. However, releases of fine DU particulates (less than 0.3 microns [µ]) to air during burning in 1981 and after, were most likely comparable to earlier years. Although the use of high efficiency particulate air (HEPA) filters on the incinerator decreased air releases of particulates greater than 0.3µ in size, fine particulates continued to be emitted. Since it is the very small DU particulates released to air that are of greatest concern from an exposure standpoint (because they can be inhaled deep into the lung), using the 1981 data representing the greatest amounts of material released from the stack and other conservative assumptions is a good estimate of dose. The maximum estimated radiation dose was not at levels which pose a public health hazard.

11)

With respect to the DU incineration at the GSA property, one respondent questioned MTL's compliance with 1966 guidelines without numerical testing to prove compliance. He was concerned that the public bares the risk if effective compliance did not exist. He also questioned ATSDR's DU air exposure assumptions in that we used the average and did not know the standard deviation of burn levels and how the wind patterns may effect exposure.

Again, sampling at the GSA property showed that radiologic contamination is primarily confined to surface soil in discrete, on-site areas. No contamination was found at the boundary of the site outside the new fence or on the southwest side of Greenough Blvd. (38)

See Comment 8 for more details on our Building 43 DU air exposure assumptions.

12)

One respondent wanted more information about the "Uncontrolled Depleted Uranium fires" listed in the report. How often and for what duration did the fires occur? What is the expected range of stack sample concentrations during these fires? Based on this data, what is the range of yearly whole body dosages which could have been received by residents living 250, 500 and 750 meters from the source?

In the past, a few uncontrolled DU fires also occurred inside the Foundry building. DU chips, that were temporarily stored in barrels, occasionally spontaneously ignited in the presence of air. The fires were contained within the barrel, inside the foundry. To prevent these types of fires from occurring, the DU chips were stored in oil or water. These fires were infrequent, and small (contained in a barrel) and no exposure outside of the foundry building was expected to have occurred from these fires. (41, 42, 43)

13)

"Why are census tracts from 1940, '50, '60 and '70 not used to determine Watertown population locations and densities? Aerial photographs from at least each decade should be used to gauge the proximity of residents to the Arsenal in these earlier years to determine the likelihood of greater contamination impacts."

Census data for those periods (1940s, 1950s and 1960s), is difficult to get and it is not available for all areas. We looked at aerial photographs from 1938, 1951, 1952, 1962, 1968, 1969, 1974, and 1990 (7) to determine the approximate distance to nearby residences.

14)

One respondent asked that we explain the significance of our dose calculations in relation to absolute risk. "In reviewing potential past exposures to radioactive materials ATSDR assumed that exposures which were equal to or less than exposures to background levels of radiation are deminimus without offering any support or documentation for this claim. We suggest that ATSDR review health effect occurrence from these potential exposures in absolute terms rather than by comparison to background exposures. For the receptors it must be assumed that these exposures are incremental in addition to their exposure to background radiation, not a part of their exposure to background radiation."

First, we have no evidence that Watertown residents have been exposed to environmental contamination from the Arsenal. During the period of DU burning on the GSA property (1955 - 1967), aerial maps and photographs show that the closest houses were on Berkley and Clarendon Streets approximately 1500 feet (~ 457 meters) away. (7) and across the Charles River, the closest buildings (unknown if they are houses), were approximately 1000 feet (~ 305 meters) away. (7) Exposures were unlikely at this distance. Further, radiologic sampling showed that contamination is primarily confined to surface soil in discrete, on-site areas. (38) Our evaluation of the DU incineration at building 43 showed that the maximum effective dose equivalent (EDE) was less than 200 mrem per year (mrem/y), whole body, to an individual living within 200 meters (or ~ 650 feet) of building 43 in 1981. As the distance one lived from the Arsenal increased, the effective dose decreased.

Absolute risk has been defined as the number of events per year above normal per million person-rem (a collective dose; a concept no longer recommended for use in radiological protection). Several national and international organizations, however, have estimated the risk of cancer with regard to whole body exposure to ionizing radiation. In the Environmental Protection Agency's (EPA) 1994 report "Estimating Radiogenic Cancer Risks", EPA estimates the annual cancer mortality risk from low dose, low dose-rate to be 509.1 x 10^-4 per gray. This is equivalent to 5.1 x 10^-4 per rad (rounding to appropriate significant figures). The International Commission on Radiological Protection (ICRP) in their report, ICRP 60, estimates the detriment of radiation exposure to be 5.92 x 10^-2 per sievert for whole body less the gonads and 1.33 x 10^-2 for the gonads for a total detriment of 7.3 x 10^-2 per sievert. (71) (Note: sievert is a unit of measure for equivalent dose) This is equivalent to 7.3 x 10^-4 per rem. The absolute risk to an individual living within 200 meters of Building 43 based on a VERY conservative estimate of release (see comment #8 for details on the assumptions) resulting in a dose of 0.2 rem (0.002 sieverts) would be 1.0 x 10^-4. However, the closest buildings were approximately 1,000 feet (~305 meters) away and exposures from DU at this distance were unlikely.

Finally, cancer is a chronic, multifactorial disease. Cancers from environmental exposure are rare (103). Cancers from uranium exposure in environmental settings are even more rare. (103) Our risk estimates reflect the possible increased risk DU exposure may have posed on a population. To say that people have a risk for a disease, does not mean they will come down with the disease.

15)	One respondent wanted us to explain the significance of dose in relation to shifting regulatory standards or guidelines.

Throughout history, the recommended dose limits have decreased. (0.2 roentgen per day (R/d); equivalent to 5 rem per month in the case of gamma radiation). The shifts in limits were primarily based on the concepts of the Linear-No Threshold (LNT) hypothesis. There is much scientific debate on the LNT hypothesis (see question 16 for additional discussion). Even with a downward shift to lower, presumably more protective standards, the estimated doses arising from the Arsenal are about 60% of the current dose limit recommendations of the International Commission on Radiological Protection (ICRP) and the public exposure regulations.

Brief History of Dose Standards and Recommendations

Early (prior to 1930s) recommendations limiting the exposure to ionizing radiation were set to control the exposure to X-rays. This early limit was called a "tolerance dose" which was not quantitative. Between 1931 to 1934, a "tolerance dose" of 0.2 R/d (approximately 0.2 rem per day for gamma radiation exposure only) was recommended by national and international organizations. In 1936, the daily limit was reduced by 50% to 0.1 R/d which is approximately 0.1 rem per day for gamma radiation exposure only. In 1949, a weekly dose of 0.3 rem was recommended. These limits were based not on hard scientific evidence, but as a matter of safety. As concern grew for the genetic effects caused by radiation, a limit of 10 R in 30 years (about 10 rem in 30 years) to the reproductive cells was adopted. This limit was recommended by the National Academy of Science in 1959. Also in 1959, the Federal Radiation Council, formed to establish federal policy, reviewed radiation health data and determined that a yearly dose of 500 millirem (mrem) (or 0.5 rem), 10% of the allowable occupational dose, would serve as the Radiation Protection Guide to the public. This limit served until the 1990s.

In the United States, the regulatory standards do not necessarily meet the recommendations or guidelines established by the NCRP or the ICRP. These organizations currently recommend that the effective dose from all sources to a member of the public not exceed 1 millisievert (mSv) (100 millirem or .1 rem) in a year, but this does not include background and radiation doses arising from medical procedures. This same limit is in use within the United States. Occupational limits, as recommended by the ICRP, however, are 20 mSv (2 rem) per year, averaged over a defined 5 year period. Within the United States, occupational limits are 5 rem per year, similar to the dose limits recommended approximately 40 years ago.

Setting the dose limits includes other issues besides the risk of lethal cancers. These issues include knowledge of the rates of fatalities (radiation or otherwise), the rates of curable cancers, and the rates of hereditary effects taking into account population and demographic statistics. In the dose limits of 20 years ago, the limits only included severe hereditary effects and did not include the curable cancer rate. Today's limit includes this and the other factors previously mentioned.

For more detailed discussions, interested parties should obtain a copy of Kathren, R.L. (1996). Pathway to a paradigm: The linear non-threshold dose-response model in historical context: The American Academy of Health Physics 1995 Radiology Centennial Hartman Oration. Health Physics 70: 621-635.

16)	One respondent wanted us to explain the significance of dose in relation to literature on low dose exposure/no threshold.

Linear-No Threshold Hypothesis

The radiation protection and scientific communities are in a state of flux discussing the significance and relevance of the Linear-No Threshold (LNT) hypothesis; a hypothesis is an unproven theory or model. The LNT hypothesis states that any exposure, no matter how small, carries a risk of detriment or harm to the exposed system. Current scientific methods have not been developed to either prove or disprove this theory of low level radiation effects.

After the second world war, interest in the LNT model expanded because of its simplicity and its representation of the upper risk level. The LNT was developed to establish practical scientifically-based radiation protection standards to minimize detriment, but also, enhance benefits of radiation exposure. (115) The LNT was first developed to explain a certain set of radiation experiments on fruit flies. Over time, the LNT, although only a model or hypothesis, has become a "fact" in spite of the apparent lack of strong scientific proof, especially in the environmental radiation dose range. However, as with any scientific thought, there are many scientific publications in the literature that do support a LNT concept for radiation effects as well as do not support such claims. In fact, there is much literature on hormesis that purports that low doses of radiation are beneficial.

The radiological doses received from operations at the Arsenal are well below the levels of exposure scientifically shown to cause an adverse health effect in humans. As they are at or near the level of exposure one receives from background radiation (about 0.3 rem or 300 mrem per year including radon) and because of the unknowns associated with LNT, the significance of these doses is unclear at the present time.

Relevant Literature

One respondent wrote that, "findings on childhood cancer and radiation exposure also present sufficient reason to be concerned about extra doses of radiation at the level of 200 mrem/year. Prenatal x-rays, averaging between 20 mrem-400 mrem, have been found to double the risk of a childhood cancer (see Stewart and McMahon)."

ATSDR is aware of Alice Stewart and coworkers findings. In another study performed over 35 years ago, the Oxford Survey of Childhood Cancers suggested a relationship between leukemia and prenatal x-irradiation for diagnostic purposes. A similar study in the New England region of the United States covering the years 1940 to 1960 reported the relative risks to prenatally exposed children was about 1.5 for leukemia and about 1.2 for other cancers. The dose to the fetus in these examinations was relatively low, on the order of 500 millirads to 5,000 millirads, but the interpretation, as discussed by the National Research Council, is complicated "by the fact that little increase in susceptibility has been evident in prenatally x-irradiated experimental animals and there is no known biological basis for such an increase in susceptibility or for the suggested equivalence in magnitude of the leukemia excess with that of other childhood cancers." These doses and the dose rates were short, acute exposures occurring during specific periods of gestation. The effects can be different from protracted, low dose, low dose-rate exposures estimated for the whole body over the entire period of gestation. Since the effects of low dose and low dose-rate to the whole body can be different from the relatively high dose and high dose-rates received by the individuals in the prenatal radiation studies or other health studies of high dose, high dose rates (116), exposure and outcome comparisons are not possible.

The same respondent wrote, "Studies of childhood leukemia and fallout from the Nevada Test Site demonstrated a statistically significant increase in leukemia from external radiation doses between 290 mrem - 3 rems. Thyroid neoplasms were also elevated in populations downwind from the Nevada Test Site (Kerber 1993) at average doses of 17 rem or 17,000 mrem (an effective dose equivalent of 500 mrem to the thyroid)."

Uranium is the major radioactive contaminant released from the Arsenal and there is no correlation between uranium and cancer in the studies of populations living downwind of atmospheric testing. (116)

Studies of populations living downwind from the Nevada Nuclear Test site have shown correlations with various diseases such as leukemia and thyroid cancer. However, the Nevada Test site had fallout from a mixed bag of isotopes; the primary dose-contributing isotopes included cesium, iodine, and strontium. Very little, if any, uranium was present. (116) The doses to the body and to specific organs in these exposures are much higher than the doses we estimated in Watertown. Additionally, the thyroid cancers experienced near the Nevada Test site were from iodine exposure. The Arsenal had no iodine releases.

Studies released in 1979 showed a reported increase in childhood leukemia for the state of Utah over the years of 1944 to 1975. In those sections of the state considered unexposed, the mortality rates were similar to the remainder of the United States. However, children in the exposed areas showed a slightly elevated rate of childhood leukemia. A further analysis, however, pointed out that the apparent increase in childhood leukemia may have been the result of a diagnostic error. (116) Additional studies of the populations impacted by the atmospheric weapons testing both in the United States as well as the South Pacific indicate that the risk of cancer is related to the amount of radioactive cesiums, strontiums, and iodides deposited in the areas and uptaken by the affected populations. (116) There is no record that the Arsenal used or released these isotopes. (7)

A recent review of the atomic bomb survivors showed that the risk of leukemia was dose-related: the risk rises slowly until a dose of 50 rads was received, then accelerates. (117) However, the type of leukemia involved differs depending on the radiation dose received. Overall, no excess leukemia cases were observed until the radiation dose received approached 5 to 10 rads (10 rads of alpha equals 200 rem) Comparing this dose to the dose estimated at MTL from the incineration of DU, the doses from MTL (maximum 200 mrem/yr or 0.2 rem/yr) are several orders of magnitude below this level known to cause leukemias. (117)

Misconception on Low Dose and Low Environmental Dose

Currently, there are several misconceptions regarding worker studies and "low exposure" versus "low environmental exposures". Worker exposures are normally at least an order of magnitude higher than the doses observed at this site. Individuals purporting an effect on workers state that these are "low level exposures", but in actuality, the exposures are much higher than environmental levels of exposure. For example, nuclear worker studies in the Department of Energy (DOE) complex show that exposures in excess of 2.5 rem (2500 mrem) begin to show correlations with adverse health effects. The maximum effective dose equivalent for people within 200 meters (or ~650 feet) of building 43 at the Arsenal before 1981 was 0.2 rem or 200 mrem. The doses at the Arsenal are at least ten times lower than doses received in the occupational low-dose DOE worker studies.

1. *Data on length of burns was not available.

Agency for Toxic Substances and Disease Registry, 1825 Century Blvd, Atlanta, GA 30345
Contact CDC: 800-232-4636 / TTY: 888-232-6348 

Department of Health and Human Services