PUBLIC HEALTH ASSESSMENT
WELLS G AND H
WOBURN, MIDDLESEX COUNTY, MASSACHUSETTS
The following sections briefly describe the compounds detected in the indoor air of threeWoburn residences and the Puddle Duck Day Care Center during the investigations cited above. Compounds of concern are those compounds with concentrations in at least one environmentalmedium exceeding the health assessment comparison values established by the ATSDR. Comparison values are media-specific concentrations that are used by health assessors to selectenvironmental contaminants for further evaluation. They are not used to predict adverse healtheffects of exposure. Media concentrations less than the comparison value are unlikely to pose ahealth threat, although other factors must also be considered in determining the health effects ofexposure (such as hypersensitivity of individuals or multimedia exposures). However, mediaconcentrations above a comparison value do not necessarily represent a health threat. Theselected contaminants will be further assessed in the Public Health Implications Section. Somecontaminants may be eliminated as a concern in the following sections due to an incomplete past,present, or future exposure pathway.
Table B-1 presents the quantitative results of all indoor air sampling and analysis conducted at three residences in the Dewey Avenue area and at the Puddle Duck Day Care Center on OlympiaAvenue.
The indoor air of the basements of three residences in the Dewey Avenue area of the Wells G andH site were sampled by the EPA (EPA, 1989). Dewey Avenue is located approximately 500 feet to the south of the Unifirst Corporation facility and 300 feet to the west of W.R. Grace. A total of nineteen eight-hour, time weighted average (TWA) air samples were collected at the residences and analyzed for VOCs.
A private environmental consulting company conducted sampling at the Woburn residencesconcurrently with the investigation conducted by the EPA (ENSR, 1989). One of the TWA indoor air samples was collected in the basement of the first residence tested, five samples werecollected in the second residence and two in the third. In addition to collecting eight 8-hour,TWA indoor air samples in the basements of the three residences, six eight-hour, TWA sampleson the first floors of the homes and five eight-hour TWA outdoor samples were collected andanalyzed for VOCs.
EPA also conducted a one day indoor air screening survey at the Puddle Duck Day Care Centerin April, 1991 (EPA, 1991a). A total of 27 grab samples were collected; 18 were collected fromvarious areas within the Puddle Duck Day Care Center, 7 were ambient background samplesfrom outside the day care in the play area, and 2 were ambient background samples from theEPA New England Regional Laboratory in Lexington, Massachusetts. Indoor air sampling wasconducted beginning at 6:50 a.m. on a Monday morning, after the day care was closed over theweekend, therefore, the air samples collected would reflect a worst case scenario. Samples wereanalyzed for VOCs.
A second indoor air survey was conducted at the Puddle Duck Day Care Center by the EPA inOctober (EPA, 1991b). A total of 14 grab samples were collected; 10 were indoor air samples from the day care and 4 were outdoor ambient background samples in the play area. Four 4-hour,TWA air samples were also collected in canisters; three were indoor air samples and one was abackground sample from outside in the play area. All samples were analyzed for VOCs.
In addition to collecting and analyzing air samples during the second investigation of the PuddleDuck Day Care Center, the disinfectant used on diaper changing surfaces in the day care centerwas sampled and a headspace analysis performed on-site utilizing the portable chromatograph. The purpose was to determine if any similarities existed between the disinfectant headspacechromatogram and the indoor air grab sample chromatograms. Analysis of the disinfectantheadspace sample produced a single peak on the chromatogram that could not be identified. Thissingle peak did not match any of the peaks appearing on the indoor air chromatograms. Therefore, the three unidentified peaks also present in the indoor air grab samples are notassociated with the disinfectant (EPA, 1991).
Each of the four investigations summarized in the previous section on environmentalcontaminants has its own quality assurance and quality control (QA/QC) procedures. Overall,the sampling, analytical techniques, and data were adequate for the purpose of the healthassessment.
The presence of physical and other hazards at the Wells G and H site was investigated during the April 27, 1995 site visit. The hazards present included the debris from area residents (tires, furniture, etc.) as well as the old fencing and concrete piping remaining from past remediation.
To determine whether nearby residents are exposed to contaminants migrating from the site, thefactors influencing human exposure were evaluated. The pathways analysis consists of fiveelements: a source of contamination, transport through an environmental medium, a point ofexposure, a route of human exposure, and an exposed population.
Pathways are identified as completed or potential. Completed pathways require that the fiveelements exist and indicate that exposure to a contaminant has occurred in the past, is currentlyoccurring, or will occur in the future. Potential pathways, however, require that at least one ofthe five elements is currently missing. Potential pathways indicate that exposure to acontaminant could have occurred in the past or could occur in the future. Pathways can beeliminated if at least one of the five elements is missing and will never be present. Thediscussion that follows incorporates only those pathways that are important and relevant to thesite.
A. Completed Exposure Pathways (Table B-3)
This addendum only evaluates recent indoor air data. Past, current, and future exposurepathways are possible from the inhalation of VOCs (the only exposure pathway reviewed in thisaddendum) present in the indoor air in the basements of residences and in buildings around thesite. Potentially exposed populations are residents and workers in the area, and children whoattend the Puddle Duck Day Care Center.
Most of the residential samples were 8-hour TWA samples, while most of the day care centersamples were grab samples. It is not possible to accurately determine whether these sampleresults are representative of longer exposure periods (i.e., whether longer-term concentrations,such as 24-hour TWA concentrations, would be higher or lower). One of the two day care centerinvestigations took grab samples early on a Monday morning, when VOC concentrations mayhave been at their highest due to the lack of indoor-outdoor air exchange that occurs when thebuilding is in use or due to reduced air flow rates that may be in effect in buildings during aweekend. Thus, these grab samples may reflect higher concentrations than would have beengathered with longer sampling times.
B. Potential Exposure Pathways
As stated above, the only exposure pathway examined was inhalation of VOCs in indoor air,which was determined to be a completed pathway.
A total of 26 compounds were detected in the indoor air investigations of Dewey Avenue arearesidences and the Puddle Duck Day Care Center. The maximum concentrations of eight compounds were above their respective comparison values (e.g., environmental media evaluation guide, or EMEG) (Table B-2). Of these eight compounds, only two compounds (i.e. trichloroethene and tetrachloroethene) are among those compounds originally identified as chemicals of greatest concern at the Wells G and H site (i.e. trichloroethene, tetrachloroethene, 1,2-dichloroethene, 1,1,1-trichloroethane, vinyl chloride) (ATSDR, 1989). The other six compounds detected in indoor air above ATSDR comparison values were methylene chloride, carbon tetrachloride, benzene, 1,2-dichloroethane, chloroform, and 1,1-dichloroethene.
While collecting indoor air samples, investigators considered household products as potentialsources of chemical contamination. At the Puddle Duck Day Care Center, samples were takenfrom a disinfectant used on diaper changing surfaces and from the headspace of its storagelocation. These samples were taken to determine if any similarities existed between analyses ofthe disinfectant and the ambient indoor air. Results of this analysis did not suggest that thedisinfectant was responsible for the contaminants detected in the ambient air of the day carecenter. Samples collected from the kitchen closet, containing bleach and other cleaning products,contained 29 ppb toluene. Because this concentration is not significantly different from thelevels detected in other samples collected in the center throughout the day, the investigatorsconcluded that these products are probably not the source of toluene. During investigations ofthree basements in Dewey Avenue area residences, investigators also noted the presence ofvarious household products including the following:
- laundry detergents
paint and paint thinners
wood stain and sealer
Although these substances were noted as being in well-sealed containers, their use within theconfines of the basement, or an inadequately sealed container, may have allowed for thevolatilization of chemicals. Through this process, these chemicals evaporate and contribute tothe concentrations of these chemicals in the ambient indoor air of the basement. Becausemaximum concentrations, in general, were detected at levels below or typical in indoor ambientair, investigators concluded that these concentrations are probably not directly associated withcontaminants present in the groundwater, but rather from the common sources in the household(EPA, 1991b).
Despite elevations above comparison values (for all compounds, the maximum concentrationwas above the ATSDR Cancer Risk Evaluation Guide (CREG), or the comparison value basedon the carcinogenicity of the constituent), the compounds detected in indoor air generally hadmaximum concentrations that are within typical ambient air or indoor air concentration ranges. Thus, indoor air concentrations detected in the Dewey Avenue area residences and in the PuddleDuck Day Care Center cannot be definitively attributed to contamination in groundwater and arenot of public health concern. Each of the eight compounds are discussed below.
The maximum concentration of benzene in residential indoor air was 2.5 ppb, with mostdetections in the 1-2 ppb range. No benzene was detected in the day care center. The maximumdetected concentration is greater than the CREG (0.03 ppb) and slightly above ATSDR's 2 ppbacute inhalation Environmental Media Evaluation Guide (EMEG), but well within typicalambient air or indoor air benzene concentrations.
ATSDR (ATSDR, 1993a) reports that benzene is ubiquitous in the atmosphere. In one study in which ambient air samples from 44 sites in 39 U.S. urban areas were collected from June through September in 1984-1986, benzene was present in every sample, with site median concentrations ranging from 4.8 to 35 ppb (overall median 12.6 ppb). Mobile sources were the major sources for benzene in ambient air. Data from 1975-1985 reported median values for suburban indoor air (1.8 ppb) and work place air (2.1 ppb) (ATSDR, 1993a). EPA's Total Exposure Methodology reported a median level of 2.2 ppb in homes with no smokers and 3.3 ppb in homes with smokers (ATSDR, 1993a).
ATSDR's acute inhalation EMEG of 2 ppb was derived from an MRL based on a mouse study that showed a reduction in several immunological parameters (considered a "less serious" effect). An uncertainty factor of 1250 was applied to the lowest-observed-adverse-effect level in the study (ATSDR, 1993a). Because the maximum concentration detected in indoor air was well within typical ambient or indoor air concentrations, and because of the large uncertainty factor (i.e., safety) factor used in deriving an acute inhalation MRL, it is not expected that residents in the homes will have adverse health effects from benzene.
Investigations of the three residential basements showed carbon tetrachloride detections from 0.2 to 0.3 ppb. The maximum concentration (0.3 ppb) is greater than ATSDR's CREG of 0.01 ppb. However, the carbon tetrachloride concentrations seen in these basements are within typical concentrations found in indoor air or ambient air. ATSDR reports that carbon tetrachloride appears to be ubiquitous in ambient air and that recent studies have shown that it is also a common contaminant in indoor air. Concentrations are usually greater in indoor air than ambient air, indicating that building materials or household products (e.g., cleaning agents) may be the source of carbon tetrachloride (ATSDR, 1994a). The average concentrations of carbon tetrachloride reported from four U.S. cities ranged from 0.14 to 0.29 ppb (ATSDR, 1994a). ATSDR also reported that typical concentrations in homes in several U.S. cities were about 0.16 ppb, with some values up to 1.4 ppb.
Chloroform was detected at a maximum concentration of 0.8 ppb; the ATSDR CREG is 0.008 ppb. It was detected in only one (6 Dewey Avenue) of the three residences tested. This maximum concentration is within typical concentrations measured in indoor or ambient air. Ambient air concentrations in U.S. cities ranged from 0.02 to 5 ppb (ATSDR, 1993b). The 1987 update of the National Ambient Volatile Organic Compounds database reported background levels of 0.02 ppb (ATSDR, 1993b). Typical median indoor air concentrations of chloroform range from about 0.2 to 4 ppb (ATSDR, 1993b), with chlorinated tap water one of the most significant indoor sources of chloroform.
1,2-Dichloroethane was detected in only one indoor air sample (i.e., in one Dewey Avenue residence) at a concentration of 0.1 ppb, higher than ATSDR's CREG of 0.01 ppb. However, this single detection was well within typical concentrations measured in ambient air or indoor air. ATSDR (1994c) reports on median values of 1,2-dichloroethane in ambient air in urban areas ranging from about 0.012 to 1.5 ppb. Results reported for indoor air concentrations from the Total Exposure Assessment Methodology (TEAM) found median concentrations in indoor air samples ranging from 0.005 to 0.72 ppb in cities located in North Carolina, Louisiana, Texas, and California (ATSDR, 1994c).
1,1-dichloroethene was estimated to be between 1.0 and 3.5 ppb at the 6 Dewey Avenueresidence. It was also detected in one of the two other residences at a concentration of 1.4 ppb;the ATSDR CREG is 0.005 ppb.
This compound has been detected in ambient air studies across the U.S. ATSDR (1994b) reports that the National Ambient Volatile Organic Compound database reported an ambient dailyaverage concentration of 4.6 ppb. Other ambient air studies have generally found average concentrations in urban areas from about 0.3 to 0.8 ppb (ATSDR 1994b). ATSDR (1994) reported on only one indoor air study (in Research Triangle Park, North Carolina), where mean concentrations ranged from 7.1 to 47.3 ppb.
Methylene chloride was detected at a maximum concentration of 130 ppb in Dewey Avenue arearesidences and 17 ppb in the day care center; both concentrations exceed ATSDR's CREG of 0.6ppb. The maximum concentration also exceeds the intermediate EMEG of 30 ppb. Among thethree residences, methylene chloride was detected only at 6 Dewey Avenue, with the highestconcentrations detected in the basement air. First floor data showed a concentration of 4.9 ppb;however, field blanks for methylene chloride, a common laboratory contaminant, showedconcentrations of 7.5 and 0.8 ppb, thereby indicating that the concentration measured on the firstfloor may be due to laboratory contamination.
As noted earlier, EPA and ENSR conducted simultaneous sampling of basement indoor airsamples at the Dewey Avenue area residences. The concentrations of methylene chloride fromthe two reports together for 6 Dewey Avenue show a range of 16-130 ppb. EPA reported thattheir data could only be estimated as showing concentrations between 16 and 29 ppb due toproblems with reproducibility or precision in the data. The 6 Dewey Avenue residence hadseveral potential sources of methylene chloride in the basement (e.g., paint thinners, paints). Theday care center concentrations ranged from 13-17 ppb; potential sources of methylene chloride inthe day care center included various paints and glues.
ATSDR (1994d) reports that methylene chloride has been detected in ambient air samples from around the world. Background levels are usually about 0.05 ppb, but concentrations in urban areas are generally one to two orders of magnitude higher (e.g., 0.5 to 5 ppb). In addition, ambient air concentrations are much lower than concentrations in buildings, where products containing methylene chloride (e.g., paints, thinners) may be used (ATSDR 1994d).
The intermediate EMEG is based on an intermediate MRL developed for methylene chloride. The intermediate inhalation MRL is based on a rat study, with cytoplasmic vacuolization in the liver (a "less serious" effect) seen in the study. An uncertainty factor of 1000 was applied to derive the MRL (ATSDR, 1994d). Given the large uncertainty factor in the MRL, the uncertainty in actual concentrations of methylene chloride in the basement of 6 Dewey Avenue (e.g., lack of reproducibility), and the intermittent exposure that would occur to constituents in basement air, it is not expected that residents would suffer adverse effects from exposure tomethylene chloride.
Tetrachloroethene (PCE) was detected at a maximum concentration of 1.9 ppb in basements ofDewey Avenue residences, a concentration higher than ATSDR's CREG for PCE (0.3 ppb). Thecompound was not detected at the day care center. The site-specific background concentrationwas 1.9 ppb, while TEAM data from three cities detected levels ranging from 0.04 to 1.3 ppb(ATSDR, 1993c).
Trichloroethene (TCE) was detected in Dewey Avenue residences at concentrations ranging from0.065 to 0.16 ppb; the maximum detected concentration is only slightly above ATSDR's CREGof 0.1 ppb. It was not detected in the day care center. The site-specific backgroundconcentration was 0.8 ppb, or higher than the maximum detected indoor air concentration.
No data are available for typical concentrations that may be present in indoor air (ATSDR 1993d). However, ambient air concentrations reported for urban areas across the United Statesranged from about 0.04 to 0.7 ppb (ATSDR 1993d). Thus, the concentrations detected in theDewey Avenue residences are typical of ambient air concentrations in urban areas and areprobably not the result of groundwater contamination.
Health concerns in Woburn have been expressed since the late 1970's. A number ofinvestigations have been conducted by the MDPH, the Centers for Disease Control (CDC), andHarvard University. Initial concerns focused on elevations in both total and male childhoodleukemia incidence (the rate of occurrence of leukemia). For the period 1969-1978, childhood(aged 0-14 years) leukemia incidence was significantly elevated for males (9 observed, 3.1expected), but not for females (3 observed, 2.2 expected). The ratio of male to female incidencewas three to one. For the area of Woburn that received the greatest proportion of water fromWells G and H (census tract #3334), childhood leukemia incidence in males was 12 times higherthan expected (5 observed, 0.4 expected, p < 0.001). A p value less than 0.001 means that theprobability that the difference between the observed and expected values was due to chance isless than 0.1 chances in 100. The remaining census tracts in Woburn did not have a significantincrease in childhood leukemia incidence for either sex. During the period 1979-1981, threecases of childhood leukemia were diagnosed in males. Childhood leukemia was not diagnosed infemales during this same period. For the period 1982-1990, seven additional cases of childhoodleukemia occurred in Woburn, from which the ratio of male (5 cases) to female (2 cases)incidence is 2.5 to 1. Five of the seven cases were diagnosed during the period 1982 - 1983. Noadditional cases of childhood leukemia were diagnosed between 1987 and 1990.
Although the focus of the cancer concerns has been childhood leukemia, kidney cancer mortalityis also significantly elevated (Standard Mortality Ratio = 145, 21.3 expected cases, 31 observedcases, p = 0.05) for the period 1969-1986. No association has been established between cancerincidence or mortality and the Woburn environment. This has partly been due to the smallnumber of cancer cases as well as difficulty in trying to characterize past exposures (ATSDR, 1989).
The MDPH convened a panel of experts from throughout the United States to assist indetermining if a link existed between environmental contamination originating from this site andhealth problems in the vicinity of the site. The panel's strongest recommendations included morestudies of the environment and the development of a system of non-invasive watchfulnessfocused on reproductive outcomes. It was hoped that reproductive outcomes would serve as amore sensitive indicator of an environmental problem and, in addition, would allow for anenvironmental association to be detected statistically.
The ATSDR awarded funds to the State of Massachusetts in support of the Woburn Environmentand Birth Study (WEBS) which was designed to address the concerns of the panel. Thisinvestigation began in the fall of 1988, and the results were released in August 1994.
WEBS was designed to establish (1) whether the occurrence of certain adverse reproductiveoutcomes was different between residents of Woburn and selected local and regional comparisonpopulations; and (2) whether the occurrence of these outcomes was different between Woburnresidents classified as "exposed" and those classified as "unexposed" to contaminated drinkingwater. The study examined the occurrence of adverse reproductive outcomes for two timeperiods: the Retrospective Study period (1969-1988) and the Prospective Surveillance Studyperiod (1989-1991). Maternal exposure to contaminated water was assessed on the basis of anestimate of the proportion of contaminated water that reached each birth residence locationduring pregnancy.
No differences in the rates of most of the outcomes evaluated between Woburn and the twelvesurrounding communities were indicated. Some types of birth defects were higher in Woburnthan in these surrounding communities; however, the discrepancies appeared to be more likelydue to methodologic differences rather than real differences in the occurrence of birth defects. Furthermore, comparisons between groups within Woburn identified as "exposed" and"unexposed" did not indicate significant differences in rates of most adverse reproductiveoutcomes evaluated. The results of WEBS were made public,and MDPH received comments,which MDPH is currently evaluating for a formal response. In addition, MDPH is currentlyevaluating the birth outcome data and their occurrence relative to tannery and greenhouse site inWoburn.
The MDPH is also conducting an expanded case control study of childhood leukemia utilizingexposure data generated as part of the WEBS. This type of study compares individuals with agiven disease to a comparison group of identical age, sex, etc. to demonstrate differencesbetween the two that may be associated with development of the disease. This WoburnChildhood Leukemia Follow-Up Study has been submitted for peer review.
Community health concerns related to the public health assessment addendum are addressed asfollows. For a more complete evaluation of community health concerns related to the site as awhole, refer to the health assessment in Appendix D.
|1.||Will the contaminants in groundwater plumes migrating beneath buildings at the Wells G and H site and the site vicinity volatilize into the indoor air of these buildings and pose a potentialhealth hazard to the building occupants?|
The VOCs detected in indoor air cannot be attributed to groundwater contamination. Householdproducts are the primary suspected sources for these chemicals. Assuming that the levelsdetected are representative of the average concentrations found in the buildings, indoor aircontaminant levels are unlikely to represent a health concern for people who live, work, or attendday care in the area buildings.
|2.||Could the contamination of indoor air be related to the increased incidence of childhood leukemia and kidney cancer in Woburn?|
This public health addendum only examined data for 3 residences and one day care center; thus,because of the limited number of buildings, the addendum cannot address questions about therelationship of leukemia or kidney cancer to indoor contamination in these four buildings. However, it is unlikely that exposure to the VOCs detected in indoor air will result in adversehealth effects. The concentrations detected were generally within typical concentrations reportedfrom many studies in the U.S. for VOC constituents in indoor or ambient air.