Skip directly to search Skip directly to A to Z list Skip directly to site content





The Vasquez Boulevard and I-70 (VBI70) site area spans approximately 450 acres in northeastDenver, Colorado and includes smaller areas that the U.S. Environmental Protection Agency(EPA) has proposed to the National Priorities List (NPL). The study area has an irregular shape,and is located primarily southeast of the interchange of Interstate 25 and Interstate 70. It includesall or part of the following five Denver neighborhoods: Clayton, Cole, Elyria, SouthwestGlobeville, and Swansea. The area is a mix of residential (approximately 17,500 people living in5,126 housing units), commercial, and industrial areas.

The EPA has taken soil samples from approximately 75% of the residential properties in theVBI70 study area and tested them for several metals, particularly arsenic and lead. This report isabout the public health significance of the soil testing.

ATSDR's Findings for Arsenic in Soil

In our daily lives, children and adults ingest small amounts of soil that cling to their hands fromwhat is called hand-to-mouth activity. The amount of soil that people ingest is somewherearound one-sixteenth to one-eighth of a teaspoon. In addition to this inadvertent or accidentalsoil ingestion, some children, particularly preschool children, will ingest large amounts of soil(for instance, a teaspoon), which is referred to as soil-pica behavior. Generally, this soil-picabehavior occurs in some 1- and 2-year-old children as part of their normal exploratory behaviorand in some 3- to 6-year-old children as part of an intentional behavior. Soil-pica behavior mightalso occur accidentally when children eat food with dirt adhering to it. The percentage ofpreschool children (ages 1 through 6) that go through a stage of soil-pica behavior is not knownprecisely, but studies have reported that this behavior occurs in as few as 4% of children or in asmany as 21% of children. One and 2-year-old children have the greatest tendency to soil-picabehavior and this tendency decreases as they get older.

ATSDR has determined that soil arsenic levels at many (but not all) of the properties in theVBI70 study area are safe for children and adults with typical soil intake levels. ATSDR,however, is concerned about soil arsenic levels in approximately 650 of the 2,986 propertiessampled so far. ATSDR is concerned that these properties have arsenic levels in soil that mightpose a public health hazard for soil-pica children who ingest unusually large amounts of soil. Based on EPA's baseline risk assessment, EPA has identified properties as a concern for childrenwith soil-pica behavior if the property has an average arsenic level in soil of 47 ppm or greater. Based on demographic information, about 300 preschool children live in these 650 householdsand somewhere between 12 to 60 of these children might have soil-pica behavior some timeduring their preschool years. Depending upon the amount of arsenic contamination in these 650properties and how much dirt soil-pica children ingest, the most likely health effects that mightoccur in soil-pica children from eating soil just one time include nausea, stomach cramps,vomiting, diarrhea, facial swelling, and headaches. No children have been diagnosed witharsenic poisoning in the VBI70 area that can be related to arsenic in soil; however, it is possiblethat cases could have been missed because the most likely symptoms (nausea, vomiting, etc.) arecommon symptoms in children that can result from a variety of causes.

Arsenic in soil at some properties is also a public health concern for long-time residents becauseof the potential increased risk of cancer from arsenic exposure. This risk is greatest for childrenwho grew up in yards with high levels of arsenic in soil and who continued to live there as adults. If EPA uses 70 ppm as their clean-up level for arsenic, about 480 properties will be affected.(1) Itshould be noted that as of spring 2001, EPA has cleaned up about 50 properties so far because ofelevated arsenic levels in soil.

Some uncertainty exists in deciding whether or not adverse health effects might occur in people at the VBI70 site. Uncertainty exists in two areas: estimating how much of a contaminant people are exposed to (that is, the dose) and determining the health effects that might occur. The uncertainty that exists in estimating the dose for soil-pica children comes from the following issues:

  • estimating the maximum arsenic level in a property based on the average arsenic level from three composite samples,
  • varying amounts of dirt soil-pica children eat,
  • variations in how often children exhibit soil-pica behavior,
  • assuming that soil-pica children eat soil from the most contaminated part of the property, and
  • uncertainty in the percentage of children with soil-pica behavior.

Therefore, a child with soil-pica behavior who lives at a property with arsenic-contaminated soil might not get sick if the child eats soil from an area in the yard with low arsenic levels; or, if the child eats only a small amount of soil, and the amount of arsenic exposure is not enough to cause health effects.

Uncertainty also exists in determining the health effects that might occur because of the inexact nature of the following:

  • uncertain estimates of how much arsenic and lead gets into the blood stream and tissues once soil-bound arsenic and lead are ingested,
  • assumptions that the harmful effects observed in people exposed to arsenic in drinking water, which is readily absorbed by the body, is similar to the harmful effects that might occur in people exposed to arsenic bound to soil, which is likely to be less absorbed by the body,
  • estimates of the dose in human studies when arsenic is found in drinking water, and
  • accuracy of the exposure estimates in human studies that were used to develop health guideline values.

ATSDR Findings for Lead in Soil

Some properties in the VBI70 site have high levels of lead in soil that are a health hazard to some preschool children living at those properties. Exposure to lead-contaminated soil at the more highly contaminated properties has the potential for increasing blood lead levels in some preschool children, and might cause harmful effects involving the brain and nervous system. Possible effects include decreased intelligence, developmental delays, decreased stature, altered vitamin D metabolism, changes in blood enzyme levels, and decreased hearing.

EPA has developed a mathematical model that uses the average soil lead levels in a property to predict the percentage of children with blood lead levels above the Centers for Disease Control and Prevention's (CDC) level of concern of 10 micrograms lead per deciliter of blood (µg/dL). For the VBI70 site, EPA's model predicts a range of soil lead levels that could result in more than 5% of the children having blood lead levels greater than 10 µg/dL. The range of soil lead levels predicted by the model vary because EPA varied certain input parameters in the model (specifically, the geometric standard deviation and dietary lead intake). The model predicted that soil lead levels ranging from as little as 208 ppm to as much as 540 ppm as being a concern for increasing blood lead levels in children depending upon which input parameters most accurately predict blood lead levels. It should be noted that 78 properties have average lead levels in soil higher than 540 ppm while about 1,350 properties have average soil lead levels higher than 208 ppm.(2)

Recent blood lead testing by the Colorado Department of Public Health and Environment (CDPHE) in summer 2000 found about 10% of the 86 preschool children tested with blood lead levels above 10 µg/dL. However, it was not possible to determine how much lead contamination in soil contributed to these blood lead levels. Studies by other researchers have shown that about 30 percent of blood lead in children comes from lead in soil. EPA's blood lead model also predicts that a significant portion of a child's blood lead comes from soil. At an average soil lead level of 195 ppm, EPA's model predicts that about 4.4 µg/dL blood lead comes from soil for a typical child. The model predicts that for some children with the highest exposure to lead in soil, their blood lead levels might be as much as 9.5 µg/dL blood lead . It is important to remember that blood lead levels in children are most likely to be the result of exposure to lead from multiple sources. Here are a few examples:

  • lead in a child's diet,
  • lead in drinking water,
  • lead from leaded paint,
  • lead in soil,
  • indoor dust,
  • lead from lead-glazed pottery,
  • other unidentified sources.

CDPHE has a blood lead program that tests children for blood lead. For more information about CDPHE's blood lead program, contact Ms. Mishelle Macias at 303-692-2622. In addition, the Denver Department of Environmental Health (DEH) within the City and County of Denver is responsible for responding to lead issues. DEH's program is managed by Mr. Gene Hook, who can be contacted at 720-865-5452. DEH follows CDC guidelines, and when a child with elevated blood lead is referred, DEH will conduct an environmental investigation to identify potential sources of lead. Typically, the investigation includes collecting environmental samples from the home environment and administering a questionnaire designed to identify lead sources. DEH also provides the family with information about the health effects of lead, ways to prevent exposure to lead, proper nutrition, access to other relevant services, and the need for follow up blood tests.


ATSDR is making a number of recommendations to address the public health issues concerning arsenic and lead contamination in the VBI70 study area. These recommendations include ATSDR activities, such as developing programs for health care provider education, community education, and health investigations. In addition, ATSDR is making recommendations to local, state, and federal agencies. For instance, ATSDR is recommending to EPA that the agency reduce exposure to arsenic in those properties where soil arsenic and lead levels are a concern for children. ATSDR is also recommending to EPA that the agency collect soil samples from the remaining thousand or so properties in the VBI70 study area. All of ATSDR's recommendations are listed in the Recommendations Section of this report.

Public Meeting

In March 2002, ATSDR released the Public Health Assessment for the VBI70 Site and held 3 days of public meetings to announce the findings of the public health assessment. These meetings were held in Swansea and Clayton. In addition to presentations from community representatives and from ATSDR and EPA staff members, ATSDR handed out plain-language fact sheets that summarized the findings of the investigation. These fact sheets can be found in Appendix L and cover the following topics:

  • Summary of the PHA,
  • Arsenic in Soil,
  • Lead in Soil, and
  • ATSDR Site Activities.

Public Comment

When ATSDR released the Public Health Assessment for the VBI70 Site in March 2002, the Agency requested that people submit comments about the assessment. Those comments have been reviewed by ATSDR and changes have been made in the report as warranted. ATSDR's response to the public comments can be found in Appendix M.

ATSDR's Plans for the Future

In addition to the investigations and community involvement activities already conducted by EPA, CDPHE, the City and County of Denver, and ATSDR, ATSDR is planning future activities for the VBI70 site. These activities include the following:

  • begin the Agency's environmental health interventions project, a project that focuses on health education for both community members (or residents) and for local health care providers,
  • survey residents about soil-pica behavior and other activities that might increase exposure to arsenic and lead in soil,
  • identify acute arsenic poisoning in children, and
  • consider other health investigations that might be appropriate for the VBI70 site.

ATSDR will continue to work with and assist the community, the City and County of Denver, CDPHE, and EPA throughout our activities at the VBI70 site.



La ubicación del Boulevard Vásquez y la I-70 abarca aproximadamente 450 acres del noreste de Denver, Colorado, e incluye otras áreas más pequeñas que la EPA (Agencia de Protección Ambiental de los EE. UU.) ha propuesto para la Lista de Prioridades Nacionales (NPL). El área estudiada tiene configuración irregular, y se encuentra principalmente al sureste del cruce de la autopista interestatal 25 y la 70. Abarca las siguientes 5 zonas de Denver, total o parcialmente: Clayton, Cole, Elyria, el suroeste de Globeville, y Swansea. El área está compuesta de zonas residenciales (aproximadamente 17,500 personas que viven en 5,126 viviendas ), comerciales, e industriales.

La EPA ha recolectado muestras de suelo de aproximadamente 75% de los terrenos residenciales en el área estudiada de la VBI70, y las ha examinado para saber si contienen varios metales, particularmente arsénico y plomo. Este informe trata de la relevancia de la investigación del suelo para la salud pública.

Los Resultados de la ATSDR con respecto al Arsénico en el suelo

En la vida cotidiana, tanto los niños como los adultos ingieren cantidades bajas de tierra que se les afierra a las manos a través de lo que se llama contacto de mano a boca. La cantidad de tierra que ingieren las personas es alrededor de la decimosexta u octava parte de una cucharadita. Además de la ingestión de tierra no intencional, algunos niños, especialmente los de edad preescolar, ingieren grandes cantidades de tierra (por ejemplo, una cucharadita), lo cual es conocido como comportamiento de pica-tierra. Por lo general, este comportamiento de pica ocurre en algunos niños de 1 ó 2 años como parte de su conducta exploratoria normal, y de manera intencional en algunos niños de 3 a 6 años. El comportamiento pica también puede ocurrir cuando los niños comen alimentos con tierra. No se sabe el porcentaje preciso de niños preescolares, o sea los que tienen de 1 a 6 años, que pasan por la etapa de pica, pero los estudios señalan que esta conducta sucede en un mínimo de 4% a un máximo de 21 % de los niños. Los niños que tienen 1 ó 2 años son los que más tienden a exhibir el comportamiento de pica-tierra, el cual disminuye mientras ellos crecen.

ATSDR ha determinado que los niveles de arsenico en el suelo en muchas (pero no todas) las propiedades en la area de estudio de VBI70 son seguros para niños y adultos con niveles típicos de ingestion de suelo. A la ATSDR le preocupa el nivel de arsénico en el suelo de aproximadamente 650 de los 2,986 terrenos analizados hasta la fecha. A la ATSDR le preocupa el que estas propiedades tienen niveles de arsénico en el suelo que podrían constituir un peligro a la salud pública en los niños con comportamiento pica que ingieren cantidades anormales de tierra. En base a la evaluación sobre riesgo, la EPA ha calificado las propiedades como preocupantes para los niños con comportamiento de pica-tierra, si la propiedad tiene un nivel de arsénico en el suelo igual o mayor a 47 partes por millón (ppm). En base a la información demográfica, alrededor de 300 niños preescolares viven en los 650 hogares, y entre 12 a 60 de éstos podrían tener comportamiento pica-tierra en algún momento durante los años preescolares. Los efectos a la salud que ocurren con mayor probabilidad en niños con comportamiento pica, debido a la ingestión de tierra en una sola ocasión, y dependiendo de la cantidad de contaminación con arsénico en aquellas 650 propiedades y de cuánta tierra ingieran estos niños, incluyen naúsea, retorcijones, vómitos, diarrhea, inflamación de la cara, y dolores de cabeza. Ningún niño de la zona VBI70 ha sido diagnosticado con intoxicación por arsénico, en relación con arsénico en el suelo; sin embargo es posible que no se hayan detectado casos, debido a que los síntomas más comunes en los niños por intoxicación con arsénico (los de vómitos, nausea, etc.), son también síntomas comunes provenientes de otras variedades de causas.

Arsenico en el suelo en algunas de las propiedades también es una preopcupación de salud pública para residentes de largo plazo porque hay potencial para aumento de riesgo de cancer por exposición al arsenico. El riesgo es mas grande para niños que crecieron en patios con altos niveles de arsenico en el suelo y que continuaron a vivir allí como adultos. Si la EPA usa 70 ppm como el nivel para limpiar el arsenico, como 480 propiedades seran afectadas.(3) Se debe notar que desde la primavera de 2001, la EPA ha limpiado como 50 propiedades por causa de niveles elevados de arsenico en el suelo.

El arsénico en la tierra también constituye una preocupación de salud en algunas personas que llevan mucho tiempo residiendo en el hogar, debido al posible aumento de riesgo de cáncer, por exposición al arsénico. El riesgo es más elevado en los niños que se criaron en patios o jardines con altos niveles de arsénico en la tierra y quienes siguen viviendo allí como adultos. La EPA ha identificado algunos 260 terrenos en los cuales el aumento de riesgo de cáncer no es aceptable. Se debe de notar que en la primavera de 2001, la EPA ya había limpiado alrededor de 50 propiedades debido a los niveles de arsénico en el suelo.

Existen algunas dudas al decidir si han podido ocurrir efectos negativos a la salud en la zona VBI70 o no. Las dudas existen en dos áreas: al estimar la cantidad de contaminante a la cual la gente está expuesta (o sea, la dosis), y al determinar cuáles efectos a la salud pueden ocurrir. La incertidumbre al estimar la dosis en los niños que tienen comportamiento pica-tierra se debe a los siguientes factores:

  • estimar el nivel máximo de arsénico en una propiedad basándose en el promedio de los niveles de arsénico en las 3 muestras compuestas,
  • cantidades de tierra variables que comen los niños con comportamiento pica-tierra,
  • variaciones en la frecuencia con que los niños demuestran comportamiento pica,
  • la presuposición de que los niños con comportamiento pica-tierra comen la tierra de la parte del terreno más contaminada, e
  • incertidumbre con respecto al porcentaje de niños con comportamiento pica

Por lo tanto, puede que un niño(a) con comportamiento pica que vive en una propiedad con terreno contaminado con arsénico no se enferme, si la tierra que come proviene de un área de la propiedad con un bajo nivel de arsénico, o si el niño solo come una baja cantidad de tierra, y la cantidad de exposición al arsénico no es suficiente para causar daños a la salud.

También existe incertidumbre al determinar los posibles efectos para la salud debido a la índole inexacta de los siguientes factores:

  • las estimaciones inexactas sobre cuánto del arsénico y del plomo se integra a la corriente sanguínea y a los tejidos, una vez que se ingiera arsénico y plomo incorporados en la tierra,
  • las presuposiciones de que los efectos dañinos observados en la gente expuesta al arsénico a través del agua potable, la cual se absorbe fácilmente en el cuerpo, son semejantes a los efectos dañinos que pueden ocurrir en las personas expuestas al arsénico a través de la tierra, aunque es probable que así se absorba menos al cuerpo,
  • los estimados de la dosis en las investigaciones humanas cuando se ha encontrado arsénico en el agua, y
  • la exactitud de las estimaciones de exposición en las investigaciones humanas que se utilizaron para establecer las normas de salud.

Los Resultados de la ATSDR con respecto al Plomo en la tierra

Algunas propiedades en la zona VBI70 tienen altos niveles de plomo en el suelo, lo cual constituye un peligro a la salud para los niños preescolares que viven allí. La exposición a tierra contaminada con plomo en los terrenos más contaminados, puede incrementar el nivel de plomo en la sangre de algunos niños preescolares, y puede culminar en efectos dañinos al cerebro y al sistema nervioso. Los posibles daños incluyen disminución de la inteligencia, atrasos en el desarrollo, alteración del metabolismo de la vitamina D, cambios en los niveles de enzimas sanguíneas, y disminución de la audición.

La EPA ha desarrollado un modelo matemático que utiliza el promedio de niveles de plomo en el terreno de una propiedad, para pronosticar el porcentaje de niños con niveles de plomo sanguíneo en exceso a la pauta aceptable establecida por el Centro de Control y de Prevención de Enfermedades (el CDC), la cual es 10 microgramos de plomo por decilitro de sangre (µg/dl). En el caso de VBI70, el modelo de la EPA predice un rango de niveles de plomo en la tierra que podría resultar en que más de 5% de los niños tengan niveles de plomo sanguíneos en exceso de 10 µg/dl. El rango de niveles de plomo en la tierra anticipado por el modelo varía debido a quee la EPA ha variado ciertos parámetros en su cálculo (específicamente, la desviación estándar geométrica y la ingestión de plomo de acuerdo a la dieta). El modelo pronosticó que los niveles de plomo en la tierra en un rango de 208 ppm a 540 ppm, constituyen una preocupación por el incremento de niveles de plomo sanguíneo en los niños, dependiendo de cuáles de los parámetros señalen con más exactitud los niveles de plomo sanguíneos. Se debe notar que el promedio de niveles de plomo en el suelo de 78 de los terrenos, es más elevado que 540 ppm, y en alrededor de 1,350 propiedades es más elevado que 208 ppm.(4)

Análisis sanguíneos recientes realizados por el Departamento de Salud Pública y Ambiental en Colorado (el CDPHE) durante el verano del 2000, descubrieron que aproximadamente 10% de los 86 niños preescolares dieron positivo a niveles de plomo sanguíneos en exceso de 10 µg/dl. Sin embargo, no fue posible determinar hasta qué punto la contaminación de plomo en la tierra contribuyó a estos niveles de plomo en la sangre. Los estudios de otros investigadores demuestran que aproximadamente 30% de plomo sanguíneo en los niños es consecuencia del plomo en la tierra. El modelo de plomo sanguíneo de la EPA también predice que una parte significante del plomo en la sangre de un niño proviene de la tierra. El modelo predice que si utilizamos como ejemplo un promedio de 195 ppm de nivel de plomo en la tierra, entonces alrededor de 4.4 µg/dl del plomo sanguíneo en el niño normal, proviene de la tierra. Es importante recordar que es probable el que los niveles de plomo sanguíneos en los niños sean la consecuencia de la exposición a plomo a múltiples fuentes. He aquí algunos ejemplos:

  • plomo en la dieta del niño,
  • plomo en el agua potable,
  • plomo proveniente de pintura con plomo,
  • plomo en la tierra,
  • polvo dentro de la casa,
  • plomo proveniente de cerámica con esmalte de plomo,
  • otras fuentes no identificadas.

El CDPHE tiene un programa a nivel estatal para examinar los niveles de plomo sanguíneo en niños. Para mayor información acerca del programa de plomo sanguíneo del CDPHE, comuníquese con la Srta. Mishelle Macias al 303-692-2622. Además, al Departamento de Salud Ambiental de Denver (DEH), de la Ciudad y el Condado de Denver, le compete responder a cuestiones de plomo. El programa del DEH es dirigido por el Sr. Gene, con quien se puede comunicar al 720-865-5452. El DEH cumple con las normas del CDC, y cuando se le informe de un niño con plomo sanguíneo elevado, el DEH dirigirá una investigación ambiental para identificar las posibles fuentes de plomo. Normalmente, la investigación incluye el recolectar muestras ambientales del entorno doméstico, y el administrar un cuestionario diseñado para identificar las fuentes de plomo. DEH también provee información a las familias sobre cuales son los efectos a la salud del plomo, cómo prevenir la exposición al plomo, la nutrición adecuada, el acceso a otros servicios relacionados, y la necesidad de análisis de sangre posteriores.


La ATSDR está planteando un número de recomendaciones en relación a las preocupaciones de salud pública concernientes a la contaminación de arsénico y plomo en la zona estudiada de la VBI70. Estas recomendaciones incluyen actividades de la ATSDR, tales como el desarrollo de programas de capacitación para proveedores de salud, la educación a la comunidad, e investigaciones de salud. Además, la ATSDR está planteando recomendaciones a las agencias locales, estatales, y federales. Por ejemplo, la ATSDR está recomendando a la EPA que la agencia reduzca la exposición al arsénico en los terrenos en donde los niveles de arsénico y plomo en el suelo son una preocupación para los niños. La ATSDR también le recomienda a la EPA que la agencia recolecte muestras de tierra de los aproximadamente mil restantes terrenos en el área de estudio de la zona VBI70. Todas las recomendaciones de la ATSDR se encuentran en la sección de recomendaciones en este informe.

Junta Pública

En Marzo 2002, ATSDR publicó la Evaluación de Salud Pública para el sitio de VBI70 y actualizó 3 dias de juntas públicas para anunciar los resultados de la evaluación de salud pública. Estas juntas se reunieron en Swansea y Clayton. Además de las presentaciones de representantes de la comunidad y de la ATSDR y la EPA, ATSDR distribuyó hojas informativas que resumen los resultados de la investigación. Estas hojas informativas se hayan en Apendix L y cubren las siguientes temas:

  • Resumen de la Evaluación de Salud Pública
  • Arsenico en el Suelo
  • Plomo en el Suelo, y
  • Actividades de la ATSDR

Comentario Público

Cuando la ATSDR publicó la Evaluación de Salud Pública para el sitio de VBI70 en Marzo de 2002, la Agencia solicitó al público que sometiera comentarios sobre la evaluación. Esos comentarios se han revisado por la ATSDR y cambios se han hecho en el reporte cuando es merecido. La respuesta de la ATSDR alos comentarios públicos se pueden hayar en la Apendix M.

Los planes de la ATSDR para el futuro

La ATSDR está planeando otras actividades en relación a la zona VBI70, además de las investigaciones y las actividades de incorporación comunitaria ya efectuadas por la EPA, el CDPHE, la Ciudad y el Condado de Denver, y la misma ATSDR. Estas actividades incluyen las siguientes:

  • empezar el proyecto de intervenciones en salud ambiental de la agencia, que se enfoca en la educación de salud tanto para los miembros de la comunidad (o residentes) como para los proveedores médicos locales,
  • encuestar a los residentes acerca de el comportamiento pica y de otras actividades que podrían incrementar la exposición al arsénico y al plomo en la tierra,
  • identificar intoxicación aguda por arsénico en los niños y
  • tomar en cuenta otras investigaciones de salud que puedan ser apropiadas para la zona VBI70.

La ATSDR continuará colaborando con, y ayudando a la comunidad, la ciudad y el condado de Denver, el CDPHE, y la EPA, a lo largo de las actividades en la zona VBI70.



The Agency for Toxic Substances and Disease Registry researches and then writes a public health assessment to evaluate a community's exposure to contaminants at hazardous wastes sites, and then decides what public health activities are needed. This evaluation may involve some or all of the following broad categories of public health activities:

  • assessing how people might be exposed to contaminants;
  • evaluating possible health effects from exposure to contaminants for a variety of appropriate public health actions;
  • recommending medical tests, health education, and health promotion;
  • making recommendations to local, state, and federal agencies; and
  • finding ways to involve the community in ATSDR's activities.

This public health assessment describes ATSDR's activities at the Vasquez Boulevard and I-70 (VBI70) site and provides what the Agency's opinion is about the public health impact of contamination at VBI70.

In order to investigate this site, ATSDR established the "VBI70 health team," referred to as the health team. Since January 1999, the health team has met regularly to discuss public health issues related to the VBI70 site. Input from team members has been invaluable to ATSDR, and they have helped the Agency evaluate chemical exposures and decide appropriate public health activities. A list of team members appears in Appendix A.

Public Health Issues

During the investigation of the VBI70 site, ATSDR and the health team identified the following public health issues that would be investigated in the public health assessment process:

  • Is arsenic contamination in soil a threat to the public's health?
  • Is lead contamination in soil a threat to the public's health?
  • Is exposure to other chemicals in the environment (e.g., in the air) a threat to the public's health?
  • Are communities of color (for example African-Americans or Hispanic people) who live at VBI70 at increased risk of harmful effects from lead and arsenic exposure because of their increased exposure or increased sensitivity?


Site Location

The VBI70 site area spans approximately 450 acres in northeast Denver, Colorado (see Appendix B, Figure 1) and includes the area that the U.S. Environmental Protection Agency (EPA) has proposed to the National Priorities List (NPL),(5),(6) EPA proposed adding VBI70 to the NPL on January 19, 1999, thus requiring ATSDR to conduct a public health assessment.

As Figure 1 shows, the study area has an irregular shape, and is located primarily southeast of the interchange of Interstate 25 and Interstate 70. The study area does not extend further east than Colorado Boulevard, further south than Martin Luther King Boulevard, further north than 52nd Avenue, and further west than the Burlington Northern rail tracks west of Interstate 25. Figure 3 shows that the VBI70 study area includes all or part of the following five Denver neighborhoods: Clayton, Cole, Elyria, Southwest Globeville, and Swansea. This area includes a mix of residential, commercial, and industrial sections.

Area History

Based on the information that is summarized in this document, there is evidence of contaminated soil in and around the VBI70 study area. Many industrial activities currently take place in and around the study area. In addition, two smelters used to operate and a third smelter still operates in the area. Some information about these smelters follows:

  • The Omaha-Grant smelter. As Figure 1 shows, the Omaha-Grant smelter was located south of Interstate 70 and west of Brighton Boulevard. The smelter operated at this location from 1882 to 1903. During this time according to government reports, it processed 2,200,000 tons of ore to produce gold, silver, copper, and lead. In 1899, the Omaha-Grant smelter became part of ASARCO, which continued to operate the plant until it closed in 1903. From 1944 to 1950, the Omaha-Grant smelter stack was used intermittently by the City of Denver as a municipal waste incinerator. The City demolished the smelter stack shortly thereafter, covered part of the area with concrete and asphalt, and built the Denver Coliseum on part of the site (Apostolopoulos 1998, ATSDR 1995).

  • The Argo smelter. As Figure 1 also shows, the Argo smelter was located near the current interchange between Interstate 25 and Interstate 70. This smelter operated from 1879 through 1910 to produce gold and silver by roasting copper ore and matte.(7) The Argo smelter no longer exists.

  • The Globe smelter. The Globe smelter is located less than 1 mile north of I-70, between Washington Street and I-25. The smelter first began operating in 1886, and was purchased by ASARCO in 1899. The operations at this facility have changed many times over the years. During the time of operations, the smelter has produced gold, silver, copper, lead, cadmium, arsenic, indium, selenium, antimony, and other metals. Current operations at the Globe plant are different from historic operations. Only a few buildings at the plant are currently in use for the production of bismuth products,(8) litharge,(9) highly purified lead, and tellurium.(10) Small amounts of highly purified "specialty metals" are also produced. Specialty metals produced in the late 1990s include cadmium telluride, cadmium sulfide, lead telluride, zinc telluride, and high purity copper cylinders.

Other industrial activities have also taken place in the study area. Those activities are presented in the Regional Geographic Initiative subsection. In addition to industrial activities, arsenic- containing pesticides (for example, herbicides to control weeds) were frequently used in the U.S. during the 1950s and 1960s. The extent to which these sources or activities have affected VBI70 study area soils has not been determined. EPA is currently investigating possible sources of arsenic and lead contamination.

Demographic Information

As part of its investigation, ATSDR considers the number and makeup of the population in the area surrounding a site. For VBI70, ATSDR reviewed the demographic information for different groups of people in the study area:

  • Demographics of the VBI70 study area: According to 2000 census data, about 17,500 people live in the VBI70 study area–an area with about 5,126 housing units. As shown in Appendix B, Figure 3, the racial composition of the area is diverse:

    • 5,442 (31%) are white,
    • 3,698 (21%) are black,
    • 450 (3 %) are American Indian, Alaskan Native, Asian, or Hawaiian, and
    • 7,952 (45%) report a race other than those listed in the census.

    The "other race" category is so high for the VBI70 study area because many Hispanic people chose this category rather than the other categories of white, black, and other racial groups. In response to a separate question in census, 12,102 people in the study area identified themselves as being of Hispanic origin. Thus, at least 69% of the people in the study area are Hispanic.

    Information on potentially sensitive populations, such as young children and older adults, is also presented in Appendix B, Figure 3. Children 6 years old or younger number about 2,400 (14%) of the population, and approximately 1,480 (8%) residents are 65 years of age or older.

  • Demographics of neighborhoods within the VBI70 study area: Appendix B, Figures 4 through 8, show the same type of population information for each of the five neighborhoods that make up the VBI70 study area: Clayton, Cole, Elyria, Southwest Globeville, and Swansea.

Actions to Reduce Exposure

After sampling soil from about 1,500 yards in 1998, EPA offered to remove soil from the most contaminated properties in the VBI70 study area. This decision was made to prevent residents from coming into contact with soil that contained potentially harmful levels of arsenic and lead. To qualify for time-critical clean-up actions by EPA, properties had to have average soil arsenic levels above 450 parts per million (ppm) or average soil lead levels above 2,000 ppm. Based on the Phase I and II sampling rounds, EPA cleaned up 18 of the 21 properties that had soil contamination above the cleanup levels; owners of the other three properties refused EPA's cleanup offer. As results for Phase III samples came in, EPA removed soil from additional properties that met their criteria.

In response to Phase III soil-sampling results that took place in late 1999 and 2000, EPA has proposed 128 ppm as an preliminary action level for arsenic.(11) About 260 properties in the VBI70 study area exceed this action level. These 260 or so properties have a composite soil sample with arsenic levels greater than 128 ppm. EPA is targeting these approximately 260 properties to protect residents from the risk of cancer. As of summer 2001, about 50 of these properties have been cleaned up by EPA because they met EPA's criteria for time-critical clean-up actions.

During ATSDR's investigations, the agency released a fact sheet about safe gardening practices. The fact sheet included information such as recommendation that residents wash garden produce grown in their yards to reduce the amount of arsenic-contaminated soil that might cling to the produce (See Appendix F for ATSDR's fact sheet on gardening.) In addition, ATSDR released a fact sheet written in English and Spanish that provides easy steps for residents to take to reduce their exposure to arsenic in soil. For example, residents should wash their hands after working or playing outside, wash their dogs periodically, and they should take their shoes off before entering their homes to prevent any arsenic-contaminated soil from being tracked inside. These and other simple steps are described in Appendix I.

Regional Geographic Initiative

In 1989, EPA reported that the Denver zip code 80216, which includes the neighborhoods of Elyria, Southwest Globeville, and Swansea, had the second highest amount of industrial emissions of hazardous pollutants of all Denver zip codes (EPA 1989). The following table shows the amount of chemicals released to the air from industries in the 80216 zip code from 1989 to 2000, the year for which the most recent TRI data are available.

  Pounds of Chemicals Released to the Air
(SIC 20-39)
Pounds of Chemicals Released to the Air
(All Industries)
1989 773,656 Not Available
1990 222,489 Not Available
1991 421,427 Not Available
1992 268,552 Not Available
1993 261,250 Not Available
1994 201,515 Not Available
1995 270,182 Not Available
1996 150,275 Not Available
1997 122,716 Not Available
1998 169,914 510,706
1999 148,888 474,194
2000 129,185 478,536

It should be noted that in 1998, EPA required additional industries to report releases to the environment, referred in the previous table as "add industries." These additional industries were metal mining, coal mining, electric utilities, chemical wholesalers, petroleum bulk terminals, and RCRA/solvent recovery. A list of industries covered by SICs 20 to 39 can be found at EPA's website.

Very few industries are located in the Clayton or Cole neighborhoods (zip codes 80207 and 80205). The combined air emissions for all industries in zip codes 80207 and 80205 in 2000 was 510 pounds, significantly less than air emissions for all industries for zip code 80216. For the adjoining zip code just north of 80205 (that is, zip code 80022 for Commerce City), the total air emissions for all industries in 2000 was 161,511 pounds.

The Cross Community Coalition (CCC), a grassroots community organization located in Swansea, is concerned about these releases and has applied for and received a grant in 1998 from EPA's Regional Geographic Initiative to study local pollution problems. Under this grant, a group of residents, industry representatives, large and small business representatives, a church representative, and staff members from federal, state, and city government have worked together to identify and characterize local sources of pollution and their potential health risks. Some of the CCC's key findings to date follow and are described in detail in Appendix C (Tables C-1 through C-5) and Appendix B (Figure 9):

  • The CCC identified numerous emission sources within zip code 80216, such as mobile sources, bakeries, manufacturing facilities, printers, metal shops, vehicle repair shops, refineries, and a major electric power plant that burns low-sulfur coal. These and other sources have been found to emit large amounts of toxic chemicals, and some emit objectionable odors.

  • The chemicals released by these sources include, but are not limited to, sulfur oxides, nitrogen oxides, carbon monoxide, particulate matter, organic compounds, and other hazardous air pollutants.

  • Four National Priorities List sites are located in or very close to the VBI70 NPL site. These sites are the ASARCO, Inc. Globe Plant, Sand Creek Industrial, Chemical Sales Company, and Broderick Wood Products. Several other industrial operations in the area have special permits from EPA to either generate, transport, or store hazardous waste.

Overall, the findings of the CCC clearly show that many sources of pollution are located in the VBI70 study area. Chemicals released from these sources are likely found at very low levels in many parts of the VBI70 study area, but the exact extent of contamination resulting from these different sources has not been quantified.

ATSDR Activities

ATSDR became involved with the VBI70 site in November 1998. One of the agency's first actions was putting together the VBI70 health team. The health team has planned several public health activities for the site, and has set a schedule for completing them. Some of the activities have been completed. Figure 2 in Appendix B has a timeline regarding the planned ATSDR activities.

ATSDR's public health assessment process–an important activity for the VBI70 site–involves ATSDR evaluating all relevant environmental data, community concerns, and sometimes health outcome data for a site. The information from this first activity is then used to decide what other activities are needed, such as medical testing, health education, and health promotion. The public health assessment for the VBI70 site focuses on evaluating environmental data and community concerns and health education activities that occurred during the investigation. Once decisions are made from evaluating environmental data, community concerns, and medical tests at the VBI70 site, other activities may take place. For instance, as a result of evaluating the extent of arsenic contamination at the site and completing discussions with residents, ATSDR decided to conduct an environmental health intervention project. This involves health education for residents and local health care providers and providing a mechanism for referrals to environmental health clinics for additional medical evaluation, if needed.

ATSDR has released two fact sheets about gardening in the VBI70 study area, has met with residents to discuss gardening issues, has held a national workshop inviting experts to provide advice to the agency about children and adults who eat soil, and will hold public meetings to answer questions concerning this health assessment once it is completed.


Colorado Department of Public Health and Environment

As a follow up to investigations at the nearby ASARCO Globe Plant Site, the Colorado Department of Public Health and Environment (CDPHE), on July 16, 1997, collected 25 soil samples, three surface water samples, and three sediment samples from what is now the VBI70 study area. The samples were analyzed in a lab for levels of inorganic metals, such as, arsenic, cadmium, and lead. The soil samples were collected in Elyria (23 samples) and Swansea (2 samples). Arsenic levels in residential yards were as high as 1,800 parts arsenic for every million parts of soil (ppm) and lead levels as high as 660 ppm (Apostolopoulos 1998; EPA 1998a). The level of arsenic is much higher than expected, which is typically around 7 ppm for Western states (ATSDR 1992). The findings of elevated levels of arsenic and lead prompted EPA to conduct more extensive soil sampling in the five neighborhoods that eventually became the VBI70 study area.

EPA Investigations

The EPA has conducted several environmental investigations at the VBI70 site. This section describes those investigations for this public health assessment.

  • Phase I and II sampling: In the spring and summer of 1998, EPA conducted what is called "Phase I" and "Phase II" soil sampling at the VBI70 site. During these sampling efforts, EPA collected soils from roughly 1,500 properties in the study area. At each property, EPA generally took two samples of surface soils (from the top 2 inches of soil) and one sample of soil from below the surface (from deeper than 6 inches). These sampling efforts identified many properties with potentially high levels of arsenic and lead in soils.

  • Confirmation sampling: Based on the results of Phase I soil samples, EPA returned to 55 properties that had some of the highest levels of contamination in order to collect additional soil samples as a "confirmation sampling." This "confirmation sampling" took place as part of Phase II sampling rounds in the summer and fall of 1998. Most of the samples that were collected are called "five-point composite samples," which means that soils from five different locations on a property were collected and mixed together to make a single, composite sample. The confirmation samples' results were used to decide which properties required immediate cleanup. Out of the 55 properties, 21 met EPA's criteria for cleanup. These 21 properties had average arsenic levels in soil above 450 ppm or average lead levels in soil above 2,000 ppm, or both.

  • Intensive sampling: EPA conducted what is called "intensive sampling" in the summer and fall of 1998 at eight properties. Some of the properties were selected because they had extremely high levels of arsenic or lead in the soils while others were selected because they had low levels of arsenic or lead in the soils. As part of this sampling effort, EPA collected soil samples from the eight selected properties, and from some properties that adjoined the eight focus properties. Using this approach, EPA collected as many as 224 soil samples from each of the eight focus properties. This provided a detailed picture of contamination at those properties. At the adjoining properties, EPA collected soil samples at the shared boundary and up to 15 feet into the neighboring properties. Therefore, the intensive sampling efforts provided information about contamination at the property, at the property line, and into some of the neighboring properties. The results of the intensive sampling effort showed that at properties with elevated arsenic contamination (for example, with several hundred ppm average arsenic levels in soil), the contamination exists throughout the property. One of the intensively sampled properties, however, showed areas of high arsenic contamination and areas of low arsenic contamination.

  • Phase III sampling: In the fall and winter of 1999 and in spring of 2000, EPA conducted "Phase III" sampling. In Phase III sampling. EPA collected surface-soil samples initially from about 1,550 properties. After this initial effort involving Phase III sampling, EPA continued to collect soil samples from another 1,440 or so properties for a total of 2,986 properties. The first 1,550 or so properties sampled were usually properties that were not sampled in Phases I and II. Many of the 1,440 or so properties that were sampled later in Phase III were properties that had been sampled earlier as part of Phases I and II (EPA 1999b).
  • EPA used a different sample design during Phase III to better estimate the average concentration of arsenic and lead at each property. The reason for using an average value to estimate lead and arsenic levels was to better predict long-term exposure. The disadvantage of using an average value, however, it that it usually cannot be used to estimate short-term (or acute) exposures. The new sample design consisted of collecting three composite soil samples from each property, and each composite consisted of 10 individual soil samples. ATSDR used the Phase III data to arrive at most of the conclusions in this public health assessment.

  • Northeast Park Hill sampling: In a separate action not related to the VBI70 NPL site, EPA provided funds and assistance to the National Association of Black Environmentalists (NABE) to investigate arsenic and lead soil contamination in Northeast Park Hill, a neighborhood east of the VBI70 study area. Working with officials from EPA, CDPHE, and the University of Colorado, NABE collected soil samples in March and August 1999 from 36 residential properties in Northeast Park Hill. ATSDR evaluated these data in order to provide recommendations to EPA that would protect the residents of this neighborhood from exposure to arsenic.

  • Other studies: In addition to the many soil sampling studies, EPA has also showed how pigs reacted to contaminated soil. Specifically, EPA fed young pigs arsenic-contaminated soil taken from yards in the VBI70 study area to determine how much arsenic the pigs could absorb from their stomach and intestines into their body. EPA plans to use this information to estimate how much arsenic will be absorbed by people who come into contact with arsenic-contaminated soil (EPA 1999c). In other types of tests, EPA conducted studies on the arsenic and lead in soil to determine the chemical and physical form of arsenic and lead that is present in soil (EPA undated).

Environmental Data Results

EPA provided ATSDR with an electronic database of Phases I, II, and III soil sampling data. Phases I and II contained soil measurements from 1,412 properties and consisted of 4,698 records. EPA provided an additional 442 records of confirmation sampling data along with 1,667 records of intensive sampling data from 8 properties. Phase III data contained surface-soil measurements for 2,986 properties.

During all of EPA's sampling efforts, levels of arsenic, lead, and other metals were measured using what is called an "X-ray fluorescent" (XRF) instrument.  In addition, 10% of the soil samples were measured using a method called "inductively coupled plasma (ICP) spectroscopy." The ICP measurements were used to check to make sure that the XRF measurements were accurate.

Arsenic in the VBI70 Study Area

According to the XRF results from EPA's Phase I and II soil samples, about 500 properties had at least one soil sample with detectable levels of arsenic in soil. The remaining 900 properties had arsenic levels below the XRF instrument's detection limit, which varied from 44 to 57 ppm. These 900 properties either had background levels of arsenic in their soils (probably around 7 parts per million)(12) or low amounts of arsenic up to the detection of 57 ppm.

A better data set to use is Phase III data, which generally has a detection limit of 11 ppm or below and thus provides more information about low as well as high levels of soil arsenic. Phase III data show that many properties in the VBI70 study area have elevated levels of arsenic. Table 1 in the text that follows shows the number of properties sampled in Phase III at different ranges of arsenic levels in soil. The value of 6 ppm is the lowest level reported. The values of 47 and 128 ppm arsenic are listed because they are levels of concern for children with soil-pica behavior and for cancer, respectively. Using the data from Phase III, which represents 2,986 properties, it is possible to know the magnitude of arsenic contamination for selected arsenic levels in soil. Table 1 shows the range of arsenic levels in soil with the corresponding number of properties in those ranges based on Phase III data. Also presented in Table 1 is the estimated number of properties in VBI70 study area that have not been sampled and what would be expected for their arsenic levels in soil. This information is important because it shows that a significant number of properties not test are likely to have arsenic levels in soil that are a health concern for children and adults. For instance, of the 924 properties not tested, about 82 properties are likely to have soil arsenic levels above 128 ppm and about 122 properties will have soil arsenic levels between 47 and 127 ppm.(13)

Table 1.

Number of properties in the VBI70 study area at different soil arsenic levels
Average arsenic level Number of properties based on Phase III samples Estimated number of properties based on the 924 properties not sampled
6 to 46 ppm 2324 720
47 to 127 ppm 394 122
greater than 128 ppm 268 82
# properties 2986 924

Legend: parts per million (ppm).

To protect children from exposure to arsenic, it is important to know the highest (or maximum) arsenic level in a yard. The maximum arsenic level can be estimated using information from the eight intensively sampled properties. Graph 1 shows a plot of the average arsenic level and the corresponding maximum arsenic level for the eight intensively sampled properties. The graph shows that as the average arsenic level in a yard increases, the maximum arsenic level in the yard increases in a very predictable way. A very good straight line relationship (as indicated by an r squared value of 0.95(14)) exists between the average soil arsenic level in a yard and the corresponding maximum soil arsenic level for that yard. This straight line relationship allows ATSDR to use Phase III data and the average arsenic level in a property (that is, the average of the three composite samples from a property) to estimate the maximum arsenic level for that yard at a single (discrete) location. The regression formula predicted by the slope is y = 6.399x. In other words, this means that the estimated maximum arsenic level in a yard equals 6.399 times the average of three composite samples for that yard. It should be pointed out, however, that this is an estimated maximum because the formula is based on only 8 properties, and it is assumed that the other properties have the same contamination patterns as these 8 properties.

Here is an example for calculating the maximum arsenic level in soil when only the average arsenic level in soil is know. As an example, let us use the highest average arsenic level from Phase III of 759 ppm arsenic in soil. This average of the three composite samples can be multiplied by 6.399 to get an estimated maximum discrete arsenic level of 4,856 ppm for that property. Similarly, a property with an average arsenic level of 200 ppm is likely to have an estimated maximum arsenic level in soil of 1,280 ppm. Table 2 shows a range of average arsenic level from various yards and the corresponding estimate of the maximum arsenic level. It is important to note that because only composite samples were collected during Phase III, any mathematical method used to estimate the maximum arsenic level in a property ha some uncertainty associated with that estimate. The true maximum arsenic level might higher or lower. It is reasonable to assume, however, that because of the excellent correlation coefficient that exists using the intensively sampled data, the formula y = 6.399x is probably very close to the true maximum arsenic level.

Graph 1

Mathematical analysis of average arsenic level and the corresponding maximum arsenic level in a property

Environmental investigations by the Colorado Department of Public Health and Environment

Before EPA started collecting soil samples from the VBI70 area, the Colorado Department of Public Health and Environment collected 25 soil samples from Elyria and Swansea in July 1997. Out of 25 properties tested, 12 had elevated arsenic levels, with the highest level being 1,800 ppm (Apostolopoulos 1998).

Table 2.

Estimated maximum discrete arsenic levels
Average arsenic level parts per million (ppm) Estimated maximum discrete arsenic level using the regression method ppm
759 4857
600 3839
500 3199
400 2559
300 1919
200 1280
150 959
100 640
50 320
30 192
20 128
15 96

Arsenic in the Northeast Park Hill neighborhood

EPA assisted the National Association of Black Environmentalists in collecting two to four individual (discrete) soil samples from 36 residential properties in the Northeast Park Hill neighborhood. Even though the number of samples from each yard is limited, the results show that some properties have elevated arsenic levels in soil. The maximum arsenic level from each property is shown in Table 3.

The limited number of soil samples does not allow ATSDR to evaluate the long-term exposures that might occur from arsenic in soil in this community. However, the high levels of arsenic in some yards can be used to determine the public health implications for arsenic exposure in children who eat soil.


Table 3.

Maximum measured arsenic levels in soil at properties in the Northeast Park Hill neighborhood
Address Arsenic level Address Arsenic level Address Arsenic level
Glencoe 1,010 Albion 143 Glencoe 20
Eudora 737 Monaco 78 Glencoe 20
Thrill Place 724 Eudora 58 Glencoe 18
Bellaire 688 Albion 36 Elm 17
Elm 658 Glencoe 36 Thrill Place 17
Dahlia 619 Leyden 32 Fairfax 16
Albion 549 Bellaire 27 Holly 15
Oneida 348 Dexter 26 Glencoe 14
Ash 280 Krameria 26 Holly 12
Hudson 172 Kemey 20 36th Avenue 12
Glencoe 155 Hudson 20 Wheeling 10
Newport 149 Pontiac 20 36th Avenue 10


EPA's X-ray fluorescent results from Phase I and II soil samples showed that most of the properties contained detectable levels of lead. The typical detection limit for the XRF instruments used by EPA was about 30 ppm, meaning that the instrument usually could not detect lead below that level. That detection limit is close to 20 ppm, the typical background level of lead in naturally occurring soils in the western United States (ATSDR 1992.) It is not unusual, however, for soil in urban and suburban areas to be contaminated with lead at several hundred parts per million. Much of this contamination is due to lead fallout from the past use of leaded gasoline in cars as well as from other sources such as exterior lead-based paint.

Phase III sample results showed that 276 properties had average soil-lead levels above 400 ppm. The highest average lead level in soil based on averaging the three composite samples from a property is 1,131 ppm of lead.

ATSDR has notified Ms. Mishelle Macias with the CDPHE lead poisoning prevention program of the lead levels in soil in the VBI70 study area. Because of the interactions between CDPHE and ATSDR, CDPHE offered blood-lead testing to preschool children in the VBI70 study area through their lead poisoning prevention program. The results of CDPHE's blood lead investigation are described in the Discussion of Health Outcome Data Section.


During Phase I and II sampling, EPA planned to use XRF instruments to measure cadmium levels in soil. These measurements, however, were often found to be inaccurate. As a result of this problem, EPA has reported that the XRF cadmium measurements from the Phase I and II sampling are not valid.

Although the XRF measurements of cadmium were unsuccessful, EPA sent 363 soil samples from their Phase I investigation to a laboratory for chemical analyses. Those results, which were found to be accurate and valid, show the average cadmium levels from the VBI70 study area to be 5 ppm in surface soil and 5.6 ppm in subsurface soil. The highest level reported was 37 ppm, from a subsurface soil sample (EPA 1998b). These levels are higher than what has been reported as the background level of cadmium in naturally occurring western soil, 0.07 to 1.1 ppm (Kabata-Pendias and Pendias 1984).(15)

While soil cadmium levels appear to be higher than background levels, the level of cadmium in soil will not cause harmful effects to people in the VBI70 study area. The estimated amount of exposure in adults and children from contact with soil is below ATSDR's chronic oral Minimal Risk Level (MRL) for cadmium and below EPA's chronic Reference Dose (RfD) for cadmium.(16) For these reasons, this health assessment report contains no further evaluation of the possibility of harmful effects from cadmium in soil.

Other Contaminants of Concern

Soil contains inorganic metals with a range of naturally occurring levels. Pollution from industrial sources and other types of activity can increase the level of metals in soil. During Phase I, EPA analyzed 44 soil samples for the metals that are most commonly found in soil. Most of these samples came from Swansea and Elyria. Except for arsenic, lead, cadmium, and zinc, the levels of inorganic metals in the 44 soil samples from VBI70 are similar to the levels that are found in soil throughout the western U.S.


The levels of zinc in surface soil at VBI70 ranged from 84 to 1,600 ppm, with an averagelevel of 629 ppm; and the levels of zinc in subsurface soil ranged from 84 to 3,300 ppm,with an average of 406 ppm. These levels are considerably higher than the average levelof 65 ppm zinc that occurs naturally in soils in the western U.S. In an earlier siteinvestigation at the ASARCO Globe Plant Site, zinc was also found at elevated levels insoil.

The levels of zinc in soil in the 44 samples are not high enough to cause harmful effectsin people. The estimated amount of exposure to zinc for children and adults from contactwith soil is below ATSDR's chronic oral MRL for zinc and EPA's chronic RfD for zinc. In addition, zinc is a nutrient, or an essential element for humans. The National Academyof Sciences has recommended that the American diet contain 10 to15 milligrams of zincper day. For these reasons, this health assessment report contains no further evaluation of the possibility of harmful effects from zinc in soil.


Thallium is another naturally occurring metal in soil. In EPA's Phase I investigation,thallium was detected in the study area at an average level of 13 ppm in surface-soilsamples and 15 ppm in subsurface soil samples. Subsequent analysis by EPA using twoother methods showed that thallium levels in soil were below 1 ppm and that the XRFinstrument was probably overestimating thallium levels in soil. The thallium levelsmeasured by the other two chemical methods are similar to background levels of thalliumin naturally occurring soil (ATSDR 1992). These background levels of thallium are notharmful to people. Therefore, thallium will not be evaluated further in this public health assessment.

Adequacy of the Data

ATSDR first reviewed the soil sampling data from different environmental investigations todetermine whether the soil sampling data were adequate for making public health decisions. Below is a summary of ATSDR's review of the adequacy of the surface soil sampling data forthe VBI70 site.

Phase I and II Samples

During Phase I and II sampling rounds, EPA collected at least three soil samples (two surface andone subsurface) at every property that was considered. The two surface-soil samplescharacterized levels of contamination at two particular points at each property. They might notprovide an accurate measure of property-wide levels of contamination, especially at residenceswhere levels of contamination changed significantly across the property.

In fact, comparison of the Phase I and II data with the more extensive data collected during theconfirmation and intensive sampling has showed some significant differences in the levels ofcontamination at selected properties. This suggests that the Phase I and II sampling did notprovide a complete account of soil contamination in some cases.

Intensive and Confirmation Samples

EPA's intensive and confirmation sampling efforts measured contamination in soils at severallocations on a property, instead of measuring contamination at just one or two locations. Therefore, these sampling efforts provide a more accurate account of contamination at the VBI70site. As a result, some of ATSDR's public health decisions for this site were drawn using theintensive and confirmation sampling, and less so from the Phase I and II sampling. More detailson this decision follow:

  • Intensive sampling: During EPA's intensive sampling, soils were collected at 5-footintervals at eight residential properties in the VBI70 study area. Additionally, EPAcollected soil samples, when possible, as far as 15 feet into the properties that adjoin theeight intensively sampled residential properties. The purpose of the intensive samplingeffort was to characterize the distribution of arsenic and lead in both contaminated andnon-contaminated yards, and in their adjoining properties. Between 89 and 224 soilsamples were collected at each of the eight properties that were included in the intensivesampling effort. Therefore, the intensive soil sampling data are sufficient to make publichealth decisions for these properties. The limited sampling of adjoining properties doesnot provide sufficient information to characterize long-term exposure but does providelimited information when evaluating very short-term exposure in children.

  • Confirmation sampling: At 55 properties, the EPA collected confirmation samples. Asnoted earlier, most of the confirmation samples were actually "five-point compositesamples," in which soils from five locations were gathered and analyzed as one sample. A composite sample was collected from the back yard and front yard of every propertyconsidered in this sampling, and discrete soil samples were collected in selected sideyards and gardens. Therefore, the confirmation sampling effort characterized levels ofcontamination at many locations at each property, thus providing a useful indicator of theproperty-wide contamination. Figures 21 and 22 in Appendix B show examples ofintensively sampled properties.

  • Intensive versus confirmation sampling: It should be noted that five properties were partof both the confirmation and intensive sampling efforts, enabling ATSDR to compare theresults for these two sampling approaches. In general, the two sampling schemes providesomewhat similar results. There are some exceptions, which show that the intensivesampling scheme provides a better estimate of property-wide contamination than the 5-point composite scheme. Because five-point composites might miss significant areas ofcontamination, the confirmation results are less reliable in making public health decisions compared to the intensive sampling.

Phase III Soil Samples

Phase III soil sample results were generated by collecting three 10-point composite samples fromeach property. This sample design will be sufficient for making public health decisions aboutlong-term exposure to contaminants in soil. Because ATSDR was able to estimate the maximumarsenic level in soil from the average level determined by the composite samples, the agency willalso use the Phase III data to determine the health hazard of very short-term exposure to arsenicin soil. In ATSDR's discussions with EPA officials in the development of Phase III sampledesign, EPA agreed to use Phase III data to decide which properties should be resampled because of possible high areas of arsenic contamination.

Air Data

The Aerometric Information Retrieval System (AIRS) is a publicly accessible database ofinformation about air pollution in the United States. EPA has many uses for AIRS, but thedatabase's main use is to track changes in air quality across the country. The information inAIRS comes primarily from states which are required to submit air quality measurements to EPAfor certain pollutants. The state of Colorado routinely provides summaries of its air qualitymeasurements to EPA, and these results are then loaded in AIRS. Currently, AIRS has extensiveair quality data for more than 20 ambient air monitoring stations throughout the Denvermetropolitan area, thus providing extensive information about this area's air quality for certainpollutants.

According to EPA, air quality throughout the Denver metropolitan area currently meets thefederal criteria for lead, nitrogen dioxide, and sulfur dioxide. In the past, Denver County andparts of Adams County were not in attainment with EPA's National Ambient Air QualityStandard for particulate matter and carbon monoxide. ATSDR has been informed by staff fromthe Denver Department of Environmental Health that Denver currently meets EPA's AmbientAir Quality Standard for particulate matter and carbon monoxide.

Prior to 1998, Denver County and parts of Adams County were not in attainment with the 1-hourozone standard. ATSDR has been informed by staff from the Denver Department ofEnvironmental Health that Denver currently meets EPA's Ambient Air Quality Standard forparticulate matter and carbon monoxide.


Completed Exposure Pathways

One of ATSDR's first goals is to identify "exposure pathways." Exposure pathways are differentways that contaminants move in the environment and the different ways that people can comeinto contact with chemicals, such as breathing them in (inhalation) or accidentally drinking oreating them (ingestion). A "completed exposure pathway" exists when information shows thatpeople have come into contact with a contaminant in soil, water, or air. Completed exposurepathways can be either in the past, the present, or could be in the future. ATSDR has identifiedtwo completed exposure pathways for the VBI70 site, as described below.

Soil Ingestion in children and adults

The most significant exposure pathway at the VBI70 site is accidental ingestion (that is,swallowing) of contaminated soil and household dust by both children and adults. This exposureoccurs when people have direct contact with soils in their environment. For instance, whenchildren play outside or crawl on floors or when adults work in yards and gardens, contaminatedsoil or dust particles cling to their hands. Residents can then accidentally swallow thecontaminants when they put their hands on or into their mouths, as children often do. Since bothpeople and pets track contaminated soils from outdoors into their homes, exposures can occurwhile people are in their homes and while they are in their yards. Factors that affect whether ornot people have contact with contaminated soil include:

  • grass cover, which is likely to reduce contact with contaminated soil when grass cover isthick but increase contact with soil when grass cover is sparse or bare ground is present,
  • weather conditions, which is likely to reduce contact with outside soil during cold monthsbecause people tend to stay indoors more often,
  • the amount of time someone spends outside playing or gardening, and
  • people's personal habits when outside, for instance, children whose play activities involve playing in the dirt are likely to have greater exposure than other children.

Unless contaminants are removed from properties, some residents will be exposed tocontaminated soils and dust as long as they live in the VBI70 study area. The contaminants ofconcern for soil ingestion at the VBI70 site are arsenic and lead.

The amount of chemicals that people are exposed to via ingestion depends on many factors, suchas the levels of contamination at their homes and the type of activities while at home. Thehighest amount of exposure to levels of soil contamination is expected to occur among peoplewho spend time at their homes. The people who live at these homes will have the mostexposure, and neighbors who are visiting these homes can be exposed to a lesser degree. Forreasons described below, preschool children, whether they live at homes with contaminated soilsor who frequently visit homes with contaminated soil, have the highest potential for exposure. On the other hand, adults and older children who visit houses with contaminated yards probablyhave much less exposure because they put their hands on or into their mouths less frequently thanpreschool children.

Another factor that greatly affects people's exposure is the amount of soils they accidentallyingest on a daily basis. Though people might not be aware of this, everyone ingests some soil ordust every day, but some people tend to swallow more soil or dust than others. Preschoolchildren, on average, swallow more soil and dust than people in any other age group. This isbecause some preschoolers often have close contact with soil and dust when they play andbecause they tend to engage frequently in hand-to-mouth activity. Children in elementary school,teenagers, and adults are also exposed to dusts and soils, but generally in much smaller amounts.


When evaluating exposures, ATSDR also considered a wide range of human activities that mightincrease exposure to arsenic and lead in soil. One activity that may increase concern, particularlyin preschool children, is a behavior called "soil-pica", the eating or ingestion of large amounts ofsoil. This behavior occurs in some preschool children as part of their normal exploratorybehavior for 1- and 2-year-old children, as part of an intentional behavior in older preschoolchildren (3 to 6 year-olds), or accidentally as they eat food dropped on the ground that picks upsoil particles. The reasons why some children engage in soil-pica behavior is not known. Scientists suspect that soil-pica behavior has something to do with nutritional deficiencies,psychological needs, and cultural factors (Danford 1982), but none of these links have beenproven or shown to be responsible for all soil-pica behavior. Soil-pica behavior is most likely tooccur in preschool children, but it can occur in older children and even in adults. The exactnumber of children who go through a stage of pica behavior is not known. Studies have reportedthat this behavior occurs in as few as 4% of children or in as many as 21% of children (Barltrop1966, Robischon 1971, Shellshear 1975, Vermeer and Frate 1979).(17) Using statistics, twoscientists have estimated as many as 33% of preschool children will have soil-pica behavior onceor twice during their preschool years (Calabrese and Stanek 1998). They admit, however, thattheir 33 percent may overestimate the percentage of children who engage in 1 to 2 days of soil-pica behavior (Calabrese and Stanek 1993, Danford 1982, EPA 1997). The percentage ofchildren in the VBI70 study area with soil-pica behavior is unknown.

Studies on children have shown that soil-pica children eat varying amounts of soil ranging from 600 mg to 5,000 or more milligrams (about 1/8 teaspoon to 1 teaspoon) (Stanek and Calabrese 2000, Calabrese and Stanek 1993, Calabrese et al. 1989, Wong 1988). Because of the limited number of such studies, some uncertainty exists in deciding what amount of soil intake should be used for soil-pica children. Therefore, for this public health assessment, ATSDR will use a range of soil intakes from 600 to 5,000 milligrams soil to estimate exposure for soil-pica children.

Limited information is also available for how often (i.e., frequency) and how long (i.e., duration) soil-pica children will have this behavior. Some preschool children might eat soil once during their preschool year while others might go through a stage of eating soil several times during a week or over several months. It is reasonable to assume that soil-pica behavior might occur for several days in a row, or a child might skip days between eating soil (Calabrese and Stanek 1998; Calabrese and Stanek 1993; Wong 1989, ATSDR 1992.) In addition, general pica behavior is greatest in 1- and 2-year-old children and decreases as children age during their preschool years (Barltrop 1966).

Since the Denver climate is relatively dry, especially in the summer, many homes have yards with exposed soil and little grass cover. Some children with soil-pica behavior who live in arsenic-contaminated properties could easily have direct access to contaminated soils because of bare spots in the yard. Moreover, since winters in Denver are generally cold, soil-pica behavior is less likely to occur during this time and probably most likely to occur during the warmer summer months, when preschool children are most likely to play outside. No studies have been conducted to determine the percentage of children in the VBI70 study area that have soil-pica behavior. As part of future activities, ATSDR is considering developing a questionnaire that would identify soil-pica behavior in children.

Studies have also shown that adults will engage in soil-pica behavior, which is often referred to as geophagy, the eating of earth. Geophagy in adults results largely as a cultural practice in the US and usually involves eating clay. Quite often pregnant women will eat clay because they believe it is beneficial. Another practice that is becoming more common in the US is the commercial marketing of bentonite clays to people who believe it cleanses the gastrointestinal tract. Geophagy has been shown to occur in African-Americans, Whites, and Hispanics. For instance, a study was conducted of culturally transmitted clay eating in African-American children and adults who lived in Holmes County, Mississippi. In this case, families used designated areas to dig clay and processed the clay using heat before eating it. What is of note in this practice is that culturally instilled geophagia involves eating processed clay that is taken from below the surface and is usually not from the person's yard. (Vermeer and Frate 1979, Reid 1992, Grigsby et al. 1999, ATSDR 2001).

In addition to people with soil-pica behavior, some workers in the VBI70 study area might accidentally come into contact with contaminated soils. As an example, contractors and utility workers might work on job sites with contaminated soils. If these workers were to get arsenic-contaminated soils on their hands, and then engage in hand-to-mouth activity, they too could be exposed to the contaminants in the area.

During our work with community members, ATSDR was told about incidents of both children and adults eating soil. In one case, a community representative reported to ATSDR that a young woman who lived in the VBI70 study area ate soil from her yard. When questioned about why she ate soil, the woman reported that she did so periodically because she liked it. In another incident, a city official reported to ATSDR that a grandmother made mud pies from yard soil and fed them to her grandchildren. These incidents were described as culturally appropriate among some groups in the area. Although they cannot be used to determine the percentage of people in the VBI70 area that practice soil-pica, they are important clues that this behavior might be present.

Soil-pica workshop

As part of the Agency's efforts to reduce the hazard of soil-pica behavior, ATSDR invited national experts to a soil-pica workshop on June 7 and 8, 2000. The purpose of the workshop was to seek advice about soil-pica behavior to help ATSDR in making public health decisions. The panelist reached the following key findings during the workshop:

  • The panelists agreed that soil-pica does exist.
  • The panelists agreed that the percentage of soil-pica behavior at given soil intake rates is poorly defined.
  • The panelists agreed that more research is needed to understand the percentage of children with soil-pica behavior and the amount of soil that soil-pica children ingest.
  • The panelists agreed that while very few studies are available, ATSDR should continue to use 5,000 mg as an estimate of soil intake for soil-pica children.
  • The panelists agreed that ATSDR should continue to evaluate the public health significance of soil-pica behavior.

ATSDR considered the advice of the expert panel in evaluating the potential for soil-pica behavior at the VBI70 site. The advice and recommendations of the panelists are reported in Summary Report for the ATSDR Soil-pica Workshop (ATSDR 2001).

Eating home-grown produce

Eating fruits, vegetables, herbs, or other produce grown locally in gardens with contaminated soil can cause exposure. This type of exposure occurs because many plants slowly absorb small amounts of the chemicals that are found in soils or because contaminated soil can adhere to the exterior surface of produce. Some of these absorbed chemicals are essential nutrients and are actually good for humans to eat, but other chemicals can present health hazards if they are found at high enough levels and are consumed on a regular basis. ATSDR and CDPHE evaluated the potential exposure from eating home-grown produce.

When reviewing this exposure pathway, ATSDR focused the evaluation on levels of arsenic in produce. The other contaminants in the VBI70 soils are either far less likely to be absorbed by plants (e.g., lead) or are much less toxic than arsenic (e.g., zinc). Using a method developed by EPA (EPA 1995b) and advice from the U.S. Department of Agriculture, ATSDR and CDPHE estimated the amount of arsenic that residents in the VBI70 study area would ingest if 30% of the produce that they ate came from their home garden. However, this analysis found that the amount of arsenic that people might ingest by eating home-grown produce is far below the amounts that are known to cause harmful effects.

Residents in the VBI70 study area have recently received two important fact sheets with public health information about eating home-grown produce. In April 1999, while ATSDR and the Colorado Department of Public Health and Environment (CDPHE) were evaluating specific health risks for the VBI70 site, CDPHE published and released the first fact sheet, which described how garden produce can absorb soil contaminants and explained how residents can protect themselves from these contaminants. A copy of this fact sheet is presented in Appendix E. In August 1999, after the public health agencies finished evaluating the risks of eating home grown produce, ATSDR published and released a fact sheet which informed residents that it was safe to eat fruits and vegetables from their home gardens, because the amount of arsenic that these plants absorb is likely far below levels that might harm the people who eat their produce.

Possible (Potential) Exposure Pathways

When important information about an exposure pathway is missing or incomplete, ATSDR classifies it as a possible (or potential) exposure pathway. In these cases, not enough information is available to conduct detailed analyses of the amount of exposures to contaminants in areas where people live, work, and play. ATSDR has identified three potential exposure pathways for the VBI70 site. The following discussion identifies these pathways and the missing information.

Ingesting or Touching Sediment and Surface Water

Rain water and snow melt can carry contaminants from the air and surface soil into local "surface waters," such as drainage ditches, creeks, streams, and rivers; some of the contaminants can then settle into the sediments at the bottom of these locations. People who play or work in these areas, in turn, can accidentally come into contact, or even swallow, small amounts of the contaminants in the water and sediments. Recognizing this route of exposure, ATSDR gathered and reviewed information on contamination in sediments and surface waters in the VBI70 study area, as described below.


ATSDR identified two studies study that measured levels of contamination in the sediments of local surface waters. One study was conducted in 1997 by CDPHE and focused on the sediments and surface water of the South Platte River–the main water way that flows through the VBI70 study area. During this study, three sediment samples were collected: one from where the river flows beneath I-70 (near the Denver Coliseum), one approximately one-half mile upstream from this location, and one approximately one-half mile downstream from this location (Apostolopoulos 1998). The samples were analyzed for concentrations of metals, including arsenic, lead, and cadmium. While these metals were detected in sediment, none were found at unusually high levels. Another study was conducted as part of the ASARCO Globe Superfund site. This study measured arsenic and lead levels in that portion of the South Platte River near the ASARCO Globe Plant. Arsenic levels in sediment ranged from 2 to 26 ppm, and lead levels in sediment ranged from 28 to 245 ppm. These levels are not a health threat. Some limited data on possible metal contamination are also available from the City and County of Denver where streams empty into the South Platte River.

Since sediment and surface water samples are not available from drainage ways in the 5 affected neighborhoods, it is not possible to determine if these areas are safe for children to play in.

Surface Water

ATSDR has identified one study–CDPHE's 1997 study, which was described above—that measured levels of contamination in the surface waters in the VBI70 study area. During this study, CDPHE collected three surface-water samples from the South Platte River, in the same locations where sediment were sampled (see previous subsection). The City and County of Denver also has a program that measures metal contamination where surface water that drains into the South Platte River.

The residents in the VBI70 study area do not come into contact regularly with water in the South Platte River. For exposure to occur, people would have to swim or wade in the South Platte River—an activity that presumably occurs only during the warmer summer months, and then probably rarely. Since arsenic and lead, the contaminants of concern at this site, do not readily pass through skin, wading in the river will likely not result in any exposure to these chemicals. To be exposed to the chemicals in the water, residents would have to swallow river water accidentally, but the likelihood of this happening is extremely low. Therefore, significant exposure to arsenic and lead detected in the South Platte River seems unlikely.

If residents come into contact with surface water in drainage ditches, streams, and puddles in the VBI70 study area, simply coming into contact with these surface waters would not result in exposure, unless the residents actually drank from these waters, which seems highly unlikely. It is unlikely that surface water could be a significant route of exposure for people who live in the VBI70 study area.

Drinking groundwater

The groundwater beneath the VBI70 study area has not been tested. EPA has stated that it plans to investigate potential groundwater contamination in this area at a later date. Until then, however, levels of contamination in the groundwater are not known, preventing ATSDR from fully evaluating the public health significance of potential groundwater contamination.

ATSDR notes, however, drinking water at all residences in the VBI70 study area is drawn from surface waters from the nearby Rocky Mountains. Therefore, even if the groundwater beneath the VBI70 study area were contaminated, it is highly unlikely that residents would ever drink the contaminated groundwater. Nonetheless, ATSDR will evaluate the public health significance of groundwater contamination, if evidence of contamination becomes available.

Breathing outdoor and indoor air

The contaminated soils and dusts in the VBI70 study area can become airborne by various processes. For example, high winds can blow fine soil and dust particles into the air, as can cars driving on roadways covered in small amounts of dust and dirt. Because the Denver area has a relatively dry climate and heavy traffic, dusts and surface soils can become airborne more easily in the VBI70 study area than in other parts of the country. These airborne contaminants can enter homes through open doors, open windows, and air intake vents. Unfortunately, ATSDR cannot evaluate the amount of contaminants in the outdoor or indoor air in the VBI70 study area because the appropriate air monitoring data measuring arsenic and lead in air are not available for this part of Denver.

As previously noted, no agency has collected indoor or outdoor air samples in the VBI70 study area that characterizes air quality throughout the site. After the EPA finishes investigating levels of contamination in soils, they will decide whether a follow-up investigation of air pollution in the study area is necessary. ATSDR will review data generated by such studies, if they are conducted.

Arsenic and Lead Patterns in the VBI70 Study Area

As previously discussed, ATSDR has found that soil contaminated with arsenic and lead present the greatest public health hazard at the VBI70 site. Focusing on the contaminated soils, ATSDR has evaluated how levels of lead and arsenic vary from one location to the next in the study area. This evaluation was necessary to determine whether the previous EPA soil sampling studies were sufficient and whether the studies should be expanded to consider soil contamination in other nearby areas. The following discussion reviews ATSDR's findings about levels of soil contamination throughout the VBI70 study area.

Lead distribution throughout the study area

The Phase III sampling data, which includes results for 2,986 properties, provides an excellent account of how levels of lead in surface soils vary throughout the VBI70 study area. As an example, Figure 10 in Appendix B illustrates the sampling results by showing the locations with the higher lead concentrations which are symbolized by the darker circles and the locations with the lower lead concentrations as the lighter circles. The higher levels of lead in soils (or the darker circles) occur more frequently in the Elyria, Cole, and Southwest Globe neighborhoods and the lower levels of lead in soils (or the lighter circles) occur more frequently in the Swansea and Clayton neighborhoods. In other words, the levels of lead in surface soils appear to increase as one travels west in the VBI70 study area.

Figure 11 in Appendix B presents a similar account of the Phase III sampling results, by looking at the distribution of lead levels below and above 400 ppm. As Figure 11 shows, the same data trend is apparent–higher levels of lead in the western portion of the study area than those in the eastern portion. The same pattern appeared when the ranges were set at 101 to 190 ppm and 191 to 282 ppm. Therefore, the somewhat arbitrary choice of concentration ranges in the figures appears to have no bearing on the data trend. The same trend is also seen when plotting Phases1 and II data.

One possible explanation for this trend is that fallout from one of several nearby smelters has raised lead levels in surface soil. Other possible explanations also exist. For instance, the western portion of the site is closer to the intersection of Interstate 70 and Interstate 25 and so fallout from leaded gasoline might have increased lead levels in nearby yards. Another possible explanation is that the homes in the western portion of the site are older and therefore more likely to have lead paint on the exterior. The implication is that years of weathering and chipping paint have contaminated the yards. This explanation seems unlikely since the percentage of older homes (for instance, homes built before 1950 when high amounts of lead were commonly added to paints) is very similar in Elyria, Cole, and Southwest Globeville compared to Swansea and Clayton (see Appendix D, Table D-1). ATSDR identified other data trends that deserve mention:

  • The five highest soil concentrations of lead observed in the VBI70 study area during the Phase I and II sampling occurred at three properties located within 1,000 feet of the former Omaha-Grant smelter. Four of the five highest levels came from samples taken below the surface (see Appendix B, Figure 12). This trend indicates that significant lead contamination in subsurface soils might occur near the former smelter.

  • Surface-soil sampling data from the Globeville community showed a distinctive north-south trend in soil-lead concentrations, in addition to the east-west trend discussed above for the VBI70 study area. As Figure 13 in Appendix B shows, relatively lower lead levels were found in the northern portions of Globeville, while relatively higher lead levels were found in the southern areas.

  • An interesting observation is that the variations in zinc concentrations in surface soils (see Figure 14 in Appendix B) throughout the study area are quite similar to the variations in lead concentrations (see Figures 10 and 11 in Appendix B).

  • The maps show that the industrial area near the center of the VBI70 study area has not been extensively sampled. Therefore, levels of contamination in this area remain unknown.

In review, the trends depicted in Figures 10, 11, and 13 indicate that significant soil-lead contamination is likely to exist south and west of the VBI70 study area. In other words, significant lead contamination might exist south of Martin Luther King Boulevard/Blake Street and west of Fox Street/Burlington Northern Railroad, though these soils have not been tested by EPA's sampling efforts. In addition, significant soil lead contamination may exist in the industrial area in the center of the VBI70 study area.

It should also be pointed out that while the frequency of elevated lead in soil is greater for western portions of the site, about 85 properties in the eastern portion of the site (that is, Swansea and Clayton) have significant lead contamination in soil. In addition, limited soil samples from Northeast Park Hill, a neighborhood east of the VBI70 site, show significant lead contamination at some properties.

Lead distribution at individual properties

Because of EPA's intensive sampling effort at eight properties in the study area, it is possible to evaluate lead distribution patterns at those properties and in some of the adjoining yards that EPA also sampled. The properties with high levels of lead generally show consistently elevated lead levels throughout the yards while lead levels in adjoining properties drop off significantly.

There appears to be, however, some migration of lead onto adjoining properties (see Appendix B, Figure 15.) This pattern of high lead levels dropping off at the property boundary is similar to the pattern for properties with high arsenic levels and may point to a similar mechanism by which lead got into the soil at properties that are highly contaminated.

A comparison of lead and arsenic levels in soil for the eight intensively sampled properties showed that as lead levels go up, arsenic levels tend to go up also. This correlation is shown in Graph 2 that follows, but it is not apparent when looking at Phase III data. ATSDR is unsure of the significance of this finding and wants to point it out as information for the other agencies to investigate.

Arsenic distribution in the study area

As part of ATSDR's normal evaluation process, the agency also evaluated how soil arsenic concentrations vary from location to location in the VBI70 study area. When looking at a plot of Phase III data, no obvious patterns in arsenic concentrations were apparent when looking at high arsenic levels (see Figure 16 in Appendix B). The high arsenic levels appeared to be scattered randomly around the study area. However, a pattern exists for low levels of arsenic.

Figure 17, Appendix B shows the distribution of arsenic levels for 0-11 ppm and 12 to 29 ppm, which is similar to the lead pattern described previously. That is, at these lower levels of arsenic, samples in the range of 12 to 29 ppm are found more frequently in the western portion of the study area, that is Cole, Elyria, and Southwest Globeville. Figure 18, Appendix B shows a similar pattern when looking at the ranges of 12 to 17 ppm and 18 to 23 ppm. These levels of arsenic in soil are unlikely to cause harmful effects in people at the VBI70 site. ATSDR points out this distribution pattern because it might point to another source of arsenic in the neighborhoods in addition to arsenic-based pesticides.

Graph 2

Comparison of average arsenic levels to average lead levels for the eight intensively sampled properties

Comparison of average arsenic levels to average lead levels for the eight intensively sampled properties

The number of samples collected from a neighborhood might visually influence what is seen when plotting the data in Figures 17 and 18. To address this potential problem, ATSDR determined the sampling density for Phase III (see Appendix B, Figure 19.) Figure 19 shows some minor differences in the number of soil samples collected from each neighborhood. For instance, it appears that based on sampling density, fewer samples were collected from Clayton. To ensure the pattern in Figures 17 and 18 was not being influenced by the difference in the number of samples from each neighborhood, ATSDR calculated the percentage of low-level arsenic samples in various ranges for each neighborhood. If no pattern exists, the percentages should be similar in each neighborhood. As Table 4 shows, the percentage of samples is different when comparing Elyria, Southwest Globeville, and Cole to Swansea and Clayton.

As can be seen in Graph 3 that follows, this difference is apparent when comparing the distribution of low arsenic levels in Elyria, Southwest Globeville, and Cole to the same distribution of low arsenic levels in Swansea and Clayton. To determine if the distribution of low arsenic levels in Elyria, Southwest Globeville and Cole is different from the distribution of low arsenic levels in Clayton and Swansea, ATSDR compared the distributions mathematically.

Graph 3

Distribution of Low Levels of Arsenic in Soil Up to 36 ppm

Table 4.

Percentage of samples in various ranges of arsenic levels
Neighborhood % less than 12 ppm % 12 to 17.9 ppm % 18 to 23.9 ppm % 24 to 29.9 ppm % 30 to 36 ppm
Elyria 28 46 17 8 0.6
SW Globeville 31 42 17 8 2
Cole 40 42 10 4 4
Clayton 64 21 6 5 4
Swansea 63 22 5 5 5

The results showed that the pattern of arsenic percentages between Elyria, Southwest Globeville, and Cole is statistically different from the pattern of arsenic percentages in Clayton and Swansea.(18) It seems reasonable to assume that this pattern may be the result of fallout from the nearby smelters.

In addition to these observations about low levels of arsenic, another important observation can be made about high levels of arsenic in soil. Since Phase III sampling detected relatively high levels of arsenic at residences very close to the boundary of the VBI70 study area, elevated arsenic concentrations might occur in areas beyond the current VBI70 study area. This conclusion is supported by a report released with EPA's assistance by the National Association of Black Environmentalists (NABE.) Collecting soil samples from 36 properties in the Northeast Park Hill neighborhood in March and August 1999, NABE found many yards with elevated levels of arsenic in soil. The highest arsenic level detected was 1,010 ppm. In addition, according to a staff member with the CDPHE, Dr. Drexler, a professor with the University of Colorado, has found high levels of arsenic in yards from other parts of Denver.

It is important to evaluate the population density in the area surrounding the VBI70 study area when making decisions about the significance of possible contamination outside the VBI70 study area. Figure 20 in Appendix B shows the population density for the neighborhoods surrounding the VBI70 study area. Using this map and previous figures showing patterns of arsenic contamination, there are areas outside the VBI70 site with unknown, but possibly significant, arsenic contamination in soils . They include:

  • Residential neighborhoods south, west, and north of the study area, south of Martin Luther King Boulevard/Blake Street, west of Fox Street/Burlington Northern Railroad, and north of southwest Globeville.

  • Residential neighborhoods east and southeast of Clayton,

  • The industrial area in the central portion of the VBI70 study area, and

  • Other possibly other residential neighborhood in the Denver area.

Arsenic at individual properties in the study area

Because of EPA's intensive sampling effort at eight properties, it is possible to understand the distribution of arsenic levels in soil at those properties, which helps in evaluating people's exposure to arsenic. Arsenic distribution in highly contaminated properties can be consistently high throughout the property, (see Figures 21 and 22, Appendix B) or it can be found at high levels in isolated spots throughout the property (see Figure 23, Appendix B). This pattern of patches of high arsenic contamination is important to understand when evaluating children's exposure to arsenic since children can play in any part of the yard.

EPA collected many soil samples from adjoining properties for six of the eight intensively sampled properties. At five of those six properties, elevated arsenic levels in soil were found at the property line of the adjoining yard, although the arsenic levels in soil from the adjoining properties were lower than those from the original properties. The arsenic might have migrated onto the adjoining property as part of surface-water runoff. In some cases, past landscaping or construction activity may have also caused the movement.

1 In March 2002 when the public release of the public health assessment took place, EPA was considering a clean-up level for arsenic in soil of 128 ppm. About 270 properties have average arsenic levels above 128 ppm. Since then, EPA has received comments from community members, and from federal, state, and local agencies. Based on these comments and on public meetings, EPA has re-evaluated its clean-up level and is now considering a clean-up level of 70 ppm, the same as the nearby Globeville ASARCO Superfund site. About 480 properties have average arsenic levels greater than 70 ppm.
2 In March 2002 when the public release of the public health assessment took place, EPA was considering a clean-up level for lead in soil of 540 ppm. Since then, EPA has received comments from community members, and from federal, state, and local agencies. Based on these comments and public meetings, EPA has re-evaluated its clean-up level and is now considering a clean-up level of 400 ppm. About 275 properties have average lead levels greater than 400 ppm.
3 En Marzo de 2002, cuando se publicó la evaluación de salud pública, la EPA estaba considerando un nivel para limpiar el arsenico en en suelo de 128 ppm. Como 270 propiedades tienen un promedio de niveles de arsenico de 128 ppm. Desde entonices, la EPA ha recibido comentarios de miembros de la comunidad, y de agencias federales, estatales y locales. Basado en estos comentarios y en juntas públicas, la EPA a re-evaluado el nivel de limpiar y ahora está considerando un nivel para limpiar de 70 ppm igual como cercano sitio de Superfondo Globeville ASARCO. Como 480 propiedades tienen niveles dearsenico mayores de 70 ppm.
4 En Marzo de 2002, cuando se publicó la evaluación de salud pública, la EPA estaba considerando un nivel para limpiar el arsenico en en suelo de 540 ppm. Desde entonices, la EPA ha recibido comentarios de miembros de la comunidad, y de agencias federales, estatales y locales. Basado en estos comentarios y en juntas públicas, la EPA a re-evaluado el nivel de limpiar y ahora está considerando un nivel para limpiar de 400 ppm. Como 275 propiedades tienen niveles de plomo mas grandes que 400 ppm.
5 EPA's NPL is a list of hazardous waste sites that may need some kind of remedial action to reduce exposure to toxic chemicals or radiation.
6 The boundaries of the the study area shown in Figure 1 are based on maps that EPA provided to ATSDR.
7 Matte is a product that has a sulfur containing metal. Common examples are copper matte and nickel matte.
8 Bismuth is a metal like lead and arsenic and is used in making pharmaceutical products (for example, Pepto Bismol). It is also used in industrial processes.
9 Litharge is an oxide of lead made by heating metallic lead.
10 Tellurium is a nonmetallic element similar to sulfur. It has a number of industrial uses, for example, as part of stainless steel and iron castings as well as a coloring agent in glass and ceramics.
11 Because of comments from the community, and federal, state, and local officials, EPA proposed in 2003 a clean-up level of 70 ppm arsenic in soil. This proposed new clean-up level is still being reviewed by EPA. At an average arsenic level of 70 ppm in soil, about 480 properties or so are a concern.
12 In a survey of U.S. western soils from urban and non-urban areas, the background level was determined to be 7 ppm, with the highest detected level in all samples being 97 ppm (ATSDR 1992).
13 If now considering 70 ppm as the clean-up level for arsenic at the VBI70 site. If EPA selects 70 ppm arsenic as the action level, about 480 properties will make the list. Based on the remaining 924 properties that have not been sampled, about 148 additional properties could have average arsenic levels above 70 ppm and could be added to the list.
14 An r square value of 1.0 indicates a perfect correlation.
15 Urban areas often contain higher levels of cadmium because of automobile traffic and possible local industrial sources. A survey by Skyline Labs, Inc., found a geometric average cadmium level in soil of 2.2 ppm in metropolitan Denver (Skyline Labs 1986).
16 Minimal Risk Levels (MRLs) and Reference Doses (RfDs) are health guidelines designed to identify exposure levels in humans below which harmful effects are unlikely. Please refer to the glossary for more explanation about MRLs and RfDs. ATSDR's chronic oral MRL for cadmium is 0.0002 mg/kg/day; EPA's chronic reference dose is 0.0005 mg/kg/day for water and 0.001 mg/kg/day for food. A list MRLs and RfDs can be found at these websites: and
17 This means that as few as 4 or as many as 21 out of every 100 children might have soil-pica behavior.
18 Using the Chi-square test, the p value was < 0.0001. This p value indicates that the difference between the two distributions is most likely real.

Next Section     Table of Contents The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #