Health Consultation - Exposure Investigation ASARCO HAYDEN SMELTER SITE

Att 7 AsarcoHaydenSmelterSite HC EI Report.pdf

ATSDR Exposure Investigations (EIs)

Health Consultation - Exposure Investigation ASARCO HAYDEN SMELTER SITE

OMB: 0923-0048

⚠️ Notice: This form may be outdated. More recent filings and information on OMB 0923-0048 can be found here:

Document [pdf]

Download: pdf | pdf

Health Consultation
Exposure Investigation
Biological Testing for Exposure to Lead and Arsenic near
ASARCO HAYDEN SMELTER SITE
HAYDEN AND WINKELMAN, ARIZONA

MARCH 27, 2017

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Agency for Toxic Substances and Disease Registry
Division of Community Health Investigations
Atlanta, Georgia 30333

Health Consultation: A Note of Explanation

An ATSDR health consultation is a verbal or written response from ATSDR to a specific
request for information about health risks related to a specific site, a chemical release, or
the presence of hazardous material. In order to prevent or mitigate exposures, a
consultation may lead to specific actions, such as restricting use of or replacing water
supplies; intensifying environmental sampling; restricting site access; or removing the
contaminated material.
In addition, consultations may recommend additional public health actions, such as
conducting health surveillance activities to evaluate exposure or trends in adverse health
outcomes; conducting biological indicators of exposure studies to assess exposure; and
providing health education for health care providers and community members. This
concludes the health consultation process for this site, unless additional information is
obtained by ATSDR which, in the Agency’s opinion, indicates a need to revise or append
the conclusions previously issued.

You May Contact ATSDR Toll Free at
1-800-CDC-INFO
or
Visit our Home Page at: http://www.atsdr.cdc.gov

HEALTH CONSULTATION

Exposure Investigation
Biological Testing for Exposure to Lead and Arsenic near
ASARCO HAYDEN SMELTER SITE
HAYDEN AND WINKELMAN, ARIZONA

Prepared By:
U.S. Department of Health and Human Services
Agency for Toxic Substances and Disease Registry (ATSDR)
Division of Community Health Investigations

Table of Contents
Executive Summary ....................................................................................................................................... 1
Introduction .................................................................................................................................................. 5
Site Location, History, and Status .................................................................................................................. 5
Environmental Contamination in Hayden and Winkelman ........................................................................... 6
Current Lead and Arsenic Exposure Pathways for Hayden and Winkelman Residents ................................. 8
Past Lead and Arsenic Biological Testing ....................................................................................................... 9
Exposure Investigation Process and Methods ............................................................................................... 9
Results ......................................................................................................................................................... 15
Discussion .................................................................................................................................................... 20
Uncertainties and Limitations ..................................................................................................................... 25
Conclusions ................................................................................................................................................. 27
Recommendations ...................................................................................................................................... 28
Public Health Action Plan ............................................................................................................................ 29
References................................................................................................................................................... 48

List of Figures
Figure 1: Asarco Hayden Smelter Site Map and Community Demographics.............................................. 31
Figure 2. Asarco Hayden Exposure Investigation participant blood lead levels by participant age ........... 32
Figure 3. Asarco Hayden Exposure Investigation participant blood lead levels by household .................. 33
Figure 4: Participant blood lead levels compared to U.S. population 2011–12 median and 95th percentile
levels and the exposure investigation follow-up level................................................................................ 34
Figure 5: Participant urinary total arsenic (creatinine corrected) results by household ............................ 35
Figure 6: Participant urinary total arsenic levels (creatinine corrected) by age group compared to 2011–
12 U.S. population median and 95th percentile levels and the exposure investigation follow-up level ... 36
Figure 7: Percentage of detected results and limits of detection for each arsenic species among exposure
investigation participants and the U.S. population .................................................................................... 37
Figure 8: Correlation of participant blood lead and urine arsenic (creatinine corrected) levels ............... 38
Figure 9: Relative contribution of smelter and background sources to 2013–15 air contamination levels
at Hayden and Winkelman monitoring stations when the smelter was operating and shut down........... 39

List of Tables
Table 1: Federal and state agency roles for the Asarco Hayden Exposure Investigation ........................... 40
Table 2. Number of participants by age group, gender, and contaminant tested ..................................... 40
Table 3. Exposure investigation participant and U.S. population (NHANES) blood lead median and 95th
percentile levels and confidence intervals ................................................................................................. 41
Table 4: Exposure investigation participant and U.S. population (NHANES) total urinary arsenic
(creatinine corrected) median and 95th percentile levels and confidence intervals ................................. 41
ii

Table 5: Arsenic levels (creatinine corrected and uncorrected) for Asarco Hayden Exposure Investigation
participants with urinary creatinine above 300 mg/dL (n = 4) compared with the investigation follow-up
level and U.S. population (NHANES) age specific 90th percentile.............................................................. 42
Table 6: Exposure investigation participant and U.S. population (NHANES) urinary inorganic-related
arsenic species 50th percentile (median) and 95th percentile levels and confidence intervals (creatinine
corrected).................................................................................................................................................... 42
Table 7: Average lead and arsenic ambient air concentrations across all Hayden and Winkelman
monitoring stations in 2015 and during smelter shutdown ....................................................................... 43
Table 8: Average lead and arsenic ambient air concentrations at each Hayden and Winkelman
monitoring station in 2015 and during smelter shutdown......................................................................... 44
Table 9: Exposure investigation participant and U.S. population urinary inorganic-related arsenic species
50th percentile (median) and 95th percentile levels and confidence intervals ......................................... 45

List of Appendices
Appendix A: Map of ambient air monitoring stations in Hayden and Winkelman, Arizona ...................... 54
Appendix B: Positive Matrix Factorization Model Results .......................................................................... 55
Appendix C: Exposure Investigation Participant and U.S. Population Urinary Creatinine Levels............... 57
Appendix D: Asarco Hayden Exposure Investigation Report Summary ...................................................... 59

iii

Abbreviations and Acronyms
As
ADEQ
ADHS
ATSDR
BLL
CDC
DMA
DLS
EI
EPA
IQ
LOD
mg/dL
μg/dL
μg/g
μg/L
µg/m3
ml
MMA
NAAQS
NHANES
NIOSH
NPL
OMB
Pb
PM10
PMF
ppb
ppm
WHO

Arsenic
Arizona Department of Environmental Quality
Arizona Department of Health Services
Agency for Toxic Substances and Disease Registry
Blood lead level
Centers for Disease Control and Prevention
Dimethylarsinic acid
Division of Laboratory Sciences
Exposure investigation
Environmental Protection Agency
Intelligence quotient
Limit of detection
Milligrams per deciliter
Micrograms per deciliter
Micrograms per gram
Micrograms per liter
Micrograms per meter cubed
Milliliter
Monomethylarsonic acid
National Ambient Air Quality Standard
National Health and Nutrition Examination Survey
National Institute for Occupational Safety and Health
National Priorities List
Office of Management and Budget
Lead
Particulate matter 10 micrometers or less in diameter
Postive Matrix Factorization
Parts per billion
Parts per million
World Health Organization

Executive Summary
Purpose

In April 2015, the Agency for Toxic Substances and Disease Registry
(ATSDR) provided lead and arsenic testing for Hayden and
Winkelman, Arizona residents most at risk for lead and arsenic
exposure or health effects. ATSDR offered the testing to determine
whether residents have elevated levels of these metals in their
bodies. This report summarizes the results.

Background

The Asarco Hayden Smelter Plant Site is in rural Arizona, about 90
miles southeast of Phoenix (Figure 1). The site includes the small
towns of Hayden and Winkelman. Historic and ongoing copper
smelting and processing caused environmental contamination in
these towns. Lead and arsenic are present in the air, mine waste
piles, and soil in some non-residential locations. In addition to
smelter-related contamination, lead may be present in old paint and
some other sources in homes (e.g., pottery), while arsenic may be in
certain foods (e.g., seafood and rice). The United States
Environmental Protection Agency (EPA) has completed residential
soil clean up at 266 Hayden and Winkelman yards and publicly
accessible areas. Though residential soil has been cleaned up,
residents may be exposed to lead and arsenic from current copper
production related emissions and other potential sources of
contamination. The populations most at risk for exposure or
negative health effects of exposure include young children, pregnant
women, and women who may become pregnant.
Community members asked they be tested for lead and arsenic
exposure. ATSDR worked with EPA, the Arizona Department of
Health Services (ADHS), and the Arizona Department of
Environmental Quality (ADEQ) (Table 1) to offer free, voluntary
blood and urine testing for residents most at risk for lead and
arsenic exposure or health effects.
Children ages 9 months to 5 years were eligible for blood lead
testing. Children and adolescents 6 to 17 years, pregnant women of
any age, and women of childbearing age (up to age 44) were eligible
for blood lead and urinary arsenic testing. Testing participants
included 83 residents (ages 1 to 40 years) of Hayden and Winkelman
from 29 different households. All participants received blood lead
testing and 58 participants received urine arsenic testing (Table 2).
ATSDR mailed results letters to participants in June 2015.

Conclusion 1

Some children in Hayden and Winkelman have been exposed to lead
at levels that could harm their health.

Basis for Conclusion 1 Two children exceeded the exposure investigation blood lead followup level [5 micrograms per deciliter (µg/dL)] (one in the 1–5 year age
group and one in the 6–11 year age group) (Figures 2, 3, and 4). In
addition, two children in the 1–5 year age group had blood lead
levels (BLLs) between 4 and 5 µg/dL. ATSDR’s exposure investigation
blood lead follow-up level is based on the Centers for Disease
Control and Prevention’s (CDC’s) current reference level.
Conclusion 2

Overall, children and adolescent participants had more lead in their
bodies than children and adolescents from across the United States.

Basis for Conclusion 2 The median blood lead levels for children and adolescent participant
age groups (1–5, 6–11, and 12–19 years) were about two times
higher than U.S. population medians for those age groups (Table 3).
No safe blood lead level in children has been identified.
Conclusion 3

ATSDR needs more information to determine how much arsenic
participants have in their bodies when air pollution levels are typical
for the community. Asarco shut down the smelter for maintenance
during the time of ATSDR’s testing, which lowered lead and arsenic
levels in the air.

Basis for Conclusion 3 Asarco shut down the smelter for maintenance in the days before
ATSDR collected blood and urine samples. As a result, participants
were exposed to about eight times less arsenic and seven times less
lead in the air than other times in 2015 (Table 7, Figure 9, and
Appendix B). Since arsenic is typically excreted from the body within
several days of exposure, the lower level of arsenic in air in the days
before testing could have led to a lower amount of arsenic in
participants’ urine. Since lead stays in blood longer than arsenic
stays in urine, ATSDR does not expect that the shutdown had a
significant effect on participants’ blood lead results.
No participant had a total urinary arsenic result (creatinine
corrected) that exceeded the exposure investigation follow-up level
(Figures 5 and 6). Median total and inorganic arsenic levels
(creatinine corrected) were similar to U.S. population age groupspecific medians (Table 4).

Recommendations

ATSDR recommends that EPA, ADEQ, Asarco, and the Gila County
Health Department take the following steps to protect the health of
the community.
• Reduce lead and arsenic air emissions at the Asarco Hayden
Smelter Plant.
• Continue environmental sampling and clean-up efforts in
Hayden and Winkelman.
o Consider resampling residential soil at a limited
number of homes in areas with higher levels of air
contamination to address community concerns that
soil may have been recontaminated since they were
cleaned up.
o Sample soil for lead at a specific Winkelman home
that was not previously sampled to ensure that
residential soil exposures did not contribute to a
participant’s elevated blood lead level.
• Incorporate ATSDR’s exposure investigation results in human
health risk assessments, as appropriate.
• Implement a home lead paint testing and abatement project,
as outlined in the 2015 EPA/Asarco settlement.
ATSDR recommends that exposure investigation participants take
part in a second round of arsenic testing. ATSDR intends to offer this
testing when the smelter is operating normally.
ATSDR recommends that Hayden and Winkelman residents
• Take the steps listed in the summary factsheet (Appendix D)
to reduce their exposure to lead and arsenic.
• Participate in the home lead paint testing project that Asarco
will develop and fund as part of the 2015 EPA/Asarco
settlement.
ATSDR recommends that parents/guardians of the two child
participants whose blood lead results were above the follow-up
level discuss the child’s result with their primary health care
provider.
ATSDR recommends that health care providers follow the Advisory
Committee for Childhood Lead Poisoning Prevention’s
recommendations for management of children with blood lead
levels above the CDC reference level.

Limitations and
Uncertainties

The results of this exposure investigation are subject to several
limitations and uncertainities. They are summarized here and
discussed in detail later in the report.
• Exposure investigation results are applicable only to the
individuals tested and cannot be generalized to other
individuals or areas.
• Test results cannot be used to determine the sources of lead
or arsenic exposures.
• Single blood lead and urinary arsenic tests are snap shots of
exposure and may not accurately represent a person’s past
or long term lead and arsenic exposures. Arsenic is excreted
within several days of exposure, while the half life of lead in
blood is about a month.
• The Asarco smelter was shut down for maintenance during
the ATSDR testing event, reducing the levels of lead and
arsenic in the air and potentially reducing participants’ urine
arsenic levels.
• ATSDR used creatinine corrected urine arsenic results to
adjust for variation in urine dilution and compare arsenic
results between participants. However, participants in this
exposure investigation had higher creatinine levels than the
U.S. population. The difference in creatinine levels creates
some uncertainty when comparing participant creatinine
corrected arsenic results with U.S. population results.
• Children less than 6 years of age were not eligible for urinary
arsenic testing because there are no national values for
comparison.
• Comparisons between adult participants (women 20–40
years old) and U.S. population adults (men and women 20
years and older) should be interpreted with caution due to
sex and age differences.

For More
Information

If you have questions about this report call ATSDR toll-free at 1-800CDC-INFO and ask for information on the Asarco Hayden Smelter
Plant site.

Introduction
In Hayden and Winkelman, Arizona, lead and arsenic are present in the air, mine waste piles,
and soil in some non-residential locations. In addition, lead may be present in old paint in
homes. Community members requested they be tested for lead and arsenic. In 2015, ATSDR
partnered with EPA, the Arizona Department of Health Services (ADHS), and the Arizona
Department of Environmental Quality (ADEQ) to offer blood and urine testing for people in the
community (ATSDR 2015b). ATSDR collected blood and urine samples in April 2015 and mailed
results to participants in June 2015. This report summarizes the results from all participants.

Site Location, History, and Status
The Asarco Hayden Smelter Plant Site is in rural Arizona, about 90 miles southeast of Phoenix
and 70 miles northeast of Tucson (Figure 1). The site includes the small towns of Hayden and
Winkelman, Arizona (populations 662 and 353, respectively) (Census 2010a). The area is dry,
windy, and sparsely vegetated. Historic and ongoing copper smelting and processing has caused
environmental contamination in these towns (EPA 2015a). Companies have processed copper
ore at several smelting operations and other facilities at this site for over 100 years (EPA
2014a). Asarco continues to operate a copper concentrator and smelter, producing copper
from copper sulfide ore (Asarco 2015a). Ore and concentrate are transported by railroad from
the nearby Ray mine and concentrator to the Hayden concentrator and smelter (Asarco 2015a).
Residential and public areas in Hayden and Winkelman are near various current and past
copper production related facilities, conveyances, and waste areas. Residents of Hayden live
within ¼ mile of the site, while residents of Winkelman live within 1 mile of the site (Figure 1).
EPA, ADEQ, and Asarco Grupo Mexico LLC (Asarco) are cleaning up lead, arsenic, and copper
contamination at the site through a Superfund alternative process (EPA 2015a).1 Historic
smelter emissions and other copper production-related activities deposited these contaminants
across the area. In addition, active copper production in Hayden contributes to elevated levels
of lead, arsenic, and copper in the air throughout the area (EPA 2015a).
Separate from the Superfund alternative process, in 2015, EPA announced a Consent Decree
(i.e., legal settlement) with Asarco to resolve Clean Air Act violations at Asarco’s Hayden facility
(EPA 2015d). The settlement requires the company to install new equipment and pollution
control technology at the Hayden smelter, fund local environmental health projects (including a
lead-based paint testing and abatement program for homes, schools, and other public buildings
in Hayden and Winkelman), replace a diesel locomotive with a cleaner model, and pay a civil
penalty (EPA 2015d).

Through the Superfund alternative approach, Asarco has agreed to complete the same investigation and cleanup
process that is used for National Priorities List (NPL) sites, without EPA listing the site on the NPL (EPA 2017).

Environmental Contamination in Hayden and Winkelman
EPA and ADEQ have been investigating environmental contamination at the site since 2002.
Environmental data indicated elevated levels of lead and arsenic in Hayden and Winkelman air
and some non-residential and residential soils. EPA’s Phase I Remedial Investigation focused on
residential soil and air contamination (EPA 2008) and led to clean up of 266 residential yards
with elevated levels of lead and/or arsenic (EPA 2015b). In the ongoing Phase II Remedial
Investigation, EPA will assess air, non-residential soils, groundwater, surface water, and
sediment contamination.
Copper production-related environmental contamination
Due to past and current copper production activities in the community, lead and arsenic are
present above EPA and ATSDR screening levels in air, some non-residential soils, and in mine
waste areas. Contaminated non-residential areas include Asarco-owned industrial areas (e.g.,
two large tailing piles), arroyos (i.e., dry creek beds), and railroad track areas, including those
located near residences. Levels of contamination in non-residential soils (largely collected as
surface soil samples) range from 3.5 to 1,230 parts per million (ppm) of lead and from 0.4 to
1,720 ppm of arsenic (EPA 2008; Table 4-3). EPA screening levels for lead in commercial and
residential soil are 800 ppm and 400 ppm, respectively. For arsenic in residential soils, ATSDR
uses 15 ppm as a screening value for determining whether to conduct a more detailed exposure
evaluation.
EPA completed soil clean up at 266 Hayden and Winkelman residential yards between 2008
and 2014 (EPA 2015b). EPA also completed soil remediation in dirt alleys and public parks.
Residential soil was cleaned up to below 400 ppm for lead and 23.4 ppm for arsenic (EPA
2015b).
Air quality monitoring indicates that smelter emissions contribute to elevated levels of metals
in the air in both the Hayden and Winkelman communities. Lead levels in local air sometimes
exceed EPA’s National Ambient Air Quality Standard (NAAQS) for lead [0.15 micrograms per
meter cubed (µg/m3)]. From 2012-2014, three month rolling average lead levels at individual
local air monitoring stations in Hayden and Winkelman ranged from 0.02 µg/m3 to 1.18 µg/m3
(EPA 2014a). EPA redesignated the Hayden area a nonattainment area for the lead NAAQS in
August 2014 (EPA 2014b) and it remains a nonattainment area as of September 2016 (EPA
2016a).2
ATSDR analyzed EPA-collected air quality data from 2015 at 10 monitoring stations located
throughout Hayden and Winkelman to learn more about air quality at the time of the April
2015 testing. Appendix A includes a map of the air monitoring stations; many are in or adjacent
to residential areas.

EPA defines nonattainment areas as “any area that does not meet (or that contributes to ambient air quality in a
nearby area that does not meet) the national primary or secondary ambient air quality standard for the pollutant.”
(EPA 2015c).

The Asarco smelter was shut down for maintence between April 6 and May 21, 2015. The
average level of lead across all stations for 2015, excluding the April 6–May 21 smelter
shutdown, was 0.11 µg/m3. Average lead levels for 2015 (excluding the shutdown timeframe) at
individual monitoring stations ranged from 0.016 µg/m3 (ST-2, Winkelman High School) to 0.49
µg/m3 (ST-14, smelter parking lot).
Arsenic levels across all monitoring stations averaged 0.06 µg/m3 in 2015, excluding the smelter
shutdown timeframe. Individual monitoring station averages ranged from 0.006 µg/m3 (ST-2,
Winkelman High School) to 0.27 µg/m3 (ST-14, smelter parking lot). EPA’s regional screening
level for arsenic in air is 0.0065 µg/m3. California’s acute, 8 hour, and chronic Reference
Exposure Levels3 for arsenic in air are 0.2, 0.015, and 0.015 µg/m3, respectively (OEHHA 2016).
In the 2008 remedial investigation, EPA used data from the the Organ Pipe National Monument
area southwest of Tucson, which is unaffected by mining or other human activity, for a
background ambient air point of comparison (EPA 2008). This area had average arsenic
concentrations of 0.0004 µg/m3 or less and average lead concentration of 0.001 µg/m3 (EPA
2008 and 2015a). During 2015, the average levels of arsenic and lead in Hayden and Winkelman
air (excluding the shutdown timeframe) were about 150 and 110 times that of the area
unaffected by smelting, respectively. Urban areas generally have mean arsenic levels in air
ranging from 0.02 to 0.03 µg/m3 (ATSDR 2007c).
Other local environmental sources of lead and arsenic
In addition to contamination from copper production, there are other sources of lead and
arsenic in the community. About 44% of the housing units were constructed before 1950
(Census 2010b), when lead was widely used in paint.
Hayden and Winkelman residents receive drinking water from two public drinking water
systems, which draw from local groundwater sources. Arsenic is often found in groundwater in
some parts of the United States, including the Southwest. EPA collected drinking water samples
in 2006 and found arsenic levels ranged from 3.6–5 µg/L across all sample locations in Hayden
and Winkelman (EPA 2008, Table 4-22). Although both systems contain low levels of arsenic,
they are below EPA’s Maximum Contaminant Level for arsenic (10 micrograms per liter µg/L).
Lead was not detected in the 2006 drinking water samples from either town. These 2006 levels
are similar to those described in the Hayden and Winkelman drinking water systems’ 2014 and
2015 water quality reports (Asarco 2014 and 2015b; Arizona Water Company 2014 and 2015).
Those reports note both systems detected arsenic concentrations up to 5 µg/L (based on 2012
and 2013 samples for Winkelman and 2013 samples for Hayden). In 2013, the highest lead level
detected by the Hayden water system was below 4 ppb lead, while the Winkelman system’s
highest detection was 1 ppb. EPA’s action level for lead in drinking water is 15 ppb (EPA 2016b).
While lead has not been detected above this level in the Hayden and Winkelman drinking water
3

Reference Expsoure Levels are airborne concentrations of a chemical that are not anticipated to result in adverse
non–cancer health effects for specified exposure durations in the general population, including sensitive
subpopulations (OEHHA 2014).

systems, the plumbing and fixtures in older buildings may contain lead, potentially increasing
lead levels in the water of some buildings.

Current Lead and Arsenic Exposure Pathways for Hayden and Winkelman Residents
The environmental data summarized above indicate that community members are at risk for
exposure to lead and arsenic. Hayden and Winkelman residents may be exposed to lead and
arsenic from copper production operations by breathing air and accidentally ingesting nonresidential soils in the community.
Hayden and Winkelman community demographics indicate several risk factors for higher blood
lead levels, including living in older housing (44% of housing units were built before 1950), and
in poverty (38% of people across both towns have a poverty income ratio4 < 1.24) (Census
2010a; Census 2010b; CDC 2013; Bernard et al. 2003; Jones et al. 2009). If deteriorating leadbased paint is present in Hayden and Winkelman homes, children in those homes are at greater
risk for higher blood lead levels. In addition, 57% of participants self-identified as MexicanAmercian, which may increase their risk of exposure to lead in products imported from Mexico
(e.g. candies, pottery and folk remedies) (Dixon et al. 2009). Drinking water is not a significant
lead exposure pathway for Hayden and Winkelman community members.
Community members may be exposed to arsenic in local air and soils. They may also be
exposed to low levels of arsenic in dietary sources, such as seafood and rice, and drinking
water. There are several types of arsenic that fall into two categories, organic and inorganic
(see Box 1). While exposure to organic arsenic is likely not associated with health concerns,
exposure to inorganic arsenic can harm people’s health (ATSDR 2007a).

Box 1: Arsenic: Sources and Types
Arsenic is an element that is widely distributed in the earth’s surface. Arsenic is released into
the environment from both human activities (e.g., mining, commercial use) and natural
processes (e.g., weathering of arsenic-containing minerals in soil and groundwater).
There are two basic types of arsenic:
Organic arsenic exposure doesn’t usually cause health problems. It is often found in fish
and seafood, so eating fish or seafood before arsenic testing may increase a person’s organic
and total arsenic level.
Inorganic arsenic exposure may cause health problems. It is found in many places in the
environment, like in soil and water, and in some foods, such as some types of rice.

A family’s income divided by their poverty threshold is their poverty income ratio. See
https://www.census.gov/hhes/www/poverty/about/overview/measure.html.

Past Lead and Arsenic Biological Testing
Previous lead and arsenic testing in Hayden and Winkelman has been limited. From 2003-2012
laboratories and physicians reported to ADHS 46 blood lead test results in Hayden and 86
results from Winkelman for children 0 to 16 years of age.5 Two children in Hayden had a blood
lead level (BLL) over 10 micrograms per deciliter (µg/dL) and six children in Winkelman had
blood lead levels between 5–10 µg/dL.
In 1999, with funding from Asarco, the University of Arizona and ADHS conducted blood lead
testing for young children (with an emphasis on children less than 3 years old) and spot urine6
arsenic testing for adults and children of any age7 in Hayden and Winkelman (Burgess et al.
2000; ADHS 2002). All fourteen children8 tested had blood lead levels below 10 µg/dL (the level
of concern at that time) and their average level was 3.6 µg/dL.9 About 77% of the 224
participants tested for arsenic were over 20 years old (Hysong et al. 2003). The average urinary
total arsenic concentration of individuals tested was 13.7 µg/L, less than the study reference
level of 30 µg/L. For the 18 participants with total arsenic concentrations exceeding 30 μg/L,
speciated analysis was used to measure inorganic arsenic. Five of those individuals had
inorganic urinary arsenic concentrations exceeding 30 µg/L, up to a maximum of 47 µg/L.
Urinary arsenic concentrations were not adjusted for creatinine and could have been
influenced by dietary sources (e.g., seafood). Results from this lead and arsenic exposure survey
are further discussed in the Discussion section of this report.

Exposure Investigation Process and Methods
In 2013, EPA requested that ATSDR conduct an exposure investigation to measure Hayden and
Winkelman residents’ lead and arsenic exposure levels. EPA had received requests from
residents for additional biological testing.10 ATSDR conducted the exposure investigation to
provide both individual residents and federal, state, and local agencies with more information
about lead and arsenic exposures in Hayden and Winkelman.
Agency roles
ATSDR, the lead agency for the investigation, collaborated with EPA, ADHS, ADEQ, and the
Centers for Disease Control and Prevention (CDC). The roles of each agency are described in
Table 1.
5

Arizona law requires physicians to report to ADHS blood lead levels ≥ 10 µg/dL, while laboratories are required to
report all blood lead test results, regardless of blood lead level (ADHS 2016).
6
Generally first morning urine samples.
7
Participants had to be able to collect urine in a cup.
8
Blood lead tests were provided to two children less than 6 months, seven children 6 months to 36 months, and
five children older than 36 months (Burgess et al. 2000).
9
The limit of detection for these blood lead tests was 1 µg/dL (Burgess et al. 2000).
10
Community members expressed interest in lead, arsenic, and copper biological testing to EPA. At the time of this
exposure investigation, ATSDR did not have established methods to conduct a biological test and interpret results
for copper,

Recruitment and participant eligibility
In March 2015, ATSDR and partner organizations visited Hayden and Winkelman to share
information, answer questions, and sign-up eligible participants for testing appointments.
Representatives worked toward these goals by holding public meetings, open house events,
and door to door conversations in Hayden and Winkelman. To raise awareness about the
testing opportunity, ATSDR developed a website on the project
(http://www.atsdr.cdc.gov/sites/HWAZ/; ATSDR 2015b), twice sent postcards to all Hayden and
Winkelman households, posted information in prominent locations in the towns, and left fact
sheets (in English and Spanish) at local businesses and institutions (ATSDR 2015a). Community
partners also distributed information. For instance, the Hayden-Winkelman Unified School
District sent fact sheets home with eligible students and posted information on their Facebook
page.
ATSDR sought to enroll people living in Hayden or Winkelman who may be at higher risk for
health effects from exposure to lead and arsenic (e.g., young children and pregnant women)
(Box 2). Initially, the following groups of Hayden or Winkelman residents were eligible to
participate in the exposure investigation.
• Children between the ages 9 months to 11 years were eligible for lead testing.
• Children between the ages of 6 years to 11 years were eligible for arsenic testing.11
• Pregnant women of any age were eligible to participate in both lead and arsenic testing.
Later in the recruitment period, ATSDR expanded the eligibility criteria to allow the following
groups to participate in both lead and arsenic testing (Box 2). 12
• Adolescents aged 12–17 years
• Women of childbearing age (up to age 44)

ATSDR did not offer arsenic testing to children ages 9 months to 5 years because (1) it is difficult to collect urine
samples from young children, especially those wearing diapers, and (2) ATSDR cannot interpret the testing results
because national comparison values do not exist.
12
During the course of recruitment ATSDR learned that parents also wanted testing for adolescents ages 12–17
living in Hayden and Winkelman. Adolescents ages 12–17 are not as likely to be exposed to lead and arsenic from
soil because they play differently than younger children do. However, because they are still growing and
developing, adolescents have more susceptibility to health effects of lead and arsenic than adults. Because of
parental interest and because resources were available to offer testing slots, ATSDR expanded the eligibility to
include adolescents ages 12–17 before the testing appointments began. During the recruitment period ATSDR also
expanded eligibility to include women of childbearing age (up to age 44) who live in Hayden and Winkelman.
ATSDR expanded to this group because ATSDR had resources available to offer testing slots and a developing baby
is sensitive to lead and arsenic in the mother’s body.

Box 2: Eligibility Criteria
People living in Hayden and Winkelman who met the following criteria
were eligible to participate.
Lead testing only
• Children ages 9 month to 5 years
Lead and arsenic testing
• Children and adolescents ages 6 years to 17 years
• Pregnant women of any age
• Women of childbearing age (up to age 44)
The exposure investigation team faced several challenges while recruiting participants. Some
people who worked for Asarco expressed concern that their family’s participation could put
their employment at risk. Others were concerned that the findings might be used as a rationale
for shutting down the smelter, negatively affecting the local economy. Finally, some parents
noted that their child’s health care provider tested them regularly for lead and/or arsenic.
Particpants and testing appointments
In April 2015, ATSDR offered free, voluntary blood lead and urine arsenic testing to Hayden and
Winkelman residents. ATSDR and ADHS representatives completed testing April 17–19, 2015 for
83 residents from 29 different Hayden and Winkelman households. All participants received
lead testing and 58 participants also received arsenic testing (see Results section for additional
information on participants).
Biologic sample collection and analysis
Participant consent and questionnaire
ATSDR administered consent, assent, and parental permission forms prior to collecting the
blood and urine samples. Blood and spot urine (generally first morning) samples were collected
April 17–19, 2015. ATSDR team members collected pertinent information from the head of each
household using an Office of Management and Budget (OMB) approved questionnaire (OMB #
0923-0048). The household questionnaire included questions on demographics, characteristics
and age of residence, and activities that might result in exposure to lead and arsenic. ATSDR
collected information on participant race and ethnicity as part of the questionnaire. This
information helped ATSDR understand differences between the participant population and the
U.S. population. It also allowed ATSDR to compare individual and aggregate participant results
to appropriate U.S. subpopulations, when necessary.
Confidentiality
Federal rules require that ATSDR maintain confidentiality of the information gathered through
interviews as well as the results of laboratory tests unless the data is aggregate and without
identifiable information. Arizona law (A.A.C. R9-4-301) requires laboratories to report all blood
lead tests to the Arizona Department of Health Services (ADHS 2016). In compliance with this
11

statute, ATSDR provided all blood lead testing results to ADHS. In addition, all participants gave
ATSDR permission to share their test results with other environmental and health government
agencies.
Blood lead sampling and laboratory analysis
Blood lead sampling is the most reliable method for measuring lead exposure from all sources
(Barbosa et al. 2005). ATSDR obtained whole blood samples by venous puncture. A
phlebotomist (medical professional who draws blood from a vein) collected three milliliters (ml)
of blood from each participant. CDC provided the collection tubes and supplies. To maintain
privacy, the samples were labeled with a unique identification number. After collection, blood
samples were maintained near four degrees Celsius throughout the collection period and
during overnight shipment. These samples were delivered for analysis to the CDC laboratory in
Atlanta, Georgia. The CDC environmental health laboratory performed blood lead analysis using
Division of Laboratory Science (DLS) method 3016.8 for blood metals (CDC 2014a).
Urine arsenic sampling and laboratory analysis
Determining urinary arsenic levels is the most reliable method to measure recent exposures to
arsenic (i.e., exposures experienced within the past few days) (Orloff et al. 2009). A 24-hour
urine collection is optimal due to fluctuations in excretion rates. However, most studies use a
first morning or random spot urine sample because it is convenient and increases compliance.
Both methods correlate well with 24-hour collection results (Orloff et al. 2009), though first
morning samples more so than random spot (Wang et al. 2016). ATSDR collected 58 spot,
generally first morning, urine samples. The collection cups were supplied by the CDC laboratory.
Most participants collected their urine sample at home on the day of their blood sample
appointment, froze the sample, and then brought it to the collection location at the time of
their blood sampling appointment. Some participants collected their urine sample at the blood
sampling location. To maintain privacy, the samples were labeled with a unique identification
number. Urine samples were kept frozen on dry ice and shipped to the CDC laboratory.
The CDC environmental health laboratory performed urinary arsenic analyses (total and
speciated arsenic for all participants) using the following methods: DLS 3018A.4 for urine total
arsenic and DLS 3000.14 for arsenic speciation (CDC 2014b and 2014c). The lab also measured
creatinine levels in urine samples to allow ATSDR to calculate creatinine corrected arsenic
levels.
Inorganic-related arsenic species
ATSDR calculated the sum of inorganic-related arsenic species for each participant because
“inorganic-related arsenic may be a more toxicologically and health relevant measure than total
urinary arsenic, which includes non-toxic organic arsenic species” (CDC 2015). Following the
methods CDC outlined in the February 2015 updated National Exposure Report tables (CDC
2015), ATSDR summed arsenic (V) acid, arsenous (III) acid, dimethylarsinic acid (DMA), and
monomethylarsonic acid (MMA) for each participant. When the value of a species was less than
the laboratory’s limit of detection (LOD), as was the case for some participants’ arsenic (V) acid
12

level, an imputed (i.e., substitute) value was used. The imputed value was calculated as the LOD
divided by the square root of two (Hornung and Reed 1990).
Creatinine correction for urinary arsenic
ATSDR used participants’ urine creatinine concentration to adjust arsenic results for urine
dilution (Barr et al. 2005). Creatinine corrected arsenic results are reported as microgram of
arsenic per gram creatinine (µg of arsenic per g creatinine). Creatinine correction allowed
ATSDR to compare arsenic results across participants who were more or less hydrated and thus
have different urine concentrations. However, creatinine concentrations also vary by age, sex,
race/ethnicity, and certain health conditions (Barr et al. 2005). To account for variation in
creatinine levels by age, ATSDR compared age group specific participant and U.S. population
(NHANES) creatinine corrected arsenic levels.13,14 ATSDR also compared participant and U.S.
population (NHANES) age group specific creatinine levels.
In addition, ATSDR used participant creatinine levels to gauge the validity of a urine sample
(Barr et al. 2005). In a state of under or over hydration, the kidney’s excretion rate of
contaminants changes, which can yield results that are not an accurate reflection of the
participant’s exposure. World Health Organization (WHO) urinary creatinine concentrations
guidelines are often used to determine valid spot urine samples for occupational monitoring.
The guidelines suggest resampling if a urine sample is too dilute (creatinine concentration < 30
mg/dL) or too concentrated (creatinine concentration > 300 mg/dL) to provide a valid measure
(Barr et al. 2005).15 While the WHO guidelines were developed for adults rather than children,
the focus of this investigation, ATSDR used them to help identify participants whose creatinine
corrected arsenic results might be biased.
Exposure investigation follow-up levels
ATSDR compared blood lead and urine arsenic test results from individual participants and
specific age groups to the U.S. population [National Health and Nutrition Examination Survey
(NHANES) results] (CDC 2015). For individual participant blood lead results, ATSDR used the
CDC’s blood lead 5 µg/dL reference value as the exposure investigation follow-up level. CDC
uses a reference level of 5 µg/dL to identify children with blood lead levels that are higher than
most children’s levels (CDC 2012a and 2012b). This level is based on the U.S. population of
children ages 1–5 years who are in the highest 2.5% of children when tested for lead in their
blood (i.e., the 97.5th percentile of the NHANES’s blood lead distribution in children). The
National Institute for Occupational Safety and Health (NIOSH) also uses 5 µg/dL as the blood
lead reference level for adults (NIOSH 2015). ATSDR compared individual total urinary arsenic
13

Comparisons between the adult participant age group (women 20–40 years old) and U.S. population adults (men
and women 20 years and older) should be interpreted with caution due to sex and age differences.
14
To further investigate participant’s inorganic-related arsenic species levels, ATSDR used both creatinine
corrected and uncorrected results.
15
Creatinine correction for a target chemical (e.g., arsenic) measured in a highly concentrated urine sample (i.e.,
elevated creatinine) tends to underestimate the concentration of the target chemical. Conversely, creatinine
correction for a target chemical measured in a very dilute urine sample (i.e., low creatinine) tends to overestimate
the concentration of the chemical.

results (creatinine corrected) to the exposure investigation follow-up level of 28.4 µg/g
creatinine. The arsenic exposure investigation follow-up level was the lowest 95th percentile
level for any age group in the 2009–10 NHANES (the 12–19 year age group). ATSDR chose this
level as a conservative screening value to identify participants with a potentially elevated
urinary arsenic level.
Individual result letters and follow-up
In June 2015, ATSDR sent results letters to individual participants along with a fact sheet on
ways to reduce exposure to lead and arsenic (ATSDR 2015c). ATSDR and ADHS also contacted
participants with elevated results to discuss their results and recommend steps to take to
protect the participant’s health.
Air monitoring data
Between 2013 and 2015, Asarco, with EPA oversight, collected air monitoring data on lead,
arsenic, and other contaminants at 10 monitoring stations located throughout Hayden and
Winkelman (Appendix A). ATSDR used this data to assess levels of lead and arsenic in the area.
Specifically, ATSDR used measurements of lead and arsenic in particulate matter with an
aerodynamic diameter of 10 micrometers or less (PM10) from 23+ hour samples collected every
six days at each monitoring station. Asarco used Thermo Scientific Partisol Plus 2025 sequential
air samplers configured with PM10 sharp cut cyclone inlets and AirMetrics™ MiniVol samplers
with teflon, quartz, and capillary pore membrane polycarbonate (0.1 micron pore size) filters to
collect the samples (EPA 2012).
Statistical analyses
ATSDR used R software (version 3.2.4) (R Core Team 2015) for statistical analyses of lead and
arsenic testing data and analysis of air monitoring data.
Statistical analysis of lead and arsenic results
ATSDR calculated statistics to compare exposure investigation participants to U.S. population
statistics. For lead and arsenic results, ATSDR estimated median (i.e., 50th percentile) and 95th
percentile levels for participants (see Box 3). ATSDR also used percentile bootstrap methods (n
= 2,000) to calculate 95% confidence intervals for lead and arsenic median and 95th percentile
levels. These confidence intervals allowed ATSDR to gauge whether exposure investigation
participant median and 95th percentile levels were statistically different (higher or lower) than
U.S. population levels (Krzywinski and Naomi 2013).
Statistical analysis of air monitoring data
For lead and arsenic air monitoring data, ATSDR calculated mean (i.e., average, see Box 3) levels
for various time periods and used the Wilcoxon rank sum test to determine whether
differences were statistically significant. ATSDR also used the EPA Positive Matrix Factorization
model to estimate the number and composition of air pollution sources and their relative
contributions to contaminant levels (all contaminants, not just lead and arsenic) at each
monitoring station (EPA 2015e).

Box 3: What are the median, mean, and 95th percentile?
The median and mean are different ways of measuring the center of a collection of numbers.
• The median is the middle value in a list of numbers. In a set of numbers it separates
the higher half from the lower half.
• The mean (or average) is the sum of a set of numbers divided by the number of
numbers in the set.
The 95th percentile is the value below which 95 percent of the values in a data set are found.

Results
Participants in the exposure investigation
Eighty-three people (ages one year to 40 years) from 29 households participated in the
exposure investigation. All 83 participants received blood lead testing, while 58 received urine
arsenic testing as well (Table 2).16 All participants were residents of the towns of Hayden or
Winkelman, Arizona [59 from Hayden (71%) and 24 from Winkelman (29%)]. As noted earlier,
ATSDR’s focus was on enrolling young children, as they are often at higher risk for lead
exposure. Thus, 65% of lead testing participants were children one year to 11 years, while 50%
of arsenic testing participants were children 6–11 years. Adolescents age 12–19 made up 20%
and 29% of lead and arsenic testing participants, respectively. All adults evaluated were women
of childbearing age (defined as less than 45 years old), including one pregnant woman. Adult
women (age 20–40) made up 14% of lead testing participants and 21% of arsenic testing
participants. Table 2 provides additional information on participants by age and sex. Based on
census estimates, approximately 37% of Hayden and Winkelman residents 9 months to 11 years
old, ATSDR’s primary target population, participated in the investigation.
Based on questionnaire responses, 90% (75 of 83) of the participants self-identified as Hispanic
or Latino and 10% (8 of 83) self-identified as Non-Hispanic. Of the self-reported Hispanic or
Latino participants, 57% (43 of 75) indicated they were of Mexican ethnicity, 8% (6 of 75)
identified as being of Puerto Rican ethnicity and 35% (29 of 75) identified themselves as “other”
Hispanic or Latino ethnicity. Five percent (4 of 75) identified themselves as having 2 or more
Hispanic ethnicities. With regards to race, 83% (70 of 83) of participants (including Hispanics or
Latinos) self-reported their race as white, one percent (1 of 83) self-reported their race as
African American and one percent (1 of 83) self-identified as more than one race. Sixteen
percent (13 of 83) of participants declined to answer with regard to race.

Blood lead results
As discussed in the methods section, CDC uses a reference level of 5 µg/dL to identify children
(1–5 years) with blood lead levels that are higher than most children’s levels (CDC 2012a and
2012b). NIOSH also uses 5 µg/dL as the blood lead reference level for adults (NIOSH 2015). For
the Asarco Hayden exposure investigation, ATSDR used 5 µg/dL as the investigation level to
identify participants for follow-up, including children older than 6 years, pregnant women, and
women of child bearing age.
Two participants from different households exceeded the exposure investigation blood lead
level (Figure 2 and Figure 3). A child in the 1–5 year old age group had the highest BLL, 5.9
µg/dL. A child in the 6–11 year old age group had a BLL of 5.3 µg/dL. In addition, two
participants, both in the 1–5 year old age group, had BLLs between 4 and 5 µg/dL. One of these
children and the 6–11 year old participant with the 5.3 µg/dL BLL were from the same
household. A third participant from that household had a lower BLL (0.96 µg/dL) (Figure 3).
In addition to comparing individual participant results to the exposure investigation follow-up
blood lead level, ATSDR compared individual participant results, and the median (50th
percentile) and 95th percentile BLL estimates for each age group of participants to age group
specific U.S. population (i.e., NHANES) median and 95th percentile BLLs. Twenty-four percent of
participants 1–5 years, 38% of participants 6–11 years, and 35% of participants 12–19 years
had BLLs above their age group specific NHANES 95th percentile level (Figure 4). Eight percent of
adult participants had BLLs above the adult NHANES 95th percentile level.
Median BLLs for children (1–5 years: 1.9 µg/dL and 6–11 years: 1.3 µg/dL) and adolescents (12–
19 years: 1.2 µg/dL) were about two times higher than U.S. population age specific comparison
groups (Table 3), though only the 1–5 year old and 12–19 year old age groups were statistically
different from the comparison age groups. The median BLL for adult participants (women age
20–40; 0.86 µg/dL) was slightly lower than the median for the adult U.S. population (1.05
µg/dL) (Table 3) and similar to the median for U.S. women of all ages (0.82 µg/dL). Age groupspecific children and adolescent 95th percentile BLL estimates ranged from 1.5 to 2.4 times
higher than U.S. population comparison age groups, though only the 12–19 year age group was
statistically different. The estimated 95th percentile BLL for adult participants (women age 20–
40; 2.61 µg/dL) was slightly lower than, but not statistically different from, the 95th percentile
for the adult U.S. population (3.36 µg/dL) (Table 3) and similar to the 95th percentile for U.S.
women of all ages (2.59 µg/dL).
Urinary arsenic results
ATSDR performed several evaluations of urinary arsenic results to understand how participant
arsenic exposure levels compare to the U.S. population. ATSDR used urine samples to evaluate
participant exposures to total arsenic, inorganic arsenic, and individual types (i.e., species) of
arsenic.

Total arsenic
First, ATSDR compared each participant’s total arsenic (creatinine corrected) results to the
exposure investigation follow-up level [the lowest 95th percentile level for any age group in the
2009–10 NHANES (28.4 µg/g creatinine, 12-19 year age group)] (Figure 5). No participants
exceeded that level. ATSDR used this comparison level in letters reporting total arsenic results
to participants. The highest urinary total arsenic level measured was 28.3 µg/g creatinine in an
11-year-old male. This participant’s inorganic-related arsenic species result was similar to the
U.S. population in his age group. Review of the participant’s organic arsenic species indicated
that arsenobetaine accounted for the majority of the participant’s total arsenic. The participant
had eaten 1–2 portions of seafood in the week before the testing, which likely contributed to
his elevated arsenobetaine level.
Second, ATSDR compared participant total arsenic results to age specific NHANES 2011-12
levels, which were released in February 2015. No participant’s total urinary arsenic (creatinine
corrected) exceeded the respective NHANES age group specific 95th percentile level (Figure 6).
Third, ATSDR estimated median and 95th percentile total arsenic exposure levels (creatinine
corrected) for each participant age group and compared these levels to NHANES age group
specific median and 95th percentile levels. Median total arsenic levels for each participant age
group were similar to U.S. population age group-specific medians, while participant 95th
percentile levels were lower than U.S. population 95th percentile levels (Table 4).
Total arsenic results for participants with creatinine concentrations outside the target range
As discussed in the methods section, ATSDR used participants’ urine creatinine concentration to
gauge the validity of a urine sample (Barr et al. 2005). WHO guidelines suggest resampling if a
urine sample is too dilute (creatinine concentration < 30 mg/dL) or too concentrated (creatinine
concentration > 300 mg/dL) to provide a valid measure (Barr et al. 2005).
Urine samples from two participants were below 30 mg/dL (29.03 and 29.97 mg/dL). Both
participants were female. Neither of these samples resulted in a creatinine corrected arsenic
result that was elevated and could potentially be confused for a falsely elevated result (7.3 and
9.0 ug/g of creatinine).
Urine samples from 4 of the 58 participants (6.8%) had a creatinine level above 300 mg/dL,
ranging from 313.9 to 419.3 mg/dL. A creatinine level above 300 mg/dL could potentially result
in an artificially low value for creatinine corrected urinary arsenic. Three of the four participants
with elevated creatinine results were Hispanic males between the ages of 12–19. The fourth
participant was a Hispanic female of child bearing age (20–40 years). Males generally have a
higher creatinine level than females (Barr et al. 2005). Hispanics and Mexican-Americans
generally report lower creatinine levels than the U.S. population (Barr et al. 2005). A prior study
found that uncorrected and creatinine corrected concentrations of inorganic urinary arsenic
were significantly correlated in a population with low-level environmental arsenic exposure as
was the case in this investigation (Hinwood et al. 2002). The urine samples from all four
participants had a creatinine corrected urinary arsenic level well below the 95th percentile and
17

median of their 2011–12 NHANES age group as well as the exposure investigation follow-up
level. In addition, the four participants’ uncorrected total urinary arsenic results were all below
the 90th percentile 2009–10 and 2011–12 NHANES levels for age, gender and race/ethnicity.
While these participants’ creatinine levels were outside of the target range, their total
uncorrected urinary arsenic values suggest that they were below the exposure investigation
follow-up level. Table 5 presents the creatinine corrected and uncorrected urinary arsenic
results for these participants with the exposure investigation follow-up level and 90th percentile
U.S. population levels.
Inorganic-related arsenic species
ATSDR estimated median and 95th percentile urinary inorganic-related arsenic species levels
(creatinine corrected) for child (6–11 years) and adolescent (12–19 years) age groups (Table 6).
For both participant age groups, inorganic-related arsenic species median levels (creatinine
corrected) were similar to U.S. population medians. Age group specific 95th percentile
inorganic-related arsenic species levels (creatinine corrected) were lower than the U.S.
population 95th percentile levels. ATSDR did not include in this report inorganic-related arsenic
species summary statistics for adult participants because 75% of adult participants had levels of
arsenic (V) acid (one type of inorganic arsenic), below the lab’s level of detection.
Individual arsenic species
ATSDR reviewed results on the individual species (or types) of arsenic measured in participants’
urine samples. As an indicator of exposure, ATSDR compared how frequently each arsenic
species was detected among participants as compared with the U.S. population in 2011–12. The
CDC laboratory’s levels of detection for each arsenic species changed between the time when
NHANES 2011–12 samples were analyzed and this exposure investigation. Thus, ATSDR
compared the percentage of detections using both the exposure investigation levels of
detection and after adjusting participant results based on the NHANES 2011–12 levels of
detection (i.e., counting as a non-detect any participant result that was below the NHANES
2011–12 level of detection, but above the exposure investigation level of detection).
Inorganic arsenic species [i.e., arsenic (V) acid, arsenous (III) acid, DMA, and MMA] were
detected more frequently in exposure investigation participants than in the U.S. population
(Figure 7). For instance, arsenous (III) acid was detected in all participants’ urine samples, but
only 32% to 39% of the U.S. population’s urine samples, depending on the age group. Arsenic
(V) acid was detected in 69% of participants 6–11 years old, but only 5% of the U.S. population
in that age group. This trend was true for results based on the exposure investigation levels of
detection and when results were adjusted based on NHANES 2011–12 levels of detection.
Correlation and spatial distribution of participant lead and arsenic results
ATSDR explored whether participants’ lead and arsenic levels were correlated (e.g., whether
participants with higher lead levels also have higher arsenic levels). ATSDR did not find a
correlation between participants lead and arsenic results (Figure 8). In addition, ATSDR mapped
participant lead and arsenic results, but did not see a clear trend in the spatial distribution of

lead or arsenic results. ATSDR does not include a map of participants’ locations and results to
protect their privacy.
Air monitoring results
After collecting participants’ blood and urine samples (from April 17–19, 2015), ATSDR learned
that the Asarco Hayden Smelter was shut down for maintenance 11 days before ATSDR started
blood and urine sampling until 31 days after sampling ended (April 6–May 21, 2015) (personal
communication with Tom Aldrich, ASARCO LLC, July 1, 2015). ATSDR used EPA air monitoring
data from 10 monitoring stations located across Hayden and Winkelman to assess whether the
shutdown changed the level of lead and arsenic in the air before and during our testing event. A
map of the ambient air monitoring network is included in Appendix A.
Average concentrations of lead and arsenic were about 7 and 8 times lower respectively during
the shutdown as compared with all of 2015 (excluding the shutdown timeframe) (Table 7). At
individual monitoring stations, average lead levels for 2015 (excluding the shutdown
timeframe) were between 3 (at ST-2 and ST-18) and 18 (at ST-16) times higher than during the
shutdown timeframe (Table 8). Similarly, average arsenic levels at individual monitoring
stations for 2015 (excluding the shutdown timeframe) were between 1.5 (at ST-2) and 18 (at
ST-16) times higher than during the shutdown timeframe (Table 8).
To better understand how the shutdown affected local air quality, ATSDR used the EPA Positive
Matrix Factorization (PMF) model to estimate the number and composition of air pollution
sources and their relative contributions to contaminant levels (all contaminants, not just lead
and arsenic) at each monitoring station. The model results indicate two primary sources of air
contamination in Hayden and Winkelman: the smelter and a background source, which includes
soil, dust, and other windblown sources.17 Further, the results show a clear reduction in the
relative contribution18 of the smelter source to air contamination levels at all Hayden and
Winkelman monitoring stations during the shutdown (Figure 9). While the background source
contribution to air contamination levels was similar throughout the 2013–15 timeframe,
including the shutdown period, the smelter source contribution was much lower during the
2015 shutdown than other periods (Appendix B; Figures B.1 and B.2).

Initially, ATSDR attempted PMF modeling with two factors but chose to include a third factor because some air
monitoring sites had a factor consisting of almost entirely chlorine. When ATSDR increased the number of factors
to four, the arsenic and lead concentrations were assigned to separate factors. ATSDR determined that three
factors gave the best fit with one factor including the majority of the lead and arsenic (smelter factor) and one
factor containing the majority of crustal elements (background factor) while a third factor included chlorine or
phosphorus but had some site dependence. This third factor is not used in this analysis.
18
The average relative importance over the entire time period for each factor is defined to be 1.0.

Discussion
Smelter shutdown before and during testing
As noted earlier, ATSDR learned after collecting blood and urine samples that the smelter had
been shut down in the days preceding and during the sample collection event. Due to
differences in the way lead and arsenic behave in the human body, the smelter shutdown likely
did not affect participant lead results, but could have affected arsenic results. Lead stays in the
blood for several weeks, so it is likely that lead exposures experienced before the shutdown
began would still be reflected in participants’ blood 11–13 days later, when ATSDR collected
samples. However, since arsenic is typically excreted from the body within several days of
exposure, the lower level of arsenic in air in the days preceding testing, could have reduced the
amount of arsenic ATSDR measured in participants’ urine.
Lead and health effects
Lead – background discussion
Lead is a naturally occurring metal. Typically found at low levels in soil, lead is processed for
many industrial and manufacturing applications, and it is found in many metallic alloys. Today,
lead can be found in all parts of our environment because of past and current human activities
including burning fossil fuels, mining, and manufacturing processes (ATSDR 2007b). Lead was
previously found in many gasoline additives, but by the mid 1970’s the U.S. began phasing out
the use of lead in gasoline and the Clean Air Act banned the sale of leaded fuel for on-road
vehicles in 1996 (EPA 1996). Lead was banned from paint in 1978.
Because lead is found throughout the environment, it is often found in the body at low levels.
Lead exposure occurs primarily via the oral route, with some contribution from the inhalation
route. The toxic effects of lead are the same regardless of the route of entry into the body.
Exposure to lead can have many health effects. Depending on the level of exposure, lead can
harm the nervous system, kidney function, immune system, reproductive system, development,
and cardiovascular system. Lead exposure also affects the oxygen carrying capacity of the
blood. The health effects of lead most commonly encountered in current populations are
neurological effects in children, and cardiovascular effects (e.g., high blood pressure and heart
disease) in adults. Infants and young children are especially sensitive to low levels of lead,
which may contribute to behavioral problems, learning deficits, and lowered IQ (ATSDR 2007b).
Lead in a pregnant woman’s body can negatively affect the health of her unborn child. Lead
exposure can also cause a miscarriage. It is not known for certain if lead causes cancer in
humans. Rats and mice fed large amounts of lead in their food developed kidney tumors. DHHS
classifies lead as “reasonably anticipated” to cause cancer and EPA considers lead a “probable”
cancer causing substance (ATSDR 2007b).
No lower threshold can be identified for some of the adverse neurological effects of lead in
children (ACCLPP 2012). Because of the absence of any clear threshold for some of lead’s more
sensitive health effects, ATSDR has not established guidelines for a low or no risk lead intake
20

dose. Blood lead levels should be kept as low as possible since no safe blood lead level in
children has been identified (ACCLPP 2012).
The half-life of lead in blood is approximately 28-36 days (ATSDR 2007b). Blood serves as the
initial repository of lead that is absorbed into the body, but typically carries only a small portion
of the total lead burden in the body. As lead moves through the body it harms various organ
systems and can be stored in the bones and teeth. Lead that is not stored in bones and teeth is
excreted from the body in urine and feces. About 99% of the amount of lead taken into the
body of an adult will leave the body in urine or feces within four to five weeks, while only about
30% of the lead taken into the body of a child will leave the body in urine or feces (ATSDR
2007b). Lead can stay in bones for decades (ATSDR 2007b). Lead can leave bones and re-enter
the blood and deposit in organs under certain circumstances; for example, during pregnancy
and lactation, after a bone is broken, and during menopause in women (due to osteoporosis).
Some biological (e.g., age, sex) and social (e.g., race, socio economic status) factors make
people either more vulnerable for lead exposure and/or susceptible to lead’s health effects.
Living in older housing (CDC 2013; Bernard et al. 2003), and poverty (CDC 2013; Jones et al.
2009), combined with being Mexican-American (Dixon et al. 2009; EPA 2013) and being nonHispanic black (Bernard et al. 2003; CDC 2013; Jones et al. 2009) are risk factors for higher
blood lead levels.
Lead - participant blood lead results discussion
Multiple environmental sources and risk factors likely contributed to the lead ATSDR measured
in participants’ blood. Some sources were community-wide (e.g., air), while others may have
been household specific (e.g., lead paint). Other risk factors relate to individual behavior (e.g., a
child putting their hands in their mouth or consuming certain Mexican candies). ATSDR
collected information from participants about some risk factors, which ATSDR used to provide
recommendations to participants with higher BLLs and to look for trends across participant
results.
ATSDR asked participants (or their parent/guardian) about the age of their homes, since lead
paint was used widely in homes built before 1950 and was phased out in 1978. ATSDR did not
observe a difference in blood lead levels for participants who reported living in homes built
before 1950 (n=26), between 1950 and 1979 (n=29), after 1979 (n=7), or did not answer (n=21).
ATSDR also collected information about the occupations of adult participants and the parents
of child participants. Results did not suggest a relationship between participant blood lead
levels and having a parent who worked at a mine, smelter, or other settings with potential lead
exposure.
As noted in the results section, two participants had BLLs above the investigation follow-up
level (5 µg/dL) and two other participants had BLLs close to the follow-up level, between 4–5
µg/dL. Two of the four participants with a BLL above 4 µg/dL resided in Winkelman and two
resided in Hayden. The two from Hayden lived in the same household. In follow-up
21

conversations with the participants’ parents, ATSDR learned that at the time of testing the two
participants from Winkelman did not reside in the same household but frequently spent time
together.
ATSDR could not identify a clear exposure source that would account for these children’s higher
blood lead levels. None of these participants were reported to routinely eat dirt, but all four
were reported to frequently put dirty hands or toys in their mouth, as is typical for toddlers and
young children. The parents/guardians of 20 of 25 participants under 6 years old noted their
child put dirty toys in their mouth. The four participants with higher BLLs are unlikely to be
exposed to high levels of lead in their residential soils. During the residential soil clean up
process, EPA sampled soil at the Hayden home and one of the two Winkelman homes. The soil
at the Hayden household required clean up, which EPA completed in 2009. The soil at the
Winkelman home that EPA sampled did not require clean up. The Winkelman home that EPA
did not sample is located in the vicinity of several properties where levels of lead and arsenic in
soil did not require clean up.
Exposure to lead-based paint may be a concern at the Hayden household, which was built in
the 1950s. Lead-based paint may also be present in the Winkelman home where EPA collected
soil samples, which the participant reported was built in the 1940s. The Winkelman home
where soil was not sampled was constructed in the 1990s, after the phase out of lead-based
paint.
The two participants with BLL above 4 µg/dL residing in Hayden lived in the same household
and are siblings. ATSDR learned in a follow-up conversation with the family that they have
spent time with family members in Mexico and that the children do routinely consume candy
obtained from Mexico. ATSDR advised the parents that consumption of this candy is a potential
source of lead exposure and that they should avoid continued consumption of candy that was
obtained in Mexico.

Arsenic and health effects
Arsenic – background discussion
Arsenic is a naturally occurring element that is found in combination with either inorganic or
organic substances to form many different compounds. Inorganic arsenic compounds are more
toxic than organic arsenic compounds (ATSDR 2007a). Arsenic is also released into the
environment from mining, ore smelting, and industrial use. In the past, inorganic arsenic was
used as a pesticide and as a preservative for wood (commonly referred to as pressure-treated
wood) (ATSDR 2007a). Inorganic arsenic compounds are found in groundwater, soils,
sediments, and some foods (e.g., rice). Groundwater in several regions of the United States,
including the southwest, have higher naturally occurring levels of inorganic arsenic than other
areas. People can be exposed to inorganic arsenic by drinking arsenic-contaminated drinking
water, ingesting arsenic after touching contaminated soil or wood preserved with arsenic,
breathing arsenic-contaminated air, and eating foods contaminated with arsenic. Fish and
shellfish commonly contain organic arsenic compounds, which can lead to organic arsenic
22

exposure in people consuming seafood. Animal studies have found that organic arsenic appears
to be less toxic than inorganic arsenic (ATSDR 2007a).
Inorganic arsenic has been linked to skin, liver, bladder, and lung cancer, and the Department of
Health and Human Services (DHHS) has designated it as a known human carcinogen (ATSDR
2007a). Arsenic also induces a wide variety of non-cancer effects in humans (ATSDR 2007a and
2016). Unusually large doses of inorganic arsenic can cause symptoms ranging from nausea,
vomiting, and diarrhea to dehydration and shock (ATSDR 2007a). Long-term exposure to high
levels of inorganic arsenic in drinking water has been associated with skin disorders (e.g.,
hyperkeratosis and hyperpigmentation) and increased risks for diabetes and high blood
pressure among other health risks (ATSDR 2007a and 2016). Long term exposure to arsenic in
air at lower concentrations can lead to skin effects, and also to circulatory and nervous system
problems (ATSDR 2007a).
Arsenic - participant urine arsenic results discussion
As noted in the methods section, ATSDR used creatinine correction to adjust arsenic results for
urine dilution. Creatinine correction allowed ATSDR to compare arsenic results across
participants who were more or less hydrated and thus have different urine concentrations.
However, urine creatinine levels can vary depending on a person’s age, sex, body mass, and
certain health conditions (Barr et al. 2005). To account for variation by age, ATSDR used
creatinine corrected age group specific U.S. population arsenic levels for comparison. ATSDR
also compared participant and U.S. population (NHANES) creatinine levels.
Participant age group specific median creatinine levels were higher than comparable age group
medians for the NHANES 2011-12 and NHANES III (1988-1994) U.S. population and Mexican
American sub-population (Appendix C) (CDC 2015 and Barr et al. 2005). Several factors may
contribute to differences between participant and U.S. population levels of urinary creatinine.
First, following ATSDR direction, most participants collected first morning urine samples,
whereas NHANES urine samples were collected at random times throughout the day. First
morning urine samples are generally more concentrated (and thus have higher creatinine
concentrations) than urine samples collected at other times (Barr et al. 2005). In addition,
participants live in a warm, arid region which may affect their hydration status. ATSDR did not
collect information on participant body mass and health conditions, but participants may have
differed from the U.S. population on such factors. While participants included more women
(59%) and persons who identified as hispanic or latino (90%) than the U.S. population, women,
hispanics, and Mexican Americans generally report lower creatinine levels than the U.S.
population (Appendix C).
To further investigate participant’s inorganic arsenic exposure levels, ATSDR compared urinary
inorganic-related arsenic species median and 95th percentile levels (without creatinine
correction) for the younger age groups to the U.S. population (Table 9). For 6–11 year old
participants, uncorrected inorganic arsenic median and 95th percentile levels were
approximately double those of the U.S. population for that age. For the 12–19 year old age
group, the median level was also double the U.S. population for that age, while the 95th
23

percentile level was similar to the U.S. population for that age. Given participants’ higher
creatinine levels, uncorrected inorganic-related arsenic species results may overestimate
exposures relative to the U.S. population, while creatinine corrected results may underestimate
exposures. However, a more detailed analysis would be required to determine whether
creatinine correction introduced any bias.
ATSDR also found that each individual inorganic arsenic species was detected more frequently
in participants’ urine than in the U.S. population (though some more than others) (Figure 7).
These results suggest higher levels, and perhaps different sources, of inorganic arsenic in the
Hayden and Winkelman environment as compared with the United States generally.
Arsenic – comparison of urinary levels to previous Hayden and Winkelman results and other
smelter communities
As noted earlier in this report, in 1999 the University of Arizona and ADHS conducted arsenic
testing for 224 Hayden and Winkelman residents, 77 percent of which were over age 20
(Burgess et al. 2000 and Hysong et al. 2003). Due to differences in the age of the study
population and analytical methods, many of the results of the 1999 exposure survey are not
directly comparable to the ATSDR exposure investigation. That said, median total urinary
arsenic levels provide one point of comparison. The median total urinary arsenic level for the
1999 exposure survey participants was 9.6 µg/L (Hysong et al. 2003). ATSDR exposure
investigation participant median total urinary arsenic levels were: 12 µg/L, 11 µg/L, and 7.2 µg/L
for the 6–11, 12–19, and 20–40 year age groups, respectively. This comparison suggests that
Hayden and Winkelman residents were exposed to similar levels of arsenic in 1999 and 2015.
A University of Arizona exposure study of 70 children age 1 to 11 in Dewey-Humboldt, Arizona,
a historic mining and smelting area (Loh et al. 2016), found arsenic exposure levels close to
Hayden and Winkelman exposure investigation participants. Differences in the age of the
Dewey-Humboldt study population make direct comparisons difficult. The median and 95th
percentile levels of inorganic arsenic related species for Dewey-Humboldt study participants
(age 1–11) were 10.4 and 28.5 µg/L respectively (Loh et al. 2016 and Personal communication
with Miranda Loh, May 5, 2016). The median and 95th percentile levels of inorganic arsenic
related species for the Hayden and Winkelman 6–11 year age group were 11.9 and 24.9 µg/L
respectively. Though both towns share a mining and smelting history, the potential arsenic
exposure sources are different. In Dewey-Humboldt, arsenic contamination of residential soils
and drinking water (especially in private wells) are of greater concern (Loh et al. 2016). In
Hayden and Winkelman, non-residential soils, ambient air (largely from active smelting
operations), and drinking water are likely the primary sources of environmental arsenic
exposure. House and wind-blown dust are likely sources of exposure in both communities.
Despite these different exposure routes, children in these two Arizona communities have
similar arsenic exposure levels.
Discussion of correlation and spatial distribution of lead and arsenic results
There are several potential reasons that ATSDR did not observe a correlation between
individual participants’ lead and arsenic levels nor a clear spatial pattern in the results. First,
24

environmental contamination in the community is not uniformly distributed. For instance, a
child playing in a non-residential area with arsenic contaminated soil may not be exposed to
lead in the soil concurrently. Second, there are several lead and arsenic exposure sources that
could contribute to a participant’s results independently. For instance, a participant may have
been exposed to lead-based paint in their home, but little arsenic from dietary sources. Finally,
because lead stays in the body longer than arsenic, the smelter shutdown during the days
before ATSDR’s testing may have affected arsenic results more than lead results.

Uncertainties and Limitations
All investigations have uncertainties and limitations. This exposure investigation has the
following uncertainties and limitations:
•

The results of this exposure investigation are applicable only to the individuals tested
and cannot be generalized to other individuals or areas.

•

The tests results cannot be used to
o determine the sources of lead or arsenic exposures, or
o predict the future occurrence of disease in individuals.

•

The single blood lead and urinary arsenic tests are snap shots of exposure. They may not
accurately represent a participant’s past or long term lead and arsenic exposures.
o Participant urinary arsenic results indicate very recent exposure to arsenic.
Arsenic is rapidly metabolized and excreted from the body (e.g., half of the
amount of ingested arsenic excreted in a 4 day period is excreted within the first
28 hours) (Orloff et al. 2009). In addition, urinary arsenic levels vary over time
(Wang et al. 2016).
o Participant blood lead results represent recent past exposure to lead. The half
life of lead in an adult’s blood is about 30 days (ATSDR 2007b).

•

The Asarco Hayden Smelter was shut down 11 days before ATSDR began holding its
three day blood and urine sample collection event, reducing the amount of lead and
aresenic in local air. As a result, in the days before testing, participants were exposed to
about seven times less lead and eight times less arsenic in local air as compared with all
of 2015 (excluding the shutdown timeframe). Since arsenic is typically excreted from the
body within several days of exposure, the lower level of arsenic in air in the days
preceding testing, could have reduced the amount of arsenic ATSDR measured in
participants’ urine. Lead stays in blood for weeks and is stored in bones for years, so
lead exposures experienced before the shutdown began were likely still reflected in
participants’ blood when ATSDR collected samples. As noted in conclusions 1 and 2,
younger participants’ blood lead levels (age groups 1–5, 6–11, and 12–19 years) were
higher than U.S. population comparison age groups, while younger participants’ urine
arsenic levels (creatinine corrected) (6-11 and 12-19 years old) were not. While various
sources likely contribute to participant lead exposures, these findings suggest that the
shutdown may have affected urine arsenic levels more so than blood lead levels.
25

•

ATSDR used creatinine corrected urine arsenic results to adjust for variation in urine
dilution. Creatinine correction helped ATSDR compare arsenic results between
participants. However, participants in this exposure investigation had higher creatinine
levels than the comparison U.S. population. The difference in creatinine levels creates
some uncertainty when comparing participant creatinine corrected arsenic results with
the U.S. population. Participants’ creatinine levels may have differed from the
comparison NHANES participant population for several reasons.
o Most exposure investigation participants provided first morning spot urine
samples, whereas NHANES samples were random spot urine samples. First
morning urine samples are generally more concentrated than other samples
throughout the day.
o Participants live in a hot, arid environment and thus may be less hydrated than
the U.S. population.
o Participants may differ from the U.S. population on certain physiological traits
(e.g., body mass) and/or health conditions (e.g., diabetes) that affect creatinine
levels.
o The participant population included more women and Hispanics than the
NHANES population. However, women and Mexican Americans typically have
lower creatinine levels than than men and other racial/ethnic groups
respectively (Barr et al. 2005).

•

Comparisons between the adult participant age group and U.S. population adults should
be interpreted with caution due to sex and age differences. Adult participants were
exclusively women 20–40 years old, while the NHANES comparison age group included
both men and women 20 years old and older.

•

Children less than 6 years of age were not evaluated for arsenic in urine because there
are no national values for comparison.

•

ATSDR did not present separate summary exposure statistics for Hayden participants
and Winkelman participants, as further subdividing each age group by town left some
age groups with too few participants to draw conclusions.

•

Due to the small number of participants in the 12–19 and 20–40 year old age groups (n
= 17 and 12, respectively), the percentile bootstrap method ATSDR used may have
produced narrow 95 percent confidence intervals for lead and arsenic median and 95th
percentile levels for these age groups.

Conclusions
ATSDR has three main conclusions from this exposure investigation.
Conclusion 1
Some children in Hayden and Winkelman have been exposed to lead at levels that could harm
their health.
Basis for Conclusion 1
Two children exceeded the exposure investigation blood lead follow-up level (5 µg/dL) (one in
the 1–5 year age group and one in the 6–11 year age group) (Figures 2, 3, and 4). In addition,
two children in the 1–5 year age group had BLLs between 4 and 5 µg/dL. ATSDR’s exposure
investigation blood lead follow-up level is based on the CDC’s current reference level.
Conclusion 2
Overall, children and adolescent participants had more lead in their bodies than children and
adolescents from across the United States.
Basis for conclusion 2
The median blood lead levels for children and adolescent participant age groups (1–5, 6–11,
and 12–19 years) were about two times higher than U.S. population medians for those age
groups (Table 3). No safe blood lead level in children has been identified.
Conclusion 3
ATSDR needs more information to determine how much arsenic participants’ have in their
bodies when air pollution levels are typical for the community. Asarco shut down the smelter
for maintenance during the time of ATSDR’s testing, reducing lead and arsenic levels in the air.
Basis for conclusion 3
In the days before ATSDR collected blood and urine samples, Asarco shut down the smelter for
maintenance. As a result, participants were exposed to about eight times less arsenic and seven
times less lead in the air than other times in 2015 (Table 7, Figure 9, and Appendix B). Since
arsenic is typically excreted from the body within several days of exposure, the lower level of
arsenic in air in the days before testing could have reduced the amount of arsenic ATSDR
measured in participants’ urine. Since lead stays in blood longer than arsenic stays in urine, the
shutdown should not have had a significant effect on lead results.
No participant had a total urinary arsenic result (creatinine corrected) that exceeded the
exposure investigation follow-up level (Figures 5 and 6). Median total and inorganic arsenic
levels (creatinine corrected) were similar to U.S. population age group-specific medians (Table
4).

Recommendations
ATSDR recommends that EPA, ADEQ, Asarco, and the Gila County Health Department take the
following steps to protect the health the community.
•
•

•
•

Reduce lead and arsenic air emissions at the Asarco Hayden Smelter Plant.
Continue environmental sampling and clean-up efforts in Hayden and Winkelman.
o Consider resampling residential soil at a limited number of homes in areas with
higher levels of air contamination to address community concerns that soil may
have been recontaminated since they were cleaned up.
o Sample soil for lead at a specific Winkelman home that was not previously
sampled to ensure that residential soil exposures did not contribute to a
participant’s elevated blood lead level.
Incorporate ATSDR’s exposure investigation results in human health risk assessments, as
appropriate.
Develop and implement a lead-based paint testing and abatement project for homes,
schools, and other public buildings, as outlined in the 2015 EPA/Asarco settlement (EPA
2015d).

ATSDR recommends that exposure investigation participants participate in a second round of
arsenic testing. ATSDR intends to offer this testing when the smelter is operating normally.
ATSDR recommends that Hayden and Winkelman residents take the following steps to reduce
their exposure to lead and arsenic.
•

•

Participate in the home lead-based paint testing and abatement project Asarco will
develop and fund as part of the 2015 EPA/Asarco settlement. Contact Amy Veek at
Asarco (520-356-3296, aveek@asarco.com) for information about the status of this
project.
Take the steps listed in the summary factsheet (Appendix D) to reduce your exposure to
lead and arsenic.

ATSDR recommends that parents/guardians of the two children whose blood lead results were
above the follow-up level discuss the child’s result with their primary health care provider.
ATSDR further recommends that health care providers follow the Advisory Committee for
Childhood Lead Poisoning Prevention’s recommendations for management of children with
blood lead levels above the CDC reference level (ACCLPP 2012).

Public Health Action Plan
The purpose of the Asarco Hayden exposure investigation was to to better understand
residents’ blood lead and urinary arsenic levels and provide a plan of action designed to
prevent or mitigate adverse human health effects from exposures. The following public health
action plan notes completed, proposed, and potential ATSDR and ADHS public health activities.
Completed actions
• Followed up with participants
o In June 2015, ATSDR sent each participant a letter with their blood lead and
urine arsenic results. ATSDR included a fact sheet on ways to reduce exposure to
lead and arsenic with each letter.
o In June 2015, ATSDR contacted the parents/guardians of participants with
elevated lead results by phone to discuss their child’s results and recommend
steps to take to protect the child’s health.
o In 2015, ADHS sent additional materials with recommendations for preventing
lead exposures to participants with elevated blood lead levels.
o In 2016, ATSDR attempted to contact the parents/guardians of the four
participants with results above or approaching the exposure investigation followup level again by phone. However, because ATSDR was not able to reach the
participants by phone, an ATSDR representative visited the three homes of these
participants in November 2016. Two of the homes are no longer occupied by
participant families. The family of two of these participants had indicated in 2015
their intention to move away from the community. That home is now occupied
by another family. Another participant home was vacant. At the third home,
ATSDR learned that one child would soon receive a follow-up blood lead test.
ATSDR also learned that the child from the now vacant home had received a
follow-up test that was “okay,” though ATSDR did not learn the test result.
• Supported the Gila County Health Department in initial preparations for a lead-based
paint testing and abatement project in Hayden and Winkelman. Under the 2015
EPA/Asarco settlement Asarco agreed to develop a lead-based paint abatement project
plan and provide funding to the Gila County Health Department to implement the
project (EPA 2015d).
o ATSDR shared participant lead results with the Gila County Health Department to
help the county prioritize its outreach efforts (after entering into a data sharing
agreement).
o ATSDR advised the county on community outreach and recruitment strategies
based on lessons learned during the exposure investigation.
o ADHS provided the county with information on Hayden and Winkelman lead test
results that have been reported to the state.

o ADHS connected Gila County with an established residential lead testing and
abatement program in another Arizona community, so Gila County could learn
from their experience.
Future, proposed, and potential actions
• ATSDR will hold a public meeting in Hayden or Winkelman to present the results of the
exposure investigation, explain steps community members can take to limit their
exposure to lead and arsenic, and answer questions.
• ATSDR plans to offer a second round of arsenic testing for exposure investigation
participants. ATSDR intends to conduct the testing at a time when the smelter is
operating normally.
• ATSDR will continue to support the development and implementation of the Hayden
and Winkelman lead-based paint testing and abatement project outlined in the 2015
EPA/Asarco settlement.
• Upon request, ATSDR is available to organize trainings for local health care providers on
testing for lead and arsenic exposure, interpreting results, and reducing exposures.

Figure 1: Asarco Hayden Smelter Site Map and Community Demographics

Figure 2. Asarco Hayden Exposure Investigation participant blood lead levels by participant age

Figure 3. Asarco Hayden Exposure Investigation participant blood lead levels by household

Figure 4: Participant blood lead levels compared to U.S. population 2011–12 median and 95th
percentile levels and the exposure investigation follow-up level

Figure 5: Participant urinary total arsenic (creatinine corrected) results by household

Figure 6: Participant urinary total arsenic levels (creatinine corrected) by age group compared to
2011–12 U.S. population median and 95th percentile levels and the exposure investigation follow-up
level

Figure 7: Percentage of detected results and limits of detection for each arsenic species among exposure
investigation participants and the U.S. population

Figure 8: Correlation of participant blood lead and urine arsenic (creatinine corrected) levels

Figure 9: Relative contribution of smelter and background sources to 2013–15 air contamination
levels at Hayden and Winkelman monitoring stations when the smelter was operating and shut
down

Relative Importance of Factors with Smelter Operation

Relative Factor Importance Averaged Across All Sites

1.20E+00

1.00E+00

8.00E-01

6.00E-01

4.00E-01

2.00E-01

0.00E+00
Operating

Shutdown

Smelter Status
Smelter

Background

Notes: Relative factors calculated using the EPA Positive Factorization Model (PMF) (EPA 2015e). The smelter was shut
down April 6–May 21, 2015. Shutdown period estimates are based on data collected April 6–May 18, 2015. Air samples
were not collected daily. Source: EPA air monitoring data (unpublished). Appendix B provides more detailed results from
the PMF model.

Table 1: Federal and state agency roles for the Asarco Hayden Exposure Investigation
Activity

Agency*

Agency Roles

ATSDR

Wrote the protocol, which included sampling and analysis
plan, fact sheets, questionnaire, consent/
assent/permission forms, and results reporting plans

Communicated with
community officials and
organizations

ATSDR, ADHS,
EPA, ADEQ

Conducted multiple conference calls, telephone calls, inperson meetings, and email briefings with local
organizations about project

Recruited participants

ATSDR, ADHS,
EPA, ADEQ

Worked as a team to conduct recruitment activities and
schedule appointments

Collected biological samples

ATSDR, ADHS,
EPA, ADEQ

Worked as a team to implement blood and urine sample
collection from participants

CDC

Used laboratory methods to analyze biological samples
and provide results to ATSDR

Developed exposure
investigation protocol

Analyzed blood and urine
samples
Reported results back to
participants

ATSDR

Prepared and mailed letters with results to individual
participants

Called participants with elevated results to discuss blood
lead and urine arsenic results
Provided health information
to participants with elevated
lead results

ADHS

Mailed information about how to reduce lead exposures
to participants with elevated lead results

Prepared the summary report

ATSDR

Analyzed data and wrote summary report

*Abbreviations: ATSDR, Agency for Toxic Substances and Disease Registry; ADHS, Arizona Department of Health Services;
EPA, Environmental Protection Agency; ADEQ, Arizona Department of Environmental Quality; CDC, U.S. Centers for
Disease Control and Prevention.

Table 2. Number of participants by age group, gender, and contaminant tested

Age group

Total

Total
1–5 yrs
6–11 yrs
12–19 yrs
20–40 yrs

83
25 30%
29 35%
17 20%
12 14%

Female
n=49
59% of
total
49
12 24%
19 39%
6 12%
12 24%

Male
n=34
41% of
total
34
13 38%
10 29%
11 32%
0 0%

Number of
participants
tested for
lead
83
25 30%
29 35%
17 20%
12 14%

Number of
participants
tested for
arsenic
58
0 0%
29 50%
17 29%
12 21%

* Italicized percentages in the body of the table are based on the totals in row 1 for each column category.

Table 3. Exposure investigation participant and U.S. population (NHANES) blood lead median and
95th percentile levels and confidence intervals

Age group

Number of
participants

Blood lead level (BLL) and 95% confidence intervals,
in micrograms per deciliter (µg/dL)
th
50 percentile (median)
95th percentile
ATSDR EI
NHANES
ATSDR EI
NHANES

ATSDR EI

NHANES

1–5 yrs

713

1.90*
(1.45–2.28)

0.95
(0.87–1.04)

4.56
(3.06–5.95)

2.91
(2.41–3.83)

6–11 yrs

1,048

1.30
(0.51–1.74)

0.64
(0.60–0.70)

3.56
(1.9–5.09)

1.89
(1.36–2.94)

12–19 yrs

1,129

1.20*
(0.92–1.47)

0.53
(0.49–0.57)

3.16*
(2.31–4.83)

1.31
(1.16–1.65)

20–40 yrs†

5,030

0.86
(0.26–1.19)

1.05
(1.00–1.12)

2.61
(1.2–4.15)

3.36
(2.98–3.93)

* This value and the corresponding U.S. population (NHANES) value are statistically different.
† Comparisons between the adult participant age group (women 20–40 years old) and NHANES adults (men and women 20
years and older) should be interpreted with caution due to sex and age differences.
Confidence intervals calculated using percentile bootstrap methods, n=2,000.
Abbreviations: BLL, blood lead level; ATSDR, Agency for Toxic Substances and Disease Registry; EI, Exposure Investigation;
NHANES, National Health and Nutrition Examination Survey (2011–12 data) (CDC 2015).

Table 4: Exposure investigation participant and U.S. population (NHANES) total urinary arsenic
(creatinine corrected) median and 95th percentile levels and confidence intervals

Age group

Number of
participants

Urinary total arsenic and 95% confidence intervals
(µg/g creatinine)
th
50 percentile (median)
95th percentile
ATSDR EI
NHANES
ATSDR EI
NHANES
8.3
6.87
16.36*
91.2
(14.03–19.00) (26.2–129.0)
(6.28–10.24) (5.84–8.00)

ATSDR EI

NHANES

6–11 yrs

401

12–19 yrs

392

5.6
(3.16–7.67)

4.69
(3.70–5.73)

24.14
(14.19–41.86)

34.9
(21.1–159.0)

20–40 yrs†

1,723

4.95
(1.75–7.4)

6.52
(5.88–7.69)

18.1*
(8.63–30.85)

49.7
(38.2–70.1)

*This value and the corresponding U.S. population (NHANES) value are statistically different.
† Comparisons between the adult participant age group (women 20–40 years old) and NHANES adults (men and women 20
years and older) should be interpreted with caution due to sex and age differences.
Confidence intervals calculated using percentile bootstrap methods, n=2,000.
Abbreviations: ATSDR, Agency for Toxic Substances and Disease Registry; EI, Exposure Investigation; NHANES, National
Health and Nutrition Examination Survey (2011–12 data) (CDC 2015).

Participant results

2009-10
NHANES
age specific,
uncorrected
total arsenic
90th percentile
(µg/L)

2011-12
NHANES
age specific,
uncorrected
total arsenic
90th percentile
(µg/L)

Creatinine
(mg/dL)

Total
arsenic,
corrected
(µg/g creat)

Total
arsenic,
uncorrected
(µg/L)

Investigation
follow-up
level*
total arsenic,
corrected
(µg/g creat)

Hispanic Female
Age 20–40

313.9

4.1

13.0

28.4

52.1

33.2

Hispanic Male
Age 12–19

390.0

3.1

12.0

28.4

25.9

Hispanic Male
Age 12–19

413.0

2.7

11.0

28.4

25.9

Hispanic Male
Age 12–19

419.3

4.1

17.0

28.4

25.9

Participant
ethnicity,
gender and age
group

* ATSDR compared individual total urinary arsenic results (creatinine corrected) to the exposure investigation follow-up level of
28.4 µg/g creatinine. The arsenic exposure investigation follow-up level was the lowest 95th percentile level for any age group
in the 2009–10 NHANES (the 12–19 year age group). ATSDR chose this level as a conservative screening value to identify
participants with a potentially elevated urinary arsenic level.

Table 6: Exposure investigation participant and U.S. population (NHANES) urinary inorganic-related
arsenic species 50th percentile (median) and 95th percentile levels and confidence intervals
(creatinine corrected)
Age Group*

Number of
participants
ATSDR EI

NHANES

6–11 yrs

401

12–19 yrs

392

Urinary inorganic-related arsenic species and 95% confidence
intervals (µg/g creatinine)
50th percentile (median)
95th percentile
ATSDR EI
NHANES
ATSDR EI
NHANES
8.9
7.33
15.52
20.2
(6.86–11.0) (6.95–8.26)
(12.58–18.9)
(16.8–22.3)
5.5
(2.98–7.26)

4.76
(4.40–5.11)

10.86
(2.53–17.09)

16.8
(11.4–28.7)

* ATSDR does not report inorganic-related arsenic species summary statistics for adult participants because 75% of adult
participants had levels of arsenic (V) acid (one type of inorganic arsenic), below the lab’s level of detection.
Participant inorganic-related arsenic species levels calculated using methods outlined in the CDC
Confidence intervals calculated using percentile bootstrap methods, n=2,000.
Abbreviations: ATSDR, Agency for Toxic Substances and Disease Registry; EI, Exposure Investigation; NHANES, National
Health and Nutrition Examination Survey (2011–12 data) (CDC 2015).
National Report on Human Exposure to Environmental Chemicals February 2015 Updated Tables (CDC 2015).

Table 7: Average lead and arsenic ambient air concentrations across all Hayden and Winkelman
monitoring stations in 2015 and during smelter shutdown
Average ambient air concentrations (µg/m3)

Contaminant

2015
(excluding shutdown)

April 6 – May 21, 2015
Smelter Shutdown*

P-value

Lead

0.114

0.016†

< 0.01

Arsenic

0.059

0.007†

< 0.01

* Shutdown period estimates are based on data collected April 6–May 18, 2015. Air samples were not collected daily.
† The difference between the shutdown timeframe average and 2015 average (excluding shutdown) is statistically
significant (p-value < 0.05). P-values calculated using Wilcoxon rank sum test.
Note: EPA air monitoring data (unpublished).

Table 8: Average lead and arsenic ambient air concentrations at each Hayden and Winkelman
monitoring station in 2015 and during smelter shutdown
Air
Monitor
Station

ST-01

Average ambient air concentration of
lead (µg/m3)
2015
April 6 – May
(excluding
21, 2015
P-value
shutdown)
Smelter
(Number of
Shutdown*
samples)
(Number of
samples)
0.066
0.010†
< 0.01
(14)
(6)

Average ambient air concentration of
arsenic (µg/m3)
2015
April 6 – May
(excluding
21, 2015
P-value
shutdown)
Smelter
(Number of
Shutdown*
samples)
(Number of
samples)
0.035
0.006†
0.01
(14)
(6)

ST-02

0.016
(44)

0.005†
(16)

< 0.01

0.006
(44)

0.004
(16)

0.08

ST-05

0.082
(44)

0.020
(16)

0.08

0.034
(44)

0.007†
(16)

0.01

ST-08

0.031
(10)

0.009†
(6)

0.02

0.018
(10)

0.005†
(6)

0.04

ST-09

0.056
(42)

0.007†
(16)

< 0.01

0.026
(42)

0.006†
(16)

< 0.01

ST-14

0.490
(43)

0.048†
(16)

< 0.01

0.270
(43)

0.016†
(16)

< 0.01

ST-16

0.092
(45)

0.005†
(16)

< 0.01

0.049
(45)

0.003†
(16)

< 0.01

ST-18

0.065
(43)

0.021†
(14)

< 0.01

0.030
(43)

0.010†
(14)

< 0.01

ST-23

0.086
(43)

0.016†
(15)

< 0.01

0.044
(43)

0.006†
(15)

< 0.01

ST-26

0.063
(42)

0.015†
(16)

< 0.01

0.028
(42)

0.007†
(16)

< 0.01

* Shutdown period estimates are based on data collected April 6–May 18, 2015. Air samples were not collected daily.
† The difference between the shutdown timeframe average and 2015 average (excluding shutdown) is statistically
significant (p- value < 0.05). P- values calculated using Wilcoxon rank sum test.
Notes: EPA air monitoring data (unpublished). Appendix A includes a map of air monitoring locations.

Table 9: Exposure investigation participant and U.S. population urinary inorganic-related arsenic
species 50th percentile (median) and 95th percentile levels and confidence intervals

Age Group†

Number of
participants
ATSDR EI

NHANES

6–11 yrs

401

12–19 yrs

392

Urinary inorganic-related arsenic species and 95% confidence
intervals (µg/L)
th
50 percentile (median)
95th percentile
ATSDR EI
NHANES
ATSDR EI
NHANES
11.9*
5.36
13.4
24.88
(9.65–15.21) (4.50–5.98)
(21.02–30.79)* (11.6–16.2)
10.6*
5.09
(8.49–12.71) (4.24–6.00)

16.74
(15.00–19.61)

15.6
(10.8–23.9)

*This value and the corresponding U.S. population (NHANES) value are statistically different.
† ATSDR does not report inorganic-related arsenic species summary statistics for adult participants because 75% of adult
participants had levels of arsenic (V) acid (one type of inorganic arsenic), below the lab’s level of detection.
Confidence intervals calculated using bootstrap methods, n=2,000.
Abbreviations: ATSDR, Agency for Toxic Substances and Disease Registry; EI, Exposure Investigation; NHANES, National
Health and Nutrition Examination Survey (2011–12 data) (CDC 2015).

Authors
Ben Gerhardstein, MPH
Environmental Health Scientist, Region 9
Division of Community Health Investigations
Bruce C. Tierney, MD
Captain, U.S. Public Health Service
Senior Medical Officer
Data Analysis and Exposure Investigation Team, Science Support Branch
Division of Community Health Investigations
Barbara Anderson, PE, MSEn E
Environmental Health Scientist
Data Analysis and Exposure Investigation Team, Science Support Branch
Division of Community Health Investigations
Jamie Rayman
Health Educator, Region 9
Division of Community Health Investigations

Contributors
James Durant, MSPH, CIH
Science Support Branch, Data Analysis and Exposure Investigation Team
Division of Community Health Investigations
Nina Dutton, MPH
ORISE Research Participant
Geospatial Research, Analysis and Ser ices Program
Division of Toxicology and Human Health Sciences
Bradley Goodwin, PhD
Lieutenant, U.S. Public Health Service
Science Support Branch, Data Analysis and Exposure Investigation Team
Division of Community Health Investigations
Aaron Grober
Lieutenant, U.S. Public Health Service
Health Service Officer
Science Support Branch, Data Analysis and Exposure Investigation Team
Division of Community Health Investigations

Michael Wellman, MS
Situation Awareness Team
Division of Emergency Operations
Office of Public Health Preparedness and Response (OPHPR)
Efomo Woghiren
Geospatial Research, Analysis and Services Program
Division of Toxicology and Human Health Sciences

Acknowledgements
This exposure investigation was a collaborative effort among federal, state, and local
organizations. ATSDR appreciates the assistance and cooperation of the Arizona Department of
Health Services, Arizona Department of Environmental Quality, U.S. Environmental Protection
Agency, Hayden-Winkelman Unified School District, City of Hayden, City of Winkelman, Central
Arizona Governments, Gila County Board of Supervisors, Gila County Health Department, Pinal
County Health Department, Pinal County Supervisor Pete Rios, Hayden Senior Center, and St.
Joseph’s Catholic Church. Finally, ATSDR thanks the Hayden and Winkelman community
members who participated in this exposure investigation.

References
[ACCLPP 2012] Advisory Committee on Childhood Lead Poisoning Prevention (ACCLPP) of the
Centers for Disease Control and Prevention (2012). Low level lead exposure harms children: a
renewed call for primary prevention. Available at
http://www.cdc.gov/nceh/lead/acclpp/final_document_030712.pdf. Last visited September 13,
2016.

[ADHS 2002] Arizona Department of Health Services (2002). Public Health Assessment, Asarco
Hayden Smelter Site. Available at
http://www.atsdr.cdc.gov/HAC/pha/PHA.asp?docid=905&pg=0. Last visited September 13,
2016.
[ADHS 2016] Arizona Department of Health Services (2016). Report Blood Lead Test Results.
Available at http://azdhs.gov/preparedness/epidemiology-disease-control/childhoodlead/index.php#blood-lead-test-results. Last visited September 13, 2016.
Arizona Water Company (2014). Annual Water Quality Report for Winkelman, Arizona, PWSID
#04-003. Available at http://azwater.com/files/water-quality/ccr-winkelman-2014.pdf. Last
visited September 13, 2016.
Arizona Water Company (2015). Annual Water Quality Report for Winkelman, Arizona, PWSID
#04-003. Available at http://azwater.com/files/water-quality/ccr-winkelman-2015.pdf. Last
visited September 13, 2016.
[Asarco 2014] Asarco Groupo Mexico, Asarco LLC, Hayden Concentrator (2014). Annual Drinking
Water Quality Report for PWS# 04-012. Available at http://www.asarco.com/wpcontent/uploads/14rpt-Consumer-Confidence-Report.pdf. Last visited September 13, 2016.
[Asarco 2015a] Asarco Groupo Mexico (2015). Hayden Operations. Available at
http://www.asarco.com/about-us/our-locations/hayden-operations/. Last visited September
13, 2016.
[Asarco 2015b] Asarco Groupo Mexico, Asarco LLC, Hayden Concentrator (2015). Annual
Drinking Water Quality Report for PWS# 04-012. Available at: http://www.asarco.com/wpcontent/uploads/2015-Annual-Drinking-Water-Quality-Report.pdf. Last visited September 13,
2016.
[ATSDR 2007a] Agency for Toxic Substances and Disease Registry (2007). Toxicological Profile
for Arsenic. Available at: http://www.atsdr.cdc.gov/toxprofiles/tp2.pdf. Last visited September
13, 2016.

[ATSDR 2007b] Agency for Toxic Substances and Disease Registry. (2007). Toxicological profile
for lead (update). Available at http://www.atsdr.cdc.gov/toxprofiles/tp13.pdf. Last visited
September 13, 2016.
[ATSDR 2007c] Agency for Toxic Substances and Disease Registry. (2007). Public Health
Statement for Arsenic. Available at https://www.atsdr.cdc.gov/phs/phs.asp?id=18&tid=3. Last
visited October 5, 2016.
[ATSDR 2015a] Agency for Toxic Substances and Disease Registry (2015). Lead and Arsenic
Testing for Hayden and Winkelman Residents Spring 2015. Available at
http://www.atsdr.cdc.gov/sites/HWAZ/docs/Hayden%20Testing%20Fact%20Sheet%20Final.pdf
. Last visited September 13, 2016.
[ATSDR 2015b] Agency for Toxic Substances and Disease Registry (2015). Asarco Hayden
Exposure Investigation. Available at http://www.atsdr.cdc.gov/sites/HWAZ/. Last visited
September 13, 2016.
[ATSDR 2015c] Agency for Toxic Substances and Disease Registry (2015). Ways to Reduce
Exposure to Lead and Arsenic and Protect Your Health in Hayden and Winkelman, Arizona
Available at
http://www.atsdr.cdc.gov/sites/HWAZ/docs/Hayden%20Ways%20To%20Reduce%20Exposure
%20Final.pdf. Last visited September 13, 2016.
[ATSDR 2015d] Agency for Toxic Substances and Disease Registry (2015). Biological Testing for
Exposure to Lead and Arsenic near Colorado Smelter, Pueblo, Colorado. Available at:
http://www.atsdr.cdc.gov/HAC/pha/ColoradoSmelter/ColoradoSmelter_%20HCEI%20(final)_%2009-10-2015_508.pdf. Last visited September 13, 2016.
[ATSDR 2016] Agency for Toxic Substances and Disease Registry (2016). Addendum to the
Toxicological Profile for Arsenic. Available at:
http://www.atsdr.cdc.gov/toxprofiles/Arsenic_addendum.pdf. Last visited September 13, 2016.
Barbosa F, Tanus-Santos JE, Gerlach RF, Parsons PJ (2005). A critical review of biomarkers used
for monitoring human exposure to lead: Advantages, limitations and future needs.
Environmental Health Perspective, 113:1669-1674.
Barr DB, Wilder LC, Caudill SP, Gonzalez AJ, Needham LL, Pirkle JL (2005). Urinary Creatinine
Concentrations in the U.S. Population: Implications for Urinary Biologic Monitoring
Measurements. Environmental Health Perspectives. 113(2): 192–200. Available at:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1277864/. Last visited September 13, 2016.
Bernard SM, McGeeing MA, Michael A (2003). Prevalence of Blood Lead Levels greater than
5μg/dL Among U.S. Children 1 to 5 Years of Age and Socioeconomic and Demographic Factors

associated with Blood Lead Levels 5 to 10μg/dL, Third National Health and Nutrition
Examination Survey, 1988–1994. Pediatrics. 112(6).
Burgess JL, Carter DE, O’Rouke MK (2000). Hayden-Winkelman Arsenic and Lead Survey.
Available at: http://s3.documentcloud.org/documents/267080/hayden-winkelman-arsenic-andlead-survey.pdf. Last visited September 13, 2016.
[Census 2010a] United States Census Bureau. (2010). 2010 Census.
[Census 2010b] United States Census Bureau. (2010). 2006–2010 American Community Survey.
[CDC 2012a] Centers for Disease Control and Prevention (2012). What Do Parents Need to
Know to Protect Their Children? Update on Blood Lead Levels in Children. Available at
http://www.cdc.gov/nceh/lead/acclpp/blood_lead_levels.htm. Last visited September 13,
2016.
[CDC 2012b] Centers for Disease Control and Prevention (2012). CDC Response to Advisory
Committee on Childhood Lead Poisoning Prevention Recommendations in “Low Level Lead
Exposure Harms Children: A Renewed Call of Primary Prevention.” Available at
https://www.cdc.gov/nceh/lead/acclpp/cdc_response_lead_exposure_recs.pdf. Last visited
September 13, 2016.
[CDC 2013] Centers for Disease Control and Prevention (2013). Blood Lead Levels in Children
Aged 1–5 Years — United States, 1999–2010. Morbidity and Mortality Weekly Report
62(13);245-248.
[CDC 2014a] Centers for Disease Control and Prevention (2014). Laboratory Procedure Manual.
Blood Metals Panel. Method DLS 3016.8. Available at
http://www.cdc.gov/nchs/data/nhanes/nhanes_13_14/PbCd_H_MET.pdf. Last visited
September 13, 2016.
[CDC 2014b] Centers for Disease Control and Prevention (2014). Laboratory Procedure Manual.
Urine Multi-Element. Method 3018A.4. Available at
http://www.cdc.gov/Nchs/Data/Nhanes/Nhanes_13_14/UM_UMS_UTAS_UTASS_H_MET.pdf.
Last visited September 13, 2016.
[CDC 2014c] Centers for Disease Control and Prevention (2014). Laboratory Procedure Manual.
Urine Arsenic Speciation. Method DLS 3000.14. Available at
http://www.cdc.gov/Nchs/Data/Nhanes/Nhanes_13_14/UAS_UASS_H_MET.pdf. Last visited
September 13, 2016.
[CDC 2015] Centers for Disease Control and Prevention. Fourth Report on Human Exposure to
Environmental Chemicals, Updated Tables (February, 2015). Atlanta, GA: U.S. Department of

Health and Human Services, Centers for Disease Control and Prevention.
http://www.cdc.gov/exposurereport/. Last visited September 13, 2016.
Dixon SL, Gaitens JM, Jacobs DE, Strauss W, Nagaraja J, Pivets T, Wilson JW, Ashley PJ (2009).
Exposure of U.S. Children to Residential Dust Lead, 1999–2004: II. The contribution of lead
contaminated dust to children’s blood lead levels. Environmental Health Perspectives.
117(3):468-474.
[EPA 1996] Environmental Protection Agency (1996). EPA Takes Final Steps in Phase-out of
Leaded Gasoline. Press Release. Available at https://archive.epa.gov/epa/aboutepa/epa-takesfinal-step-phaseout-leaded-gasoline.html. Last visited September 13, 2016.
[EPA 2008] Environmental Protection Agency (2008). Asarco Hayden Plant. Phase 1 Remedial
Investigation Report. Available at
http://yosemite.epa.gov/r9/sfund/r9sfdocw.nsf/3dc283e6c5d6056f88257426007417a2/4c339
2df16ad6d4c882574b8006f1dbc!OpenDocument. Last visited September 13, 2016.
[EPA 2012] Environmental Protection Agency (2012). Final Phase II Remedial
Investigation/Feasibility Study Work Plan Part 1 or 2 (Air), Appendix B – Sampling and Analysis
Plan Ambient Air and Source Characterization, Part 2 – Field Sampling Plan.
[EPA 2013] Environmental Protection Agency (2013). Integrated Science Assessment for Lead.
Office of Research and Development, National Center for Environmental Assessment.
Research Triangle Park, NC. EPA/600/R-10/075F.
[EPA 2014a] Environmental Protection Agency (2014). Designation of Areas for Air Quality
Planning Purposes; State of Arizona; Pinal County and Gila County; Pb. Federal Register Notice:
Final Rule. Available at http://www.regulations.gov/document?D=EPA-R09-OAR-2014-02660041. Last visited September 13, 2016.
[EPA 2014b] Environmental Protection Agency (2014). Redesignation of the Hayden, Arizona
Area to Nonattainment for the Lead (Pb) Standards, August 20, 2014. Available at
http://www.epa.gov/region9/air/actions/pdf/az/hayden/epa-r09-oar-2014-0266-factsheethayden-nfr-2014-08-20.pdf. Last visited October 5, 2016.
[EPA 2015a] Environmental Protection Agency (2015). Asarco Hayden Plant. Available at
http://yosemite.epa.gov/r9/sfund/r9sfdocw.nsf/ViewByEPAID/AZD008397127. Last visited,
September 13, 2016.
[EPA 2015b] Environmental Protection Agency (2015). Residential Soil Removal Action Report,
Towns of Hayden and Winkelman, Arizona.
[EPA 2015c] Environmental Protection Agency (2015). Green Book Designations. Available at
http://www.epa.gov/airquality/greenbook/define.html. Last visited September 13, 2016.
51

[EPA 2015d] Environmental Protection Agency (2015). ASARCO LLC Settlement. Available at
http://www.epa.gov/enforcement/asarco-llc-settlement. Last updated on November 3, 2015.
[EPA 2015e] Environmental Protection Agency (2015). Positive Matrix Factorization Model for
environmental data analyses. Available at https://www.epa.gov/air-research/positive-matrixfactorization-model-environmental-data-analyses. Last updated on August 20, 2015.
[EPA 2016a] Environmental Protection Agency (2016). Lead (2008) Designated Area
State/Area/County Report. Available at
https://www3.epa.gov/airquality/greenbook/mbcs.html#AZ. Last visited October 5, 2016.
[EPA 2016b] Environmental Protection Agency (2016). Lead and Copper Rule. Available at
https://www.epa.gov/dwreginfo/lead-and-copper-rule. Last visited October 5, 2016.
[EPA 2017] Environmental Protection Agency (2017). Superfund Alternative Approach. Available
at https://www.epa.gov/enforcement/superfund-alternative-approach. Last updated January
27, 2017.
Haley & Aldrich, Inc. (2014). Ambient Air Quality Network Design Plan Phase II Remedial
Investigation/Feasibility Study Asarco Hayden Plant Site Addendum No. 1.
Hinwood AL, Sim MR, de Klerk N, Drummer O, Gerostamoulos J, Bastone EB (2002). Are 24-hour
urine samples and creatinine adjustment required for analysis of inorganic arsenic in urine in
population studies? Environ Res Section A. 88, 219–224.
Hornung RW and Reed LD (1990). Estimation of average concentration in the presence of
nondetectable values. Appl Occup Environ Hyg 5(1):46-51.
Hysong TA, Burgess JL, Garcia MEC, O’Rourke MK (2003). House dust and inorganic urinary
arsenic in two Arizona mining towns. J. Expo. Anal. Environ. Epidemiol. 13,211–218.
Jones RL, Homa DM, Meyer PA, Brody DJ, Caldwell KL, Pirkle JL, Brown MJ (2009). Trends in
Blood Lead Levels and Blood Lead Testing Among U.S. Children aged 1- 5 Years, 1988-2004.
Pediatrics. 123(3).
Krzywinski M and Altman N. (2013) Error Bars. Nature Methods. Vol. 10 No. 10.
Loh MM, Sugeng A, Lothrop N, Klimecki W, Cox M, Wilkinson ST, Lu Z, Beamer PI (2016).
Multimedia exposures to arsenic and lead for children near an inactive mine tailings and
smelter site. Environ. Res. 146, 331–339

[NIOSH 2015] National Institute for Occupational Safety and Health. Adult Blood Lead
Epidemiology and Surveillance (ABLES). (2015). Available at:
http://www.cdc.gov/niosh/topics/ables/description.html. Last visited October 19, 2016.
[OEHHA 2014] Office of Environmental Health Hazard Assessment. California Environmental
Protection Agency. 2014. Notice of Adoption of Revised Reference Exposure Levels for Benzene.
Available at http://oehha.ca.gov/air/crnr/notice-adoption-revised-reference-exposure-levelsbenzene. Last visited October 5, 2016.
[OEHHA 2016] Office of Environmental Health Hazard Assessment. California Environmental
Protection Agency. 2016. Arsenic (inorganic). Available at
http://oehha.ca.gov/air/chemicals/arsenic-inorganic. Last visited October 5, 2016.
Orloff K, Mistry K, Metcalf S. (2009). Biomonitoring for Environmental Exposures to Arsenic.
Journal of Toxicology and Environmental Health, Part B: Critical Reviews. 12:7, 509524.
R Core Team. (2015). R: A Language and Environment for Statistical Computing. Vienna, Austria:
R Foundation for Statistical Computing. Available at https://www.R-project.org/. Last visited
September 13, 2016.
Wang YX, Feng W, Zeng Q, Sun Y, Wang P, You L, Yang P, Huang Z, Yu SL, Lu WQ (2016).
Variability of Metal Levels in Spot, First Morning, and 24-Hour Urine Samples over a 3-Month
Period in Healthy Adult Chinese Men. Environmental Health Perspectives. 124(4). Available at
http://ehp.niehs.nih.gov/1409551/. Last visited September 13, 2016.

Appendix A: Map of ambient air monitoring stations in Hayden and Winkelman, Arizona

Adapted from Haley & Aldrich (2014).
Note: Air monitor station 8 is not shown on this map.

Appendix B: Positive Matrix Factorization Model Results
Figure B.1 Relative contribution of the smelter source to PM10 air contaminant levels at each Hayden and Winkelman air monitoring station
July 2013–June 2015
20

Relative Smelter Factor Contribution

Shutdown

Station 14 Minivol

Station 14 Partisol

Station 1 Partisol

Station 2 Minivol

Station 2 Partisol

Station 5 Minivol

Station 5 Partisol

Station 9 Minivol

Station 9 Partisol

Station 16 Minivol

Station 16 Partisol

Station 18 Minivol

Station 18 Partisol

Station 23 Minivol

Station 23 Partisol

Station 26 Minivol

Station 26 Partisol

Notes: Relative factors calculated using the EPA Positive Factorization Model (EPA 2015e). EPA air monitoring data (unpublished).

Figure B.2: Relative contribution of background sources to PM10 air contaminant levels at each Hayden and Winkelman air monitoring station
July 2013–June 2015
20

Relative Background Factor Contribution

Shutdown

Station 14 Minivol

Station 14 Partisol

Station 1 Partisol

Station 2 Minivol

Station 2 Partisol

Station 5 Minivol

Station 5 Partisol

Station 9 Minivol

Station 9 Partisol

Station 16 Minivol

Station 16 Partisol

Station 18 Minivol

Station 18 Partisol

Station 23 Minivol

Station 23 Partisol

Station 26 Minivol

Station 26 Partisol

Notes: Relative factors calculated using the EPA Positive Factorization Model (EPA 2015e). EPA air monitoring data (unpublished).

Appendix C: Exposure Investigation Participant and U.S. Population Urinary Creatinine Levels
Figure C.1: Comparison of exposure investigation participant and U.S. population (2011–12) median urinary
creatinine levels by age group

Figure C.2 U.S. population (2011–12) median urinary creatinine levels by age, race, and gender

Exposure investigation participant and historical (1988–94) U.S. population creatinine levels
Exposure investigation participant’s creatinine levels were also higher than historical U.S. population levels.
Median age group specific creatinine levels for the NHANES III (1988-94) U.S. population were 98.1, 150.2,
153.8, and 128.8 mg/dL for the 6–11, 12–19, 20–29, and 30–39 age groups respectively. Median age group
specific creatinine levels for the NHANES III (1988-1994) Mexican American population were 88.0; 140.0, 148.9,
and 132.4 mg/dL for the 6–11, 12–19, 20–29: and 30–39 age groups respectively (Barr et al. 2005).
58

Appendix D: Asarco Hayden Exposure Investigation Report Summary

Asarco Hayden Smelter
Exposure Investigation

March 2017

A Summary of Findings

Hayden and Winkelman, Arizona
Overview
People in Hayden and Winkelman might be exposed
to (come in contact with) unhealthy levels of lead
and arsenic in the outdoor air, in mine waste piles,
and in soil in some non-residential locations.
Additionally, they may be exposed to lead from
paint in older housing.
In April 2015, the Agency for Toxic Substances and
Disease Registry (ATSDR) worked with federal and
state agencies, and local leaders to test people in
Hayden and Winkelman for levels of lead and arsenic
in their bodies. Residents most at-risk for negative
health effects from exposure (children, pregnant
women, and women of childbearing age) were
eligible for testing. ATSDR sent participants their
individual results in June 2015.
This is a summary of the full ATSDR report.

Conclusions
Some children in Hayden and Winkelman have
been exposed to lead at levels that could harm
their health.
Overall, the children and adolescents ATSDR
tested in Hayden and Winkelman had higher
levels of lead in their bodies than children and
adolescents from across the U.S.
ATSDR needs more information to determine
how much arsenic participants have in their
bodies when air pollution levels are typical for
the community. Asarco shut down the smelter
for maintenance during the time of ATSDR’s
testing, reducing lead and arsenic levels in the air.

Agency for Toxic Substances and Disease Registry
Division of Community Health Investigations
CS274068-A

A Note About the Tests
These tests tell us how much
lead and arsenic were in a
participant’s blood and urine
at the time of testing. They
don’t tell us where the lead and
arsenic came from. The amount
of lead and arsenic in a person’s
body can change over time.

Asarco Hayden Smelter Exposure Investigation

Background
The Asarco Hayden Smelter Plant Site is in rural Arizona, about 90 miles southeast of Phoenix and 70 miles
northeast of Tucson. The site includes the towns of Hayden and Winkelman (population 662 and 353,
respectively). Past and current copper smelting and processing caused environmental contamination in
these towns. Copper ore has been processed here for over 100 years. Asarco continues to operate a
copper concentrator and smelter.
The U.S. Environmental Protection Agency (EPA), Arizona Department of Environmental Quality (ADEQ),
and Asarco Grupo Mexico LLC (Asarco) are cleaning up contamination at the site through a Superfund
alternative process. Between 2008 and 2014, EPA completed residential soil clean up at 266 Hayden
and Winkelman yards and publicly accessible areas. Separate from this process, in 2015 EPA and Asarco
announced a legal settlement to resolve Clean Air Act violations at the facility.

ATSDR Exposure Investigation Process
Based on requests from the community, EPA asked
ATSDR to provide lead and arsenic testing to Hayden
and Winkelman residents. In April 2015, ATSDR offered
testing to at-risk residents with support from the
Arizona Department of Health Services (ADHS), ADEQ,
the Centers for Disease Control and Prevention (CDC),
and EPA. ATSDR mailed individual results letters to
participants in June 2015 and made follow up phone
calls to participants whose blood lead results were
higher than the investigation follow up level (see Box 1).

Box 1. Lead Follow Up Level
ATSDR followed up with participants
with blood lead levels above 5. CDC uses
5 to identify children with blood lead
levels that are higher than most children’s
levels. The units are micrograms per
deciliter of blood, abbreviated µg/dL.

Exposure Investigation
Report
What did ATSDR do?
ATSDR offered free, voluntary blood lead and urine
arsenic testing to children, pregnant women, and
women of childbearing age living in Hayden and
Winkelman. These people are most at-risk from
exposure because they are growing and developing
or may become pregnant. ATSDR tested a total of 83
participants from 29 households. We tested:

ʶʶ 25 children ages 1 – 5 years for lead;
ʶʶ 29 children ages 6 – 11 years for lead and arsenic;
ʶʶ 17 adolescents ages 12 – 19 years for lead and arsenic;
ʶʶ 12 women ages 20 – 40 years for lead and arsenic.
ATSDR also looked at air monitoring data for Hayden
and Winkelman from 2013 and 2015 to learn how a
shutdown of the smelter affected air quality at the time
of the testing. The air data were collected by Asarco
with EPA oversight.
Page 2 of 6

Asarco Hayden Smelter Exposure Investigation
What did ATSDR find?

Lead

Figure 1: Comparison of blood lead levels in
children and adolescents in ATSDR’s exposure
investigation to those in the U.S. population

Two children had blood lead levels
above 5.
ʶʶ ATSDR found 2 children in Hayden and

Winkelman with blood lead levels above the
investigation follow up level [(5 micrograms
of lead per deciliter of blood (µg/dL)].

ʶʶ One child was in the age range 1 – 5 and
one child was in the age range 6 – 11.

ʶʶ Two other children had blood lead levels

between 4 – 5 µg/dL, near the investigation
follow up level.

Children’s (including adolescents’) blood
lead levels were above the U.S. median.
ʶʶ The median blood lead levels by age group
for children and adolescent participants
were about two times higher than the U.S.
population age groups. See Figure 1 and Box 3.

Adult blood lead levels were lower than
the U.S. median.
ʶʶ Median blood lead levels of adult participants
(women age 20-40) in Hayden-Winkelman were
slightly lower than adults 20 years and older
from across the U.S.

*Centers for Disease Control and Prevention. Fourth Report
on Human Exposure to Environmental Chemicals, Updated
Tables, (February, 2015). Atlanta, GA: U.S. Department of
Health and Human Services, Centers for Disease Control
and Prevention. www.cdc.gov/exposurereport/

Box 2. Lead and Your Health
Lead exposure can cause learning and
behavior problems in children and many
other health effects. Some of the effects
of exposure to lead may never go away.
Lead can stay in your body for many years
after exposure.

Box 3. What is the median?
The median is the middle value in a list of
numbers. In a set of numbers it separates
the higher half from the lower half.

Page 3 of 6

Asarco Hayden Smelter Exposure Investigation

Arsenic
ATSDR needs more information to determine
participants’ urinary arsenic levels when air
pollution is typical for the area.
ʶʶ The smelter was shut down before and during the urine

testing, so participant arsenic levels may have been lower
than they would be typically.

Urinary arsenic levels for all participants were
similar to U.S. population results.
ʶʶ No participants had urinary arsenic levels above the follow

up level (28.4 micrograms of arsenic per gram of creatinine).
You can read more about the follow up level ATSDR chose
for arsenic in the methods section of the full report.

Air Quality
Due to the smelter shutdown, outdoor air
pollutant levels were lower before and
during the testing.
ʶʶ In 2015, typical outdoor air pollutant levels in Hayden
and Winkelman were
• 7 times higher for lead and
• 8 times higher for arsenic than during the smelter
shutdown.
ʶʶ The smelter was shut down for maintenance from
April 6 – May 21, 2015.

ʶʶ ATSDR collected participants’ blood and urine samples
April 17 – 19, 2015.

Other possible lead and arsenic sources
ʶʶ Housing: About 44% of the housing units were built

before 1950. Prior to 1955 there were no limits on lead in
paint. Lead was widely used in house paint until the early
1980s. If paint in older housing is deteriorating, children in
those homes are at greater risk for higher blood lead levels.

Box 4. Arsenic and Your Health
Exposure to low levels of arsenic for
more than 1 year can cause dark
patches of “warts” or “corns” on the
skin. Arsenic exposure over many
years also raises the risk of cancer
of the skin, bladder, lung, and liver.
Arsenic stays in your urine for about
3 days after exposure.

Box 5. The smelter shutdown
and ATSDR’s testing results
Since arsenic leaves the body within
a few days, the lower level of arsenic
in air before testing could have led to
less than typical amounts of arsenic
in participants’ urine during testing.
Since lead stays in blood longer, we
expect the shutdown did not have
much of an effect on lead results.

ʶʶ Hayden and Winkelman drinking water systems:
• Arsenic: Although both systems contain low levels of arsenic, they are below the EPA limits and at typical
•

levels seen in other Arizona public water systems.
Lead: Both systems are well below EPA’s action level for lead. Still, lead can get into drinking water from
pipes and fixtures in your home. Homes built before 1986 are more likely to have plumbing with lead.

ʶʶ Mexican imports: Some types of imported Mexican pottery and candies may contain lead.
ʶʶ Foods: Some foods such as rice and seafood contain arsenic.
Page 4 of 6

Asarco Hayden Smelter Exposure Investigation
What can I do if my child has high blood lead levels?
ATSDR recommends that parents/guardians of the two children
whose blood lead results were above the follow up level discuss
the child’s result with their primary health care provider. Follow
the tips below to reduce your family’s exposure to lead.
ATSDR further recommends that health care providers follow the
Advisory Committee for Childhood Lead Poisoning Prevention’s
recommendations for management of children with blood lead
levels above the CDC reference level.

How can my family reduce exposure to lead
and arsenic?
Families and people in Hayden and Winkelman can take the
following steps to protect their health.

Keep dirt and dust from getting into your body.
Outside

ʶʶ Don’t play in arroyos or on waste piles; stay away from railroad
tracks in Hayden; do not trespass.

At Home

ʶʶ Wipe shoes on a doormat and remove shoes before entering
your house.
ʶʶ Wet-mop or wet-wipe floors, windowsills, counters and
hard-surface furniture every 2 – 3 weeks.
ʶʶ Make sure your child does not chew on surfaces painted
with lead-based paint.

Keep things clean

ʶʶ Wash things children put into their mouths, such as pacifiers,

bottles, and toys whenever they fall on the floor or ground.
ʶʶ Wash your hands and your children’s hands before eating and
after being outside.
ʶʶ Wash fruits, vegetables, and root crops (like potatoes) before
preparing them to eat.

At work

ʶʶ If you could be exposed to lead or arsenic in your workplace,
change your clothes at work before returning home or
immediately after arriving home.
ʶʶ Wash your work clothes separately from the clothes of other
family members.

Pets

ʶʶ Wash pets that spend time outside and inside your home at
least every 2-3 weeks.

Page 5 of 6

Asarco Hayden Smelter Exposure Investigation
Maintain healthy eating habits for
your family.
ʶʶ Give your family healthful meals rich in iron,

calcium, zinc, and vitamin C. Children who eat
healthy diets absorb less lead.

Participate in the home lead-based
paint testing project Asarco will develop
and fund as part of their 2015 legal
settlement with EPA.
To learn about the status of this project, contact Amy
Veek at Asarco (520-356-3296, aveek@asarco.com).

What will happen next?
To make sure the community is safe, ATSDR
recommends that EPA, ADEQ, Gila County Health
Department, and Asarco:

ʶʶ Make changes to the smelter to reduce lead

and arsenic in outdoor air.
ʶʶ Continue environmental sampling and cleanup
efforts in Hayden and Winkelman.
ʶʶ Incorporate these exposure investigation
results in future human health risk assessments,
as appropriate.
ʶʶ Implement a home lead testing and abatement
project for local residents, as outlined in the
legal settlement between EPA and Asarco.

ATSDR will also:
ʶʶ Plan to offer another round of arsenic testing for
existing participants at a time when the smelter
is expected to be operating normally.
ʶʶ Continue to support the development and
implementation of the Hayden and Winkelman
lead-based paint testing and abatement project
outlined in the 2015 EPA/Asarco settlement.
ʶʶ Give information about lead and arsenic testing
to local doctors or nurses, upon request.

Where can I learn more?

ʶʶ Visit the Asarco Hayden exposure investigation

webpage. http://www.atsdr.cdc.gov/sites/HWAZ/

ʶʶ Check out the full report. http://www.atsdr.cdc.
gov/HAC/PHA/HCPHA.asp?State=AZ

ʶʶ Learn more about lead. http://www.atsdr.cdc.
gov/toxfaqs/tf.asp?id=93&tid=22

ʶʶ Read about arsenic. http://www.atsdr.cdc.gov/
toxfaqs/tf.asp?id=19&tid=3

Contacts
ATSDR
Dr. Bruce Tierney, Medical Officer,
btierney@cdc.gov & (770) 488-0771
Ben Gerhardstein, Environmental Health
Scientist, bgerhardstein@cdc.gov &
(415) 947-4316
Jamie Rayman, Health Educator,
jrayman@cdc.gov & (415) 947-4318

Page 6 of 6

File Type	application/pdf
File Title	Asarco Hayden Smelter Site Biological Testing for Exposure to Lead and Arsenic
Subject	Asarco, Arizona, Lead exposure, Hayden, Winkelman
Author	ATSDR
File Modified	2017-03-24
File Created	2017-03-24