CDC Standards for Nationally Consistent Data Measures

Att12_NCDMrequirementsDocument.pdf

Environmental Public Health Tracking Network (Tracking Network)

CDC Standards for Nationally Consistent Data Measures

OMB: 0920-1175

Document [pdf]
Download: pdf | pdf
Centers for Disease Control and Prevention Standards
for Nationally Consistent Data and Measures within
the Environmental Public Health Tracking Network
Version 3.0
June 20, 2013

Environmental Health Tracking Branch
Division of Environmental Hazards and Health Effects
National Center for Environmental Health
Centers for Disease Control and Prevention

Foreword
This document was first published in March, 2008, setting the standards for the first Nationally
Consistent Data and Measures (NCDMs) for the National Environmental Health Tracking
Program. The purpose of these NCDMS was to ensure compatibility and comparability of data
and measures useful for understanding the impact of our environment on our health. Version 2.0
 reflect the lessons learned in implementing the first NCDMs across local, state, and
national tracking networks
 improve the utility of specific measures
 identify recommended temporal and spatial resolution, specifically for health outcomes,
based on confidentiality protection needs and data steward requests
Specific updates included in version 2 include:






Clarified description of process for creating and adopting the first set of NCDMs
Clarified the meaning of indicator, measure, and data within the Tracking Network
Added columns to the table summarizing the indicators and measures in order to identify
o minimum temporal and geographic resolution
o data source
o grantee requirements
Updated indicator templates to reflect minimum temporal and geographic resolution at
which measures are to be displayed on public portals

Version 3.0 includes a change from required to optional for the Fertility indicator and
documentation for NCDMs adopted since the release of version 2 in August 2011.





Hospitalizations and ED visits for heat
ED visits for asthma
Blood lead levels by birth cohort and annual blood lead levels
Updates to drinking water NCDMs

NCDM Recommendations Version 3.0
Page 2 3/19/2013

Table of Contents
Introduction
Section 1: Summary of Environmental Public Health Tracking’s Standards for Nationally
Consistent Data and Measures (NCDMs)
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.

Acute Myocardial Infarction
Air Quality
Asthma
Birth Defects
Cancer
Carbon Dioxide Poisoning
Childhood Lead Poisoning
Drinking Water
Heat
Reproductive Health Outcomes

Section 2: Indicator Templates
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.

Acute Myocardial Infarction
Air Quality
Asthma
Birth Defects
Cancer
Carbon Dioxide Poisoning
Childhood Lead Poisoning
Drinking Water
Heat
Reproductive Health Outcomes

NCDM Recommendations Version 3.0
Page 3 3/19/2013

Introduction
Environmental Public Health Tracking is the ongoing collection, integration, analysis,
interpretation, and dissemination of data from environmental hazard monitoring, human
exposure, and health effects surveillance. In financial year 2002, Congress appropriated funds to
the Centers for Disease Control and Prevention (CDC) to develop a national environmental
public health tracking network and to improve environmental health capacity at the state and
local level.
CDC established its National Environmental Public Health Tracking Program with the following
goals:
1. Build a sustainable national environmental public health tracking network (Tracking
Network);
2. Enhance environmental public health tracking workforce and infrastructure;
3. Disseminate information to guide policy, practice, and other actions to improve the
Nation’s health;
4. Advance environmental public health science and research;
5. Foster collaboration among health and environmental programs.
In 2006, CDC transitioned from a piloting and planning phase to implementation. The network
was envisioned as a web-based, secure, distributed network of standardized electronic health and
environmental data. Sixteen states and New York City were funded in August 2006 to construct
state-wide (city-wide) networks that will be components of the national network and to
participate in a collaborative process to develop network standards development process.
Additional funding from Congress allowed CDC to add 6 more states in 2009 and 1 in 2010.
As part of the implementation process, CDC established a Content Work Group (CWG) to:
1. Identify and recommend core measures for the Tracking Network;
2. Examine the availability and applicability of existing data and identify approaches for
deriving or collecting needed data;
3. Identify and adapt standards and guidelines to facilitate nationally consistent data
collection and ensure compatibility with existing standards efforts;
4. Recommend metadata elements to describe data quality;
5. Identify and recommend methods and tools for data integration, analysis and
presentation.
The CWG structure included a steering group made up of the principal investigators for grantee
health departments and academic partners. Content-specific teams advised the steering group
These teams included content experts from: grantee states, cities and academic partners; nonfunded states and cities; CDC; other government agencies including the Environmental
Protection Agency (EPA), the National Aeronautics and Space Administration (NASA), the US
Geological Survey (USGS) and the National Institutes of Health (NIH); and non-governmental
organizations including the American Association of Poison Control Centers (AAPCC), the
National Birth Defects Prevention Network (NBDPN), the National Association of Health Data
Organizations (NAHDO), the National Association for Public Health Statistics and Information
NCDM Recommendations Version 3.0
Page 4 3/19/2013

Systems (NAPHSIS) and the North American Association of Central Cancer Registries
(NAACCR).
Eight content teams were established, and each provided recommendations to CDC via the
steering group for an initial set of Nationally Consistent Data and Measures (NCDMs)( Figure
1). NCDMs consist of measures, grouped by indicators, and the data required to generate them.
A measure is a summary characteristic or statistic, such as a sum, percentage, or rate. There may
be several measures of a specific indicator which when considered in conjunction fully describe
the indicator. An indicator is one or more items, characteristics or other things that will be
assessed and that provide information about a population's health status, their environment, and
other factors with the goal allowing us to monitor trends, compare situations, and better
understand the link between environment and health. It is assessed through direct and indirect
measures (e.g. levels of a pollutant in the environment as a measure of possible exposure) that
describe health or a factor associated with health (i.e., environmental hazard, age) in a specified
population. In general, content teams focused on developing measures specific to one of these
areas, but they also considered both proven and potential linkages to the other areas.

NCDM Recommendations Version 3.0
Page 5 3/19/2013

Figure 1: Content Work Group (CWG) Structure and Process, 2006 - 2010

CONTENT TEAMS

Submission to CDC

CWG
STEERING
GROUP

Air Quality

Birth Defects

Drinking Water
Quality

Vital Statistics
(Birth Outcomes)

Childhood Lead
Exposure

Hospitalization
Data

Cancer

Carbon Monoxide
Poisoning

Provide Direction
Draft Recommendations
Feedback
Final Recommendations
Review and Vote

Content Workgroup = Steering Group + Teams

CDC Tracking Program
Recommendations from content teams were separated into two parts; the first part concerned
indicators, measures, and how-to-guides which described the methods for extracting necessary
data and generating the measures. The second part was a data dictionary which described the
data to be shared with CDC. Recommendations were reviewed by the CWG Steering Group for
scientific rigor, utility for Tracking, and feasibility of each grantee generating the measures and
where specified providing data to CDC for use on the National Tracking Portal.
This document provides an updated summary of the NCDMs adopted by CDC as Tracking
standards. Section One of this document includes tables that summarize the indicators and
measures and identify the requirements of Tracking grantees for creating measures and providing
data to CDC. These Tracking standards incorporate discussions among the CWG steering group
as well as the recommendations of content teams concerning the use of existing national datasets,
where relevant.
Section Two includes the indicator templates originally developed by the teams and updated by
CDC. An indicator template describes the indicator’s measures and their deviations, uses, and
limitations. Although teams generally adhered to the template there was some minor variation in
NCDM Recommendations Version 3.0
Page 6 3/19/2013

the submitted documents. In creating this document original recommendations were modified to
ensure compatibility with the National Network and consistency across NCDMs.
Details regarding the data needed to generate the measures are provided in the how-to-guides,
data dictionaries, and schemas available from the CDC Tracking Program. Each set of
documentation represents a data feed needed to generate one or more measures.

NCDM Recommendations Version 3.0
Page 7 3/19/2013

SECTION ONE: SUMMARY OF NATIONALLY CONSISTENT DATA AND MEASURES
This section lists all NCDMs for the Tracking Network by indicator and measure name. The minimum temporal and geographic
resolutions are provided for the display of each required measure. These resolutions were selected to provide the most granular view
of the measure possible while considering the rarity of the outcome being measured and data steward requirements. Grantees able to
publish more temporally or geographically resolved measures are encouraged to do so. Grantees unable to publish at least the
minimum temporal and geographic resolutions should provide written documentation to CDC Tracking Program. The temporal and
geographic resolutions of the measures in this document are not necessarily the temporal and spatial resolution of the data
requirements. Information about the required fields and resolution of the data to generate the measures are provided in the
how-to-guides and data dictionaries. The source of the data required to generate each measure at the national level is provided in
the summary table. Some data are provided by state and local grantees while other data are provided by national partners. Each
measure is also listed as either required or optional for Tracking Grantees. Required means the grantees must (1) provide the data to
CDC Tracking Program if the data are not available nationally and (2) publish the measure on their state or local portals.

NCDM Recommendations Version 3.0
Page 8 3/19/2013

Content Domain: Heart Attacks or Acute Myocardial Infarction (AMI)
Temporal
Resolution

Geographic
Resolution

Source of Data
for National
Network

Grantee
Required

Indicator

Measure

Heart Attacks

Number of hospitalizations for
heart attack
Average daily number of
hospitalizations for heart attack,
by month
Maximum daily number of
hospitalizations for heart attack
by month

Annual

State and county

Grantee Provided

Required

Annual

State and county

Grantee Provided

Optional

Annual

State and county

Minimum daily number of
hospitalizations for heart attack
by month

Annual

State and county

Rate of hospitalization for heart
attack among persons 35 and over
by age group (total, 35-64, 65+)
per 10,000 population

Annual

State and county

Grantee Provided

Required

Age-adjusted rate of
hospitalization for heart attack
persons 35 and over per 10,000
population

Annual

State and county

NCDM Recommendations Version 3.0
Page 9 3/19/2013

Content Domain: Air Quality
Indicator

Measure

Temporal
Resolution

Geographic
Resolution

Source of
Data for
National
Network

Grantee
Required

Ozone—Days
Above
Regulatory
Standard

Number of days with maximum 8-hour
average ozone concentration over the
National Ambient Air Quality Standard

Annual

County

Nationally
Derived

Required

Number of person-days with maximum
8-hour average ozone concentration over
the National Ambient Air Quality
Standard
Percent of days with PM2.5 levels over
the National Ambient Air Quality
Standard (NAAQS)

Annual

County

Annual

County

Nationally
Derived

Required

Number of person-days with PM2.5 over
the National Ambient Air Quality
Standard (NAAQS)

Annual

County

Average ambient concentrations of PM
2.5 in micrograms per cubic meter (based
on seasonal averages and daily
measurement)
Percent of population living in counties
exceeding the National Ambient Air
Quality Standard (compared to percent of
population living in counties that meet
the standard and percent of population
living in counties without PM2.5
monitoring)

Annual

County

Nationally
Derived

Required

Annual

State

Fine Particle
(PM2.5)—
Days Above
Regulatory
Standard

Annual PM2.5
Level

NCDM Recommendations Version 3.0
Page 10 3/19/2013

Content Domain: Asthma
Temporal
Resolution

Geographic
Resolution

Source of Data
for National
Network

Grantee
Required

Annual

State and county

Grantee Provided

Required

Average daily number of
hospitalizations for asthma, by month
Maximum daily number of
hospitalizations for asthma by month

Annual

State and county

Grantee Provided

Optional

Annual

State and county

Minimum daily number of
hospitalizations for asthma by month

Annual

State and county

Rate of hospitalization for asthma by
age group (total, 0-4, 5-14, 15-34, 3564, and 65+) per 10,000 population
Age-adjusted rate of hospitalization for
asthma per 10,000 population

Annual

State and county

Grantee Provided

Required

Annual

State and county

Annual number of emergency
department visits for asthma
Average number of emergency
department visits for asthma as primary
diagnosis per month
Annual crude rate of emergency
department visits for asthma by age
group (total, 0–4, 5–14, 15–34, 35–64,
and 65+) per 10,000 population by age
group
Annual age-adjusted rate of emergency
department visits for asthma by age
groups ( total, 0–4, 5–14, 15–34, 35–64,
and 65+) per 10,000 population

Annual

State and county

Grantee Provided

Required

Annual

State and county

Annual

State and county

Annual

State and county

Indicator

Measure

Hospitalizatio
ns for Asthma

Number of hospitalizations for asthma

Emergency
Department
Visits for
Asthma

NCDM Recommendations Version 3.0
Page 11 3/19/2013

Content Domain: Birth Defects

Indicator

Measure

Prevalence
of Birth
Defects

Prevalence of Anencephaly per 10,000
live births
Prevalence of Spina Bifida (without
Anencephaly) per 10,000 live births over
Prevalence of Hypoplastic Left Heart
Syndrome per 10,000 live births
Prevalence of Tetralogy of Fallot per
10,000 live births
Prevalence of Transposition of the Great
Arteries (vessels) per 10,000 live births
Prevalence of Cleft Lip with or without
Cleft Palate per 10,000 live births
Prevalence of Cleft Palate without Cleft
Lip per 10,000 live births
Prevalence of Hypospadias per 10,000 live
male births
Prevalence of Gastroschisis per 10,000
live births
Prevalence of Upper Limb Deficiencies
per 10,000 live births
Prevalence of Lower Limb Deficiencies
per 10,000 live births
Prevalence of Trisomy 21 per 10,000 live
births by maternal age at delivery (<35
and >/=35)

NCDM Recommendations Version 3.0
Page 12 3/19/2013

Temporal
Resolution

Geographic
Resolution

Source of
Data for
National
Network

Grantee
Required

5 year

State and county

Grantee
Provided

Required

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

5 year

State and county

Content Domain: Cancer

Indicator
Incidence of
Selected
Cancers

Temporal
Resolution

Geographic
Resolution

Source of
Data for
National
Network

Grantee
Required

Number of cases of Mesothelioma

5 year

State

Nationally
Derived

Required

Age-adjusted incidence rate of Mesothelioma per 100,000
population

5 year

State

Number of cases of Melanoma of the Skin

Annual

State

5 year

State and county

Age-adjusted incidence rate of Melanoma of the Skin per
100,000 population

Annual

State

5 year

State and county

Number of cases of Liver and Intrahepatic Bile Duct
Cancer

Annual

State

5 year

State and county

Age-adjusted incidence rate of Liver and Intrahepatic Bile
Duct Cancer per 100,000 population

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Age-adjusted incidence rate of Kidney and Renal Pelvis
Cancer per 100,000 population

Annual

State

5 year

State and county

Number of cases of Breast Cancer in females by Age group
(<50, ≥50, total)

Annual

State

5 year

State and county

Age-adjusted incidence rate of Breast Cancer in females
per 100,000 population by Age group (<50, ≥50, total)

Annual

State

Measure

Number of cases of Kidney and Renal Pelvis Cancer

NCDM Recommendations Version 3.0
Page 13 3/19/2013

Number of cases of Lung and Bronchus Cancer

Age-adjusted incidence rate of Lung and Bronchus Cancer
per 100,000 population

Number of cases of Bladder Cancer (including in situ)
Age-adjusted incidence rate of Bladder Cancer (including
in situ) per 100,000 population
Number of cases of Brain and other nervous systems
Cancer
Age-adjusted incidence rate of Brain and other nervous
systems Cancer per 100,000 population
Number of cases of Brain and Central Nervous System
Cancer in children (<15 years and <20 years)

5 year

State and county

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Annual
5 year

State
State and county

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Annual

State

Annual

State

Annual

State

5 year

State and county

Annual

State

Age-adjusted incidence rate of Brain and Central Nervous
System Cancer in children (<15 years and <20 years) per
1,000,000 population

Number of cases of Thyroid Cancer

Age-adjusted incidence rate of Thyroid Cancer per 100,000

NCDM Recommendations Version 3.0
Page 14 3/19/2013

population

5 year

State and county

Number of cases of Non-Hodgkin’s Lymphoma

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Annual
5 year
Annual
5 year

State
State and county
State
State and county

Annual

State

Age-adjusted incidence rate of Leukemia in children (<15
years and <20 years) per 1,000,000 population

Annual

State

Number of cases of Chronic Lymphocytic Leukemia

Annual

State

Age-adjusted incidence rate of Chronic Lymphocytic
Leukemia per 100,000 population

Annual

State

Number of cases of Acute Myeloid Leukemia

Annual

State

Age-adjusted incidence rate of Acute Myeloid Leukemia
per 100,000 population

Annual

State

Number of Acute Myeloid Leukemia in children (<15 years Annual
and <20 years)

State

Age-adjusted incidence rate of Non-Hodgkin’s Lymphoma
per 100,000 population
Number of cases of Leukemia
Age-adjusted incidence rate of Leukemia per 100,000
population
Number of Leukemia in children (<15 years and <20 years)

NCDM Recommendations Version 3.0
Page 15 3/19/2013

Incidence of
Selected
Cancers

Age-adjusted incidence rate of Acute Myeloid Leukemia in
children (<15 years and <20 years) per 1,000,000
population

Annual

State

Number of cases of Acute Lymphocytic Leukemia in
children (<15 years and <20 years)

Annual

State

Age-adjusted incidence rate of Acute Lymphocytic
Leukemia in children (<15 years and <20 years) per
1,000,000 population

Annual

State

Number of cases of Oral Cavity and Pharynx Cancer

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Age-adjusted incidence rate of Larynx Cancer per 100,000
population

Annual

State

5 year

State and county

Number of cases of Esophagus Cancer

Annual

State

5 year

State and county

Age-adjusted incidence rate of Esophagus Cancer per
100,000 population

Annual

State

5 year

State and county

Number of cases of Pancreas Cancer

Annual

State

5 year

State and county

Annual

State

5 year

State and county

Age-adjusted incidence rate of Oral Cavity and Pharynx
Cancer per 100,000 population
Number of cases of Larynx Cancer

Age-adjusted incidence rate of Pancreas Cancer per
100,000 population

NCDM Recommendations Version 3.0
Page 16 3/19/2013

Nationally
Derived

Optional

Content Domain: Carbon Monoxide
Temporal
Resolution

Geographic
Resolution

Source of Data Grantee
for National
Required
Network

Number of hospitalizations for CO
poisoning by cause/intent (unintentional
fire-related, unintentional non-fire
related, and unknown intent)
Crude rate of hospitalization for CO
poisoning per 100,000 population by
cause/intent (unintentional fire-related,
unintentional non-fire related, and
unknown intent)

Annual

State

Grantee
Provided

Annual

State

Age-adjusted rate of hospitalization for
CO poisoning per 100,000 population by
cause/intent (unintentional fire-related,
unintentional non-fire related, and
unknown intent)

Annual

State

Number of emergency department visits
for CO Poisoning by cause/intent
(unintentional fire-related, unintentional
non-fire related, and unknown intent)

Annual

State

Crude rate of emergency department
visits for CO poisoning per 100,000
population by cause/intent (unintentional
fire-related, unintentional non-fire
related, and unknown intent)

Annual

State

Age-adjusted rate of emergency
department visits for CO poisoning per
100,000 population by cause/intent
(unintentional fire-related, unintentional

Annual

State

Indicator

Measure

Hospitalizations
for Carbon
Monoxide (CO)
Poisoning

Emergency
Department Visits
for CO Poisoning

NCDM Recommendations Version 3.0
Page 17 3/19/2013

Required

Optional
Grantee
Provided

non-fire related, and unknown intent)
CO Poisoning
Mortality

Reported
Exposure to CO

Home CO
Detector Coverage

Number of deaths from CO poisoning by
cause/intent (unintentional fire-related,
unintentional non-fire related, and
unknown intent)

Annual

State

Crude rate of death from CO poisoning
per 100,000 population by cause/intent
(unintentional fire-related, unintentional
non-fire related, and unknown intent)

Annual

State

Age-adjusted rate of death from CO
poisoning per 100,000 population by
cause/intent (unintentional fire-related,
unintentional non-fire related, and
unknown intent)

Annual

State

Number of unintentional CO exposures
reported to poison control centers by
resulting health effect and treatment in a
healthcare facility

Annual

State

Crude rate of unintentional CO exposures
reported to poison control centers per
100,000 population by resulting health
effect and treatment in a healthcare
facility

Annual

State

Percent of Behavioral Risk Factor
Surveillance System (BRFSS)
respondents reporting at least one CO
detector in their household

Annual

State

NCDM Recommendations Version 3.0
Page 18 3/19/2013

Nationally
Derived

Required

Nationally
Derived

Optional

Nationally
Derived

Optional

Content Domain: Childhood Lead Poisoning
Indicator

Measure

Temporal
Resolution

Geographic
Resolution

Source of Data
for National
Network

Grantee
Required

Testing and
Housing Age

Number of children born in the same
year and tested
Percent of children born in the same
year and tested
Number of homes built before 1950
(as measured in the 2000 Census)
Percent of homes built before 1950
(as measured in the 2000 Census)

Annual

State and county

Nationally
Derived

Required

Annual

State and county

Annual

State and county

Annual

State and county

Number of children younger than 5
years living in poverty (as measured
in 2000 census)
Percent of children younger than 5
years living in poverty (as measured
in 2000 census)

Annual

State and county

Annual

State and county

Number of children born in the same
year and tested
Percent of children born in the same
year and tested
Number of children born in the same
year and tested with confirmed blood
lead levels ≥ 10 μg/dL
Percent of children born in the same
year and tested with confirmed blood
lead levels ≥ 10 μg/dL
Number of children born in the same
year and tested with confirmed blood
lead levels ≥ 10 μg/dL, by blood lead
level category
Percent of children born in the same

Annual

State and county

Annual

State and county

Annual

State and county

Annual

State and county

Annual

State

Annual

State

Blood Lead
Levels by Birth
Cohort

NCDM Recommendations Version 3.0
Page 19 3/19/2013

Optional

Nationally
Derived

Required

Annual Blood
Lead Levels

year and tested with confirmed blood
lead levels ≥ 10 μg/dL, by blood lead
level category
PROPOSED *Number of children
born in the same year and tested with
blood lead levels between 5 and <10
μg/dL
PROPOSED*Percent of children
born in the same year and tested with
blood lead levels between 5 and <10
μg/dL
Number of children tested, by age
group
Percent of children tested, by age
group
Number of children tested with
confirmed blood lead levels ≥ 10
μg/dL, by age group
Percent of children tested with
confirmed blood lead levels ≥ 10
μg/dL, by age group
Number of children tested with
confirmed blood lead levels ≥ 10
μg/dL by blood lead level category,
by age group
Percent of children tested with
confirmed blood lead levels ≥ 10
μg/dL, by blood lead level category,
by age group
PROPOSED *Number of children
tested with blood lead levels between
5 and <10 μg/dL
PROPOSED*Percent of children
tested with blood lead levels between
5 and <10 μg/dL

NCDM Recommendations Version 3.0
Page 20 3/19/2013

Annual

State and county

Annual

State and county

Annual

State and county

Annual

State and county

Annual

State and county

Annual

State and county

Annual

State

Annual

State

Annual

State and county

Annual

State and county

Nationally
Derived

Required

Content Domain: Climate Change
Indicator

Measure

Number of hospitalizations for heat stress
Heat Stress
Hospitalizations

Heat Stress
Emergency
Department
Visits for Heat
Stress

Temporal
Resolution

Annual from
May–
September
Crude rate of hospitalization for heat stress Annual from
by age groups (total, 0–4, 5–14, 15–34,
May–
35–64, and 65+) per 100,000 population
September
Age-adjusted rate of hospitalization for
Annual from
heat stress (by age groups 0–4, 5–14, 15–
May–
34, 35–64, and 65+) per 100,000
September
population
Annual number of emergency department Annual from
visits for heat stress
May–
September
Annual crude rate of emergency
Annual from
department visits for heat stress by age
May–
group (total, 0–4, 5–14, 15–34, 35–64, and September
65+) per 100,000
Age-adjusted rate of emergency
Annual from
department visits for heat stress by age
May–
groups (total, 0–4, 5–14, 15–34, 35–64,
September
and 65+) per 100,000 population

NCDM Recommendations Version 3.0
Page 21 3/19/2013

Geographic
Resolution

Source of Data for Grantee Required
National Network

State and
national

Grantee Provided

Required

Grantee Provided

Required

State and
national
State and
national

State and
county
State and
county

State and
county

Content Domain: Drinking Water
Temporal
Resolution

Geographic
Resolution

Source of Data
for National
Network

Grantee
Required

Grantee Provided

Required

Grantee Provided

Required

Indicator

Measure

Atrazine Level and
Potential
Population
Exposures

Distribution of number of Community Water
Systems (CWS) by mean atrazine concentration
(micrograms per liter)
Distribution of number of CWS by maximum
atrazine concentration (micrograms per liter)
Distribution of number of CWS by mean atrazine
concentration (micrograms per liter)
Mean concentration of atrazine (micrograms per
liter) at CWS-level
Distribution of number of people served by CWS
by mean atrazine concentration (micrograms per
liter)
Distribution of number of people served by CWS
by maximum atrazine concentration (micrograms
per liter)
Distribution of number of people served by CWS
by mean atrazine concentration (micrograms per
liter)
Distribution of number of community water
systems by mean arsenic concentrations
(micrograms per liter)
Distribution of number of people served by
community water systems by mean arsenic
concentrations (micrograms per liter)

Quarterly

County

Annual

County

Annual

County

Annual

County

Quarterly

County

Annual

County

Annual

County

Annual

State

Annual

State

Distribution of number of community water
systems by maximum arsenic concentrations
(micrograms per liter)

Annual

State

Arsenic Level and
Potential
Population
Exposures

NCDM Recommendations Version 3.0
Page 22 3/19/2013

Di (2-Ethylhexyl)
phthalate (DEHP)
Level and Potential
Population
Exposures

Nitrate Level and
Potential
Population
Exposures

Distribution of number of people served by
community water systems by maximum arsenic
concentrations (micrograms per liter)

Annual

State

Mean concentration of Arsenic (micrograms per
liter) at CWS-level

Annual

State

Distribution of number of Community Water
Systems (CWS) by maximum DEHP
concentration (micrograms per liter)
Distribution of number of CWS by mean DEHP
concentration (micrograms per liter)
Mean concentration of DEHP (micrograms per
liter) at CWS-level
Distribution of number of people served by CWS
by maximum DEHP concentration (micrograms
per liter)
Distribution of number of people served by CWS
by mean DEHP concentration (micrograms per
liter)
Distribution of number of community water
systems by mean nitrate concentrations
(milligrams per liter)
Distribution of number of people served by
community water systems by mean nitrate
concentrations (milligrams per liter)

Annual

County

Annual

County

Annual

County

Annual

County

Annual

County

Annual

State

Annual

State

Distribution of number of community water
systems by maximum nitrate concentrations
(milligrams per liter)

Annual

State

Distribution of number of people served by
community water systems by maximum nitrate
concentrations (milligrams per liter)

Annual

State

NCDM Recommendations Version 3.0
Page 23 3/19/2013

Grantee Provided

Required

Grantee Provided

Required

Disinfection
Byproducts (DBP)
Level and Potential
Population
Exposure (TTHM)

Distribution of number of community water
systems by mean nitrate concentrations
(milligrams per liter)

Quarterly

State

Distribution of number of people served by
community water systems by mean nitrate
concentrations (milligrams per liter)

Quarterly

State

Mean concentration of nitrate (milligrams per
liter) at CWS-level

Annual

State

Distribution of number of community water
systems by mean trihalomethane (THM)
concentrations (micrograms per liter)

Annual

State

Distribution of number of people served by
community water systems by mean
trihalomethane (THM) concentrations
(micrograms per liter)
Distribution of number of community water
systems by maximum trihalomethane (THM)
concentrations (micrograms per liter)

Annual

State

Annual

State

Distribution of number of people served by
community water systems by maximum
trihalomethane (THM) concentrations
(micrograms per liter)

Annual

State

Distribution of number of community water
systems by mean trihalomethane concentrations
(micrograms per liter)

Quarterly

State

Distribution of number of people served by
community water systems by mean
trihalomethane (THM) concentrations
(micrograms per liter)

Quarterly

State

NCDM Recommendations Version 3.0
Page 24 3/19/2013

Grantee Provided

Required

Disinfection
Byproduct: Levels
and Potential
Population
Exposures (HAA5)

Public Water Use
Combined
Radium-226 and 228 Levels and
Potential
Population

Distribution of number of community water
systems by mean haloacetic acids (HAA5)
concentrations (micrograms per liter)

Annual

State

Mean concentration of HAA5 (micrograms per
liter) at CWS-level
Distribution of number of community water
systems by maximum haloacetic acids (HAA5)
concentrations (micrograms per liter)
Distribution of number of CWS by maximum
TTHM concentration (micrograms per liter)
Distribution of number of people served by
community water systems by mean haloacetic
acids (HAA5) concentrations (micrograms per
liter)
Distribution of number of CWS by mean TTHM
concentrations (micrograms per liter)

Annual

State

Annual

State

Annual

State

Quarterly

State

Quarterly

State

Distribution of number of CWS by mean TTHM
concentration (micrograms per liter)

Annual

State

Mean concentration (micrograms per liter) of
TTHM at CWS-level

Annual

State

Number of people receiving water from
community water systems
Distribution of number of Community Water
Systems (CWS) by maximum Radium
concentration picoCuries per Liter
Distribution of number of CWS by mean Radium
concentration picoCuries per Liter

Annual

NCDM Recommendations Version 3.0
Page 25 3/19/2013

Grantee Provided

Required

State

Grantee Provided

Required

Annual

County

Grantee Provided

Required

Annual

County

Exposure

Tetrachloroethene
(PCE) Levels and
Potential
Population
Exposure

Trichloroethene
(TCE) Levels and
Potential
Population
Exposure

Mean concentration of Radium picoCuries per
Liter at CWS-level
Distribution of number of people served by CWS
by maximum Radium concentration picoCuries
per Liter
Distribution of number of people served by CWS
by mean Radium concentration picoCuries per
Liter
Distribution of number of Community Water
Systems (CWS) by maximum PCE concentration
(micrograms per liter)
Distribution of number of CWS by mean PCE
concentration (micrograms per liter)

Annual

County

Annual

County

Annual

County

Annual

County

Annual

County

Mean concentration of PCE (micrograms per
liter) at CWS-level
Distribution of number of people served by CWS
by maximum PCE concentration (micrograms per
liter)
Distribution of number of people served by CWS
by mean PCE concentration (micrograms per
liter)
Distribution of number of CWS by maximum
TCE concentration (micrograms per liter)

Annual

County

Annual

County

Annual

County

Annual

County

Distribution of number of CWS by mean TCE
concentration (micrograms per liter)

Annual

County

Mean concentration of TCE (micrograms per
liter) at CWS-level
Distribution of number of people served by CWS
by maximum TCE concentration (micrograms
per liter)
Distribution of number of people served by CWS
by mean TCE concentration (micrograms per
liter)

Annual

County

Annual

County

Annual

County

NCDM Recommendations Version 3.0
Page 26 3/19/2013

Grantee Provided

Required

Grantee Provided

Required

Uranium Levels
and Potential
Population
Exposure

Distribution of number of Community Water
Systems (CWS) by maximum Uranium
concentration (micrograms per liter)
Distribution of number of CWS by mean
Uranium concentration (micrograms per liter)

Annual

County

Annual

County

Mean concentration of Uranium (micrograms per
liter) at CWS-level
Distribution of number of people served by CWS
by maximum Uranium concentration
(micrograms per liter)
Distribution of number of people served by CWS
by mean Uranium concentration (micrograms per
liter)

Annual

County

Annual

County

Annual

County

NCDM Recommendations Version 3.0
Page 27 3/19/2013

Grantee Provided

Required

Content Domain: Reproductive Health Outcomes
Indicator

Measure

Temporal
Resolution

Geographic
Resolution

Source of Data
for National
Network

Grantee
Required

Prematurity

Percent of preterm (less than 37 weeks
gestation) live singleton births

Annual

State and
county

Nationally
Derived

Required

Percent of very preterm (less than 32
weeks gestation) live singleton births

5 year Annual
Average

State and
county

Percent of low birthweight (less than
2500 grams) live term singleton births

Annual

State and
county

Nationally
Derived

Required

Percent of very low birthweight (less than
1500 grams) live singleton births

5 year Annual
Average

State and
county

Average Infant (less than 1 year of age)
Mortality Rate per 1000 live births

5 year Annual
Average

State and
county

Nationally
Derived

Required

Average Neonatal (less than 28 days of
age) Mortality Rate per 1000 live births

5 year Annual
Average

State and
county

Average Perinatal (equal to or greater
than 28 weeks gestation to less than 7
days of age) Mortality Rate per 1000 live
births (plus fetal deaths equal to or
greater than 28 weeks gestation)
Average Postneonatal (equal to or greater
than 28 days to less than 1 year of age)
Mortality Rate per 1000 live births
Total Fertility Rate per 1000 women of
reproductive age

5 year Annual
Average

State and
county

5 year Annual
Average

State and
county

Annual

State and
county

Nationally
Derived

Optional

Male to Female sex ratio at birth (term
singletons only)

Annual

State and
county

Nationally
Derived

Required

Low
Birthweight

Mortality

Fertility
Sex Ratio at
Birth

NCDM Recommendations Version 3.0
Page 28 3/19/2013

SECTION TWO: INDICATOR TEMPLATES
This section contains an indicator template for each indicator and corresponding measures listed
in section one. The indicator template provides basic information about the indicator including:
1. Measures
2. Derivations of the measures
3. Units
4. Geographic Scope
5. Geographic Scale
6. Time Period
7. Time Scale
8. Rationale
9. Use of the Measure
10. Limitations of the Measure
11. Data Sources
12. Limitations of Data Sources
13. References

Additional information about the underlying data needed for the indicator and steps for
extracting the data and generating the measures can be found in the how-to-guides and data
dictionaries.

NCDM Recommendations Version 3.0
Page 29 3/19/2013

CONTENT DOMAIN: HEART ATTACK
INDICATOR: HOSPITALIZATIONS FOR HEART ATTACK
Type of EPHT Indicator

Measures

Health Outcome
1. Number of hospitalizations for acute myocardial infarction (AMI)
2. Minimum daily number of hospitalizations for AMI by month
3. Maximum daily number of hospitalizations for AMI by month
4. Average daily number of hospitalizations for AMI by month
5. Crude rate of hospitalizations for AMI among persons 35 and older
by age group (total, 35-64, 65+) per 10,000 population
6. Annual age-adjusted rate of hospitalizations for AMI among
persons 35 and older per 10,000 population
When supported by sufficient data volume, the measures may also be
reported stratified by sex, race, and ethnicity.
Numerator:
Resident hospitalizations for AMI, ICD-9-CM: 410.00–410.92 by
gender and total for state and by county

Derivation of Measures

Unit
Geographic Scope
Geographic Scale
Time Period
Time Scale

Rationale

Denominator:
Midyear resident population by gender, for state and by county
Adjustment:
Age-adjustment by the direct method to Year 2000 U.S. Standard
population
Hospital admission (categorized by discharge diagnosis)
State and national (tracking network states)
State and county
Hospital admissions from January 1 through December 31 for each
year, 2000–current
Daily, monthly, and annually (as appropriate for the measure)
There currently is no single AMI surveillance system is in place in the
United States, nor does such a system exist for coronary heart disease
(CHD) in general. Mortality is the sole descriptor for national data for
AMI. Estimates of incidence and prevalence of AMI and CHD are
largely based on survey samples (e.g., NHANES) or large cohort
studies such as the Atherosclerosis Risk in Communities (ARIC)
study.
In 2007, the American Heart Association estimated 565,000 new
attacks and 300,000 recurrent attacks of MI annually (National Heart,
Lung, and Blood Institute: based on unpublished data from the ARIC
study and the Cardiovascular Health Study [CHS]). Among

NCDM Recommendations Version 3.0
Page 30 3/19/2013

Americans aged ≥20 years, new and recurrent MI prevalence for both
men and women represented 3.7% of the U.S. population, or 7,900,000
(4.9 million men and 3.0 million women). Corresponding prevalence
by race and ethnicity is 5.4% for white men, 2.5% for white women,
3.9% for black men, and 3.3% for black women.
The well-documented risk factors for AMI include diabetes,
hypertension, obesity, hypercholesterolemia, and cigarette smoking.
Increasingly, investigators both in the United States and abroad have
shown significant relationships between air pollutants and increased
risk of AMI and other forms of CHD. Studies have often focused on
persons aged >65 years. A number of epidemiologic studies have
reported associations between air pollution (ozone, PM10 , CO, PM
2.5, SO2 ) and hospitalizations for AMI and other forms of heart
disease. Models have demonstrated increases in AMI hospitalization
rate in relation to fine particles (PM2.5), particularly in sensitive
subpopulations such as the elderly, patients with pre-existing heart
disease, and particularly persons who are survivors of MI or persons
with COPD. An increase of 10 ug/m3 in PM 2.5 was associated with a
4.5% elevation in risk of acute ischemic coronary events (unstable
angina and AMI) (95% CI, 1.1–8.0). Mortality statistics have been
linked for a 16-year period to chronic exposure of multiple air
pollutants in 500,000 adults residing throughout the United States.
Each 10 ug/m3 in annual PM2.5 was related to a 12% increased
mortality risk.
Developing a standardized analytic method for AMI hospital
admissions among residents in each state will provide more uniform
information for multiple users at the national, state, and local levels.
These measures will allow monitoring of trends over time, identify
high risk groups, and inform prevention, evaluation, and program
planning efforts.
These measures will address the following surveillance functions:

Use of the Measures









Examination of time trends in AMI hospitalizations.
Identification of seasonal trends.
Assessment of geographic differences in hospitalizations.
Evaluation of differences in AMI hospitalizations by age, gender,
and race/ethnicity.
With further analysis … evaluation of disparities in AMI
hospitalizations by factors such as age, race/ethnicity, gender,
education, and/or income.
Determination of populations in need of targeted interventions.
Identification of possible environmental relationships that warrant
further investigation or environmental public health action when
AMI data are linked with environmental variables.

NCDM Recommendations Version 3.0
Page 31 3/19/2013

Hospitalization data for AMIs omit persons who do not receive
medical care or who are not hospitalized, including those who die in
emergency rooms, in nursing homes, or at home without being
admitted to a hospital, and those treated in outpatient settings.
Differences in rates by time or area may reflect differences or changes
in diagnostic techniques and criteria and in the coding of AMI or in
medical care access.
Differences in rates by area may be due to different sociodemographic
characteristics and associated behaviors.
When rates across geographic areas are compared, a variety on nonenvironmental factors, such as access to medical care and diet, can
affect the likelihood of persons hospitalized for AMI.
Reporting rates at the state and/or county level will not show the true
AMI burden at a more local level (i.e., neighborhood).
Limitations of the
Measures

Reporting rates at the state and/or county level will not be resolved
geographically enough to be linked with many types of environmental
data.
When looking at small geographic levels (e.g., ZIP code), users must
consider appropriate cell suppression rules imposed by the data
providers or individual state programs.
Although duplicate records and transfers from one hospital to another
are excluded, the measures are based upon events, not individuals,
because no unique identifier is always available. When multiple
admissions are not identified, the true prevalence will be
overestimated.
Even at the county level, the measures generated will often be based
upon numbers too small to report or present without violating state and
federal privacy guidelines and regulations. Careful adherence to cell
suppression rules in cross tabulations is necessary, and methods to
increase cell sizes by combining data across time (e.g., months, years)
and geographic areas may be appropriate.
Numerator:
State inpatient hospitalization data (using admission date)

Data Sources

Limitations of Data
Sources

Denominator:
U.S. Census Bureau population data
State hospital discharge data:
Using a measure of all AMI hospitalizations will include some

NCDM Recommendations Version 3.0
Page 32 3/19/2013

transfers between hospitals for the same person for the same AMI
event. Variations in the percentage of transfers or readmissions for the
same AMI event may vary by geographic area and impact rates.
However, efforts were made to identify and exclude transfers based on
unique identifiers consisting of date of birth, zip code, gender, and
encrypted social security number when available.
Without reciprocal reporting agreements with abutting states,
statewide measures and measures for geographic areas (e.g., counties)
bordering other states may be underestimated because of health care
utilization patterns.
Each state must individually obtain permission to access and, in some
states, provide payment to obtain the data.
Veterans Affairs, Indian Health Services, and institutionalized (prison)
populations are not usually included in hospitalization datasets.
Practice patterns and payment mechanisms may affect diagnostic
coding and decisions by health care providers to hospitalize patients
Street address is not available in many states.
Sometimes mailing address of patient is listed as the residence address
of the patient.
Patients may be exposed to environmental triggers in multiple
locations, but hospital discharge geographic information is limited to
residence.
Since the data capture hospital discharges (rather than admissions),
patients admitted toward the end of the year and discharged the
following year will be omitted from the current year dataset.
Data will need to be de-duplicated (i.e., remove duplicate records for
the same event).
There is usually a two-year lag period before data are available from
the data owner.
Census data:
Available only every 10 years; thus, postcensal data will be estimated
for calculating rates for years following the census year.
Postcensal estimates at the ZIP code level are not available from the
Census Bureau. These estimates should be extrapolated or purchased
NCDM Recommendations Version 3.0
Page 33 3/19/2013

from a vendor.
1. Rosamond, W., et al., Heart disease and stroke statistics--2007
update: a report from the American Heart Association Statistics
Committee and Stroke Statistics Subcommittee. Circulation, 2007.
115(5): p. e69–171.
2. Boland, L.L., et al., Occurrence of unrecognized myocardial
infarction in subjects aged 45 to 65 years (the ARIC study). Am J
Cardiol, 2002. 90(9): p. 927–31.
3. Thom, T., et al., Cardiovascular disease in the United States and
preventive approaches, in Hurst’s The Heart, Arteries and Veins,
V. Fuster, R. Alexander, and R. O’Rourke, Editors. 2001,
McGraw-Hill: New York, NY.
4. Jones, D.W., et al., Risk factors for coronary heart disease in
African Americans: the atherosclerosis risk in communities study,
1987–1997. Arch Intern Med, 2002. 162(22): p. 2565–71.
References

5. Kannel, W.B., et al., Menopause and risk of cardiovascular
disease: the Framingham study. Ann Intern Med, 1976. 85(4): p.
447–52.
6. Pope, C.A., 3rd, et al., Cardiovascular mortality and long-term
exposure to particulate air pollution: epidemiological evidence of
general pathophysiological pathways of disease. Circulation,
2004. 109(1): p. 71–7.
7. Vermylen, J., et al., Ambient air pollution and acute myocardial
infarction. J Thromb Haemost, 2005. 3(9): p. 1955–61.
8. Pope, C.A., 3rd, et al., Ischemic heart disease events triggered by
short-term exposure to fine particulate air pollution. Circulation,
2006. 114(23): p. 2443–8
9. von Klot, S., et al., Ambient air pollution is associated with
increased risk of hospital cardiac readmissions of myocardial
infarction survivors in five European cities. Circulation, 2005.
112(20): p. 3073–9.

NCDM Recommendations Version 3.0
Page 34 3/19/2013

CONTENT DOMAIN: AIR QUALITY
INDICATOR: OZONE-DAYS ABOVE REGULATORY
STANDARD
Type of EPHT Indicator
Measures

Derivation of Measures

Hazard
1. Number of days with maximum 8-hour average ozone concentration
over the National Ambient Air Quality Standard (NAAQS)
2. Number of person-days with maximum 8-hour average ozone
concentration over the National Ambient Air Quality Standard
(NAAQS)
This overview provides the key technical points in how EPA and CDC
processed EPA’s air quality data for use in the EPHT air indicators.
Processing raw data
First, EPA extracts the air quality data from the Air Quality System
(AQS). EPA uses the following steps in developing the air data and
measures for EPHT air quality indicators.
Step 1: EPA accesses daily maximum 8-hour average ozone
concentrations (ppm) (parameter code ‘44201’ and duration code ‘W’)
and supplemental data fields (e.g. latitude, longitude, elevation) for all
the monitoring sites across the US from the EPA’s Data Mart. The data
are obtained only from monitors that are designated as Federal Reference
Methods or equivalent. The data include any flagged values associated
with exceptional events (high winds, fires, construction, etc) regardless
of concurrence by the EPA Regional Office. EPA retains data from
monitors that meet the minimum data completeness criteria set forth in
the national air quality standard (i.e. if valid 8-hour averages are
available for at least 75% of possible hours in a day or the maximum 8hour average is above ozone 8-hr NAAQS).
Step 2: For each monitoring site, retain the maximum concentration at
the site for each monitored day. The pollutant occurrence code (poc)
which distinguishes multiple monitors at a single site is listed in the
output data set.
Step 3: Site-level daily monitoring data are used to create ozone 8-hr
maximum daily county-level dataset. Daily county-level dataset is
created by retaining the maximum concentration among all monitors
within the county for each monitored day. The county-level daily dataset
is used to create number of days and number of person-days with ozone
levels over the daily NAAQS measures.

NCDM Recommendations Version 3.0
Page 35 3/19/2013

Creating Measures
Step 3: Ozone levels decrease significantly in the colder parts of the year
in many areas, ozone is required to be monitored at monitoring sites only
during the ozone season, which is defined on a state by state basis. Only
counties that have at least 75% of the days monitored during the ozone
seasons are considered complete. The measures are computed only for
counties that satisfy the completeness criteria.
Number of days with Ozone levels over the NAAQS:
Step 4: Select counties which pass the completeness criteria mentioned
in Step 3.
Step 5: To calculate the annual number of days over the daily NAAQS,
sum the number of days with ozone levels over the daily 8-hr NAAQS
for the entire year.
Number of person-days with ozone levels over the NAAQS:
Step 4: To calculate Person-days with ozone levels over the daily 8-hr
NAAQS, multiply the number of days over the daily NAAQS by the
total population of the county.
Units
Geographic Scope
Geographic Scale
Time Period
Time Scale
Rationale

1. Exceedance days
2. Population-weighted exceedance days
United States
County (where monitors exist)
2001-current
Calendar year
According to the published literature, air pollution is associated with
premature death, increased rates of hospitalization for respiratory and
cardiovascular conditions, adverse birth outcomes, and lung cancer (2,
3). Air pollution places a large economic burden on the country. In a
report prepared for the American Lung Association,(2) estimated that air
pollution related illness was estimated to cost approximately $100 billion
annually (2) (1988 dollars) in the United States, with an estimated
number of excess deaths ranging from 50,000 to 100,000 annually (3).
More than half of the U.S. population, approximately 159 million
persons, live in counties with unhealthy levels of air pollution in the
form of either ozone or particulate matter (1). Elevated pollution levels
depend on sources, transport, season geography, and atmospheric
conditions. Each part of the country has its own level of pollution
concentrations that can be exacerbated by many conditions, including
stagnation, fire, or wind. The seasons for peak concentrations also vary
between geographical regions. (4)
The Clean Air Act, which was last amended in 1990, requires EPA to set
NAAQS for widespread pollutants from numerous and diverse sources

NCDM Recommendations Version 3.0
Page 36 3/19/2013

considered harmful to public health and the environment. The Clean Air
Act established two types of national air quality standards. Primary
standards set limits to protect public health, including the health of
"sensitive" populations such as asthmatics, children, and the elderly.
Secondary standards set limits to protect public welfare, including
visibility impairment and damage to animals, crops, vegetation, and
buildings. (5)

Use of Measure

Our indicator is based on comparing measured levels of ozone by county
to the primary ozone 8-hr NAAQS, which is set at 75 ppb The Clean Air
Act requires periodic review of the science upon which the standards are
based and the standards themselves. Primary air quality standards
indicate the acceptable level of substances in the air before harm will
occur based on proven scientific and medical research. State
governments also set air quality standards. In several cases, California's
standards or other benchmarks are more stringent than the EPA NAAQS.
The indicator for the number of days with maximum 8-hour average
ozone concentration over the standard is similar to EPA’s analyses on
number of days with air quality index (AQI) levels higher than 100 (for
ozone) – see www.epa.gov/airtrends/aqi_info.html. This measure is
consistent with the EPA and state AQI program efforts to communicate
an area’s air quality levels to the public. In addition, this indicator can
be used to inform policy makers and the public of the degree of hazard
within a state (by county or MSAs with monitors) during a year. For
example, the number of days per year that ozone is higher than the
NAAQS can be used to communicate to sensitive populations (such as
asthmatics) the number of days that they may be exposed to unhealthy
levels of ozone; this is the same level used in the air quality alerts that
inform these sensitive populations when and how to reduce exposure.
See http://www.epa.gov/air/airtrends/2007/report/groundlevelozone.pdf
and http://www.epa.gov/air/airtrends/aqtrnd00/pdffiles/aqioz.pdf.
In the use of the measure, it is important to explain that not all counties
have monitors although most populated areas are monitored.

NCDM Recommendations Version 3.0
Page 37 3/19/2013

Limitations of The
Measure

Since ozone levels decrease significantly in the colder parts of the
year in many areas, ozone is required to be monitored only during the
ozone season., which are designated on a State by State basis.(6)
The number of high ozone days per year varies, which makes tracking
trends over time difficult to analyze or interpret. The variability results
from the following: a) the number of high ozone days is related to
temperature; there will be more high days in hotter summers; and b)
there are a small number of events per year, so for statistical reasons this
type of measure will bounce around more than an average. c) When
creating measures, we only consider monitors with 75% completeness
during the ozone season and ozone seasons are designated on a state by
state basis.
Variation within counties may exist but will not be captured in this
measure. Within these areas, the monitor with the highest reading on
any day is used in the measure. Larger areas will have a broader range
of pollution values and perhaps more monitors that may measure a high
value on a given day. Thus, day and person-day estimates for larger
areas may be biased higher than estimates for smaller areas. The relative
variation among county populations in many states may be large enough
relative to the variation in the number of days greater than the ozone
NAAQS that the population component can dominate the calculation of
the number of person-days. Thus, careful investigation of the underlying
data to properly identify changes in population and air quality is needed
when comparing person-days in space and time.
The data for this indicator represent only counties that have air monitors;
thus the data tend to reflect urban air quality (where most people live).
Although populations in areas without monitors also may be exposed to
ozone that exceeds the standard, they are not counted. The number of
days that exceed the EPA NAAQS or other health benchmarks does not
provide information regarding the severity (max concentrations) of
potential exposures. The relationship between ambient concentrations
and personal exposure is largely unknown and variable depending upon
pollutant, activity patterns, and microenvironments.

Data Sources
Limitations of Data
Sources
References

This indicator is not for use compliance determination with NAAQS or
reasonable further progress toward attaining compliance.
Air quality data: EPA Air Explorer http://epa.gov/mxplorer/index.htm
The AQS monitoring data, which are used in the calculation of measures,
are not present for all counties and days.
1. American Lung Association. State of the Air 2004; 2004 [cited
2008 Dec 4]. Available from:
http://lungaction.org/reports/sota04_full.html
2. Cannon J. The Health Costs of Air Pollution: A Survey of

NCDM Recommendations Version 3.0
Page 38 3/19/2013

Studies Published 1984– 1989. New York: American Lung
Association; 1990.
3. Dockery DW and Pope CA. Acute respiratory effects of
particulate air pollution. Annu Rev Public Health 1994;15:107–
132.
4. US Environmental Protection Agency. US EPA general site on
ozone effects. Available from:
http://www.epa.gov/air/ozonepollution/health.html
5. Criteria document for ozone NAAQS:
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=149923
6. Ozone Season definition by state:
http://www.epa.gov/ttn/naaqs/ozone/ozonetech/40cfr58d.htm

NCDM Recommendations Version 3.0
Page 39 3/19/2013

CONTENT DOMAIN: AIR QUALITY
INDICATOR: PM2.5—DAYS ABOVE REGULATORY STANDARD
Type of EPHT Indicator
Measures

Hazard
1. Percent of days with PM2.5 levels over the National Ambient Air
Quality Standard (NAAQS)
2. Number of person-days with PM2.5 over the National Ambient Air
Quality Standard (NAAQS)

Derivation of Measures
This overview provides the key technical points in how EPA and CDC
processed EPA’s air quality data for use in the EPHT air indicators.
Processing raw data:
First, EPA extracts the air quality data from the Air Quality System
(AQS). EPA uses the following steps in developing the air data and
measures for EPHT air quality indicators.
3
Step 1: EPA accesses PM2.5
) (parameter
code ‘88101’ and duration code ‘7’) and daily maximum 8-hour
average ozone concentrations (ppm) (parameter code ‘44201’ and
duration code ‘W’) and supplemental data fields (e.g. latitude,
longitude, elevation) for all the monitoring sites across the US from
the EPA’s Data Mart. The data are obtained only from monitors that
are designated as Federal Reference Methods or equivalent. The data
include any flagged values associated with exceptional events (high
winds, fires, construction, etc) regardless of concurrence by the EPA
Regional Office.

Step 2: For each monitoring site, retain the maximum concentration at
the site for each monitored day. The pollutant occurrence code (poc)
which distinguishes multiple monitors at a single site is listed in the
output data set.
Step 3: Site-level daily monitoring data are used to create 24-hr
maximum daily county-level PM2.5 dataset. Daily county-level dataset
is created by retaining the maximum concentration among all monitors
within the county for each monitored day. The county-level daily
dataset is used to create percent of days and number of person-days
with PM2.5 levels over the daily NAAQS measures.
Creating Measures
Percent of days with PM2.5 levels over the NAAQS:
Step 4: To calculate the annual percent of days over the daily NAAQS,
sum the number of days with PM2.5 levels over the daily NAAQS and
NCDM Recommendations Version 3.0
Page 40 3/19/2013

divide by the total number of monitored days. Multiply this
exceedance fraction by 100 to get percent of days.
Number of person-days with PM2.5 levels over the NAAQS:
Step 5: To calculate person-days with PM2.5 levels over the NAAQS
multiply the exceedance fraction from Step 4 by 365 to get the annual
days and then multiply by the total population of the county.
For PM2.5 - days above regulatory standard indicator, tracking portal
only displays counties that have year-round monitoring.
Unit
Geographic Scope
Geographic Scale
Time Period
Time Scale
Rationale

1. Exceedance days
2. Population weighted exceedance days
Contiguous United States
County (where monitors exist)
2001-current
Calendar year
According to the published literature, air pollution is associated with
premature death, increased rates of hospitalization for respiratory and
cardiovascular conditions, adverse birth outcomes, and lung cancer
(2,3,4). Air pollution places a large economic burden on the country.
In a report prepared for the American Lung Association, (2) estimated
that air pollution related illness was estimated to cost approximately
$100 billion annually (2) (1988 dollars) in the United States, with an
estimated number of excess deaths ranging from 50,000 to 100,000
annually (3). More than half of the U.S. population, approximately
159 million persons, live in counties with unhealthy levels of air
pollution in the form of either ozone or particulate matter (1). Elevated
pollution levels depend on sources, transport, season geography, and
atmospheric conditions. Each part of the country has its own level of
pollution concentrations that can be exacerbated by many conditions,
including stagnation, fire, or wind. The seasons for peak
concentrations also vary between geographical regions.
The Clean Air Act, which was last amended in 1990, requires EPA to
set NAAQS for widespread pollutants from numerous and diverse
sources considered harmful to public health and the environment. The
Clean Air Act established two types of national air quality standards.
Primary standards set limits to protect public health, including the
health of "sensitive" populations such as asthmatics, children, and the
elderly. Secondary standards set limits to protect public welfare,
including visibility impairment and damage to animals, crops,
vegetation, and buildings.
Our indicator is based on comparing measured levels of PM2.5 by
3
county to the 24-hr NAAQS for PM2.5,
. The

NCDM Recommendations Version 3.0
Page 41 3/19/2013

Use of the Measure

Clean Air Act requires periodic review of the science upon which the
standards are based and the standards themselves. Primary air quality
standards indicate the acceptable level of substances in the air before
harm will occur based on proven scientific and medical research. State
governments also set air quality standards. In several cases,
California's standards or other benchmarks are more stringent than the
EPA NAAQS. (5)
This indicator can be used to inform the public and policy makers of
the degree of potential exposures within a state (for counties with
monitors) during a year. For example, the percentage of days per year
that PM2.5 is higher than the NAAQS can be used to communicate to
sensitive populations (such as asthmatics) the percentage of days that
they may be exposed to unhealthy levels of PM2.5; this is similar to the
level used in the Air Quality Alerts that inform these sensitive
populations when and how to reduce exposure.

Limitations of the Measure

The number of person-days may be directed toward policy makers
who are interested in roughly comparing population exposure between
areas, to determine the areas most in need of prevention and pollution
control activities.
The data for this indicator represent highly populated counties that
have PM2.5 monitors. As a result, the data tend to reflect urban air
quality and longer-term average air quality levels. Populations in
counties without monitors may also be exposed to concentrations that
exceed a standard.
The percentage of days during which the EPA NAAQS or other health
benchmarks are exceeded does not provide information regarding the
severity (maximum concentrations) of potential exposures. Even with
these limitations, trends in PM2.5 levels are a useful measure to
describe public health concerns within these areas. We identify several
limitations with this indicator below.
This indicator is based on the percentage of high days rather than the
total number of high days to highlight the fact that PM2.5 monitors
follow different operating schedules. Most operate on a once-everythird day schedule, but a small proportion operates on a daily or onceevery-sixth day schedule. Because most of the monitors do not take
measurements every day, the number of short-term events (e.g., days
in which the NAAQS is exceeded ) is uncertain, and except where
PM2.5 levels vary uniformly throughout the year, estimating short-term
measures that are representative of short-term exposures over a year is
complex. To address this limitation, the measure can be based on the
percentage of monitored days. It should be noted that state air
programs will be evaluating the daily PM2.5 NAAQS by using a
frequency-based analysis to determine whether areas within the state

NCDM Recommendations Version 3.0
Page 42 3/19/2013

attain this NAAQS.

Populations in counties without monitors may be exposed to
concentrations that exceed a standard. Person-day estimates for larger,
highly populated counties may be biased higher than estimates for
smaller and lower populated counties. The indicator uses the highest
value of all monitors in the area so that larger counties with more
monitors may have a broader range of pollution values and greater
potential to measure a high day than smaller counties with fewer
monitors
The relationship between ambient concentrations and personal
exposure is largely unknown, and it varies depending upon pollutant,
activity patterns, and microenvironments.

Data Sources

Limitations of Data
Sources

References

Because the number of high PM2.5 days per year can vary considerably,
tracking trends over time needs to be done carefully. The variability
results because: the number of high PM2.5 days is related to
meteorological factors (e.g., temperature and mixing heights), and few
events occur per year, so that this type of extreme value measure will
vary considerably for statistical reasons. When creating measures, we
only consider monitors, which have atleast 11 observations per
calendar quarter.
Air–quality data: EPA Air Explorer http://epa.gov/mxplorer/index.htm
Population data: county population data can be found at
http://www.census.gov/popest/counties/CO-EST2006-01.html
Air–monitoring data provides information regarding concentrations
around the specific location of each monitor. For PM2.5 this can be a
rather large area, except when unusual local emissions (agricultural
fires) occur. Within-county variation in concentrations will likely
exist but will not be captured in this measure. Many PM2.5 monitors
operate once-every third day (some once-every-sixth day); a few
monitors operate every day.
1. American Lung Association. State of the Air 2004; 2004 [cited
2008 Dec 4]. Available from:
http://lungaction.org/reports/sota04_full.html
2. Cannon J. The Health Costs of Air Pollution: A Survey of
Studies Published 1984– 1989. New York: American Lung
Association; 1990.
3. Dockery DW and Pope CA. Acute respiratory effects of
particulate air pollution. Annu Rev Public Health 1994;15:107–
132.

NCDM Recommendations Version 3.0
Page 43 3/19/2013

4. Schwartz, J. Air pollution and hospital admissions for heart
disease in eight U.S. counties. Epidemiology 1999;10:17–22.
5. U.S. Environmental Protection Agency. U.S. EPA Criteria
Document for PM. Available from: Volume 1
VOL_I_FINAL_PM_AQCD_OCT2004.PDF and Volume 2
VOL_II_FINAL_PM_AQCD_OCT2004.PDF

NCDM Recommendations Version 3.0
Page 44 3/19/2013

CONTENT DOMAIN: AIR QUALITY
INDICATOR: ANNUAL PM2.5 LEVEL
Type of EPHT Indicator
Measure

Derivation of Measure

Hazard
1. Annual average ambient concentrations of PM2.5 in micrograms per
cubic meter (based on seasonal averages and daily measurement)
2. Annual percent of population living in counties exceeding the
National Ambient Air Quality Standard (compared to percent of
population living in counties that meet the standard and percent of
population living in counties without PM2.5 monitoring)
First, EPA extracts the air quality data from the Air Quality System (AQS).
EPA uses the following steps in developing the air data and measures for
EPHT air quality indicators.
Processing raw data
Step 1: EPA accesses PM2.5 daily concentrations (mcg/m3) (parameter code
‘88101’ and duration code ‘7’) and supplemental data fields (e.g. latitude,
longitude, elevation) for all the monitoring sites across the US from the
EPA’s Data Mart. The data are obtained only from monitors that are
designated as Federal Reference Methods or equivalent. The data include
any flagged values associated with exceptional events (high winds, fires,
construction, etc) regardless of concurrence by the EPA Regional Office.
Step 2: For each monitoring site, retain the maximum concentration at the
site for each monitored day. The pollutant occurrence code (poc) which
distinguishes multiple monitors at a single site is listed in the output data
set.
Creating Measures
Step 3: The annual average measures of PM2.5 are created using the sitelevel daily monitoring data. Only monitors that have at least 11
observations for each of the four calendar quarters are considered complete.
The annual averages are computed only for monitors that satisfy the
completeness criteria.
Annual average ambient concentrations of PM2.5 measure:
Step 4: Select monitors with complete quarterly and annual data using the
site-level monitoring data.
Step 5: Calculate the quarterly average for each calendar quarter and then
compute the annual average for each monitor with four valid quarters by
averaging the quarterly averages. If a county has more than one monitor
then the maximum annual average among monitors with complete (4 valid
quarters) data is assigned as the annual average for that county.

NCDM Recommendations Version 3.0
Page 45 3/19/2013

Annual percent of population living in counties exceeding the NAAQS
(compared to percent of population living in counties that meet the
standard and percent of population living in counties without PM2.5
monitoring) measure:
Step 6a: This is a state-level measure and uses the county-level annual
average concentrations calculated in step 3.
Step 6b: To calculate the annual percent of population living in counties
that exceed the annual NAAQS, sum the population of all counties that
exceed the annual NAAQS and divide by the total population of the state.
Multiply this fraction by 100 to get percent.
Step 6c: To calculate the annual percent of population living in counties that
meet the annual NAAQS, sum the population of all counties that meet the
annual NAAQS and divide by the total population of the state. Multiply this
fraction by 100 to get percent.

Unit
Geographic Scope
Geographic Scale
Time Period
Time scale
Rationale

Step 6d: To calculate the annual percent of population living in counties
that do not have complete monitors, sum the population of all counties that
do not have complete monitors and divide by the total population of the
state. Multiply this fraction by 100 to get percent.
1. Microgram per cubic meter (μg/m3)
2. Population proportion by hazard level
Contiguous United States
County (where monitors exist)
2001- current
Calendar year
According to work conducted by Pope et al. (1), long-term exposure to
PM2.5 is related to many adverse health conditions. Each 10 ug/m3 elevation
in PM2.5 is related to an 8% increase in lung cancer mortality, a 6% increase
in cardiopulmonary mortality, and a 4% increase in death from general
causes.(2)
The annual average provides an indication of the long-term trends in overall
PM2.5 burden, relevant to its long-term effects.
The percent of the population living in counties that exceed the standard
provides an indication of the population at risk for long-term exposure.

Use of The Measure

Note: these indicators are similar to indicators developed by EPA and state
air quality agencies for use in air quality stats and trends analyses and
reports (see www.epa.gov/airtrends)
This indicator can be used to inform policy makers and the public about the
degree of potential exposures to fine particles within a state during a year
and over time (trends). This is appropriate, as many existing health studies
have found the strongest association with health outcomes based on long-

NCDM Recommendations Version 3.0
Page 46 3/19/2013

Limitations of the
Measure

term studies; thus, EPA developed the annual NAAQS at
15 ug/m3. The indicator (annual average PM2.5 concentrations) can be
compared to the National Ambient Air Quality Standard (NAAQS) level of
15 ug/m3 or other health-based standards (although not in a regulatory
manner) to communicate the degree of public health concern to policy
makers and the general public. (3)
This measure provides a general indication of the overall trend in annual
PM2.5 concentrations. It may be affected by density and placement of
monitors, and coverage will vary across the country and within states. It
does not directly reflect exposure. Certain geographic areas, such as those
near busy roads, are likely to have higher values.
When creating measures we only consider monitors that have at least 11
observations per calendar quarter. It is important to understand that this
indicator is not for use–compliance determination with NAAQS or
reasonable further progress toward attaining compliance.
The relationship between ambient concentrations and personal exposure is
largely unknown, and it varies depending upon pollutant, activity patterns,
and microenvironments.

Data Sources
Limitations of Data
Sources

References

The percent of state population living in counties with no PM2.5
measurements must always be considered when attempting to estimate the
proportion of population at risk.
EPA Air Quality System Monitoring Data, State Air Monitoring Data.
http://www.epa.gov/air/data/aqsdb.html
Air monitoring data provides information regarding concentrations around
the specific location of each monitor. For PM2.5 this can be a rather large
area, except when unusual local emissions (agricultural fires) occur.
Within-county variation in concentrations will likely exist but will not be
captured in this measure. Many PM2.5 monitors operate once-every-third
day (some once-every-sixth day) and a few measure every day
Dockery DW and Pope CA. Acute respiratory effects of particulate air
pollution. Annu Rev Public Health 1994;15:107–132.
Cannon J. The Health Costs of Air Pollution: A Survey of Studies Published
1984– 1989. New York: American Lung Association; 1990.
U.S. Environmental Protection Agency. U.S. EPA Criteria Document for
PM. Available from: Volume 1
VOL_I_FINAL_PM_AQCD_OCT2004.PDF and Volume 2
VOL_II_FINAL_PM_AQCD_OCT2004.PDF

NCDM Recommendations Version 3.0
Page 47 3/19/2013

CONTENT DOMAIN: ASTHMA
INDICATOR: HOSPITALIZATIONS FOR ASTHMA
Type of EPHT Indicator

Measures

Derivation of Measures

Unit
Geographic Scope
Geographic Scale
Time Period
Time Scale

Rationale

Health Outcome
1. Number of hospitalizations for asthma
2. Minimum daily number of hospitalizations for asthma by month
3. Maximum daily number of hospitalizations for asthma by month
4. Average daily number of hospitalizations for asthma by month
5. Crude rate of hospitalization for asthma by age group (total, 0-4, 514, 15-34, 35-64, and 65+) per 10,000 population
6. Age-adjusted rate hospitalizations for asthma per 10,000
population (all ages)
When supported by sufficient data volume, the measures may also be
reported stratified by sex, race, and/or ethnicity.
Numerator:
Resident hospitalizations for asthma, ICD-9-CM: 493.XX.
Denominator:
Midyear resident population.
Adjustment:
Age-adjustment by the direct method to Year 2000 U.S. Standard
population
Hospital admission (categorized by discharge diagnosis)
State and national (tracking network states)
State and county
Hospital admissions from January 1 through December 31 for each
year, 2000–current
Daily, monthly, and annually (as appropriate for the measure)
In 2004, 20.5 million people in the United States reported having
asthma. In 2003, there were more than 574,000 hospitalizations for
asthma. In 2002, there were more than 4,200 deaths in which asthma
was the underlying cause. Asthma is the leading chronic health
condition among children. There are also large racial, income, and
geographic disparities in poor asthma outcomes. Asthma causes lower
quality of life, preventable undesirable health outcomes, and large
direct and indirect economic costs. Environment attributable fractions
of the 1988–1994 economic costs for asthma were 39.2% for children
aged <6 years and 44.4% for children aged 6–16 year, costing more
than $400 million for each age group.
A number of epidemiologic studies have reported associations between
air pollution exposures and asthma. The association between ambient

NCDM Recommendations Version 3.0
Page 48 3/19/2013

air particulate matter (PM) concentrations and asthma, including
increased hospital admissions, is well documented. Models
demonstrate 5–20% increases in respiratory-related hospital
admissions per 50µg/m3 of PM10 and 5–15% per 25µg/m3 of PM2.5,
with the largest effect on asthma admissions.
In the eastern United States, summer ozone pollution was associated
with more than 50,000 hospital admissions per year for asthma and
other respiratory emergencies. Large multi-city and individual city
studies found a positive association between ozone and total
respiratory hospital admissions, including asthma, especially during
the warm season. Among U.S. and Canadian studies, the ozoneassociated increase in respiratory hospital admissions ranged from 230% per 20 ppb (24 hour), 30 ppb (8-hour) or 40 ppb (1-hour)
increment of ozone in warm seasons.
In 2000, the Institute of Medicine concluded that allergens produced
by cats, cockroaches, and house dust mites exacerbates asthma, as
does exposure to environmental tobacco smoke (ETS) in pre-school
aged children. A 2005 California Air Resources Board report
concluded that ETS exacerbates asthma in children and adults (CARB,
2005). That report also estimated 202,300 childhood asthma episodes
occur each year in the United States as a result of exposure to ETS.
Developing a standardized analytic method for asthma hospital
admissions among residents in each state will provide more uniform
information for multiple users at the national, state, and local levels.
These measures will allow monitoring of trends over time, identify
high risk groups, and inform prevention, evaluation, and program
planning efforts.
These measures will address the following surveillance functions:

Use of the Measures



How many hospitalizations for asthma occur in every month?



Is there a seasonal or temporal trend of asthma hospitalizations?



What’s the distribution of asthma hospitalizations by place of
residence?



How do hospitalizations for asthma differ between geographic
areas (e.g., ZIP code, county, state, region)?



With further analysis … Are there disparities in asthma
hospitalizations by factors such as age, race, ethnicity, gender,
education, and/or income?

NCDM Recommendations Version 3.0
Page 49 3/19/2013



Which populations need targeted interventions?



When asthma data are linked with environmental variables, do the
linked measures identify environmental relationships that warrant
further investigation or environmental public health action?
Hospitalization data, by definition, do not include asthma among
individuals who do not receive medical care or who are not
hospitalized, including those who die in emergency rooms, in nursing
homes, or at home without being admitted to a hospital, and those
treated in outpatient settings.
Differences in rates by time or area may reflect differences or changes
in diagnostic techniques and criteria and in the coding of asthma.
Reporting rates at the state and/or county level will not show the true
asthma burden at a more local level (i.e., neighborhood).
Differences in rates by area may be due to different sociodemographic
characteristics and associated behaviors.
When rates across geographic areas are compared, many nonenvironmental factors, such as access to medical care and diet, can
affect the likelihood of a person being hospitalized for asthma.
Limitations of the
Measures

Reporting rates at the state and/or county level will not be resolved
geographically enough to be linked with many types of environmental
data.
When looking at small geographic levels (e.g., ZIP code), users must
consider appropriate cell suppression rules imposed by the data
providers or individual state programs.
Although duplicate records and transfers from one hospital to another
are excluded, the measures are based upon events, not individuals,
because no unique identifier is always available. When multiple
admissions are not identified, the true prevalence will be
overestimated.

Data Sources

Even at the county level, the measures generated will often be based
upon numbers too small to report or present without violating state and
federal privacy guidelines and regulations. Careful adherence to cell
suppression rules in cross tabulations is necessary, and methods to
increase cell sizes by combining data across time (e.g., months, years)
and geographic areas may be appropriate.
Numerator:
State inpatient hospitalization data (using admission date)

NCDM Recommendations Version 3.0
Page 50 3/19/2013

Denominator:
US Census Bureau population data
State hospital discharge data:
The use of a measure of all asthma hospitalizations will include some
transfers between hospitals for the same person for the same asthma
event. Variations in the percentage of transfers or readmissions for the
same asthma event may vary by geographic area and impact rates.
However, efforts were made to identify and exclude transfers based on
unique identifiers consisting of date of birth, zip code, gender, and
encrypted social security number when available.
Without reciprocal reporting agreements with abutting states,
statewide measures and measures for geographic areas (e.g., counties)
bordering other states may be underestimated because of health care
utilization patterns.
Each state must individually obtain permission to access and, in some
states, provide payment to obtain the data.
Veterans Affairs, Indian Health Services, and institutionalized (prison)
populations are excluded.
Limitations of Data
Sources

Practice patterns and payment mechanisms may affect diagnostic
coding and decisions by health care providers to hospitalize patients
Street address is not available in many states.
Sometimes mailing address of patient is listed as the residence address
of the patient.
Patients may be exposed to environmental triggers in multiple
locations, but hospital discharge geographic information is limited to
residence.
Since the data capture hospital discharges (rather than admissions),
patients admitted toward the end of the year and discharged the
following year will be omitted from the current year dataset.
Data will need to be de-duplicated (i.e., remove duplicate records for
the same event).
There is usually a two-year lag period before data are available from
the data owner.

NCDM Recommendations Version 3.0
Page 51 3/19/2013

Census data:
Available only every 10 years; thus, postcensal data must be estimated
when rates for years following the census year are calculated.
Postcensal estimates at the ZIP code level are not available from the
Census Bureau. These need to be extrapolated or purchased from a
vendor.
1. Centers for Disease Control and Prevention. Behavioral Risk
Factor Surveillance System (BRFSS) Prevalence Data. 1999–2010
November 16, 2011 [cited 2012 July 2]; Available from:
http://www.cdc.gov/asthma/brfss/default.htm#00.
2. Mannino, D.M., et al., Surveillance for asthma—United States,
1960–1995. MMWR CDC Surveill Summ, 1998. 47(SS-1): p. 1–
28.
3. Mannino, D.M., et al., Surveillance for asthma—United States,
1980–1999. MMWR Surveill Summ, 2002. 51(1): p. 1–13.
4. Britton, J. and S. Lewis, Epidemiology of Childhood Asthma, in
Asthma: Epidemiology, Anti-Inflammatory Therapy and Future
Trends, M. Giembycz and B. O'Connor, Editors. 2000, Birkhäuser
Basel: Switzerland. p. 25–56.
5. Gold, D.R. and R. Wright, Population disparities in asthma. Annu
Rev Public Health, 2005. 26: p. 89–113.

References

6. Lanphear, B.P., et al., Residential exposures associated with
asthma in US children. Pediatrics, 2001. 107(3): p. 505–11.
7. Lanphear, B.P., et al., Contribution of residential exposures to
asthma in us children and adolescents. Pediatrics, 2001. 107(6): p.
E98.
8. Redd, S.C., Asthma in the United States: burden and current
theories. Environ Health Perspect, 2002. 110 Suppl 4: p. 557–60.
9. Arif, A.A., J.E. Rohrer, and G.L. Delclos, A population-based
study of asthma, quality of life, and occupation among elderly
Hispanic and non-Hispanic whites: a cross-sectional investigation.
BMC Public Health, 2005. 5: p. 97.
10. Jorres, R.M.H., Atmospheric pollutants, in Asthma: Basic
Mechanisms and Clinical Management, P. Barnes, I. Rodger, and
N. Thomson, Editors. 1998, Academic Press: London. p. 589–596.
11. Trasande, L. and G.D. Thurston, The role of air pollution in
asthma and other pediatric morbidities. J Allergy Clin Immunol,
2005. 115(4): p. 689–99.
12. Jaffe, D.H., M.E. Singer, and A.A. Rimm, Air pollution and
emergency department visits for asthma among Ohio Medicaid

NCDM Recommendations Version 3.0
Page 52 3/19/2013

recipients, 1991–1996. Environ Res, 2003. 91(1): p. 21–8.
13. U.S. Environmental Protection Agency, Air Quality Criteria for
Particulate Matter (Final Report, Oct 2004), 2004, U.S.
Environmental Protection Agency. EPA 600/P-99/002aF-bF:
Washington, DC.
Institute of Medicine, Committee on the Assessment of Asthma and
Indoor Air. Division of Health Promotion. Disease Prevention.
Clearing the Air: Asthma and Indoor Air Exposures 2000,
Washington, DC: The National Academies Press.

NCDM Recommendations Version 3.0
Page 53 3/19/2013

Indicator Template
Content Area: Asthma
Indicator: Emergency Department Visits for Asthma
Environmental Public Health Tracking
Type of EPHT
Indicator

Measures

Health outcome
1. Annual age-adjusted rate of emergency department visits for asthma per
10,000 population
2. Annual crude rate of emergency department visits for asthma per 10,000
population
3. Annual number of emergency department visits for asthma
4. Average Number of emergency department visits for asthma as primary
diagnosis per month
Numerator:
 Emergency Department Visits during a calendar year with asthma (ICD9-CM 493) as the primary diagnosis (includes records for ED Visits
resulting in a hospitalization)
 Both inpatient and outpatient records with duplicates removed and
transfers to other hospitals included

Derivation of
Measure(s)

Unit

Denominator:
 Annual population estimates for state and county from U.S. Census
Bureau
Adjustment:
 Age-adjustment by the direct method to the Year 2000 US Standard
population
 U.S. 2000 standard population by age categories from Surveillance
Epidemiology and End Results (SEER), National Cancer Institute
1. Age-adjusted rate per 10,000 population
2. Rate per 10,000 population
3. Number
4. Number

Geographic Scope State and national
Geographic Scale Residents of jurisdiction – State, County
Hospital admissions between January 1 to December 31, inclusive, for each
Time Period
year, 2000–
Daily, monthly, and annually (as appropriate for the measure)
Time Scale
Asthma continues to be a serious public health problem that affects over 23
Rationale
million people including 7 million children in the United States. In 2008,
NCDM Recommendations Version 3.0
Page 54 3/19/2013

there were 456,000 hospitalizations and 1.8 million emergency department
visits (ED) for asthma.3 Asthma is the leading chronic health condition
among children.4 There are also large racial, income, and geographic
disparities in poor asthma outcomes.5 Asthma causes lower quality of life,
preventable undesirable health outcomes, and large direct and indirect
economic costs.
As a chronic respiratory disease, asthma attacks interfere with everyday
activities According to NCHS National Health Interview Survey, there were
10.5 million missed school days among children age 5–17 years and over
14.5 million missed work days in adult’s age 18 years or over in 2008. In
2007, there were over 3,400 deaths in which asthma was the underlying
cause.
Environment Attributable Fractions of the 1988-1994 economic costs for
asthma were 39.2% for children <6 years of age and 44.4% for 6- to 16-yearolds, costing more than $400 million for each age group. According to a more
recent estimation 30% of asthma exacerbations among children were related
to the environment. This was associated with an annual cost of $2.0 billion.
Despite the availability of effective prevention measures, asthma associated
costs are increasing.
Associations between environmental exposures and asthma have been
consistently demonstrated. Many outdoor air pollutants have been associated
with increased asthma ED visits. There is strong scientific evidence for
direct associations between increased ozone concentrations and increases in
asthma ED visits, in children and adults. In one study, asthma ED visits
increased by 33 percent when daily 1-hour maximum ozone concentrations
exceeded 75 ppb. Associations between asthma-related ED visits and ambient
air particulate matter—both PM10 and PM2.5—have been repeatedly
confirmed, and are especially robust for children. Other pollutants related to
higher asthma ED visit totals include carbon monoxide (CO), nitrogen
dioxide (NO2), and pollution from coal and petrochemical sources. Other
outdoor environmental triggers for asthma ED visits in children include weed
and tree pollen, and ambient temperature. Increased asthma ED visits has also
been associated with environmental tobacco smoke (ETS). Asthma ED visits
in children are consistently higher in the fall, co-occurring with the start of
the school year; increases in asthma ED visits in children have been shown to
be related to increased respiratory viral infections. The state emergency
department visit data is electronically maintained and is available in almost
every state in the U.S. Data stewards for 18 grantees maintain ED data.
The data has comparable basic information about each visit and can provide a
better tracking measure of asthma burden than inpatient hospitalization data
on its own. These measures can be used to evaluate the impact of ambient air
pollution on respiratory health of children and adults. Also, the measures can
be used for better resource management to further reduce the asthma related
NCDM Recommendations Version 3.0
Page 55 3/19/2013

expenditures. Combined with inpatient asthma data, emergency department
data will provide more complete spatial and temporal trends for asthma.
Additionally, emergency department visits are believed to be largely
preventable if managed properly through the use of Asthma Action Plans and
avoiding environmental triggers. This offers an outcome that may be a more
measurable indicator of environmental events and of public health
intervention
The development of a single analytic method for asthma emergency
department visits among persons living in state will inform multiple users:

Use of the
Measure

Limitations of the
Measure

State:
 May be linked with other risk factors such as air pollution to identify
susceptible populations and explore ecologic relationships
 Allows for a better understanding of what the asthma surveillance data
represents when interpreting number of inpatient hospitalizations
 Permits the monitoring of trends temporally and spatially
National:
 It will allow for comparison across states which can be used to target
interventions (especially for CDC and EPA).
Public:
 Public and concerned community members will be able to view the
Tracking Network webpage and learn the annual rate of asthma
emergency department visits and burden of asthma is high in their
community from.
 Numbers may be too small in rural areas to calculate stable rates.
 These measures do not account for other causes (triggers) of asthma or
other reasons for visiting the ED.
 The timing of the exposure may not correspond with the timing of the
asthma exacerbation leading to the ED visit.
 Individuals may have asthma exacerbations due to exposure to an
environmental risk factor that does not result in an ED visit and thus are
not captured in this measure.
 Cannot combine counts from asthma ED visit measure with counts from
asthma hospitalization measure because records for ED patients who are
subsequently hospitalized are already counted as hospitalizations (i.e.,
would result in double-counting of events).



Differences in rates by time or area may reflect differences or changes in
diagnostic techniques and criteria and in the coding of asthma.
Reporting rates at the state and/or county level will not show the true
asthma burden at a more local level (i.e. neighborhood).

NCDM Recommendations Version 3.0
Page 56 3/19/2013



Data Sources

Limitations of
Data Sources

Differences in rates by area may be due to different socio-demographic
characteristics and associated behaviors.
 When comparing rates across geographic areas, a variety on nonenvironmental factors, such as access to medical care and diet, can
impact the likelihood of persons hospitalized for asthma.
 Reporting rates at the state and/or county level will not be geographically
resolved enough to be linked with many types of environmental data.
 When looking at small geographic levels (e.g. ZIP code), users must take
into consideration appropriate cell suppression rules imposed by the data
providers or individual state programs.
 Although duplicate records and transfers from one hospital to another are
excluded, the measures are based upon events, not individuals, because
no unique identifier is always available. When multiple admissions are
not identified, the true prevalence will be overestimated.
 Even at the county level it can be expected that the measures generated
will often be based upon numbers too small to report or present without
violating state and federal privacy guidelines and regulations. Careful
adherence to cell suppression rules in cross tabulations is necessary and
methods to increase cell sizes by combining data across time (e.g.,
months, years) and geographic areas may be appropriate.
Numerator: State inpatient emergency department data
Denominator: US Census Bureau population data
State emergency department data:
 State emergency department data
 Need to obtain permission to use; not publicly available
 ED visits for asthma are only one piece of a larger picture that
describes asthma burden.
 Veteran’s Administration, Indian Health Service and institutionalized
(e.g. prison) populations are excluded
 In-state residents who visit in surrounding states would not be
included unless states have emergency department data sharing
agreements.
 Practice patterns and payment mechanisms may affect diagnostic
coding and decisions by health care providers.
 Do not have a zip code for all patients.
 Sometimes mailing address of patient (e.g., P.O. Box) is listed as the
residence address of the patient
 Patients may be exposed to environmental triggers in multiple
locations, but ED geographic information is limited to residence.
 Data will need to be de-duplicated using a standardized method.
Census data:
 Only available every 10 years, thus postcensal estimates are needed
when calculating rates for years following the census year.
 Postcensal estimates at the ZIP code level are not available from the

NCDM Recommendations Version 3.0
Page 57 3/19/2013

Census Bureau. These need to be extrapolated or purchased from a
vendor.
Related
Indicators

 Hospitalizations for Asthma
 Asthma Prevalence among Adults and Children
1. Summary Health Statistics for U.S. Adults: National Health Interview
Survey, 2008, Tables 3 and 4.
http://www.cdc.gov/nchs/data/series/sr_10/sr10_242.pdf
2. Summary Health Statistics for U.S. Children: National Health Interview
Survey, 2008, Table 1.
http://www.cdc.gov/nchs/data/series/sr_10/sr10_244.pdf
3. Akinbami LJ, Moorman JE, Liu X. Asthma Prevalence, Health Care
Use, and Mortality: United States, 2005–2009. National Health Statistics
Reports; No 32. Hyattsville, MD: National Center for Health Statistics,
2011.
4. Britton JR, Lewis SA, Epidemiology of childhood asthma. In Asthma:
Epidemiology, Anti-Inflammatory Therapy and Future Trends; MA
Giembycz and BJ O’Connor (Eds.),. Switzerland: Birkhäuser Verlag,
2000, pp. 25-56.
5. Gold DR, Wright R, Population disparities in asthma. Annu. Rev. Public
Health 2005; 26: 89-113.
6. Lanphear BP, Aligne CA, Auinger P, et al., Residential exposures
associated with asthma in US children. Pediatrics 2001; 107: 505-511.

References

7. Lanphear BP, Kahn RS, Berger O, et al., Contribution of residential
exposures to asthma in US children and adolescents. Pediatrics 2001;
107: e98.
8. Redd SC. Asthma in the United States: Burden and current theories.
Environ Health Perspect 2002; 110 (Suppl 4): 557-60.
9. Arif AA, Rohrer JE, Delclos GL. A population-based study of asthma,
quality of life, and occupation among elderly Hispanic and non-Hispanic
whites: a cross-sectional investigation. BMC Public Health 2005; 5: 97.
10. Pruss-Ustun A, Corvalan C. Preventing disease through health
environments. Towards an estimate of the environmental burden of
disease. World Health Organization. 2006.
11. Landrigan PJ, Schechter CB, et al. Environmental Pollutants and Disease
in American Children: Estimates of Morbidity, Mortality, and Costs for
Lead Poisoning, Asthma, Cancer, and Developmental Disabilities.
Environ Health Perspect. 2002:110:721-728.
12. Bahadori K, Doyle-Waters MM, Marra C, Lynd L, Alasaly K, Swiston J,
FitzGerald JM. Economic burden of asthma: a systematic review. BMC
Pulm Med. 2009 May 19;9:24.

NCDM Recommendations Version 3.0
Page 58 3/19/2013

13. Babin SM, Burkhom HS, Holtry RS, et al. Pediatric patient asthmarelated emergency department visits and admissions in Washington, DC,
from 2001-2004, and associations with air quality, socio-economic status
and age group. Environ Health. 2007; 6: 9.
14. Peel JL, Tolbert PE, Klein M, et al. Ambient air pollution and respiratory
emergency department visits. Epidemiology. 2005; 16: 164-174.
15. Stieb DM, Burnett RT, Beveridge RC, et al. Association between ozone
and asthma emergency department visits in Saint John, New Brunswick,
Canada. Environ Health Perspect. 1996; 104: 1354-60.
16. Tolbert PE, Mulholland JA, Macintosh DL, et al. Air quality and
pediatric emergency room visits for asthma in Atlanta, Georgia. Am J
Epidemiol. 2000; 151: 798-810.
17. Norris G, VoungPong SN, Koenig JQ, et al. An association between fine
particles and asthma emergency department visits for children in Seattle.
Environ Health Perspect. 1999; 107: 489-93.
18. Sun HL, Chou MC, Lue KH. The relationship of air pollution to ED
visits for asthma differs between children and adults. Am J Emerg Med.
2006; 24: 709-13.
19. Slaughter JC, Kim E, Sheppard L, et al. Association between particulate
matter and emergency room visits, hospital admissions and mortality in
Spokane, Washington. J Expo Anal Environ Epidemiol. 2005; 15: 153-9.
20. Villeneuve PJ, Chen L, Rowe BH, et al. Outdoor air pollution and
emergency department visits for asthma among children and adults: A
case-crossover study in northern Alberta, Canada. Environ Health. 2007;
6:40.
21. Teach SJ, Crain EF, Quint DM, et al. Indoor environmental exposures
among children with asthma seen in an urban emergency department.
Pediatrics. 2006; 117: S152-8.
22. Vargas PA, Brenner B, Clark S, et al. Exposure to environmental
tobacco smoke among children presenting to the emergency department
with acute asthma: A multicenter study. Pediatr Pulmonol. 2007; 42:
646-55.
23. Baibergenova A, Thabane L, Akhtar-Danesh N, et al. Effect of gender,
age, and severity of asthma attack on patterns of emergency department
visits due to asthma by month and day of the week. Eur J Epidemiol.
2005; 20: 947-56.
24. Silverman RA, Ito K, Stevenson L, et al. The relationship of fall school
opening and emergency department asthma visits in a large metropolitan
area. Arch Pediatr Adolesc Med. 2005; 159: 818-23.
25. Clark NA, Demers PA, Karr CJ, Koehoorn M, Lencar C, Tamburic L,
Brauer M. Effect of early life exposure to air pollution on development
NCDM Recommendations Version 3.0
Page 59 3/19/2013

of childhood asthma. Environ Health Perspect. 2010;118(2):284-90.
26. Expert Panel Report 3. Guidelines for the Diagnosis and Management of
Asthma. Bethesda, MD: US Department of Health and Human
Services, National Institutes of Health, National Heart, Lung, and Blood
Institute, National Asthma Education and Prevention Program, 2007.

NCDM Recommendations Version 3.0
Page 60 3/19/2013

CONTENT DOMAIN: BIRTH DEFECTS
INDICATOR: PREVALENCE OF BIRTH DEFECTS
Type of EPHT Indicator
Measure

Health Outcome
Five year prevalence rates of 12 birth defects per 10,000 live births.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.

Derivation of Measure(s)

Anencephaly
Spina bifida (without anencephaly)
Hypoplastic left heart syndrome
Tetralogy of Fallot
Transposition of the great arteries (vessels)
Cleft lip with or without cleft palate
Cleft palate without cleft lip
Hypospadias (male births only)
Gastroschisis
Upper limb deficiencies
Lower limb deficiencies
Trisomy 21
o
Among mothers <35 years of age at delivery
o
Among mothers ≥35 years of age at delivery

Five year prevalence rates at the state level are reported stratified by
maternal age at delivery, maternal ethnicity/race, and infant sex. Five
year prevalence rates at the county level are reported stratified by one
demographic variable at a time: maternal age at delivery, maternal
ethnicity/race, or infant sex.
Denominator is composed of all live-born infants in geographic region
of interest during a calendar year.
Numerator is composed of all live-born infants, fetal deaths (where
available), and terminations (where available) with birth defect ‘X’ in
the geographic region of interest during a calendar year.
For states that ascertain fetal deaths and/or terminations, two sets of
birth prevalence estimates are to be calculated for each birth defect—
one including and one excluding fetal deaths and/or terminations.

Unit
Geographic Scope
Geographic Scale
Time Period
Time Scale

Diagnosis of cases may be made up to one year of age—ascertainment
may be at any time.
Defect present at birth
State and National (tracking network states)
State, county
1998-current
Five year

NCDM Recommendations Version 3.0
Page 61 3/19/2013

Rationale

Use of the Measure

Birth defects pose a significant public health problem. One in 33
babies is born with a structural birth defect in the United States. Birth
defects are a leading cause of infant mortality; they are also
responsible for considerable morbidity and disability with enormous
economic and social costs. A lifetime of medical care and special
education for a single child can cost more than $500,000.
Approximately 60% of birth defects are of unknown etiology. The
ambient environment remains a source of great public concern, but
few environmental exposures have been well-studied. Most birth
defects likely will be explained by a complex interaction between
genetic predispositions and environmental factors. However, before
the ability to conduct studies to explore these interactions is achieved,
linking birth defects–outcome data with environmental hazard or
exposure data is critical. The first step in effecting successful linkages
of these data is the existence of high-quality birth defects prevalence
data for which the geospatial and temporal patterns and distributions
can be monitored. The environmental public health tracking (EPHT)
initiative is well-positioned to bring together birth prevalence data
from its state partners to begin analyses of these patterns, which will
provide important clues to public health officials and researchers.
The basic procedure for calculating birth prevalence is the same for all
the suggested birth defects. Once the input data are appropriately
prepared, birth prevalence will be calculable for all defects at the same
time.
State
Allow for consistent and rapid method for calculating and displaying
(using GIS) prevalence at selected geographical areas (i.e., county
level).
Allow for a better understanding of spatial and temporal patterns of
selected birth defects.
National
Allow for comparison of birth prevalence across states, which can be
used to target interventions. Any comparison of birth prevalence,
however, will need to account for the variability in data collection
methods between state surveillance systems. (See “Limitations of Data
Sources” below and introductory text in appended team
recommendations).
Local
Concerned community members will be able to view the tracking
network Web page to see the birth prevalence of selected birth defects
(while protecting confidentiality) at specified geographical areas. A

NCDM Recommendations Version 3.0
Page 62 3/19/2013

Limitations of the Measure

Data Sources

Limitations of Data
Sources

public health message will help interpret the results and provide more
information on selected birth defects and prevention measures (i.e.,
folic acid for prevention of neural tube defects, smoking and clefts,
alcohol and fetal alcohol syndrome, and known teratogenic
medications). A link to a list of known teratogens can be provided to
users.
Ideally, incidence rates would be used instead of birth prevalence to
measure birth defects occurrence. The numerator of the incidence
would be the number of new cases of birth defect A in an area and
time period and the denominator would be the number of conceptions
at risk for developing birth defect A in that area and time period.
Because both the number of conceptions and the number of cases
“lost” through spontaneous abortions (as well as terminations and later
fetal losses depending on the source of ascertainment for the specific
surveillance system) is unknown, incidence cannot be calculated. Birth
prevalence is the only appropriate measure that can be reported for
birth defects occurrence.
It is not feasible, at this time, to recommend that individual-level birth
defects surveillance data be made available on even a secure national
portal. Most states have strict guidelines with respect to
confidentiality, and even the publication of birth prevalence data based
on <5 cases in a geographic region is generally not done.
State birth defects surveillance systems: The data sources that
contribute to birth defects surveillance systems include the following
(this varies by system type):
 Vital records
 Hospital records (discharge summaries or disease indices, nursery
logs, NICU logs)
 Administrative databases (Medicaid, state hospital discharge,
HMO)
 Specialty data sources (specialty clinics, programs for children
with special health care needs)
 Prenatal diagnostic centers or genetics clinics
 Clinical examination
 Local or national laboratories for cytogenetic testing
Denominator data will come from state vital records—number of live
births, by year, by maternal age, and by race/ethnicity. These data may
be aggregated and provided to the birth defects surveillance system for
calculating birth prevalence, or it may be made available on an
individual level to the birth defects surveillance system. This varies by
state.
All states in the US do not have a birth defects surveillance program.
Among those that do, there is significant variability between
surveillance systems. These include:

NCDM Recommendations Version 3.0
Page 63 3/19/2013








Ascertainment method (active, passive, passive with followup/verification)
o
Primary differences are with data sources, coding,
availability of verbatim description, and case verification
Ascertainment of spontaneous fetal deaths and variability in
gestational age for inclusion.
Ascertainment of prenatally diagnosed cases and elective
terminations
Case definitions
Classification as isolated, multiple, or syndromic

Data for specific birth defects may not be collected by each state or
may only have been collected recently, limiting historical data for that
birth defect.
Address data tend to be based on address at delivery, not conception
(more relevant time period for birth defects-related exposure).
Approximately 50% of birth defects surveillance systems do not
geocode their address data.

NCDM Recommendations Version 3.0
Page 64 3/19/2013

CONTENT DOMAIN: CANCER
INDICATOR: INCIDENCE OF SELECTED CANCERS
Type of EPHT Indicator
Measure

Derivation of Measure(s)

Health Outcome
1. Annual number of cases for selected cancers, by state
2. Annual age-adjusted incidence rate for selected cancers per
100,000 population or per 1,000,000 for childhood cancers (<15 &
<20 years of age), by state
3. Average annual number of cases for selected cancers over five
year period, by county
4. Age-adjusted incidence rate for selected cancers per 100,000
population over a five year period, by county
Measures for each of the selected cancer types are provided by sex
and race/ethnicity groups. Some measures are also provided by age
group as defined below.
Numerator is composed of counts of unique invasive primary incident
cases of cancer “x” (bladder cancer also includes in situ) diagnosed
during a specified calendar year or five year period within residents
of a specified geographic region. Incident cancer data were originally
collected by state and regional cancer registries. It is proposed that
data for the National EPHT Network be obtained from the NCI and
CDC joint venture, State Cancer Profiles.
Denominator is composed of counts of the population residing in the
geographic region of interest during a specified calendar year or five
year period. Population data were originally collected by the U.S.
Census. For these national cancer indicators, population data is
obtained from the NCI and CDC’s State Cancer Profiles, which use
U.S. Census data as modified by SEER.
Rates will be age-adjusted to year 2000 U.S. standard population.
Cancer types:
Mesothelioma: SEER Recode B 36010. ICD-O-3 codes: histologies
9050-9055. Malignant cases: ICD behavior code ‘3’.
Melanoma of the skin*: SEER Recode B 25010. ICD-O-3 codes:
primary site C440-C449, histologies 8720-8790. Invasive melanoma
(behavior code ‘3’).

NCDM Recommendations Version 3.0
Page 65 3/19/2013

Liver & Intrahepatic Bile Duct: SEER Recode B 21071, 21072.
ICD-O-3 codes: primary sites C220, C221; excludes histologies:
9590-9989, 9050-9055, and 9140. Malignant cases: ICD behavior
code ‘3’.
Kidney & Renal Pelvis: SEER Recode B 29021, 29022. ICD-O-3
codes: C649, C659; excludes histologies: 9050-9055, 9140, 95909989. Malignant cases: ICD behavior code ‘3’.
Oral Cavity & Pharynx: SEER Recode B Site Groups 20010-20100
(20010, 20020, 20030, 20040, 20050, 20060, 20070, 20080, 20090,
20100). ICD-O-3 site codes: C000-C009, C019-C069, C079-C119,
C129-C140, C142-C148; excludes histologies 9050-9055, 9140,
9590-9989.
Esophageal: SEER Recode B 21010. ICD-O-3 site codes: C150C159; excluding histologies 9050-9055, 9140, 9590-9989.
Pancreas: SEER Recode B 21100. ICD-O-3 codes: C250-C259;
excluding histologies 9050:9055, 9140, 9590:9989.
Larynx: SEER Recode B 22020. ICD-O-3 codes: C320-C329;
excluding histologies 9050:9055, 9140, 9590:9989.
Lung & Bronchus: SEER Recode B 22030. ICD-O-3 Site codes
C340-C349; excludes histologies 9050-9055, 9140, 9590-9989.
Breast** (female): SEER Recode B 26001. ICD-O-3 Site codes
C500-C509; excludes histologies 9050-9055, 9140, 9590-9989.
Bladder: SEER Recode B 29010. ICD-O-3 Site codes C670-C679;
excludes histologies 9050-9055, 9140, 9590-9989. [includes invasive
and in-situ]
Brain & ONS***: SEER Recode B 31010, 31040. ICD-O-3 Site
codes C700-C709, C710-C719, C720-C729; excludes histologies
9050-9055, 9140, 9590-9989.
Thyroid: SEER Recode B 32010. ICD-O-3 Site codes C739;
excludes histologies 9050-9055, 9140, 9590-9989.
Non-Hodgkin Lymphoma: SEER Recode B 33041, 33042. ICD-O-3
codes: histology 9590-9596, 9670-9671, 9673, 9675, 9678-9680,
9684, 9687, 9689-9691, 9695, 9698-9702,9705,9708-9709, 97149719, 9727-9729; histology 9823 or 9827 in all sites except C420,
C421, C424.
NCDM Recommendations Version 3.0
Page 66 3/19/2013

Leukemia: SEER Recode B 35011, 35012, 35013, 35021, 35022,
35023, 35031, 35041, 35043. ICD-O-3 codes: ALL – histology
9826,9835-9837; Other lymphocytic – histology 9820, 9832-9834,
9940; Acute monocytic – histology 9891; CML – histology 9863,
9875, 9876, 9945, 9946; Other – histology 9860, 9930, 9801, 9805,
9931, 9733, 9742, 9800, 9831, 9870, 9948, 9963, 9964. Site codes
C420, C421, C424 – histology 9827. (Also include codes for CLL
and AML.)
Chronic Lymphocytic Leukemia (CLL): SEER Recode B 35012.
ICD-O-3 codes: C420, C421, C424 with histology 9823.
Acute Myeloid Leukemia (AML): SEER Recode B 35021. ICD-O-3
codes: histology 9840, 9861, 9866, 9867, 9871-9874, 9895-9897,
9910, 9920.
Child cancers: SEER ICCC3 childhood cancer codes
http://seer.cancer.gov/iccc/iccc3.html
NOTE: SEER Recode B (Dec 2003)
http://seer.cancer.gov/siterecode_b/icdo3_d12192003/
Tobacco-related cancers: consistent with SEER Recode B, CWG Cancer Team
NCDM specifies Histology Exclusions 9050-9055 (Mesothelioma), 9140 (Kaposi
Sarcoma), 9590-9989 (Lymphoma, Leukemia, Miscellaneous).
* Grantee portals may choose to additionally display In-situ cases, both
disaggregated and aggregated with invasive cases (“All combined”).
** Breast – Malignant/invasive only: The NEPHTN Metadata state “Counts and
rates for in situ breast cancer cases among women are presented; these are reported
separately and are not included in counts or rates for the "All Sites" category.”
(CDC-EHTB plans to delete this sentence from national portal Metadata.) The
NCDM states “Numerator is composed of counts of unique invasive primary
incident cases of cancer …” (in “Derivation of Measure”). Grantee portals may
choose to additionally display In-situ cases, both disaggregated and aggregated with
invasive cases (“All combined”).
*** Brain/ONS – Malignant/invasive only: The NEPHTN Metadata state
“Incidence data on nonmalignant primary brain and central nervous system (CNS)
tumors are available on this Web site.” (CDC-EHTB plans to delete this sentence
from national portal Metadata.) The NCDM states “Numerator is composed of
counts of unique invasive primary incident cases of cancer …” (in “Derivation of
Measure”).

Unit
Geographic Scope
Geographic Scale
Time Period
Time Scale
Rationale

Newly reported cancer case
State and national (tracking network states)
State and county.
2000-current
Annual and 5 year period
Approximately 1.4 million Americans are expected to be diagnosed
with cancer during 2007. The National Cancer Institute (NCI)
estimated that in January 2003, there were approximately 10.3 million
living Americans with a history of cancer. The risk of being

NCDM Recommendations Version 3.0
Page 67 3/19/2013

diagnosed with cancer increases as a person ages, and 77 % of all
cancers are diagnosed in Americans age 55 years or older. Cancer, a
diverse group of diseases characterized by the uncontrolled growth
and spread of abnormal cells, is believed to be caused by both
external and internal risk factors.
Major risk factors for cancer include tobacco use, diet, exercise, and
sun exposure (Clapp, Howe, Jacobs). For example, male smokers are
about 23 times more likely to develop lung cancer than male nonsmokers. Researchers have also identified genetic risks for cancer.
Female first degree relatives (mother, sisters, and daughters) of
women with breast cancer are about twice as likely to develop breast
cancer as women who do not have a family history of breast cancer
(Cancer Facts and Figures, 2007; ACS, 2007).
However, the etiology of many cancer types is not well established.
The physical environment (e.g., air quality, chemical pollution, and
water quality) remains a source of great public concern but few
community-level environmental exposures have been well-studied.
Studies of occupational cohorts have identified numerous suggestive
epidemiological associations between certain occupational exposures
and elevated cancer rates. After reviewing the evidence regarding the
causes of cancer in the United States, Doll and Peto published a
seminal article in 1981 estimating that 35% of all U.S. cancer deaths
were attributable to diet, 30% to smoking, 4% to occupation, and 2%
to pollution. While some authors have agreed with Doll and Peto
(Ames and Gold 1998), and others have cautioned against their
approach: “there is substantial evidence that occupational and
environmental exposures contribute to the burden of cancer” (Clapp,
Howe, and Jacobs 2006).
One way to assess cancer burden is to study geographic variation. In
recent years, geographic information systems (GIS) have become an
important tool for health and environmental research. GIS can extend
the analysis of data beyond simple mapping by enabling the linkage,
visualization, and analysis of multiple layers of health and
environmental data from both spatial and temporal perspectives.
One important use of geographic analysis of health data is in the
analysis of regional variations in cancer mortality and incidence. The
National Cancer Institute’s Atlas of Cancer Mortality for U.S.
Counties: 1950–1969 (Mason et al. 1975), represented the first effort
to map cancer mortality data at the county level throughout the
United States. In 1999, the national level analysis of cancer mortality
was updated by the NCI (Atlas of Cancer Mortality in the United
States, 1950–94, Devesa et al. 1999). More recently, multiple WebNCDM Recommendations Version 3.0
Page 68 3/19/2013

Use of the Measure

based data query systems have made U.S. cancer incidence and
mortality datasets and or maps available at the county (NCI/CDC
State Cancer Profiles: http://statecancerprofiles.cancer.gov/; NCI
SEER data: http://seer.cancer.gov/data/; NJ DHSS cancer online:
http://www.cancer-rates.info/nj/ ) and/or state level (NAACCR
CINA+ Online: http://www.cancer-rates.info/naaccr/ ; CDC U.S.
Cancer Statistics: http://apps.nccd.cdc.gov/uscs/ ).
At the local and state levels, the EPHT Network will:
Allow interested persons to obtain information on environmental
exposures (air pollution and drinking water quality) and cancer or
other health outcomes (birth defects, asthma, and birth weight) for a
selected geographic area and time interval. Standard suppression
rules will be used to prevent the release of information that might
reveal the identity of any person diagnosed with cancer. Public
health messages will help interpret the results and provide linkages to
additional information on cancer prevention, cancer etiology, and
cancer treatment options. While many of these diverse health and
environmental datasets are already available to the public, they are
not currently available through “one-stop-shopping” via the Internet.
Improve access to metadata regarding multiple health outcome
datasets and environmental exposure datasets for public health
practitioners and researchers. Enhanced access will provide better
understanding of the strengths and limitations of the available
datasets and may increase the use of the collected data.
Allow for a better understanding of spatial and temporal patterns of
selected cancers suggested to be linked to environmental exposures
within states.

Limitations of the Measure

At the national level, the EPHT Network will:
Enhance the opportunity for multi-state epidemiological research by
improving access to cancer incidence rates and environmental
exposure information. This could be particularly helpful for
uncommon cancer types or sub-types whereby incidence is too small
for meaningful ecological studies in individual states.
Counts and rates will be calculated based upon residential address at
time of diagnosis. No information is available on prior residences.
Geocoding accuracy, level of geocoding, and geocoding
completeness may vary by time and space. This could potentially
create geographically non-random errors in calculated rates of cancer.
No personal exposure information will be available, including
smoking history, diet, lifestyle, or history of cancer.

NCDM Recommendations Version 3.0
Page 69 3/19/2013

Data that will reveal the identity of any individual diagnosed with
cancer can not be released. Suppression rules will govern the release
of small case counts.

Data Sources
Strengths and Limitations
of Data Sources

No information will be available on the latency of cancer cases.
National Cancer Institute, Surveillance Epidemiology and End
Results; CDC National Program of Cancer Registries
All of the 16 states and the 1 city participating in the EPHT Network
are working with their state and/or regional cancer registry
program(s). Registry training, data collection, data coding, data
cleaning, and quality control programs are highly standardized and
subject to annual evaluation. Documentation is available online from
the North American Association of Centralized Cancer Registries
(NAACCR).
(http://www.naaccr.org/index.asp?Col_SectionKey=7&Col_ContentI
D=135).
State cancer registry programs may vary, however, regarding the
availability and quality of residential address information collected
and completeness of geocoding efforts.

NCDM Recommendations Version 3.0
Page 70 3/19/2013

CONTENT DOMAIN: CARBON MONOXIDE
INDICATOR: HOSPITALIZATIONS FOR CARBON MONOXIDE
POISONING
Type of EPHT Indicator

Health Outcome/Exposure
1.
2.

Measures
3.

Derivation of measure

Number of hospitalizations for carbon monoxide (CO) poisoning
Crude rate of hospitalization for CO poisoning per 100,000
population
Age-adjusted rate of hospitalization for CO poisoning per 100,000
population

Numerator:
Resident hospitalizations for CO poisoning that meet the 1998 CSTE
case definition for public health surveillance for a “Confirmed” or
“Probable” case of acute CO poisoning in administrative data sets.
Frequencies for three unique groups:
1.
Unintentional, non-fire related
2.
Unintentional, fire-related
3.
Unknown intent
Denominator: Midyear resident population
Adjustment: Age-adjustment by the direct method to year 2000 US
Standard Population

Unit

Hospital admission (categorized by discharge diagnosis)

Geographic Scope

State and national (tracking network states)

Geographic Scale

State; county when feasible

Time Period

2000-current

Time Scale

Calendar year

NCDM Recommendations Version 3.0
Page 71 3/19/2013

Rationale

Carbon monoxide (CO) is an odorless, colorless gas that usually remains
undetectable until exposure results in injury or death. Each year in the
United States, an estimated 10,000 persons seek medical attention or
lose at least one day of normal activity because of CO intoxication.
There is limited information on CO hospitalization. In Florida, 1,494
were hospitalized with a diagnosis of CO poisoning from 1999–2007.
Out of which 10% (n=143) were unintentional fire-related, 33% (n=493)
were unintentional non-fire-related, and 17% (n=256) were from
unknown cause of CO poisoning. During 2000–2009, a total of 68,316
CO exposures were reported to poison centers across United States.
Persons hospitalized with CO poisoning are among the most
severely poisoned cases. Unintentional CO poisoning is almost entirely
preventable. These data are available in most states.

Use of the Measure

These data can be used to assess the burden of severe CO poisoning,
monitor trends over time, identify high-risk groups, and enhance
prevention, education, and evaluation efforts.

Limitations of the
Measure

Hospitalization data, by definition, do not include: persons treated in
outpatient settings (e.g., emergency departments, urgent care clinics,
clinicians’ offices or hyperbaric chambers but not hospitalized); persons
who call poison control centers and are managed at the scene, and/or
receive medical care but are not hospitalized; persons who do not seek
any medical care; or persons who die immediately from CO exposure
without medical care.

Data Sources

Numerator:
State inpatient hospital discharge data
Denominator:
U.S. Census Bureau population data

Limitations of the Data
Source

The use and quality of ICD9-CM coding varies across jurisdictions; this
is especially true of the codes used to describe how an injury occurs,
indicated as E-codes. Examples of this variation include:

The number of diagnostic fields available to specify cause of the
injury;

Whether E-codes are mandated;

The completeness and quality of E-coding; for example, the
reliability of ICD-9-CM coding to distinguish between cases of
CO poisoning that are intentional or unintentional, and/or fire-or
non-fire related
The toxic effects of CO exposure are nonspecific and easily
misdiagnosed when CO exposure is not suspected. These misdiagnosed
cases will not be counted.
These data usually do not include data from federal facilities such as
Veteran's Administration hospitals.

NCDM Recommendations Version 3.0
Page 72 3/19/2013

These data usually include only cases of state residents treated within
the state. Health-care access is not restricted to these political
boundaries so patients hospitalized for CO poisoning in another state
may not be counted in their own state. Likewise, they may not be
counted in the jurisdiction in which they were treated. Currently, few
states have access to, or agreements to obtain, hospital discharge data
from other states where their state residents may be hospitalized. To the
extent that patients are treated out of state, there is undercounting of the
rate of state residents poisoned by CO.
Differences in rates between jurisdictions may reflect differences in
hospital admissions practices for treating persons with severe CO
poisoning. For example, some facilities may routinely admit all
patients treated with hyperbaric oxygen; other facilities may release
patients treated with hyperbaric oxygen after the treatment is completed
if they are in stable condition.
Race and ethnicity are important risk factors for CO poisoning, yet,
many hospitalization data sets do not contain these data. Those that do
may have data quality issues.
Census data:
 Only available every 10 years, thus postcensal estimates are needed
when calculating rates for years following the census year.
 Postcensal estimates at the ZIP code level are not available from the
Census Bureau. These need to be extrapolated or purchased from a
vendor.
References

1. Centers for Disease Control and Prevention, Perspectives in Disease
Prevention and Health Promotion Carbon Monoxide Intoxication—
A Preventable Environmental Health Hazard MMWR, 1982.
31(39): p. 529–31.
2. Centers for Disease Control Prevention, Carbon monoxide
exposures—United States, 2000–2009. MMWR, 2011. 60(30): p.
1014–7.
3. Harduar-Morano, L. and S. Watkins, Review of unintentional nonfire-related carbon monoxide poisoning morbidity and mortality in
Florida, 1999–2007. Public Health Rep, 2011. 126(2): p. 240–50.
4. King, M.E. and S.A. Damon, Attitudes about carbon monoxide
safety in the United States: results from the 2005 and 2006 Health
Styles Survey. Public Health Rep, 2011. 126 Suppl 1: p. 100–7.

NCDM Recommendations Version 3.0
Page 73 3/19/2013

CONTENT DOMAIN: CARBON MONOXIDE
INDICATOR: EMERGENCY DEPARTMENT VISITS FOR CARBON
MONOXIDE POISONING
Type of EPHT Indicator

Health Outcome
1.

Measures

Derivation of measure

2.
3.

Number of emergency department (ED) visits for CO poisoning
Crude rate of ED visits for CO poisoning per 100,000 population
Age-adjusted rate of ED visits for CO poisoning per 100,000
population

Numerator:
Resident emergency department visits for CO poisoning that meet the
1998 CSTE case definition for public health surveillance for a
“Confirmed” or “Probable” case of acute CO poisoning in
administrative data sets.
Frequencies for three unique groups:
1.
Unintentional, non-fire related
2.
Unintentional, fire-related
3.
Unknown intent
Denominator: Midyear resident population
Adjustment: Age-adjustment by the direct method to year 2000 US
Standard Population

Unit

Emergency department visit

Geographic Scope

State and national (tracking network states)

Geographic Scale

State

Time Period

2000-current

Time Scale

Calendar year

NCDM Recommendations Version 3.0
Page 74 3/19/2013

Rationale

Carbon Monoxide (CO) poisoning is preventable; nonetheless,
unintentional, non-fire-related CO poisoning is responsible for
approximately 15,000 emergency department visits and nearly 500
deaths annually in the United States. During 2004–2006, an estimated
average of 20,636 ED visits for nonfatal, unintentional, non-firerelated CO exposures occurred each year. Approximately 73% of these
exposures occurred in homes, and 41% occurred during winter months
(December–February). Prevention efforts targeting residential and
seasonal CO exposures can substantially reduce CO-related morbidity.
During 2000–2009, a total of 68,316 CO exposures were reported to
poison centers across United States.
Persons admitted to emergency departments and diagnosed with CO
poisoning range from suspected exposure to severe poisonings that
may result in treatment and release, hospitalization, or death.
Emergency department visits represent patients not counted in other
clinical settings. Unintentional CO poisoning is usually preventable.
Emergency department data are available in more than 50% of the
states and that number is increasing.

Use of the Measure

These data can be used to assess the burden of CO poisoning and to
monitor trends over time as well as to identify high risk groups, and
enhance prevention, education, and evaluation efforts.

Limitations of the Measure

Measures based on emergency department data alone may
underestimate its prevalence because these data may not include
persons that are managed at the scene, persons who do not seek any
medical care, persons admitted without first visiting an emergency
department, or persons who die immediately from CO exposure
without medical care.
Numerator:
State emergency department visit data

Data sources
Denominator:
U.S. Census Bureau population data

Limitations of the Data
Source

Emergency department data have limitations for comparisons across
jurisdictions because the use and quality of ICD-9-CM coding may
vary across jurisdictions; this is especially true of the codes used to
describe how an injury occurs, indicated as E-codes. Examples of this
variation include:

The number of diagnostic fields available to specify cause of
the injury vary from nine to unlimited (in some states reaching
more than 100);

E-codes are mandated in some jurisdiction but not in others;

NCDM Recommendations Version 3.0
Page 75 3/19/2013





The completeness and quality of E-coding vary by hospital as
well as jurisdiction. In addition, the reliability of ICD-9-CM
coding to distinguish between cases that are intentional or
unintentional, fire-related, or of unknown intent is
undocumented;
States are inconsistent in the use of intent codes.

The toxic effects of CO exposure are nonspecific and easily
misdiagnosed when CO exposure is not suspected. These
misdiagnosed cases will not be counted.
These data usually do not include data from federal facilities such as
Veteran's Administration hospitals.
These data usually include only cases of state residents who were
treated within the state. Health care access is not restricted to these
political boundaries so people discharged from the emergency
department for CO poisoning in another state will neither be counted
in their own state nor in the jurisdiction in which they were treated.
Currently, few states have access to, or agreements to obtain, their
emergency department data from other states in which their residents
may have received treatment. To the extent that patients are treated
out of state, there is undercounting of the rate of residents poisoned
by CO.
Regional variation between emergency departments in diagnosing CO
poisoning may exist.
Many emergency department visit data sets do not contain race or
ethnicity information and those that do may have data quality issues.
Yet, these characteristics are known risk factors for CO poisoning.
Census data:
 Only available every 10 years, thus postcensal estimates are
needed when calculating rates for years following the census year.
 Postcensal estimates at the ZIP code level are not available from
the Census Bureau. These need to be extrapolated or purchased
from a vendor.
References

1. Centers for Disease Control and Prevention. Perspectives in
Disease Prevention and Health Promotion Carbon Monoxide
Intoxication—A Preventable Environmental Health Hazard
MMWR Morb Mortal Wkly Rep 1982;31(39):529–31.
2. Centers for Disease Control Prevention. Nonfatal, unintentional,

NCDM Recommendations Version 3.0
Page 76 3/19/2013

non-fire-related carbon monoxide exposures—United States,
2004-2006. MMWR Morb Mortal Wkly Rep 2008;57(33):896–9.
3. Hampson NB. Emergency department visits for carbon monoxide
poisoning in the Pacific Northwest. J Emerg Med
1998;16(5):695–8.
4. Kao LW, Nanagas KA. Carbon monoxide poisoning. Emerg Med
Clin North Am 2004;22(4):985–1018.
5. Partrick M, Fiesseler F, Shih R, Riggs R, Hung O. Monthly
variations in the diagnosis of carbon monoxide exposures in the
emergency department. Undersea Hyperb Med 2009;36(3):161–7.

NCDM Recommendations Version 3.0
Page 77 3/19/2013

CONTENT DOMAIN: CARBON MONOXIDE
INDICATOR: CARBON MONOXIDE POISONING MORTALITY
Type of EPHT Indicator

Health Outcome
1.

Measures

2.
3.

Number of deaths from CO poisoning
Crude rate of death from CO poisoning per 100,000 population
Age-adjusted rate of death from CO poisoning per 100,000
population

Numerator:
Resident deaths from CO poisoning for three unique groups:
1.
Unintentional, non-fire related
2.
Unintentional, fire-related
3.
Unknown intent
Derivation of measure
Denominator: Midyear resident population
Adjustment: Rates age-adjusted by the direct method to the Year
2000 U.S. Standard Population
Unit

Deaths due to CO poisoning

Geographic Scope

State and National

Geographic Scale

State

Time Period

2000-current

Time Scale

Calendar year

NCDM Recommendations Version 3.0
Page 78 3/19/2013

Rationale

CO is an odorless, colorless gas that usually remains undetectable
until exposure results in injury or death. Carbon monoxide (CO)
poisoning is a leading cause of unintentional poisoning deaths in the
United States. CO poisoning is preventable; nonetheless,
unintentional, non–fire-related CO poisoning is responsible for
approximately 15,000 emergency department visits and nearly 500
deaths annually in the United States. During 1999–2004, CO
poisoning was listed as a contributing cause of death on 16,447 death
certificates in the United States and 2,631 (16%) were classified as
both unintentional and non-fire-related deaths. The annual average
age-adjusted death rate in the U.S. was 1.5 deaths per million
persons. The US Consumer Product Safety Commission’s historical
data indicate that there is a statistically significant increasing trend in
non-fire CO fatalities from 1999 through 2007. In 2007, 183
unintentional consumer product–related, non–fire-related CO deaths
were reported. Out of which heating systems were associated with the
largest percentage of non-fire CO poisoning fatalities at 38 percent
(estimated 70 deaths); Engine-Driven Tools-related CO fatalities
were also associated with 38 percent (69 deaths), and the remaining
six product categories [Charcoal Grills or Charcoal (7 deaths);
Ranges, Ovens (7 deaths); Water Heaters (3 deaths); Grills, Camp
Stoves (3 deaths); Other Products (1 death); and Multiple Products
(24 deaths)] combined were associated with a total of 25 percent.
Death is the most severe outcome of CO poisoning. Unintentional
CO poisoning deaths are almost entirely preventable. Most localities
have access to data on their resident deaths.

Use of the Measure

These data can be used to assess the burden of severe CO poisoning,
monitor trends over time, and enhance prevention, education, and
evaluation efforts.

Limitations of the Measure

This measure understates the burden of CO poisoning because most
cases do not result in death. Rates can be misleading (i.e., do not
reflect risk of occurrence) if a relatively large proportion of deaths
occur to non-residents poisoned within the jurisdiction (they are
excluded from the rate calculation). Death investigation laws vary by
locale.
Numerator:
Death certificate records from vital statistics agency

Data Sources
Denominator:
Population counts or estimates from the U.S. Bureau of the Census

NCDM Recommendations Version 3.0
Page 79 3/19/2013

Limitations of the Data
Source

References

Death investigation laws vary by locale. In addition, variations may
occur between localities in how medical
examiners/coroners/physicians assign intentionality. Thus an area
where the ME/coroner/physician is disinclined to attribute a CO
poisoning to suicide will have a higher unintentional CO poisoning
death rate than a comparable locale. Finally, CO poisonings that are
unrecognized by the ME/coroner/physician will be attributed to
other causes.
1. Centers for Disease Control Prevention, Carbon monoxide-related deaths--United States, 1999-2004. MMWR Morb Mortal
Wkly Rep, 2007. 56(50): p. 1309-12.
2. Centers for Disease Control Prevention, Unintentional non-firerelated carbon monoxide exposures--United States, 2001-2003.
MMWR Morb Mortal Wkly Rep, 2005. 54(2): p. 36-9.
3. Mott, J.A., et al., National vehicle emissions policies and
practices and declining US carbon monoxide-related mortality.
JAMA, 2002. 288(8): p. 988-95.
4. Hnatov, MV. Non-Fire Carbon Monoxide Deaths Associated
with the Use of Consumer Products 2007 Annual Estimates.
Bethesda, MD: US Consumer Product Safety Commission.
Available at:
http://www.cpsc.gov/library/foia/foia11/os/co10.pdf. Accessed
July 18, 2012

NCDM Recommendations Version 3.0
Page 80 3/19/2013

CONTENT DOMAIN: CARBON MONOXIDE
INDICATOR: REPORTED EXPOSURE TO CARBON MONOXIDE
Type of Indicator

Exposure, Health Outcome

Measures

1. Number of unintentional CO exposures reported to poison control
centers by resulting health effect and treatment in a healthcare
facility
2. Crude rate of unintentional CO exposures reported to poison
control centers per 100,000 population by resulting health effect
and treatment in a healthcare facility

Derivation of measures

Number of reported cases of unintentional carbon monoxide exposure
stratified by presence of subsequent health effect and consequential
treatment in a healthcare facility
Denominator used is Midyear resident population

Unit

Reported exposure to CO

Geographic Scope

State and national (tracking network states)

Geographic Scale

County

Time Period

2000- current

Time scale

Annual

Rationale

PCCs serve the public and healthcare providers in the management of
actual or potential exposure to hazardous substances, including CO.
PCC calls are fielded by certified specialists in poisoning information
(SPIs), and recorded in a standard electronic format. Regional PCC
data are centralized nationally by AAPCC annually.
PCC calls provide information about CO exposure that may not
otherwise be captured in hospital discharge data or emergency
department data. These include events where CO exposure was
detected but did not result in symptoms, where symptoms were mild
and did not require follow-up in a health care facility, and where the
event resulted in symptoms but the patient refused to seek medical
treatment. Two state-based evaluations (Connecticut [1] and
Wisconsin [2]) found minimal overlap between persons using PCCs

NCDM Recommendations Version 3.0
Page 81 3/19/2013

and persons treated in emergency departments. As such, tracking of
PCC calls in addition to indicators of mortality, hospitalizations, and
emergency room visits provides a more complete picture of the
public health burden of CO exposure.
Use of the Measure

These data may be used to estimate the population's exposure to CO
and to monitor trends over time. They may also be used to estimate
symptomatic CO exposures among exposed persons who may not be
treated in a health care facility and therefore would not be captured in
other health outcome datasets.

Limitations of the Measure

Exposure status should not be considered confirmed. In some cases,
ambient air sampling results or the patient’s lab results may be
reported in the case notes but only when this information is available
or provided to the SPI. In addition, it should be noted that because
they may contain identifiable and sensitive information, SPI notes are
removed from case records by regional PCCs before submitting to the
AAPCC and are therefore unavailable at the national level.
Not all potentially hazardous CO exposures will be captured by PCC
calls. For example, cases of moderately elevated exposure in the
home are unlikely to be recognized if there are no acute symptoms
and a CO alarm is not installed. Moreover, knowledge, attitudes, and
practices around the use of PCCs likely vary both within and across
jurisdictions. In the event of suspected exposure, callers may first
notify their local fire department or call 911 or even their utility
provider; in either case, the regional PCC may not be simultaneously
notified. Practices by health care providers that use PCCs are also
likely to vary from one jurisdiction to another. Generally speaking,
healthcare providers use the PCC as a resource in the diagnosis and
treatment of poisonings; in addition, in New York City, where CO
poisoning was designated as an immediately reportable condition in
2004, the PCC plays an integral role in the management of reports
from healthcare providers and in the rapid referral of the fire
department for investigation at the site of exposure for the prevention
of secondary cases (3). For these reasons, caution should be
exercised in comparing rates of reported exposure across states.

Data Sources

Numerator:
PCC calls (usually in standard Toxicall database)
Denominator:
U.S. Census Bureau population data

Limitations of the Data
Sources

SPIs are not required to collect patient state/ZIP code unless the
patient is the caller. Using caller state/ZIP code to determine

NCDM Recommendations Version 3.0
Page 82 3/19/2013

residency may cause the number of calls pertaining to state residents
to be overestimated—for example, when the caller is an out-of-state
health care provider.
The number of cases may differ slightly between datasets obtained
directly from the state’s PCC and the national AAPCC dataset for
that state; this is typically due to calls that are re-routed to another
state when the state’s PCC is overloaded. The AAPCC national
dataset is corrected for such instances.
Age adjustment is not recommended since age is often estimated
(such as "Adult > 19" or “50s”).

References

1. Toal B. Comparison of Three CO Databases in Connecticut
[PowerPoint presentation]. EPHT Web Seminar; 2006 June.
2. Bekkedal M, Sipsma K, Stremski ES, Malecki KC, Anderson
HA. Evaluation of five data sources for inclusion in a
statewide tracking system for accidental carbon monoxide
poisonings. WMJ. 2006 Mar;105(2):36-40.
3. Wheeler K, Kass D, Hoffman R, Vecchi M, Allocca A.
Preventing CO poisoning: tracking the impact of legislative
and regulatory changes in New York City [PowerPoint
Presentation]. Annual Meeting of the Council of State and
Territorial Epidemiologists; 2006 June.

NCDM Recommendations Version 3.0
Page 83 3/19/2013

CONTENT DOMAIN: CARBON MONOXIDE
INDICATOR: HOME CARBON MONOXIDE DETECTOR
COVERAGE
Type of Indicator

Intervention

Measure

Percent of Behavioral Risk Factor Surveillance System (BRFSS)
respondents reporting at least one CO detector in their household
Numerator:
The number of respondents reporting CO detector in household

Derivation of Measure

Denominator:
The number of respondents reporting CO detector in household plus
respondents reporting no CO detector in household
Proportion is adjusted using the survey’s household weight

Unit

CO detector presence

Geographic Scope

State and national (tracking network states)

Geographic Scale

State

Time Period

2004; States’ BRFSS surveys should include this question every 3–5
years and/or when implementing interventions, such as new
legislation, to increase the use of CO alarms

Time Scale

Annual

Rationale

Correctly installed and maintained CO detectors can prevent injury
and death from exposure to CO.

Use of the Measure

Collected data will determine the occurrence of CO detectors in
homes. These data also can be combined with other data collected by
the BRFSS survey, including respondent demographics (e.g., age, sex,
and race of survey respondents and age and sex composition of
household), socioeconomic characteristics (e.g., insurance status), and
relevant health and prevention risk factors (e.g., smoking status,
presence of fire alarms). The results of these analyses can be used to
target and evaluate public health prevention strategies.
Notes about conducting the analysis:

NCDM Recommendations Version 3.0
Page 84 3/19/2013

BRFSS data should be analyzed by experts in analysis of sample
survey data and the software available to conduct this type of analysis
(e.g., SUDAAN and SAS survey procedures).
The BRFSS survey is designed so that the primary sampling unit is the
respondent. As such, BRFSS data are typically directly weighted to
account for sampling error based on data collected at the individual
level. However, the question about CO detectors is based on the
household rather than the individual as the sampling unit. Using the
weighting designed for individuals may bias the prevalence estimate
of household risk factors. The indicator will therefore use a weight
based on the potential error associated with sampling the household
rather than the individual.
Carbon monoxide alarms must be properly installed and maintained to
be effective; a single question does not capture information about
either. Maine has developed two questions that can be asked to get
supplemental information on maintenance:
1. Is your carbon monoxide detector battery powered or have a
battery for back-up power?
Limitations of the Measure
Response categories: Yes; No; Don’t Know; Refused
2. When was the last time you checked the batteries?
Response categories (Read only if needed): Within the past
year; More than a year; Don’t know/Not sure; Refused
BRFSS state-added question from the Indoor Air Pollution Module,
question number 4:
Data Sources

Limitations of the Data
Resources

A carbon monoxide or CO detector checks the level of carbon
monoxide in your home. It is not a smoke detector. Do you have a
carbon monoxide detector in your home?
While the data collection methods are standardized to allow
comparisons between states, there may still be bias introduced by
“house-effects”—that is, the variation introduced by different
organizations and individuals implementing the survey for different
states.
The BRFSS questionnaire is available in English or Spanish language
versions; persons who are not conversationally fluent in English (or
Spanish in the states that offer the Spanish-language option) are not
eligible. This population of non-English speakers may differ
systematically from English speakers in health and behavior

NCDM Recommendations Version 3.0
Page 85 3/19/2013

characteristics, including the presence of a CO detector in their homes.
The BRFSS is a telephone survey. While the effect of telephone noncoverage on estimates derived from BRFSS is small, the population
without telephones is not likely representative of the general
population. In particular, this population is less likely to have a CO
detector in the household; therefore, these results should not be
generalized to populations without telephone coverage.
An increasing number of households use telephone technology that
may result in changes in the population sampled and therefore may
make the survey results less reliably generalized and introduce other
bias. Two examples are:
1. Households with cellular telephones and no traditional
telephone. These households are not in the sampling frame for
the BRFSS
2. Households that use Caller ID to screen calls; their members
may be less likely to pick up the call.
Surveys based on self-reported information are likely less accurate
than those based on physical measurements. However, when
measuring change over time, this type of bias is likely to be constant
and therefore not a factor in trend analysis.

NCDM Recommendations Version 3.0
Page 86 3/19/2013

CONTENT DOMAIN: CHILDHOOD LEAD POISONING
INDICATOR: TESTING COVERAGE AND HOUSING AGE
Type of EPHT Indicator
Measures

Derivation of Measure(s)

Hazard /Intervention
1. Number of children born in the same year and tested for lead
before age 3
2. Percent of children born in the same year and tested before age 3
3. Number of homes built before 1950 (as measured in the 2000
Census)
4. Percent of homes built before 1950 (as measured in the 2000
Census)
Use birth year cohort to calculate the percentage of children with at
least one test prior to age 36 months.

Unit
Geographic Scope
Geographic Scale
Time Period

Use 2000 Census, Summary file 3, to calculate the percentage of pre1950 housing units
Proportion of houses by age-based hazard assessment
State and national
county and state
2000-

Time Scale

annual; birth cohort

Rationale

Elevated BLLs in young children have been associated with adverse
health effects ranging from learning impairment and behavioral
problems to death. Because children may have elevated BLLs and not
have any specific symptoms, CDC recommends a blood-lead test for
young children at risk for lead poisoning. Risk factors identified in the
National Health and Nutrition Examination Surveys (NHANES)
include living in housing built before 1950, especially deteriorating
condition, being African American and living in a family in poverty.
Many states have adopted a targeted testing strategy (test children at
high risk), and some states recommend universal testing (test all
young children). Nevertheless, studies have documented low bloodlead testing rates among children at high risk. CDC recommends that
state and local childhood lead poisoning prevention programs
(CLPPPs) evaluate testing among high-risk populations. All CLPPPs
have assessed testing in their states but many methods have been used
and it is not possible to compare across states.
CLPPPs also administer education campaigns for physicians and
parents about childhood lead poisoning to enable them to identify

NCDM Recommendations Version 3.0
Page 87 3/19/2013

children at risk.

Use of the Measure

For both universal testing plans and targeted testing plans, children
should be tested at least once before the age of 3 years. Some states
require more than one test between the ages of 6 and 36 months.
Using a birth cohort, the number of children born in a specific year
tested before the age of 36 months can be determined.
State
Identify populations that are not being tested adequately and improve
testing
Allow for a better understanding of what the blood-lead surveillance
data represent
National
Allow for comparison across states; such comparison can be used to
target interventions (especially CDC, EPA, HUD)
Public/parents
Determine if their community is at risk and the percentage of children
being tested. There will be a public health message which will help
interpret the results and provide more information on lead sources and
prevention.

Limitations of the Measure

Health care providers
Identify children who should be tested for lead by identifying highrisk communities
This measure estimates testing rates in children living in communities
which may be at greater risk of exposure due to older housing. It is a
surrogate for a child’s risk of lead poisoning due to lead paint in the
home. A more direct measure would be based on individual children
and the actual age of their housing.
Some tested children’s addresses are not in the CLPPP data system,
while only the provider’s address is provided for other children. This
can result in some tests being attributed to the wrong county or not
being counted at all.
Counties are not homogenous with respect to the distribution of lead
hazards or risk factors for lead exposure.
Using number of pre-1950s housing from Census does not account for
houses that have been renovated or have had lead removed.
This measure does not account for other lead sources in the
community.

NCDM Recommendations Version 3.0
Page 88 3/19/2013

Children may be exposed to lead paint in neighboring counties
(visiting family, day care)

Data Sources

Limitations of Data
Sources

Many states require children be tested more than once. This indicator
does not determine how many children are tested more than once to
meet such state requirements.
 Childhood Blood Lead Surveillance Data
 US Census (Summary file 3) for total number of housing units and
number of pre-1950 units
 Vital statistics birth data for number of births
Childhood Blood Lead Surveillance Data

Surveillance data are not randomly sampled or representative
of the population.

Addresses for all children tested are not included.

Address of the treating clinic is listed sometimes as the address
of the child.

De-duplication by a standardized method will be required

Race and ethnicity are not always captured.
Census data

Data are available only every 10 years.

Does not have information on renovation of pre 1950 housing
is not available.

Does not have information on the condition of the housing is
not available.

Address level information on the year the housing was built is
not available.
Vital Statistics Birth Data

Children may move to another county after birth

NCDM Recommendations Version 3.0
Page 89 3/19/2013

CONTENT DOMAIN: CHILDHOOD LEAD POISONING
INDICATOR: BLOOD LEAD LEVELS BY BIRTH COHORT
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator

Exposure

Measure(s)

1. Number of children born in the same year and tested , by county
and state
2. Percent of children born in the same year and tested, by county and
state
3. Number of children born in the same year and tested with
confirmed blood lead levels ≥ 10 μg/dL 2, by county and state
4. Percent of children born in the same year and tested with confirmed
blood lead levels ≥ 10 μg/dL 2, by county and state
5. Number of children born in the same year and tested with
confirmed blood lead levels ≥ 10 μg/dL2, by blood lead level
category3, by state
6. Percent of children born in the same year and tested with confirmed
blood lead levels ≥ 10 μg/dL 2, by blood lead level category3, by state
1

The current blood lead reference level is 5 μg/dL based on National
Health and Nutrition Examination Survey (NHANES) 2007 – 2008
and 2009 – 2010 data published in the Fourth National Report on
Human Exposure to Environmental Chemicals, and updated in 2012.
Blood Lead Levels (BLLs) are confirmed if there is either: (1) one
elevated venous test or (2) two elevated capillary and/or unknown
tests at least 1 day but less than 12 weeks apart.
2

Details about selecting the appropriate test to classify a child are in
the “How-To-Guide for Creating CLP-2 datasets.”
3

BLL categories (in units of μg/dL) are <10, 10-<15, 15-<20, 20-<25,
25-<45, 45-<70, and ≥ 70. An additional category for unconfirmed
single capillary or unknown specimen tests is used to calculate the
total number of children tested. Data are presented by categories at the
state level only.
NCDM Recommendations Version 3.0
Page 90 3/19/2013

Derivation of Measure(s)

Create CLP-2 (county level) dataset using the “How-To-Guide for
Creating CLP-2 datasets.”
 Select children’s records from childhood lead poisoning
database.
 Classify test results.
 Aggregate by county of residence and birth cohort.
 Merge with total number of county to obtain the denominator.

From CLP-2 dataset, calculate the measures:

1. Number of children born in the same year and tested, by county and
state
 Sum all BLL categories including the unconfirmed
2. Percent of children born in the same year and tested, by county and
state
 Divide number of children tested by the total number of
children in the birth cohort
3. Number of children born in the same year and tested with
confirmed blood lead levels ≥ 10 μg/dL 2, by county and state
 Sum number of children in BLL categories ≥ 10 μg/dL
(BLLs10_14,…,BLLs70), excluding unconfirmed
4. Percent of children born in the same year and tested with confirmed
blood lead levels ≥ 10 μg/dL 2, by county and state
 Divide number of children tested with BLLs ≥ 10 μg/dL by the
total number of children tested and multiply by 100
5. Number of children born in the same year and tested with
confirmed blood lead levels ≥ 10 μg/dL2, by blood lead level
category3, by state
 Sum number of children by BLL categories ≥ 10 μg/dL
(BLLs10_14,…,BLLs70), excluding unconfirmed
6. Percent of children born in the same year and tested with confirmed
blood lead levels ≥ 10 μg/dL2, by blood lead level category3, by state
 BLL Categories = Divide number of children for each BLL
category by the total number of children tested and multiply by
100

Unit

Number and percent

Geographic Scope

State or National

Geographic Scale

County or State (measures 1-4 available by county and state; measures

NCDM Recommendations Version 3.0
Page 91 3/19/2013

Time Period

5 and 6 available by state)
2000 (or first available) to current

Time Scale

Annual birth cohort

Rationale

Blood lead levels in young children have been associated with adverse
health effects ranging from learning impairment and behavioral
problems to death. No threshold for adverse effects has been
identified. Because children may have elevated BLLs and not have
any specific symptoms, CDC recommends blood lead testing for
young children at risk for lead poisoning. The risk factors identified
by the National Health and Nutrition Examination Surveys
(NHANES) include living in housing built before 1950, especially
housing in deteriorating condition, being African American, and living
in poverty.
Many states have adopted a targeted testing strategy (i.e., test children
at high risk), whereas some states recommend universal testing (i.e.,
test all children), either statewide or within high-risk counties and
cities. For both universal and targeted testing strategies, children
should be tested at least once before the age of 3 years. Some states
require more than one test between the ages of 6 and 36 months. In all
states, a blood lead test is required for Medicaid-eligible children at 12
and 24 months of age.
CDC updated its recommendations on children’s blood lead levels in
May 2012. The new recommendation is based on the U.S population
of children aged 1-5 years who are in the top 2.5% of children tested
for lead in their blood. This reference value is the 97.5th percentile,
which is currently 5 μg/dL based on NHANES 2007 – 2008 and 2009
– 2010 data (CDC, 2012). The recommendation that chelation therapy
should be considered for children with BLLs ≥45 μg/dL has not
changed. BLL results ≥70 μg/dL represent a medical emergency.
Many states initiate environmental investigations at either BLLs ≥20
μg/dL or persistent BLL results that are 15-19 μg/dL
This indicator uses a birth cohort approach. Using these measures, it is
possible to determine how many children born in a specific year were
tested before the ages of 3 and how many of those tested had an
elevated BLL. For children with more than one test before the age of
3, this indicator uses the highest venous specimen result or if there is
no venous specimen the highest confirmatory capillary/unknown
result. Using the highest results allows for examination of the peak
BLLs for the birth cohort. Inclusion of multiple cohorts will allow for
the evaluation of trends in testing and BLLs greater than the reference
value.

NCDM Recommendations Version 3.0
Page 92 3/19/2013

Use of the Measure(s)









Limitations of the
Measure(s)

Data Sources

Limitations of Data Sources

To identify and monitor temporal and spatial changes in BLL
testing and -BLLs by birth cohort.
To better understand BLL surveillance data when interpreting
number of -BLLs.
To compare testing and BLLs within and across states for the
purpose of targeting interventions. Comparisons should only be
made between areas with similar testing and reporting rules.
To link data on risk factors and compare risk factors within and
across states.
To guide interventions and allocation of resources related to BLL
testing and prevention of lead exposure in young children..
To develop and support public health policy and legislation related
to BLL testing and prevention of childhood lead poisoning.
To monitor progress towards eliminating BLLs ≥5 μg/dL, the
current reference value (NHANES 2007 – 2008 and 2009 – 2010
data).



The analysis uses the county of the child’s residence at the time of
the test, which may be different from the county where the child
was exposed to lead.
 Counties are not homogenous with respect to the distribution of
lead hazards or risk factors for lead exposure.
 Number and percent of BLLs cannot be interpreted as prevalence
or incidence for the population.
 State to state comparisons must be made cautiously and require
additional information about the states’ testing practices,
confirmatory testing practices, and reporting laws.
 Because the capillary test is subject to contamination it can result
in a false positive BLL. The number and percent of BLLs may be
overestimated when non-venous test results are used.
Childhood Blood Lead Surveillance Data
Vital Statistics Birth Data
Childhood Blood Lead Surveillance Data
 Surveillance data are not randomly sampled or representative
of the population.
 Complete residential addresses are not available for all
children tested.
 Sometimes the address of the provider or another address is
listed as the child’s address when the data is not provided by
the reporting authority.
Vital Statistics Birth Data
 The number of children born from Vital Statistics does not
include children who have moved in or out of the area since
birth. Therefore, as a denominator, it may under or over

NCDM Recommendations Version 3.0
Page 93 3/19/2013

estimate the number of children in a birth cohort.
Presentation

Small numbers of children tested, births, or BLLs may exist when the
measures are calculated at the county levels. These small numbers are
not accurate estimates for childhood lead poisoning in these polygons.
In addition, these small numbers will require additional data
processing steps to preserve confidentiality. One or more of the
following methods can be used:
 Suppression of small numbers,
 Aggregation of neighboring geographic units.
 Aggregation to a lower resolved geographic level unit,
 Aggregation of successive birth cohorts.
Data on blood lead levels are presented by categories at the state level
only.
This indicator should be displayed with information about the lead
testing program, including:
 State and/or local testing policies or strategies (i.e., targeted or
universal)
 CDC-funded Childhood Lead Poisoning Prevention Program
 Minimum BLL reported by laboratories to state or local lead
program

Related Indicators
References

Blood Lead Testing and Housing Age
Annual Blood Lead Levels
Centers for Disease Control and Prevention (CDC). 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”.

NCDM Recommendations Version 3.0
Page 94 3/19/2013

CONTENT DOMAIN: CHILDHOOD LEAD POISONING
INDICATOR: ANNUAL BLOOD LEAD LEVELS
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator

Exposure

Measure(s)

1. Number of children tested, by age group1, by county and state
2. Percent of children tested, by age group1, by county and state
3. Number of children tested with confirmed blood lead levels ≥ 10
μg/dL3,4, by age group1, by county and state
4. Percent of children tested with confirmed blood lead levels ≥ 10
μg/dL 3,4, by age group1, by county and state
5. Number of children tested with confirmed blood lead levels ≥ 10
μg/dL by blood lead level category2,3,4, by age group1, by state
6. Percent of children tested with confirmed blood lead levels ≥ 10
μg/dL, by blood lead level category2,3,4, by age group1, by state
1

Measures are available stratified by two age groups: <36 months and
36 to <72 months
2

The current blood lead reference level is 5 μg/dL based on National
Health and Nutrition Examination Survey (NHANES) 2007 – 2008 and
2009 – 2010 data published in the Fourth National Report on Human
Exposure to Environmental Chemicals, and updated in 2012. Blood
Lead Levels (BLLs) ≥ 10 μg/dL are confirmed if there is either: (1) one
elevated venous test or (2) two elevated capillary and/or unknown tests
at least 1 day but less than 12 weeks apart.
3

Details about selecting the appropriate test to classify a child are in the
“How-To-Guide for Creating CLP-4 datasets.”
4

BLL categories (in units of μg/dL) are <10, 10-14, 15-19, 20-24, 2544, 45-69, and ≥ 70. An additional category for unconfirmed elevated
capillary or unknown specimen tests is used to calculate the total
number of children tested. Confirmed BLLs ≥ 10µg/dL and BLLs 59µg/dL, reflecting the NHANES reference value, will be included by
Spring 2013. Data on confirmed BLLs ≥ 10µg/dL will be presented by
blood lead categories at the state level only.

NCDM Recommendations Version 3.0
Page 95 3/19/2013

Derivation of Measure(s)

Create CLP-4 (county level) dataset using the “How-To-Guide for
Creating CLP-4 datasets.”
 Select children’s records from childhood lead poisoning
database.
 Classify test results.
 Aggregate by county of residence and year
 Merge with total number of children by county to obtain the
denominator.

From CLP-4 dataset, calculate the measures:
1. Number of children tested
 Sum all BLL categories including the unconfirmed
2. Percent of children tested
 Divide number of children tested by the total number of children
3. Number of children tested with confirmed blood lead levels ≥ 10
μg/dL4
 Sum number of children in BLL categories ≥ 10 μg/dL (BLLs
10-14,…,BLLs70), excluding unconfirmed
4. Percent of children tested with confirmed blood lead levels ≥ 10
μg/dL4
 Divide number of children tested with blood lead levels ≥ 10
μg/dL by the total number of children tested and multiply by 100
5. Number of children tested with confirmed blood lead levels ≥ 10
μg/dL4
 Sum number of children for each BLL category
6. Percent of children tested with confirmed blood lead levels ≥ 10
μg/dL4
 Divide number of children for each BLL category by the total
number of children tested and multiply by 100
Unit

Number and percent

Geographic Scope

State or National

Geographic Scale
Time Period

County or State (measures 1-4 available at county and state; measures 5
and 6 available only at state)
2000 to current

Time Scale

Annual

Rationale

Blood lead levels in children have been associated with adverse health
effects ranging from learning impairment and behavioral problems to
death. Lead can affect almost every organ and system in your body. The

NCDM Recommendations Version 3.0
Page 96 3/19/2013

effects of lead are the same whether it enters the body through breathing
or swallowing. Small children can be exposed by eating lead-based
paint chips, chewing on objects painted with lead-based paint or
swallowing house dust or soil that contains lead. Children are more
vulnerable to lead poisoning than adults. The main target for lead
toxicity is the nervous system in young children. A child who swallows
large amounts of lead may develop blood anemia, severe stomachache,
muscle weakness, and brain damage. If a child swallows smaller
amounts of lead, much less severe effects on blood and brain function
may occur. Even at much lower levels of exposure, lead can affect a
child’s mental and physical growth.
Since children may have higher BLLs and not display any specific
symptoms, CDC recommends blood lead testing for young children at
risk for lead poisoning. The risk factors identified by the National
Health and Nutrition Examination Surveys (NHANES) include living in
housing built before 1950, especially housing in deteriorating condition,
being African American, and living in poverty.
States have developed and implemented assessment protocols for
children to determine the need for a blood lead test. For both universal
and targeted testing strategies, children should be tested at least once
before the age of 3 years. Some states require more than one test
between the ages of 6 and 36 months. Children not tested before the age
of 3 should be tested at least once before the age of 6. In all states, a
blood lead test is required for Medicaid-eligible children at 12 and 24
months.
CDC updated its recommendations on children’s blood lead levels in
May 2012. The new recommendation is based on the U.S population of
children aged 1-5 years who are in the top 2.5% of children tested for
lead in their blood. This reference value is the 97.5th percentile, which is
currently 5 μg/dL based on NHANES 2007 – 2008 and 2009 – 2010
data (CDC, 2012). The recommendation that chelation therapy should
be considered for children with BLLs ≥45 μg/dL has not changed. BLL
results ≥70 μg/dL represent a medical emergency. Many states initiate
environmental investigations at either BLLs ≥20 μg/dL or persistent
BLL results that are 15-19 μg/dL
This indicator provides information on the number of children tested
each year and the number of those children tested with confirmed blood
lead levels above 10 μg/dL. This information is used to direct resources
for testing and management of elevated cases and be linked with
environmental or the risk factor data to monitor trends over time.

NCDM Recommendations Version 3.0
Page 97 3/19/2013

Use of the Measure(s)

Limitations of the
Measure(s)

Data Sources

Limitations of Data Sources

Related Indicators
References



To identify and monitor temporal and spatial changes in BLL testing
and confirmed BLLs ≥ 10µg/dL4 by year.
 To better understand BLL surveillance data when interpreting
number of confirmed BLLs ≥ 10µg/dL4.
 To compare testing and BLLs within and across states for the
purpose of targeting interventions. Comparisons should only be
made between areas with similar testing and reporting rules.
 To link data on risk factors and compare risk factors within and
across states.
 To guide interventions and allocation of resources related to BLL
testing and prevention of EBLLs in children.
 To develop and support public health policy and legislation related
to BL testing and prevention of childhood lead exposure.
 To monitor progress towards eliminating BLLs ≥5 μg/dL, the
current reference value (NHANES 2007 – 2008 and 2009 – 2010
data).
 The analysis uses the county of the child’s residence at the time of
the test, which may be different from the county where the child was
exposed to lead.
 Counties are not homogenous with respect to the distribution of lead
hazards or risk factors for lead exposure.
 Number and percent of EBLLs through surveillance data cannot be
interpreted as prevalence or incidence for the population as a whole
 State to state comparisons must be made cautiously and require
additional information about the states’ testing practices,
confirmatory testing practices, and reporting laws.
 Because the capillary test is subject to contamination it can result in
a false positive EBLL. The number and percent of EBLLs would be
overestimated if unconfirmed, non-venous test results are used.
Childhood Blood Lead Surveillance Data
Census Population Data: Vintage bridged-race post-censal population
estimates: http://www.cdc.gov/nchs/nvss/bridged_race.htm
Childhood Blood Lead Surveillance Data
 Surveillance data are not randomly sampled or representative of
the population.
 Complete residential addresses are not available for all children
tested.
 If the child’s address is not provided the address of the provider
may be used.
Blood Lead Testing and Housing Age
Blood Lead Levels by Birth Cohort
Centers for Disease Control and Prevention (CDC). 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”.

NCDM Recommendations Version 3.0
Page 98 3/19/2013

INDICATOR TEMPLATE
CONTENT AREA: CLIMATE AND HEALTH
INDICATOR: HEAT STRESS HOSPITALIZATIONS
Type of EPHT
Indicator
Measures

Health outcome

Derivation of
Measure(s)

Numerator:
Hospital admissions having any ICD-9 code in the range of 992.0-992.9, or cause of
injury code E900.0 or E900.9, EXCLUDING cases with a code of E900.1 (manmade source of heat) anywhere in the record.

1. Age-adjusted rate of hospitalization for heat stress per 100,000 population
2. Crude rate of hospitalization for heat stress per 100,000 population
3. Number of hospitalizations for heat stress

Denominator:
Midyear resident population, by gender, for state and by county

Unit

Geographic Scope
Geographic Scale
Time Period
Time Scale
Rationale

Adjustment:
Age-adjustment by the direct method to year 2000 US standard population
1. Age-adjusted rate per 100,000 population
2. Rate per 100,000 population
3. Number
State and national
Residents of jurisdiction – State
Hospital admissions between May 1 to September 30, inclusive, for each year,
2000–
May–September of each data year
The Intergovernmental Panel on Climate Change (IPCC) projects with “virtual
certainty” suggest that climate change will cause more frequent, more intense, and
longer heat waves (1). Any individual, regardless of age, sex or health status can
develop heat stress if engaged in intense physical activity and/or exposed to
environmental heat (and humidity). Physiologic mechanisms maintain the core body
temperature (i.e., the operating temperature of vital organs in the head or trunk) in a
narrow optimum range around 37 °C (98.6 °F). When core body temperature rises,
the physiologic response is to sweat and circulate blood closer to the skin's surface
to increase cooling. If heat exposure exceeds the physiologic capacity to cool, and
core body temperature rises, then a range of heat-related symptoms and conditions
can develop. Heat stress or Heat-related illness ranges from mild heat edema and
rash, heat syncope, heat cramps, to the most common type, heat exhaustion (2).
Heat-related cramps, rash, and edema are relatively minor readily treatable
conditions; however, they should be used as important warning signs to immediately
remove the affected individual from the exposure situation.

NCDM Recommendations Version 3.0
Page 99 3/19/2013

Heat cramps are brief, intermittent, and often severe muscular cramps occurring
typically in muscles that are fatigued by heavy work (2). Individuals with heat
cramp can also exhibit hyponatremia, hypochloremia (which are low serum sodium
and chloride levels).
Heat syncope is a temporary loss of consciousness as a result of prolonged heat
exposure (2). Individuals adapt to hot, humid environment by dilation of cutaneous
vessels in the skin to radiate heat. Peripheral vasodilation along with blood volume
loss, results in lowering the blood pressure which can result in inadequate central
venous return and cerebral perfusion, causing light-headedness and fainting.
Heat exhaustion is a consequence of extreme depletion of blood plasma volume,
which may be coincident with hyponatremia and/or peripheral blood pooling (2).
Heat exhaustion often does not present with definitive symptoms and may be
misdiagnosed, often as an acute viral illness. Symptoms include mild disorientation,
generalized malaise, weakness, nausea, vomiting, headache, tachycardia (rapid
beating of the heart), and hypotension. Because untreated heat exhaustion can
progress to heat stroke, the most serious form of heat-related illness, treatment
should begin at the first signs of heat exhaustion (3).
Heat stroke is an extreme medical emergency that if untreated can result in death or
permanent neurological impairment (2). Heat stroke occurs when a person’s core
body temperature rises above 40 °C (104 °F) as a result of impaired
thermoregulation. High core body temperature and disseminated intravascular
coagulation results in cell damage in vital organs, such as the brain, liver, and
kidneys, which can lead to serious illness and death (3). Death may occur rapidly
due to cardiac failure or hypoxia, or it can occur days later as a result of renal failure
due to dehydration and/or rhabdomyolysis (i.e., the breakdown of muscle fibers with
release into the circulation of muscle fiber contents, some of which are toxic to the
kidney and can cause kidney damage) (4). Heat stroke is typically divided into two
types. The two types are in general clinically the same, except that the
individuals/population groups affected require medical interventions specific to their
unique physiology and medical status (3). “Exertional Heat Stroke,” as the name
implies, involves strenuous physical activity under high temperature conditions to
which the heat stroke victim was not acclimatized, and usually affects healthy young
adults, such as athletes, outdoor laborers and soldiers. “Classic” heat stroke, by
definition does not involve exertion, and usually affects susceptible individuals, such
as infants and young children, the elderly, or people with chronic illness. Because
heat stroke, even if treated, can have a death rate as high as 33%, and up to 17% of
heat stroke survivors suffer permanent damage, measures should be taken to prevent
heat-related illness, especially among vulnerable populations.
The relationship between extreme heat and increased daily morbidity and mortality
is well established. This indicator captures hospital admissions directly attributed to
heat stress (e.g., heat illness, heat stroke, and hyperthermia). It is a measure that can
NCDM Recommendations Version 3.0
Page 100 3/19/2013

be tracked easily and consistently across geography and time, and acts as a sentinel
for the broader range of heat-related illness that is not recognized and/or coded as
such.

Use of the Measure

Heat stress can manifest in a number of clinical outcomes, and people with chronic
health problems (e.g., cardiovascular disease, diabetes, obesity) are more susceptible
to the effects of heat than healthy individuals. For these reasons, heat stress may not
be listed as the primary diagnosis. This indicator therefore includes all cases where
heat stress is explicitly listed as the primary diagnosis or any other diagnosis.
Increases in the rates of hospital admission for heat stress are one potential impact of
rising global temperatures. Tracking these data can help document changes over
place and time, monitor vulnerable areas, and evaluate the results of local climateadaptation strategies.

Limitations of the
Measure

Periods of extreme heat are frequently associated with increases in hospital visits
and admissions for many causes. This measure does not capture the full spectrum of
heat stress, especially where exposure to excess heat is not explicitly documented.
Numerator: State inpatient hospital discharge data (using admission date)

Data Sources

Denominator: US Census Bureau population data
Limitations of Data State hospital discharge data:
Sources
 Using a measure of all heat stress hospitalizations will include some transfers
between hospitals for the same individual for the same heat stress event.
Variations in the percentage of transfers or readmissions for the same heat
stress event may vary by geographic area and impact rates.
 Without reciprocal reporting agreements with abutting states, statewide
measures and measures for geographic areas (e.g., counties) bordering other
states may be underestimated because of health care utilization patterns.
 Each state must individually obtain permission to access and, in some states,
provide payment to obtain the data.
 Veterans Affairs, Indian Health Services and institutionalized (e.g. Prison)
populations are excluded.
 Practice patterns and payment mechanisms may affect diagnostic coding and
decisions by health care providers to hospitalize patients
 Street address is currently not available in many states.
 Sometimes mailing address of patient is listed as the residence address of the
patient
 Patients may be exposed to environmental triggers in multiple locations, but
hospital discharge geographic information is limited to residence.
 Since the data captures hospital discharges (rather than admissions), patients
admitted toward the end of the year and discharged the following year will
be omitted from the current year dataset
NCDM Recommendations Version 3.0
Page 101 3/19/2013




Related Indicators

References

Data will need to be de-duplicated (i.e. remove duplicate records for the
same event)
There is usually a two year lag period before data are available from the data
owner.

Census data:
 Only available every 10 years, thus postcensal estimates are needed when
calculating rates for years following the census year.
 Postcensal estimates at the ZIP code level are not available from the Census
 Heat vulnerability
 Heat-related mortality
 Temperature distribution
 Emergency department visits for heat stress
1. Confalonieri U, Menne B, Akhtar R, Ebi KL, Hauengue M, Kovats RS, et al.
2007. Human health In: Parry ML, Canziani OF, Palutikof JP, van der Linden
PJ, Hanson CE. , editors. Climate Change 2007: Impacts, Adaptation and
Vulnerability Contribution of Working Group II to: Fourth Assessment Report
of the Intergovernmental Panel on Climate Change. New York: Cambridge
University Press. pp. 391–431.
2. Rosen’s Emergency Medicine: Concepts and Clinical Practice. 2010. Chapter
139: Heat illness. In JA Marx Editor-in-Chief; RS Hockberger & RM Walls
Senior Editors; JG Adams … [et al] Editors (7th ed). Philadelphia: Mosby
Elsevier.
3. American Medical Association. Heat-related Illness During Extreme Weather
Emergencies (Report 10 of the Council on Scientific Affairs (A97), 1997;
www.ama-assn.org/ama/pub/category/13637.html).
4. Centers for Disease Control and Prevention. Heat-related deaths--Los Angeles
County, California, 1999-2000, and United States, 1979-1998. MMWR
2001;50(29):623-6.

NCDM Recommendations Version 3.0
Page 102 3/19/2013

INDICATOR TEMPLATE
CONTENT AREA: CLIMATE AND HEALTH
INDICATOR: EMERGENCY DEPARTMENT VISITS FOR HEAT STRESS
Type of EPHT
Indicator

Measures

Derivation of
Measure(s)

Unit
Geographic Scope
Geographic Scale
Time Period
Time Scale

Rationale

Health outcome
1. Annual age-adjusted rate of emergency department visits for heat stress per 100,000
population
2. Annual crude rate of emergency department visits for heat stress per 100,000
population
3. Annual number of emergency department visits for heat stress
Numerator:
 Patients treated in an Emergency Department (ED) having any ICD-9 code in the
range of 992.0-992.9, or cause of injury code E900.0 or E900.9.
 Cases with a code of E900.1 (man-made source or heat) anywhere in the record are
excluded.

Denominator:
Midyear resident population, by gender, for state and by county
Adjustment:
 Age-adjustment by the direct method to the Year 2000 US Standard population
 U.S. 2000 standard population by age categories from Surveillance Epidemiology
and End Results (SEER), National Cancer Institute
5. Age-adjusted rate per 100,000 population
6. Rate per 100,000 population
7. Number
State and national
State (annual), County (aggregate years)
Hospital admissions between May 1 to September 30, inclusive, for each year, 2000–
May–September of each data year
The Intergovernmental Panel on Climate Change (IPCC) projects with “virtual certainty”
suggest that climate change will cause more frequent, more intense, and longer heat
waves (1). Any individual, regardless of age, sex or health status can develop heat stress
if engaged in intense physical activity and/or exposed to environmental heat (and
humidity). Physiologic mechanisms maintain the core body temperature (i.e., the
operating temperature of vital organs in the head or trunk) in a narrow optimum range
around 37 °C (98.6 °F). When core body temperature rises, the physiologic response is to
sweat and circulate blood closer to the skin's surface to increase cooling. If heat exposure
exceeds the physiologic capacity to cool, and core body temperature rises, then a range
of heat-related symptoms and conditions can develop. Heat stress or Heat-related illness
ranges from mild heat edema, rash, heat syncope, heat cramps, to the most common type,

NCDM Recommendations Version 3.0
Page 103 3/19/2013

heat exhaustion (2). Heat-related cramps, rash, and edema are relatively minor readily
treatable conditions; however, they should be used as important warning signs to
immediately remove the affected individual from the exposure situation.
Heat cramps are brief, intermittent, and often severe muscular cramps occurring
typically in muscles that are fatigued by heavy work (2). Individuals with heat cramp can
also exhibit hyponatremia, hypochloremia, and low serum sodium and chloride levels.
Heat syncope is a temporary loss of consciousness as a result of prolonged heat exposure
(2). Individuals adapt to hot, humid environment by dilation of cutaneous vessels in the
skin to radiate heat. Peripheral vasodilation along with blood volume loss, results in
lowering the blood pressure which can result in inadequate central venous return and
cerebral perfusion, causing light-headedness and fainting.
Heat exhaustion is a consequence of extreme depletion of blood plasma volume, which
may be coincident with hyponatremia and/or peripheral blood pooling (2). Heat
exhaustion often does not present with definitive symptoms and may be misdiagnosed,
often as an acute viral illness. Symptoms include mild disorientation, generalized
malaise, weakness, nausea, vomiting, headache, tachycardia (rapid beating of the heart),
and hypotension. Because untreated heat exhaustion can progress to heat stroke, the
most serious form of heat-related illness, treatment should begin at the first signs of heat
exhaustion (3).
Heat stroke is an extreme medical emergency that if untreated can result in death or
permanent neurological impairment (2). Heat stroke occurs when a person’s core body
temperature rises above 40 °C (104 °F) as a result of impaired thermoregulation. High
core body temperature and disseminated intravascular coagulation results in cell damage
in vital organs, such as the brain, liver, and kidneys, which can lead to serious illness and
death (3). Death may occur rapidly due to cardiac failure or hypoxia, or it can occur days
later as a result of renal failure due to dehydration and/or rhabdomyolysis (i.e., the
breakdown of muscle fibers with release into the circulation of muscle fiber contents,
some of which are toxic to the kidney and can cause kidney damage) (4). Heat stroke is
typically divided into two types. The two types are in general clinically the same, except
that the individuals/population groups affected require medical interventions specific to
their unique physiology and medical status (3). “Exertional Heat Stroke,” as the name
implies, involves strenuous physical activity under high temperature conditions to which
the heat stroke victim was not acclimatized, and usually affects healthy young adults,
such as athletes, outdoor laborers and soldiers. “Classic” heat stroke, by definition does
not involve exertion, and usually affects susceptible individuals, such as infants and
young children, the elderly, or people with chronic illness. Because heat stroke, even if
treated, can have a death rate as high as 33%, and up to 17% of heat stroke survivors
suffer permanent be taken to prevent heat-related illness, especially among vulnerable
populations.
The relationship between extreme heat and increased daily morbidity and mortality is
well established. This indicator captures hospital admissions directly attributed to heat
NCDM Recommendations Version 3.0
Page 104 3/19/2013

stress (e.g., heat illness, heat stroke, and hyperthermia). It is a measure that can be
tracked easily and consistently across geography and time, and acts as a sentinel for the
broader range of heat-related illness that is not recognized and/or coded as such.

Use of the Measure

Limitations of the
Measure
Data Sources

Limitations of Data
Sources

Related Indicators

References

Heat stress can manifest in a number of clinical outcomes, and people with chronic
health problems (e.g., cardiovascular disease, diabetes, obesity) are more susceptible to
the effects of heat than healthy individuals. For these reasons, heat stress may not be
listed as the primary diagnosis. This indicator therefore includes all cases where heat
stress is explicitly listed as the primary diagnosis or any other diagnosis.
Increases in the rates of ED visits for heat stress are one potential impact of rising global
temperatures. Tracking these data can help document changes over place and time,
monitor vulnerable areas, and evaluate the results of local climate-adaptation strategies.
Periods of extreme heat are frequently associated with increases in hospital visits and
admissions for many causes. This measure does not capture the full spectrum of heatstress, where exposure to excess heat is not explicitly documented.
Numerator: State emergency department data
Denominator: US Census Bureau population data
Emergency Department data:
 Data are not available for all states.
 Number of diagnostic fields in hospital records varies from state to state. Utilization
of EDs varies geographically.
Census data:
 Only available every 10 years, thus postcensal estimates are needed when
calculating rates for years following the census year.
 Heat vulnerability
 Heat-related mortality
 Temperature distribution
 Heat stress hopitalizations
1. Confalonieri U, Menne B, Akhtar R, Ebi KL, Hauengue M, Kovats RS, et al. 2007.
Human health In: Parry ML, Canziani OF, Palutikof JP, van der Linden PJ, Hanson
CE. , editors. Climate Change 2007: Impacts, Adaptation and Vulnerability
Contribution of Working Group II to: Fourth Assessment Report of the
Intergovernmental Panel on Climate Change. New York: Cambridge University
Press. pp. 391–431.
2. Rosen’s Emergency Medicine: Concepts and Clinical Practice. 2010. Chapter 139:
Heat illness. In JA Marx Editor-in-Chief; RS Hockberger & RM Walls Senior
Editors; JG Adams … [et al] Editors (7th ed). Philadelphia: Mosby Elsevier.
3. American Medical Association. Heat-related Illness During Extreme Weather
Emergencies (Report 10 of the Council on Scientific Affairs (A97), 1997; www.amaassn.org/ama/pub/category/13637.html).
4. Centers for Disease Control and Prevention. Heat-related deaths--Los Angeles
County, California, 1999-2000, and United States, 1979-1998. MMWR 2001;
50(29):623-6.

NCDM Recommendations Version 3.0
Page 105 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: ATRAZINE
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT
Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
1. Quarterly distribution of number of Community Water Systems (CWS) by mean
atrazine concentration (cut-points: 0-1, >1-3, >3-4, >4 µg/L atrazine).
2. Yearly distribution of number of CWS by maximum atrazine concentration (cutpoints: 0-1, >1-3, >3-4, >4 µg/L atrazine).
3. Yearly distribution of number of CWS by mean atrazine concentration (cutpoints: 0-1, >1-3, >3-4, >4 µg/L atrazine).
4. Mean concentration of atrazine at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
1. Quarterly distribution of number of people served by CWS by mean atrazine
concentration (cut-points: 0-1, >1-3, >3-4, >4 µg/L atrazine).
2. Yearly distribution of number of people served by CWS by maximum atrazine
concentration (cut-points: 0-1, >1-3, >3-4, >4 µg/L atrazine).
3. Yearly distribution of number of people served by CWS by mean atrazine
concentration (cut-points: 0-1, >1-3, >3-4, >4 µg/L atrazine).
Atrazine measures will be developed from water system attribute and water quality
data stored in state Safe Drinking Water Act (SDWA) databases such as the Safe
Drinking Water Information System (SDWIS/State). Data will be cleaned and
transformed to a standard format. Analytical results of drinking water samples
(usually taken at entry points to the distribution system or representative sampling
points after treatment) will be used in conjunction with information about each
CWS (such as service population and latitude and longitude of representative
location of the CWS service area) to generate the measures.
µg/L of Atrazine

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Atrazine and Public Health

Derivation of
Measures

Atrazine is a widely used herbicide active against broadleaf and grassy weeds. Atrazine was
first registered as an herbicide in 1958. More than 70 million pounds have been applied

NCDM Recommendations Version 3.0
Page 106 3/19/2013

annually in recent years, with about 75% of corn cropland receiving treatment. In addition
to agricultural uses, atrazine is used in residential turf applications and on golf courses and
sod farms to control weeds. Atrazine and its degradation products are the most commonly
detected pesticides in ground and surface waters (Barr et al., 2007). The frequent detection
of atrazine and its degradation products in streams, rivers, groundwater, and reservoirs is
related directly to the volume of its use, its persistence in soils due to its resistance to
photolysis and hydrolysis, and its ability to travel within water systems (Nelson et al.,
2001). In water systems, atrazine is transformed over time by various chemical reactions
into other compounds or its degradation products or metabolites, including dealkylated
compounds such as desethylatrazine (DEA), desisopropylatrazine (DIA), and
diaminochlorotriazine (DACT). In soil, atrazine degrades slowly to dealkylated
compounds, which have half-lives of several months. Bacteria and plants can metabolize
atrazine to hydroxylated products. In plants, atrazine is absorbed by the root system and
tends to form hydroxylated metabolites that cannot be removed by washing contaminated
vegetables (Nelson et al., 2001). Atrazine does not bioaccumulate. Studies suggest that in
animals, the degradation products that retain the chlorine have biologic activity similar to
that of atrazine, while the hydroxylated metabolites do not retain its biologic activity
(Nelson et al., 2001). Use of atrazine in the presence of nitrogen fertilizers, has raised a
possibility of N-nitrosation in soil (DeMarini and Zahm, 1999). There may also be
endogenous formation of N-nitrosoatrazine from precursors ingested in the diet and
drinking water. For the general population, drinking water is an infrequent source of
atrazine exposure, but estimates of seasonal intakes from drinking water in a small number
of communities have exceeded the recommended limits (U.S. EPA, 2003). As a result,
atrazine use has progressively been restricted in an effort to reduce surface and ground
water contamination.
In an analysis of occurrence data from the EPA 6 Year Review of National Primary
Drinking Water Regulations, atrazine was detected in 888 systems serving greater than 34
million people (EPA, 2009). Concentrations of atrazine were greater than the MCL in 98
systems serving 3.1 million people. Atrazine was the second highest occurring regulated
synthetic organic chemical found based on the percent of detections found from the 6 Year
Review data (EPA, 2009).
While it is used on many crops, atrazine has not been found in many food samples, and then
only at very low levels. Therefore, it is very unlikely that people would be exposed to
atrazine by eating crops from atrazine-accumulated soil.
Most people are not exposed regularly to atrazine. People living near areas where atrazine
was applied to crops may be exposed through contaminated drinking water. Atrazine has
been found at about 20 Superfund sites in the United States. People living near those sites
may be exposed to higher levels of atrazine. Factory workers who work with atrazine may
be exposed to higher amounts of atrazine than other workers. The government has estimated
that approximately 1,000 people may be exposed to atrazine in this way (ATSDR, 2003).
Applicators of atrazine may be exposed dermally and by inhalation. Atrazine is well
absorbed orally, metabolized, and then eliminated in the urine over a few days (Bradway et
al., 1982; Catenacci et al., 1993; Timchalk et al., 1990).
Metabolism of atrazine and its degradation products is complex and results in many
potential metabolites (Barr et al., 2007). As many as 8-12 metabolites of atrazine have been
identified in animals and humans, with recent studies showing DACT as the primary

NCDM Recommendations Version 3.0
Page 107 3/19/2013

metabolite (Barr et al., 2007); therefore, earlier biomonitoring studies measuring atrazine
mercapturate alone misrepresent and underestimate total atrazine exposure. Panuwet et al.,
(2008) developed an analytical method that measures the seven primary urinary metabolites
of atrazine, which are: hydroxyatrazine, DACT, DIA, DEA, desethylatrazine mercapturate,
atrazine mercapturate, and atrazine itself.
Human health effects of atrazine at environmental doses or at biomonitored levels from
environmental exposure are unknown. In mammalian studies, atrazine is rated as having
low acute toxicity. Atrazine product formulations can be mild skin sensitizers and irritants.
Some human ecologic and epidemiologic studies of reproductive and cancer outcomes have
shown either positive or no associations, but effects are difficult to attribute due to lack of
exposure markers or due to mixed chemical or pesticide exposures (ATSDR, 2003;
Gammon et al., 2005; Sathiakumar and Delzell, 1997). Studies of couples living on farms
that use atrazine for weed control found an increase in the risk of pre-term delivery. These
studies are difficult to interpret because most of the farmers were men who may have been
exposed to several types of pesticides. A meta-analysis linked hypospadias to parental
exposure to pesticides with possible endocrine-mediated effects (Rocheleau et al., 2009).
Some epidemiological studies that looked at the potential impact of prenatal exposure to
atrazine or its products of environmental degradation on pregnancy outcomes in the general
population observed higher rates of babies born small-for gestational age (SGA) (Munger et
al., 1997, Villanueva et al., 2005; Ochoa-Acuna et al., 2009). They also linked exposure of
mothers who lived closer to sites with high atrazine concentrations with a higher risk of
gastroschisis (Waller et al., 2010). Most of these studies were retrospective and relied on
ecological assessment of exposure to atrazine. However, the most recent study that
measured urinary biomarkers of prenatal atrazine exposure and was based on a prospective
population-based cohort found associations between environmental exposure to atrazine and
adverse effects on fetal growth, specifically birth weight, birth length, and small head
circumference (Chevrier et al., 2011). Atrazine is not mutagenic and is not considered
genotoxic. The International Agency for Research on Cancer (IARC) considers atrazine not
classifiable with respect to human carcinogenicity, and the EPA considers atrazine unlikely
to be a human carcinogen. However, IARC recommends future research to characterize the
ability of atrazine to interfere with the hypothalamic-pituitary-ovarian axis in women. This
research would help determine whether atrazine is a mammary carcinogen in women.
Another area for future research is to explore atrazine’s ability to alter immune and
aromatase function in humans. Additional information is available from U.S. EPA at:
http://www.epa.gov/pesticides/ ; from ATSDR at: http://www.atsdr.cdc.gov/toxpro2.html,
and IARC at http://www.iarc.fr/
Children are likely to be exposed to atrazine in the same way as adults, primarily through
contact with dirt that contains atrazine or by drinking water from wells that are
contaminated with the herbicide. Little information is available about the effects of atrazine
in children. Maternal exposure to atrazine in drinking water has been associated with low
fetal weight and heart, urinary, or limb defects in humans. It is not known whether atrazine
or its metabolites can be transferred from a pregnant mother to a developing fetus through
the placenta or from a nursing mother to her offspring through breast milk.
Biomonitoring Information
Urinary levels of atrazine mercapturate reflect recent exposure. In the NHANES 2001–2002
subsample, levels of atrazine mercapturate were generally not detectable (CDC, 2005). In
small studies of Maryland residents in 1995–1996 (MacIntosh et al., 1999) and 83

NCDM Recommendations Version 3.0
Page 108 3/19/2013

Minnesota children with multiple urine collections during 1997 (Adgate et al., 2001),
atrazine mercapturate was infrequently detected at the detection limit of 0.3 µg/L. In a study
of 60 farm worker children, atrazine was detected in only four children (Arcury et al.,
2007). Using immunoassay atrazine equivalents (detected mostly as atrazine mercapturate),
the urinary geometric mean levels for herbicide applicators in Ohio and Wisconsin were
about 6 µg/L (Hines et al., 2003; Perry et al., 2000). The geometric mean of urinary atrazine
mercapturate was 1.2 µg/L in 15 farmers studied several days after spraying the pesticide
(Curwin et al., 2005). In a small number of field workers, urinary concentrations ranged
from 5-1756 µg/L (Lucas et al., 1993). However, biomonitoring studies that have
evaluated only one urinary metabolite of atrazine (such as atrazine mercapturate) probably
underestimated exposure (Barr et al, 2007).
Finding measurable amounts of atrazine or its metabolites in urine does not mean that the
levels of atrazine and its metabolites (e.g., atrazine mercapturate) cause an adverse health
effect. Biomonitoring studies on levels of atrazine mercapturate provide physicians and
public health officials with reference values so that they can determine whether people have
been exposed to higher levels of atrazine than are found in the general population.
Biomonitoring data can also help scientists plan and conduct research on exposure and
health effects.
Sources of Atrazine
Atrazine is the common name for an herbicide that is widely used to kill weeds. It is used
mostly on farms. Pure atrazine—an odorless, white powder—is not very volatile, reactive,
or flammable. It will dissolve in water. Atrazine is made in the laboratory; it does not occur
naturally.
Atrazine is used on crops such as sugarcane, corn, pineapples, sorghum, and macadamia
nuts, and on evergreen tree farms and for evergreen forest re-growth. It has also been used
to keep weeds from growing on both highway and railroad rights-of-way. Some of the trade
names of atrazine are Aatrex®, Aatram®, Atratol®, and Gesaprim®. The scientific name
for atrazine is 6-chloro-N-ethyl-N'-(1-methylethyl)-triazine-2,4-diamine. Atrazine is a
Restricted Use Pesticide , which means that only certified herbicide users may purchase or
use it. Certification for the use of atrazine is obtained through the appropriate state office
where the herbicide user is licensed. Atrazine is usually used in the spring and summer
months. For it to be active, atrazine needs to dissolve in water and enter the plants through
their roots. It then acts in the shoots and leaves of the weed to stop photosynthesis. Atrazine
is taken up by all plants, but in plants not affected by atrazine, it is broken down before it
can affect photosynthesis. The application of atrazine to crops as an herbicide accounts for
almost all of the atrazine that enters the environment, but some may be released from
manufacture, formulation, transport, and disposal.
Any atrazine that is washed from the soil into streams and other bodies of water will stay
there for a long time, because chemical breakdown is slow in rivers and lakes. It also will
persist for a long time in groundwater. This is one reason why atrazine is found commonly
in the water collected from drinking water wells in some agricultural regions.
If atrazine enters the air, it can be broken down by reactions with other reactive chemicals
in the air. However, sometimes atrazine is on particles such as dust. When this happens,
breakdown is not expected. Atrazine is removed from air mainly by rainfall. When atrazine
is on dust particles, the wind can blow it long distances from the nearest application area.

NCDM Recommendations Version 3.0
Page 109 3/19/2013

For example, atrazine has been found in rainwater more than 180 miles (300 kilometers)
from the nearest application area.
Atrazine does not tend to accumulate in living organisms such as algae, bacteria, clams, or
fish, and, therefore, does not tend to build up in the food chain.
Atrazine Regulation and Monitoring
Congress established the Safe Drinking Water Act in 1974, which set enforceable
Maximum Contaminant Levels (MCLs) and non-enforceable Maximum Contaminant Level
Goals (MCLGs) for certain, specified contaminants. In the case of atrazine in drinking
water, EPA has set an MCL of 3 µg/L. Atrazine is designated as a Restricted Use Pesticide,
which means that only certified pesticide applicators can use atrazine. The Occupational
Safety and Health Administration (OSHA) has set a limit of 5 milligrams of atrazine per
cubic meter of workplace air (5 mg/m3) for an 8_hour workday and 40-hour work week.
EPA has determined maximum levels allowed in foods of 0.02-15 parts atrazine per million
parts of food (0.02-15 ppm).

Use of Measure

These measures assist by providing data that can be used for surveillance purposes.




Distribution measures provide information on the number of CWS and the
number of people potentially exposed to atrazine at different concentrations.
Maximum concentrations provide information on the peak potential
exposure to atrazine at the state level.
Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.

Limitations of the
Measure

The current measures are derived for CWS only. Private wells are another important
source of population exposure to atrazine in some agricultural regions. Transient
non-community water systems, which are regulated by EPA, may also be an
important source of atrazine exposure. Measures do not account for the variability
in sampling, numbers of sampling repeats, and variability within systems.
Concentrations in drinking water cannot be converted directly to exposure, because
water consumption varies by climate, level of physical activity, and between people
(EPA 2004). Due to errors in estimating populations, the measures may
overestimate or underestimate the number of affected people.

Data Sources
Limitations of Data
Sources

State grantee
Ground water systems may have many wells with different atrazine concentrations
that serve different parts of the population. Compliance samples are taken at each
entry point to the distribution system. In systems with separate wells serving some
branches or sections of the distribution system, the system mean would tend to
underestimate the atrazine concentration of people served by wells with higher
atrazine concentrations.
Exposure may be higher or lower than estimated if data from multiple entry points
for water with different atrazine levels are averaged to estimate levels for the PWS.

NCDM Recommendations Version 3.0
Page 110 3/19/2013

Related Indicators

Public Water Use

References

1. Adgate JL, Barr DB, Clayton CA, Eberly LE, Freeman NC, Lioy PJ, et al. Measurement
of children's exposure to pesticides: analysis of urinary metabolite levels in a probabilitybased sample. Environ Health Perspect 2001;109(6):583-590.
2. Agency for Toxic Substances and Disease Registry (ATSDR). 2003. Toxicological
Profile for Atrazine. Atlanta, GA: U.S. Department of Health and Human Services, Public
Health Service.
3. Arcury TA, Grzywacz JG, Barr DB, Tapia J, Chen H, Quandt SA. Pesticide urinary
metabolite levels of children in eastern North Carolina farmworker households. Environ
Health Perspect 2007;115(8):1254-1260.
4. Barr D.B., P. Panuwet, J.V. Nguyen, S. Udunka, L.L. Needham. Assessing exposure to
atrazine and its metabolites using biomonitoring. Environmental Health Perspectives 2007,
Vol. 115, No. 10, 1474-1478.
5. Bradway DE, Moseman RF. Determination of urinary residue levels of the N-dealkyl
metabolites of triazine herbicides. J Agric Food Chem 1982;30(2):244-247.
6. Catenacci G, Barbieri F, Bersani M, Ferioli A, Cottica D, Maroni M. Biological
monitoring of human exposure to atrazine. Toxicol Lett 1993;69(2):217-222.
7. Centers for Disease Control and Prevention (CDC). Third National Report on Human
Exposure to Environmental Chemicals. Atlanta (GA). 2005. 3/11/09
8. Chevrier C., G. Limon. C. Monfort, f. Rouget, R. Garlantezec, C. Petit, G. Durand, S.
Cordier. Urinary biomarkers of prenatal atrazine exposure and adverse birth outcomes In
the PELAGIE Birth Cohort. Environmental Health Perspectives 2011, March 2
(doi:10.1289/ehp.1002775)
9. Curwin BD, Hein MJ, Sanderson WT, Barr DB, Heederik D, Reynolds SJ, et al. Urinary
and hand wipe pesticide levels among farmers and nonfarmers in Iowa. J Expo Anal
Environ Epidemiol 2005;15(6):500-508.
10. DeMarini DM, Zahm SH. Atrazine IARC Monographs 73, 1999.
11. Gammon DW, Aldous CN, Carr WC Jr, Sanborn JR, Pfeifer KF. A risk assessment of
atrazine use in California: human health and ecological aspects. Pest Manag Sci
2005;61(4):331-355.
12. Hines CJ, Deddens JA, Striley CA, Biagini RE, Shoemaker DA, Brown KK, et al.
Biological monitoring for selected herbicide biomarkers in the urine of exposed custom
applicators: application of mixed-effect models. Ann Occup Hyg 2003;47(6):503-517.
13. Lucas AD, Jones AD, Goodrow MH, Saiz SG, Blewett C, Seiber JN, et al.
Determination of atrazine metabolites in human urine: development of a biomarker of
exposure. Chem Res Toxicol 1993;6(1):107-116.
14. MacIntosh DL, Needham LL, Hammerstrom KA, Ryan PB. A longitudinal investigation
of selected pesticide metabolites in urine. J Expo Anal Environ Epidemiol 1999;9(5):494501.
15. Munger R., P. Isacson, S. Hu, T., Burns, J. Hanson, C. F. Lynch et al., Intrauterine
growth retardation in Iowa communities with herbicide-contaminated drinking water
supplies. Environmental Health Perspectives, 1997; Vol., 105, 308-314.
16. Ochoa-Acuna H., J. Frankenberger J., L. Hahn, C. Carbajo. Drinking-water herbicide
exposure in Indiana and prevalence of small-for-gestational-age and preterm delivery.
Environmental Health Perspectives 2009; Vol. 117, 10, 1619-1624.
17. Panuwet R, J. V. Nguyen, P. Kuklenyik, S. O. Udunka, L.L. Needham, D. B. Barr.
Quantification of atrazine and its metabolites in urine by on-line solid-phase extractionhigh-performance liquid chromatography-tandem mass spectrometry. Anal Bioanal Chem
2008; 391: 1931-1939.

NCDM Recommendations Version 3.0
Page 111 3/19/2013

18. Rocheleau C. M., P. A. Romitti, L. K. Dennis. Pesticides and hypospadias: a metaanalysis. J Pediatr Urol 2009; Vol. 5, 17-24.
19. Perry M, Christiani D, Dagenhart D, Tortorelli J, Singzoni B. Urinary biomarkers of
atrazine exposure among farm pesticide applicators. Ann Epidemiol 2000;10(7):479.
20. Sathiakumar N, Delzell E. A review of epidemiologic studies of triazine herbicides and
cancer. Crit Rev Toxicol 1997;27(6):599-612.
21. Timchalk C, Dryzga MD, Langvardt PW, Kastl PE, Osborne DW. Determination of the
effect of tridiphane on the pharmacokinetics of [14C]-atrazine following oral administration
to male Fischer 344 rats. Toxicology 1990;61(1):27-40.
22. U.S. Environmental Protection Agency (U.S. EPA). Interim Reregistration Eligibility
Decision For Atrazine. Case No. 0062. 2003. Available at URL:
http://www.epa.gov/oppsrrd1/REDs/atrazine_ired.pdf . 3/11/09
23. U.S. Environmental Protection Agency (U.S. EPA). The Analysis of Regulated
Contaminant Occurrence Data from public Water Systems in Support of the Second Sixyear Review of National Primary Drinking Water Regulations. EPA-815-B-09-006,
October 2009.
24. Villanueva C. M., G. Durand, M. B. Coutte, C. Chevrier, S. Cordier. Atrazine in
municipal drinking water and risk of low birth weight, preterm delivery, and small-forgestational-age status. Occupational Environ Med 2005; Vol. 62, 6, 400-405.
25. Waller S. A., K. Paul, S. E. Peterson, J. E. Hitti. Agricultural-related chemical
exposures, season of conception, and risk of gastroschisis in Washington State. Am J Obstet
Gynecol 2010, Vol. 202, 241.

NCDM Recommendations Version 3.0
Page 112 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: ARSENIC
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
1. Yearly distribution of number of Community Water Systems (CWS) by
maximum arsenic concentration (cut-points: 0-5, >5-10, >10-30, >30
µg/L arsenic).
2. Yearly distribution of number of CWS by mean arsenic concentration
(cut-points: 0-5, >5-10, >10-20, >20-30, >30 µg/L arsenic).
3. Mean concentration of arsenic at CWS-level, by year.
Potential Population Exposure to Contaminants in Finished Water
1. Yearly distribution of number of people served by CWS by maximum
arsenic concentration (cut-points: 0-5, >5-10, >10-20, >20-30, >30 µg/L
arsenic).
2. Yearly distribution of number of people served by CWS by mean arsenic
concentration (cut-points: 0-5, >5-10, >10-20, >20-30, >30 µg/L
arsenic).

Units

Arsenic measures will be developed from water system attribute and water
quality data stored in state Safe Drinking Water Act (SDWA) databases such as
the Safe Drinking Water Information System (SDWIS/State). Data will be
cleaned and transformed to a standard format. Analytical results of drinking
water samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
Concentration of arsenic, µg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Arsenic and Public Health

Derivation of Measures

Exposures to higher than average levels of arsenic can come from elevated
localized soil and ground water concentrations from application and runoff of
NCDM Recommendations Version 3.0
Page 113 3/19/2013

arsenical pesticides and leachate from coal ash and landfills (ATSDR 2005).
Exposure to hundreds of micrograms per liter of arsenic found in drinking
water of Taiwan, Chile, Argentina, Mexico, Bangladesh, and India has been
associated with many adverse health effects including lung, bladder, liver and
skin cancers (NRC, 1999; Rahman et al. 2005; Salazar et al. 2004; Fazal et al.,
2001). Arsenic has been identified as a human carcinogen by the International
Agency for Research in Cancer (IARC) (IARC, 2004). Other adverse health
effects include nausea, cardiovascular disease, (Chen et al., 2007; Chih-Hao et
al., 2007; Bunderson et al., 2004), developmental and reproductive effects
(Hopenhayn et al., 2003; Ahmad et al., 2001)), Diabetes Mellitus (Rahman et
al., 1998), and skin keratosis and hyperpigmentation (Kapaj et al., 2006).
Measured arsenic concentrations in finished drinking water can be used to
understand the distribution of potential arsenic exposure levels for populations
served by community water supplies. These measures allow for comparison of
potential for arsenic exposures between the populations served by different
water systems and water sources over time, and potentially across demographic
groups.
Sources of Arsenic
Arsenic compounds (As (III) and As (V)) are found in both ground water and
surface waters. The primary sources are geologic formations from which
arsenic can be dissolved. Higher levels of arsenic tend to be found in ground
water (e.g. aquifers) as compared to surface waters (e.g., lakes, rivers).
Arsenic Regulation and Monitoring
In 2001 EPA reduced the regulatory drinking water standard Maximum
Contaminant Level (MCL) to 10 μg/L from 50 μg/L (effective January 23,
2006) on the basis of bladder and lung cancer risks (EPA 2001a). The cancer
risks were extrapolated from the Taiwanese (Chen et al. 1985) study to U.S.
risks. Lowering the MCL from 50 to 10 ppb statistically reduces bladder and
lung cancer mortality and morbidity by 37-56 cancers a year in the U.S. (EPA
2001b). Based on the current understanding of the health impacts from arsenic
exposure, the potential for adverse health effects from drinking water exposure
to arsenic is very low for most municipal drinking water systems.
Use of Measure

These measures assist by providing data that can be used for surveillance
purposes.




Distribution measures provide information on the number of CWS and
the number of people potentially exposed to arsenic at different
concentrations.
Maximum concentrations provide information on the peak potential
exposure to arsenic at the state level.
Mean concentrations at the CWS level provide information on potential

NCDM Recommendations Version 3.0
Page 114 3/19/2013

exposure at a smaller geographic scale.
Limitations of The
Measure

Measures do not account for the variability in sampling, numbers of sampling
repeats, and variability within systems. Concentrations in drinking water
cannot be directly converted to exposure, because water consumption varies by
climate, level of physical activity, and between people (EPA 2004). Due to
errors in estimating populations, the measures may overestimate or
underestimate the number of affected people.

Data Sources
Limitations of Data
Sources

State grantee
Samples are taken once a year (surface sources), once every three years
(groundwater sources), or once every nine years (for sources with a waiver).
Frequency of sampling is based on compliance with the MCL; the lower the
measured concentration the fewer samples will be taken and some years there
may be no sampling for arsenic.
Ground water systems may have multiple wells with different arsenic
concentrations that serve different parts of the population. Compliance samples
are taken at each entry point to the distribution system. In systems with
separate wells serving some branches or sections of the distribution system, the
system mean would tend to underestimate the arsenic concentration of people
served by wells with higher arsenic concentrations.

Related Indicators
References

Exposure may be higher or lower than estimated if data from multiple entry
points for water with different arsenic levels are averaged to estimate levels for
the PWS.
Public Water Use
1. Ahmad SA, Sayed MH, Barua S, Khan MH, Faruquee MH, Jalil A, Hadi SA,
Talukder HK., 2001. Arsenic in drinking water and pregnancy outcomes.
Environmental Health Prospectives; 109(6):629-31.
2. ATSDR 2005. Agency for Toxic Substances and Disease Registry.
Toxicological Profile for Arsenic. Draft for Public comment. September 2005.
Available at http://www.atsdr.cdc.gov/toxprofiles/tp2.html
3. Bunderson M, Brooks DM, Walker DL, Rosenfeld ME, Coffin JD, Beall
HD., 2004. Arsenic exposure exacerbates atherosclerotic plaque formation and
increases nitrotyrosine and leukotriene biosynthesis. Toxicology and Applied
Pharmacology 2004 Nov 15;201(1):32-9.
4. Chen C-J, Chuang Y-C, Lin T-M, Wu H-Y. Malignant neoplasms among
residents of a blackfoot disease-endemic area in Taiwan: high-arsenic well
water and cancers. Cancer Res. 1985;45:5895–5899.
5. Chen Y., Factor-Litvak P., Howe GR., Graziano JH., Brandt-Rauf P., Parvez
F., van Geen A., Ahsan H., 2007. Arsenic exposure from drinking water,
dietary intakes of B vitamins and folate, and risk of high blood pressure in
Bangladesh: a population-based, cross-sectional study. American Journal of
Epidemiology, Mar 1;165(5):541-52
6. Chih-Hao Wang, Chuhsing Kate Hsiao, Chi-Ling Chen, Lin-I Hsu, Hung-Yi

NCDM Recommendations Version 3.0
Page 115 3/19/2013

Chiou, Shu- Yuan Chen, Yu-Mei Hsueh, Meei-Maan Wu, Chien-Jen Chen
2007. A review of the epidemiologic literature on the role of environmental
arsenic exposure and cardiovascular diseases. Toxicology and Applied
Pharmacology 222 (2007) 315–326.
7. Fazal MA, Kawachi T, Ichion E. Extent and severity of groundwater arsenic
contamination in Bangladesh. Water International (2001) 26:370–79
8. Hopenhayn C, Ferreccio C, Browning SR, Huang B, Peralta C, Gibb H,
Hertz-Picciotto I., 2003. Arsenic exposure from drinking water and birth
weight.
Epidemiology.;14(5):593-602.
9. IARC 2004. Monographs on the Evaluation of Carcinogenic Risks to
Humans. Volume 84. Some Drinking Water Disinfectants and Contaminants,
including Arsenic. September Lyon, pp. 39-267. Available at
http://monographs.iarc.fr/ENG/Monographs/vol84/volume84.pdf
10. Kaltreider RC, Davis AM, Lariviere JP, Hamilton JW. 2001. Arsenic
alters the function of the glucocorticoid receptor as a transcription factor.
Environ. Health Perspect 109:245-251.
11. Kapaj S., Peterson H., Liber K., Bhattacharya P., 2006. Human health
effects from chronic Arsenic poisoning-A review, Journal of Environmental
Science and Health, Part A;41(10):2399-2428.
12. Karagas M, Stukel TA, Tosteson TD. 2002. Assessment of cancer risk and
environmental levels of arsenic in New Hampshire. Int. J. Hyg. Environ.
Health 205, 85-94.
13. NCI .2003. Ries LAG, Eisner MP, Kosary CL, Hankey BF, Miller BA,
Clegg L, Mariotto A, Fay MP, Feuer EJ, Edwards BK (eds). SEER Cancer
Statistics Review, 1975-2000, National Cancer Institute (NCI). Bethesda, MD,
http://seer.cancer.gov/csr/1975_2000/, 2003.
http://seer.cancer.gov/csr/1975_2000/results_single/sect_01_table.01.pdf
14. National Research Council. Arsenic in Drinking Water, National Academy
Press. 1999.
15. Rahman M., Tondel M., Ahmad SA., Axelson O., 1998. Diabetes Mellitus
associated with Arsenic exposure in Bangladesh. American Journal of
Epidemiology, Vol 48, no2, pp 198-203.
16. Rahman MM, Sengupta MK, Ahamed S, et al. Arsenic contamination of
groundwater and its health impact on residents in a village in West Bengal,
India. Bulletin of the World Health Organization (2005) 83:49–57.
17. Salazar et al. 2004. p53 Expression in circulating lymphocytes of nonmelanoma skin cancer patients from an arsenic contaminated region in Mexico.
A pilot study. Molecular and Cellular Biochemistry 255:25-31.
18. U.S. EPA 2000a. Arsenic in drinking water proposal, 65 FR 38888 at
38894 and 38897, June 22, 2000, available at
http://www.epa.gov/fedrgstr/EPA-WATER/2000/June/Day-22/w13546.htm
and http://frwebgate.access.gpo.gov/cgibin/getdoc.cgi?dbname=2000_register&docid=00-13546-filed.pdf
19. U.S. EPA 2000b. Arsenic in Drinking Water Rule Economic Analysis
December 2000, available at http://www.epa.gov/safewater/arsenic/history.html
NCDM Recommendations Version 3.0
Page 116 3/19/2013

20. U.S. EPA 2001a. Arsenic in drinking water final rule, 66 FR 6976.
Available at http://www.epa.gov/fedrgstr/EPA-WATER/2001/January/Day22/w1668.htm
21. U.S. EPA 2001b. Quick Reference Guide for Arsenic. EPA 816-F-01-004.
Available at http://www.epa.gov/safewater/arsenic/pdfs/quickguide.pdf
22. U.S. EPA 2001c. Arsenic in drinking water Technical fact sheet. Available
at http://www.epa.gov/ogwdw/arsenic/regulations_techfactsheet.html
23. U.S. EPA 2004. Estimated per Capita Water Ingestion and Body Weight in
the United States – an Update. EPA-822-R-00-001. October 2004. Available
at http://epa.gov/waterscience/criteria/drinking/percapita/2004.pdf
24. U.S. EPA 2005. Factoids: Drinking Water and Ground Water Statistics for
2004. EPA Office of Water. EPA 816-K-05-001. May 2005. Available at
http://www.epa.gov/safewater/data/pdfs/data_factoids_2004.pdf

NCDM Recommendations Version 3.0
Page 117 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: DI(2-ETHYLHEXYL)PHTHALATE (DEHP)
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
1. Yearly distribution of number of Community Water Systems (CWS) by
maximum DEHP concentration (cut-points: 0-2, >2-4, >4-6, >6-10, >10
µg/L DEHP).
2. Yearly distribution of number of CWS by mean DEHP concentration
(cut-points: 0-2, >2-4, >4-6, >6-10, >10 µg/L DEHP).
3. Mean concentration of DEHP at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
4. Yearly distribution of number of people served by CWS by maximum
DEHP concentration (cut-points: 0-2, >2-4, >4-6, >6-10, >10 µg/L
DEHP).
5. Yearly distribution of number of people served by CWS by mean DEHP
concentration (cut-points: 0-2, >2-4, >4-6, >6-10, >10 µg/L DEHP).
DEHP measures will be developed from water system attribute and water
quality data stored in state Safe Drinking Water Act (SDWA) databases such as
the Safe Drinking Water Information System (SDWIS/State). Data will be
cleaned and transformed to a standard format. Analytical results of drinking
water samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
DEHP, µg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Di (2-ethylhexyl)phthalate and Public Health

Derivation of Measures

DEHP is the most commonly used of a group of related chemicals called
phthalates or phthalic acid esters. Some people who drink water containing
DEHP well in excess of the maximum contaminant level (MCL) for many
years may have problems with their livers or could experience reproductive
difficulties and may have an increased risk of getting cancer. (U.S.EPA, 2010)
NCDM Recommendations Version 3.0
Page 118 3/19/2013

In an analysis of occurrence data from the EPA 6 Year Review of National
Primary Drinking Water Regulations, DEHP was detected in 3,098 systems,
which collectively serve more than 45 million people (EPA, 2009).
Concentrations of DEHP were greater than the MCL in 460 systems serving
11.5 million people. DEHP was the highest occurring regulated synthetic
organic chemical found based on the percent of detections found from the 6
Year Review data. This contamination could be due, in part, to sample
contamination from older generation laboratory and field sampling equipment
made of plastics that contained and released phthalates (EPA, 2009).
Most of what we know about the health effects of DEHP comes from studies of
rats and mice given high amounts of DEHP. Brief oral exposure to very high
levels of DEHP damaged sperm in mice. Although the effect reversed when
exposure ceased, sexual maturity was delayed in the animals. High amounts of
DEHP damaged the liver of rats and mice. Whether or not DEHP contributes to
human kidney damage is unclear.
The Department of Health and Human Services has determined that DEHP may
reasonably be anticipated to be a human carcinogen. The EPA has determined
that DEHP is a probable human carcinogen. These determinations were based
entirely on liver cancer in rats and mice. The International Agency for Research
on Cancer has stated that DEHP cannot be classified as to its carcinogenicity to
humans.
People are exposed through ingestion, inhalation, and, to a lesser extent, dermal
contact with products that contain phthalates. For the general population,
dietary sources have been considered as the major exposure route, followed by
inhaling indoor air. Infants may have relatively greater exposures from
ingesting indoor dust containing some phthalates (Clark et al., 2003). Human
milk can be a source of phthalate exposure for nursing infants (Calafat et al.,
2004; Mortensen et al., 2005). The intravenous or parenteral exposure route can
be important in patients undergoing medical procedures involving devices or
materials containing phthalates. In settings where workers may be exposed to
higher air phthalate concentrations than the general population, urinary
metabolite and air phthalate concentrations are roughly correlated (Liss et al.,
1985; Nielsen et al., 1985; Pan et al., 2006). Phthalates are metabolized and
excreted quickly and do not accumulate in the body (Anderson et al., 2001).
Biomonitoring Information
Four metabolites of DEHP were measured for the Fourth National Report on
Human Exposure to Environmental Chemicals: mono-(2-ethyl-5-hexyl)
phthalate (MEHP), mono- (2-ethyl-5-oxohexyl) phthalate (MEOHP), mono-(2ethyl- 5-hydroxyhexyl) phthalate (MEHHP) and mono-(2-ethyl- 5carboxypentyl) phthalate (MECPP). MEHP is primarily formed by the
hydrolysis of DEHP in the gastrointestinal tract and then absorbed. By contrast,
DEHP present in medical devices and parenteral delivery systems results in the
NCDM Recommendations Version 3.0
Page 119 3/19/2013

diester parent compound, rather than the monoester metabolite. being directly
introduced into the blood. After parenteral administration hydrolysis of DEHP
most likely also occurs in the blood, and subsequent metabolism is similar to
that following ingestion (Koch et al., 2005a, 2005b, 2005c). MEOHP, MEHHP,
and MECPP are produced by the oxidative metabolism of MEHP and are
present at roughly three- to five-fold higher concentrations than MEHP in urine
(Barr et al., 2003; Fromme et al., 2007; Koch et al., 2003). MEHP is the
putative toxic metabolite of DEHP. Liver toxicity, decreased testicular weight,
and testicular atrophy have been observed in rodents fed high doses over a short
term or with chronic dosing (McKee et al., 2004; NTP-CERHR, 2000c, 2006).
In contrast, marmoset monkeys fed high dose DEHP for longer than a year did
not demonstrate testicular or liver toxicity (NTP-CERHR, 2006). Very high
doses of DEHP have suppressed estradiol production in female rats
(Lovecamp-Swan and Davis, 2003). The U.S. Food and Drug Administration
determined that in adults, the amounts of DEHP or MEHP received from
intravenous delivery systems or blood transfusions (DEHP is hydrolyzed to
MEHP in stored blood) would result in short-term elevations similar to
background levels (FDA, 2001). However, critically ill neonates and infants
receiving selected or multiple intensive procedures, such as exchange
transfusions, extracorporeal membrane oxygenation, and parenteral nutrition,
could receive higher exposures than the general population (Calafat et al., 2004;
FDA, 2001; Loff et al., 2000; Weuve et al., 2006).
The levels of MEHP reported in NHANES 1999-2000, 2001-2002, and 20032004 appear roughly comparable to those reported previously in several small
U.S. studies involving adults (Blount et al., 2000), pregnant women in New
York City (Adibi et al., 2003), and low income African-American women in
Washington, DC (Hoppin et al., 2002). In another sample of men attending an
infertility clinic, the median and 95th percentile values of urinary MEHP were
similar, but MEHHP and MEOHP were about three to five times higher than
comparable values found in males in two NHANES survey periods (1999-2000,
2001-2002) (CDC, 2005; Hauser et al., 2007). In separate analyses of
NHANES 1999-2000 and NHANES 2001-2002, the adjusted geometric mean
levels of urinary MEHP were significantly higher in children compared with
adolescents and adults, and in females compared with males (CDC, 2005; Silva
et al., 2004). Studies of hospitalized neonates have reported urinary geometric
mean levels of MEHP, MEOHP, and MEHHP that were two to five times
higher, or more (depending on the intensity of DEHP-product exposure), than
the geometric means of children in the NHANES subsamples for all three
survey periods (Calafat et al., 2004; Weuve et al., 2006). Small studies of
plasma and platelet donors have reported very high levels of MEHP, MEOHP,
MEHHP and MECPP in urine collected shortly after these procedures (Koch et
al., 2005b, 2005c). Finding a measurable amount of one or more DEHP
metabolites in urine does not mean that the levels of the metabolites or the
parent compound cause an adverse health effect. Biomonitoring studies on
levels of urinary DEHP metabolites provide physicians and public health
NCDM Recommendations Version 3.0
Page 120 3/19/2013

officials with reference values so that they can determine whether people have
been exposed to higher levels of DEHP than are found in the general
population. Biomonitoring data can also help scientists plan and conduct
research on exposure and health effects.
Sources of DEHP
Phthalates are industrial chemicals, often called plasticizers, that are added to
plastics make them more flexible and resilient. Phthalates are also used in other
applications as solubilizing and stabilizing agents. Numerous products contain
phthalates: adhesives; automotive plastics; detergents; lubricating oils; some
medical devices and pharmaceuticals; plastic raincoats; solvents; vinyl tiles and
flooring; and personal-care products, such as soap, shampoo, deodorants,
lotions, fragrances, hair spray, and nail polish. Phthalates are often used in
polyvinyl chloride-type plastics, such as plastic bags, garden hoses, inflatable
recreational toys, blood product storage bags, intravenous medical tubing, and
toys (ATSDR, 2001, 2002). Because they are not chemically bound to the
plastics to which they are added, phthalates can be released into the
environment during use or disposal of the product. Various phthalate esters
have been measured in specific foods, indoor and ambient air, indoor dust,
water sources, and sediments (Clark et al., 2003).
DEHP is primarily used to produce flexibility in plastics, mainly polyvinyl
chloride, which is used for many consumer products, toys, packaging film, and
blood product storage and intravenous delivery systems. Concentrations in
plastic materials may reach 40% by weight. DEHP has been removed from or
replaced in most toys and food packaging in the United States. Following
ingestion, DEHP is metabolized to more than 30 metabolites which are rapidly
eliminated in urine, and in humans, as glucuronide conjugates (Albro et al.,
1982; Albro and Lavenhar, 1989; ATSDR, 2002; Peck and Albro, 1982). The
major source of di(2-ethylhexyl) phthalate in drinking water is discharge from
rubber and chemical factories (U.S. EPA, 2010).
DEHP Regulation and Monitoring
The EPA limits the amount of DEHP that may be present in drinking water to 6
parts of DEHP per billion parts of water (6 ppb), or 6 ug/L.
The Occupational Safety and Health Administration (OSHA) sets a maximum
average of 5 milligrams of DEHP per cubic meter of air (5 mg/m3) in the
workplace during an 8-hour shift. The short-term (15-minute) exposure limit is
10 mg/m3.
Use of Measure

These measures assist by providing data that can be used for surveillance
purposes.
• Distribution measures provide information on the number of CWS and the

NCDM Recommendations Version 3.0
Page 121 3/19/2013

Limitations of The
Measure

Data Sources
Limitations of Data
Sources

Related Indicators
References

number of people potentially exposed to DEHP at different
concentrations.
• Maximum concentrations provide information on the peak potential
exposure to DEHP at the state level.
• Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.
The current measures are derived for CWS only. Private wells may be another
source of population exposure to DEHP. Transient non-community water
systems, which are regulated by EPA, may also be an important source of
DEHP exposure. Measures do not account for the variability in sampling,
numbers of sampling repeats, and variability within systems. Concentrations in
drinking water cannot be directly converted to exposure, because water
consumption varies by climate, level of physical activity, and between people
(EPA 2004). Due to errors in estimating populations, the measures may
overestimate or underestimate the number of affected people.
State grantee
Ground water systems may have many wells with different DEHP
concentrations that serve different parts of the population. Compliance samples
are taken at each entry point to the distribution system. In systems with
separate wells serving some branches or sections of the distribution system, the
system mean would tend to underestimate the DEHP concentration of people
served by wells with higher DEHP concentrations.
Exposure may be higher or lower than estimated if data from multiple entry
points for water with different DEHP levels are averaged to estimate levels for
the PWS.
Public Water Use
1. Adibi JJ, Perera FP, Jedrychowski W, Camann DE, Barr D, Jacek R, et al. Prenatal
exposures to phthalates among women in New York City and Krakow, Poland.
Environ Health Perspect 2003;111(14):1719-1722.
2. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological profile
for di-n-butyl phthalate update [online]. 2001. Available at URL:
http://www.atsdr.cdc.gov/toxprofiles/ tp135.html. 4/20/09
3. Agency for Toxic Substances and Disease Registry (ATSDR). 2002. Toxicological
Profile for Di(2-ethylhexyl) phthalate. Update. Atlanta, GA: U.S. Department of
Health and Human Services, Public Health Service.
4. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological
profile for di(2-ethylhexyl)phthalate update [online]. 2002. Available at URL:
http://www.atsdr.cdc.gov/ toxprofiles/tp9.html. 4/20/09.
5. Albro PW, Corbett JT, Schroeder JL, Jordan S, Matthews HB. Pharmacokinetics,
interactions with macromolecules and species differences in metabolism of DEHP.
Environ Health Perspect 1982;45:19-25.
6. Albro PW and Lavenhar SR. Metabolism of di(2-ethylhexyl) phthalate. Drug
Metab Rev 1989;21:13-34.
7. Anderson WA, Castle L, Scotter MJ, Massey RC, Springall C. A biomarker
approach to measuring human dietary exposure to certain phthalate diesters. Food

NCDM Recommendations Version 3.0
Page 122 3/19/2013

Addit Contam 2001;18(12):1068- 1074.
8. Barr DB, Silva MJ, Kato K, Reidy JA, Malek NA, Hurtz D, et al. Assessing human
exposure to phthalates using monoesters and their oxidized metabolites as biomarkers.
Environ Health Perspect 2003;111(9):1148-1151.
9. Blount BC, Silva MJ, Caudill SP, Needham LL, Pirkle JL, Sampson EJ, et al.
Levels of seven urinary phthalate metabolites in a human reference population.
Environ Health Perspect 2000;108(10)979-982.
10. Calafat AM, Slakman AR, Silva MJ, Herbert AR, Needham LL. Automated solid
phase extraction and quantitative analysis of human milk for 13 phthalate metabolites.
J Chromatogr B 2004;805:49-56.
11. Calafat AM, Needham LL, Silva MJ, Lambert G. Exposure to di- (2-ethylhexyl)
phthalate among premature neonates in a neonatal intensive care unit. Pediatrics
2004;113:e429-e434.
12. Centers for Disease Control and Prevention (CDC). Third National Report on
Human Exposure to Environmental Chemicals. Atlanta (GA). 2005.
13. Clark K, Cousins IT, Mackay D. Assessment of critical exposure pathways. In
Staples CA (ed), The Handbook of Environmental Chemistry, Vol.3, Part Q: Phthalate
Esters. 2003;New York, Springer, pp. 227-262.
14. Food and Drug Administration (FDA). Safety assessment of di(2ethylhexyl)phthalate (DEHP) released form PCV medical devices. 2001 [online].
Available at URL: http://www.fda.gov/ cdrh/ost/dehp-pvc.pdf. 4/20/09
15. Fromme H, Bolte G, Koch HM, Angerer J, Boehmer S, Drexler H, et al.
Occurrence and daily variation of phthalate metabolites in the urine of an adult
population. Int J Hyg Environ Health 2007;210:21-33.
16. Hauser R, Meeker JD, Singh NP, Silva MJ, Ryan L, Duty S, et al. DNA damage in
human sperm is related to urinary levels of phthalate monoester and oxidative
metabolites. Hum Reprod 2007; 22(3):688-695.
17. Hoppin JA, Brock JW, Davis BJ, Baird DD. Reproducibility of urinary phthalate
metabolites in first morning urine samples. Environ Health Perspect 2002; 110(5):515518.
18. Koch HM, Bolt HM, Preuss R, Angerer J. New metabolites of di(2ethylhexyl)phthalate (DEHP) in human urine and serum after single oral doses of
deuterium-labeled DEHP. Arch Toxicol 2005a; 79:367-376.
19. Koch HM, Bolt HM, Preuss R, Eckstein R, Weisback V, Angerer J. Intravenous
exposure to di(2-ethylhexyl)phthalate (DEHP) : metabolites of DEHP in urine after a
voluntary platelet donation. Arch Toxicol 2005b; 79:689-693.
20. Koch HM, Angerer J, Drexler H, Eckstein R, Weisbach V. Di(2ethylhexyl)phthalate (DEHP) exposure of voluntary plasma and platelet donors. Int J
Hyg Environ Health 2005c;208:489-498. Koch HM, Drexler H, Angerer J. Internal
exposure of nursery- school children and their parents and teachers to di(2-ethylhexyl)
phthalate (DEHP). Int J Hyg Environ Health. 2004; 207:15-22.
21. Koch HM, Rossbach B, Drexler H, Angerer J. Internal exposure of the general
population to DEHP and other phthalates - determination of secondary and primary
phthalate monoester metabolites in urine. Environ Res 2003; 93:177-185.
22. Liss GM, Albro PW, Hartle RW, Stringer WT. Urine phthalate determinations as
an index of occupational exposure to phthalic anhydride and di (2-ethylhexyl)
phthalate. Scand J Work Environ Health 1985;11(5):381-387.
23. Loff S, Kabs F, Witt K, Sartoris J, Mandl B, Niessen KH, et al. Polyvinylchloride
infusion lines expose infants to large amounts of toxic plasticizers. J Pediatr Surg
2000; 35(12):1775-1781.
24. Lovekamp-Swan T, Davis BJ. Mechanisms of phthalate ester toxicity in the

NCDM Recommendations Version 3.0
Page 123 3/19/2013

female reproductive system. Environ Health Perspect 2003; 111(2):139-145.
25. McKee RH, Butala JH, David RM, Gans G. NTP center for the evaluation of risks
to human reproduction reports on phthalates: addressing the data gaps [review].
Reprod Toxicol 2004; 18(1):1- 22.
26. Mortensen GK, Main KM, Andersson A-M, Leffers H, Skakkebaek NE.
Determination of phthalate monoesters in human milk, consumer milk, and infant
formula by tandem mass spectrometry (LC-MS-MS). Anal Bioanal Chem
2005;382:1084- 1092.
27. Nielsen J, Akesson B, Skerfving S. Phthalate ester exposure—air levels and health
of workers processing polyvinylchloride. Am Ind Hyg Assoc J 1985;46(11):643-647.
28. NTP-CERHR. National Toxicology Program-Center for the Evaluation of Risks to
Human Reproduction. Monograph on the Potential Human Reproductive and
Developmental Effects of Di (2-ethylhexyl) Phthalate (DEHP). Research Triangle Park
(NC). 2000c [online]. Available at URL: http://cerhr.niehs.nih.
gov/chemicals/dehp/dehp-eval.html. 6/2/09
29. NTP-CERHR. National Toxicology Program-Center for the Evaluation of Risks to
Human Reproduction. Draft Update Monograph on the Potential Human Reproductive
and Developmental Effects of Di (2-ethylhexyl) Phthalate (DEHP). Research Triangle
Park (NC). 2006 [online]. Available at URL:
http://cerhr.niehs.nih.gov/chemicals/dehp/dehp-eval.html. 6/2/09
30. Pan G, Hanaoka T, Yoshimura M, Zhang S, Wang P, Tsukino H, et al. Decreased
serum free testosterone in workers exposed to high levels of di-n-butyl phthalate
(DBP) and di-2-ethylhexylphthalate (DEHP): a cross-sectional study in China. Environ
Health Perspect 2006; 114(11):1643-1648.
31. Peck CC, Albro PW. Toxic potential of the plasticizer di (2-ethylhexyl) phthalate
in the context of its disposition and metabolism in primates and man. Environ Health
Perspect 1982; 45:11-17.
32. Silva MJ, Barr DB, Reidy JA, Malek NA, Hodge CC, Caudill SP, et al. Urinary
levels of seven phthalate metabolites in the U.S. population from the National Health
and Nutrition Examination Survey (NHANES) 1999-2000 [published erratum appears
in Environ Health Perspect 2004; 112(5):A270]. Environ Health Perspect 2004;
112(3):331-338.
33. U.S. Environmental Protection Agency (U.S. EPA). The Analysis of Regulated
Contaminant Occurrence Data from public Water Systems in Support of the Second
Six-year Review of National Primary Drinking Water Regulations. EPA-815-B-09006, October 2009.
34. U.S. EPA, Basic Information about Di (2- ethylhexyl) phthalate in Drinking Water.
2010. http://water.epa.gov/drink/contaminants/basicinformation/di_2ethylhexyl_phthalate.cfm
35. Weuve J, Sanchez GN, Calafat AM, Schettler T, Green RA, Hu H, et al. Exposure
to phthalates in neonatal intensive care unit infants: urinary concentrations of
monoesters and oxidative metabolites. Environ Health Perspect 2006; 114(9):14241431.

NCDM Recommendations Version 3.0
Page 124 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: DISINFECTION BYPRODUCTS
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
1. Quarterly distribution of number of Community Water Systems (CWS)
by mean HAA5 concentration (cut-points: (0-15), (>15-30), (>30-45),
(>45-60), (>60-75), (>75) mg/L HAA5).
2. Yearly distribution of number of CWS by maximum HAA5
concentration (cut-points: (0-15), (>15-30), (>30-45), (>45-60), (>6075), (>75) mg/L HAA5).
3. Yearly distribution of number of CWS by mean HAA5 concentration
(cut-points: (0-15), (>15-30), (>30-45), (>45-60), (>60-75), (>75) mg/L
HAA5).
4. Mean concentration of HAA5 at CWS-level, by year.
5. Quarterly distribution of number of CWS by mean TTHM concentration
(cut-points: (0-20), (>20-40), (>40-60), (>60-80), (>80-100), (>100)
mg/L TTHM).
6. Yearly distribution of number of CWS by maximum TTHM
concentration (cut-points: (0-20), (>20-40), (>40-60), (>60-80), (>80100), (>100) mg/L TTHM).
7. Yearly distribution of number of CWS by mean TTHM concentration
(cut-points: (0-20), (>20-40), (>40-60), (>60-80), (>80-100), (>100)
mg/L TTHM).
8. Mean concentration of TTHM at CWS-level, by year.
Potential Population Exposure to Contaminants in Finished Water
9. Quarterly distribution of number of people served by CWS by mean
HAA5 concentration (cut-points: (0-15), (>15-30), (>30-45), (>45-60),
(>60-75), (>75) mg/L HAA5).
10. Yearly distribution of number of people served by CWS by maximum
HAA5 concentration (cut-points: (0-15), (>15-30), (>30-45), (>45-60),
(>60-75), (>75) mg/L HAA5).
11. Yearly distribution of number of people served by CWS by mean
HAA5 concentration (cut-points: (0-15), (>15-30), (>30-45), (>45-60),
(>60-75), (>75) mg/L HAA5).
12. Quarterly distribution of number of people served by CWS by mean
TTHM concentration (cut-points: (0-20), (>20-40), (>40-60), (>60-80),
(>80-100), (>100) mg/L TTHM).
13. Yearly distribution of number of people served by CWS by maximum
TTHM concentration (cut-points: (0-20), (>20-40), (>40-60), (>60-80),
(>80-100), (>100) mg/L TTHM).
14. Yearly distribution of number of people served by CWS by mean
TTHM concentration (cut-points: (0-20), (>20-40), (>40-60), (>60-80),

NCDM Recommendations Version 3.0
Page 125 3/19/2013

Derivation of Measures

Units

(>80-100), (>100) mg/L TTHM).
Disinfection byproducts measures will be developed from water system
attribute and water quality data stored in state Safe Drinking Water Act
(SDWA) databases such as the Safe Drinking Water Information System
(SDWIS/State). Trihalomethanes comprise chloroform, bromodichloromethane,
dibromochloromethane, bromoform and their sum, denoted total
trihalomethanes (TTHM). Haloacetic acids comprise trichloroacetic acid,
dichloroacetic acid, monochloroacetic acid, dibromoacetic acid,
monobromoacetic acid, and their sum, denoted HAA5. Data will be cleaned
and transformed to a standard format. Analytical results of drinking water
samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
concentration of HAA5, µg/L
concentration of TTHM, µg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

2002 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Disinfection By Products and Public Health
Disinfection byproducts (DBP) are formed when disinfectants used to
inactivate microbial contaminants in water react with materials, primarily
organic matter, in the water (Bellar et al. 1974, Rook 1974, Cedergren et al.
2002, Sadiq and Rodriguez 2004). Several hundred DBPs in over a dozen
chemical classes have been identified (Woo et al. 2002, Krasner et al. 2006).
Most commonly, DBPs form when chlorine reacts with naturally occurring
organic matter in the source water.
DBPs have been associated with both cancer and adverse pregnancy outcomes.
High DBP levels, mainly for THMs, have been linked to bladder, colon and
rectal cancer (King and Marrett 1996, Cantor et al. 1998, Amy et al. 2005,
Villanueva et al. 2004, Villanueva et al. 2007), with bladder cancer reported
most frequently. Although findings about adverse pregnancy outcomes have
been less definitive, DBPs have been implicated in fetal loss (Swan et al. 1998,
Waller et al. 1998, King et al. 2000, Dodds et al. 2004) and a variety of adverse
birth outcomes involving growth (Bove et al. 1995, Gallagher et al. 1998,
Wright et al. 2004, Infante-Rivard 2004, Toledano et al. 2005) and birth defects
(Dodds et al. 1999, Klotz and Pyrch 1999, Dodds and King 2001, Cedergren et

NCDM Recommendations Version 3.0
Page 126 3/19/2013

al. 2002, Shaw et al. 2003). In contrast, however, other research has found little
effect on birth outcomes (Savitz et al., 2006).
Animal, microbial, in vitro and modeling studies have also pointed to toxicity
or carcinogenicity of a wide variety of DBPs (Boorman 1999, Komulainen
2004). Numerous studies have indicated that different DBPs among the THMs
and HAAs have different health effects. A number of studies have suggested
that iodinated and brominated DBPs are more toxic than their chlorinated
counterparts (Plewa et al. 2002, 2004, Richardson 2005). It is therefore
appropriate that the tracking network follow individual DBP species and not
just class totals (c.f. Singer 2006).
Sources of DBPs
DPB levels tend to be highest in water derived from surface sources because
ground water generally has little organic matter (Symons et al. 1975, Whitaker
et al. 2003). Ground water can, however, produce relatively high levels of the
more brominated DBPs when the water, due either to geological circumstances
(Whitaker et al. 2003) or salt water intrusion in coastal areas (von Gunten
2003), has elevated levels of bromide.
Bromate and chlorite are formed primarily after disinfection by ozone and
chlorine dioxide, respectively. Sampling for these DBPs is required only for
treatment plants that use the disinfectants that form them. Ozonation and
chlorine dioxide are less common mechanisms of disinfection so these two
DBPs will not be tracked initially. The disinfection processes that produce
these two byproducts are likely to be used more often in the future so bromate
and chlorite should be considered for eventual incorporation into the tracking
network.
DBP Regulation and Monitoring
Safe Drinking Water Act (SDWA) regulation of DBPs began with the 1979
Total Trihalomethane Rule. This rule set an interim MCL for total
trihalomethanes (TTHM), defined as the sum of four trihalomethanes, of 0.10
mg/L for community water systems (CWS) serving 10,000 or more people and
using a disinfectant. The Stage 1 Disinfectants and Disinfection Byproducts
Rule of 1998 (US EPA 1998) reduced the MCL for TTHM to 0.080 mg/L,
added MCLs for the sum of five haloacetic acids (HAA5) of 0.060 mg/L,
bromate of 0.010 mg/L and chlorite of 1.0 mg/L, and increased the scope of the
rule to cover all CWS that disinfect. The rule had phased compliance with a
date of 1 January 2002 for public water systems (PWS) with 10,000 or more
people with a surface water or ground water under direct influence source and a
date of 1 January 2004 for all other affected PWSs. The Stage 2 Disinfectants
and Disinfection Byproducts Rule of 2006 (US EPA 2006) did not alter MCLs
but did change how compliance with MCLs will be calculated and requires that
PWSs evaluate their distribution systems for appropriate sampling locations.
The results of this evaluation may affect the number and location of samples.
NCDM Recommendations Version 3.0
Page 127 3/19/2013

The scope of the rule also increased to cover consecutive systems that receive
finished water from other systems. The first reporting deadline for compliance
with the Stage 2 rule was in 2006 but it will be a number of years before the
rule requires the new compliance calculations based on routine DBP samples.
Currently, therefore, Safe Drinking Water Act standards exist for two classes of
halogenated organic DBPs, trihalomethanes (THM) and haloacetic acids
(HAA), and for two inorganic compounds, bromate and chlorite (US EPA,
2007). Given the near ubiquity of chlorine disinfection, the THMs and HAAs
are useful indicators of risk for other DBPs because they occur at high levels
and are easily measured.

Use of Measure

In summary, evidence suggests that disinfection byproducts adversely affect
human health. The THMs and HAAs are the most commonly formed DBPs that
are routinely tracked in state Safe Drinking Water Act databases. Measures
based on these contaminants thus provide a window into potential human
exposure to DBPs in publicly provided drinking water. They show where
people are potentially exposed to high levels of DBPs. These water supply
systems are candidates for enhancement of source water quality, infrastructure
improvements or other interventions to reduce DBP exposure.
These measures assist by providing data that can be used for surveillance
purposes.




Limitations of The
Measure

Data Sources
Limitations of Data
Sources

Distribution measures provide information on the number of CWS and
the number of people potentially exposed to nitrate at different
concentrations.
Maximum concentrations provide information on the peak potential
exposure to nitrate at the state level.
Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.

The current measures are derived for CWS only. Transient non-community
water systems, which are regulated by EPA, may also be an important source of
DBPs exposure. Measures do not account for the variability in sampling,
numbers of sampling repeats, and variability within systems. Concentrations in
drinking water cannot be directly converted to exposure, because water
consumption varies by climate, level of physical activity, and between people
(EPA 2004). Due to errors in estimating populations, the measures may
overestimate or underestimate the number of affected people.
State grantee
Safe Drinking Water Act compliance data include only a handful of the
hundreds of known DBPs (Weinberg et al. 2002), most of which occur in
chemical classes other than THMs and HAAs. While compliance sampling for
THMs and HAAs is directed at the DBPs thought to be most commonly
produced by chlorination, non-regulated DBPs exist even among the THMs and
HAAs.

NCDM Recommendations Version 3.0
Page 128 3/19/2013

Concern has also been expressed about iodinated THMs and HAAs which,
while present in lower concentrations than the brominated and chlorinated
THMs, are thought to be toxic at lower doses (e.g. Plewa et al. 2004).
THMs and HAAs may not be the most satisfactory indicators of DBP levels in
waters subject to alternative disinfection methods that produce different DBPs
in different proportions than chlorination (Richardson 2002, Weinberg et al.
2002) and may result in high levels of unregulated DBPs. Little is known about
the quantitative occurrence of these DBPs in the distribution system
(Richardson et al. 2002, Krasner et al. 2006). While the health effects of
different DBPs may vary, with some suspected to be hazardous, few have been
characterized for their effects on human health (Woo et al. 2002).
Correlations among different DBPs can be relatively low (King et al. 2004,
Rodriguez et al. 2004a) so that the measured concentrations of THMs and
HAAs may not be good predictors of exposure to other DBPs or overall DBP
exposure. THM4 or HAA5, which are the only available data in some state
databases, may therefore tell little about the relative concentrations of the
THMs or HAAs.
DBP levels vary seasonally (Singer et al. 1981, Whitaker et al. 2003, Rodriguez
et al. 2004b). Quarterly samples may not capture maximum levels and may not
even adequately reflect short term levels. They may therefore be inadequate for
estimating exposure during critical periods of a pregnancy, which may be as
short as tow to three weeks, especially if peak exposure matters more than
average exposure. Furthermore, these fluctuations make it difficult to
characterize levels with a single number such as an annual average and thus
pose challenges to the development of meaningful synopses of patterns and
trends.
DBP levels are spatially and temporally labile within a distribution system
(Rodriguez et al. 2004b). THM levels increase with time after disinfection and
therefore with distance from the treatment plant (Chen and Weisel 1998,
Rodriguez and Sérodes 2001). HAA levels may increase or decrease (Chen and
Weisel 1998, Rodriguez et al. 2004b), depending upon distribution system
conditions. Rechlorination further increases DBP levels. For all but small
distribution systems it is therefore impossible to adequately characterize DBP
levels with a single value. DBP sampling locations may change over time,
making it more difficult to compare measurements from year to year. Better
estimation of DBP levels will require spatial and hydraulic modeling of
distribution systems.
Water supply systems sample for DBPs on different schedules that range from
quarterly to triennially. Different sampling frequencies complicate comparisons
among different water supply systems. Long intervals between samples,
although allowed only where THM and HAA levels have been found to be well
under the MCL, create greater uncertainty about levels between sampling dates
NCDM Recommendations Version 3.0
Page 129 3/19/2013

and require stronger assumptions when estimating exposure during short term
events such as pregnancies. When allowed, annual or triennial monitoring takes
place during the month of warmest weather and may therefore overestimate
average DBP levels.
Water supply systems that have disinfection waivers generally have no DBP
sample results. While the default assumption that these water supply systems
have DBP concentrations of zero is generally reasonable, low levels of DBPs
can be found in raw ground water, e.g., from surface contamination or from
movement of chlorinated water from onsite wastewater treatment systems into
ground water.
Human behavior greatly influences exposure, complicating efforts to estimate
exposure from tap water measurements (Nieuwenhuijen et al. 2000, Kaur et al.
2004, Nuckols et al. 2005). Among the influences on exposure are showering
and bathing time, consumption of tap water, use of bottled water, and exposure
to water at workplaces or other locations outside the home. Moreover,
ascertaining DBP levels in drinking water does not address other routes of
exposure such as swimming (Villanueva et al. 2007, Zwiener et al. 2007). This
consideration is not strictly a limitation of the measure but pertains to using the
measure as an indicator of exposure.

Related Indicators
References

Some state SDWA databases may contain only totals for THMs and HAAs and
may not record sample results for individual DBPs. Measures involving
individual THMs and HAAs cannot be calculated for these states.
Public Water Use
1. Amy G., G. Craun, S. W. Krasner, K. P. Cantor, M. Hildesheim, P. Weyer, and W.
D. King. 2005. Improved exposure assessment on existing cancer studies. Awwa
Research Foundation, Denver, Colorado.
2. Bellar, T. A., J. J. Lichtenberg, and R. C. Kroner. 1974. The occurrence of
organohalides in chlorinated drinking waters. Journal American Water Works
Association 66:703-706.
3. Boorman, G. A., V. Dellarco, J. K. Dunnick, R. E. Chapin, S. Hunter, F.
Hauchman, H. Gardner, M. Cox, and R. C. Sills. 1999. Drinking water disinfection
byproducts: Review and approach to toxicity evaluation. Environmental Health
Perspectives 107 (suppl 1):207-217.
4. Bove, F. J., M. C. Fulcomer, J. B. Klotz, J. Esmart, E. M. Dufficy, and J. E.
Savrin. 1995. Public drinking water contamination and birth outcomes. American
Journal of Epidemiology 141:850-862.
5. Cantor, K. P., C. F. Lynch, M. Hildesheim, M. Dosemeci, J. Lubin, M. Alavanja,
and G. Craun. 1998. Drinking water source and chlorination byproducts I. Risk of
bladder cancer. Epidemiology 9:21-28.
6. Cedergren, M. I., A. J. Selbing, O. Lofman, and B. A. J. Kallen. 2002.
Chlorination byproducts and nitrate in drinking water and risk for congenital
cardiac defects. Environmental Research 89:124-130.
7. Chen, W. J., and C. P. Weisel. 1998. Halogenated DBP concentrations in a
distribution system. Journal American Water Works Association 90:151-163.
8. Dodds, L., W. King, C. Woolcott, and J. Pole. 1999. Trihalomethanes in public

NCDM Recommendations Version 3.0
Page 130 3/19/2013

water supplies and adverse birth outcomes. Epidemiology 10:233-237.
9. Dodds, L., and W. D. King. 2001. Relation between trihalomethane compounds
and birth defects. Occupational & Environmental Medicine 58:443-446.
10. Dodds, L., W. D. King, A. C. Allen, B. A. Armson, D. B. Fell, and C. Nimrod.
2004. Trihalomethanes in public water supplies and risk of stillbirth.
Epidemiology 15:179-186.
11. Gallagher, M. D., J. R. Nuckols, L. Stallones, and D. A. Savitz. 1998. Exposure to
trihalomethanes and adverse pregnancy outcomes. Epidemiology 9:484-489.
12. Infante-Rivard, C. 2004. Drinking water contaminants, gene polymorphisms, and
fetal growth. Environmental Health Perspectives 112:1213-1216.
13. Kaur, S., M. J. Nieuwenhuijsen, and P. Steer. 2004. Exposure of pregnant women
to tap water related activities. Occupational and Environmental Medicine 61:454460.
14. King, W. D., L. Dodds, and A. C. Allen. 2000. Relation between stillbirth and
specific chlorination by-products in public water supplies. Environmental Health
Perspectives 108:883-886.
15. King, W. D., L. Dodds, B. A. Armson, A. C. Allen, D. B. Fell, and C. Nimrod.
2004. Exposure assessment in epidemiologic studies of adverse pregnancy
outcomes and disinfection byproducts. Journal of Exposure Analysis and
Environmental Epidemiology 14:466-472.
16. King, W. D., and L. D. Marrett. 1996. Case-control study of bladder cancer and
chlorination by-products in treated water (Ontario, Canada). Cancer Causes &
Control 7:596-604.
17. Klotz, J. B. and L. A. Pyrch. 1999. Neural tube defects and drinking water
disinfection by-products. Epidemiology 10:383-90.
18. Komulainen, H. 2004. Experimental cancer studies of chlorinated by-products.
Toxicology 198(1-3):239-48.
19. Krasner, S. W., H. S. Weinberg, S. D. Richardson, S. J. Pastor, R. Chinn, M. J.
Sclimenti, G. D. Onstad, and A. D. Thruston, Jr. 2006. The occurrence of a new
generation of disinfection byproducts. Environmental Science & Technology
40:7175-7185.
20. Nieuwenhuijsen, M. J., M. B. Toledano, and P. Elliott. 2000. Uptake of
chlorination disinfection by-products; a review and a discussion of its implications
for exposure assessment in epidemiological studies. Journal of Exposure Analysis
& Environmental Epidemiology 10:586-599.
21. Nuckols, J. R., D. L. Ashley, C. Lyu, S. M. Gordon, A. F. Hinckley, and P. Singer.
2005. Influence of tap water quality and household water use activities on indoor
air and internal dose levels of trihalomethanes. Environmental Health Perspectives
113:863-870.
22. Plewa, M. J., Y. Kargalioglu, D. Vankerk, R. Minear, and E. D. Wagner. 2002.
Mammalian cell cytotoxicity and genotoxicity analysis of drinking water
disinfection by-products. Environmental & Molecular Mutagenesis 40:134-142.
23. Plewa, M. J., E. D. Wagner, S. D. Richardson, A. D. Thruston, Jr., Y.-T. Woo, and
A. B. McKague. 2004. Chemical and biological characterization of newly
discovered iodoacid drinking water disinfection byproducts. Environmental
Science & Technology 38:4713-4722.
24. Richardson, S. D. 2005. New disinfection by-product issues: Emerging DBPs and
alternative routes of exposure. Global NEST Journal 7:43-60.
25. Richardson, S. D., J. E. Simmons, and G. Rice. 2002. Disinfection byproducts:
The next generation. Environmental Science & Technology 36:198A-205A.
26. Roberts, M. G., P. C. Singer, and A. Obolensky. 2002. Comparing total HAA and

NCDM Recommendations Version 3.0
Page 131 3/19/2013

27.

28.

29.

30.
31.

32.

33.

34.

35.

36.

37.

38.

39.
40.
41.

42.

total THM concentrations using ICR data. Journal American Water Works
Association 94:103-114.
Rodriguez, M. J., and J. B. Serodes. 2001. Spatial and temporal evolution of
trihalomethanes in three water distribution systems. Water Research 35:15721586.
Rodriguez, M. J., M. Huard, and J. B. Sérodes. 2004a. Experimental study of the
formation of chlorination by-products in potable water of Quebec City, Canada.
Bulletin of Environmental Contamination & Toxicology 72:211-218.
Rodriguez, M. J., J. B. Sérodes, and P. Levallois. 2004b. Behavior of
trihalomethanes and haloacetic acids in a drinking water distribution system.
Water Research 38:4367-4382.
Rook, J. J. 1974. Formation of haloforms during chlorination of natural waters.
Water Treatment and Examination 23:234-243.
Sadiq, R., and M. J. Rodriguez. 2004. Disinfection by-products (DBPs) in drinking
water and predictive models for their occurrence: a review. Science of the Total
Environment 321:21-46.
Savitz, D. A., P. C. Singer, A. H. Herring, K. E. Hartmann, H. S. Weinberg, and C.
Makarushka. 2006. Exposure to drinking water disinfection by-products and
pregnancy loss. American Journal of Epidemiology 164:1043-1051.
Shaw, G. M., D. Ranatunga, T. Quach, E. Neri, A. Correa, and R. R. Neutra. 2003.
Trihalomethane exposures from municipal water supplies and selected congenital
malformations. Epidemiology 14:191-199.
Singer, P. C. 2006. DBPs in drinking water: Additional scientific and policy
considerations for public health protection. Journal American Water Works
Association 98:73-80.
Singer, P. C., J. J. Barry III, G. M. Palen, and A. E. Scrivner. 1981.
Trihalomethane formation in North Carolina drinking waters. Journal American
Water Works Association 73:392-401.
Swan, S. H., K. Waller, B. Hopkins, G. Windham, L. Fenster, C. Schaefer, and R.
R. Neutra. 1998. A prospective study of spontaneous abortion: relation to amount
and source of drinking water consumed in early pregnancy. Epidemiology 9:126133.
Symons, J. M., T. A. Bellar, J. J. Carswell, J. DeMarco, K. L. Kropp, G. G.
Robect, D. R. Seeger, C. J. Slocum, B. L. Smith, and A. A. Stevens. 1975.
National organics reconnaissance survey for halogenated organics. Journal
American Water Works Association 67:634-647.
Toledano, M. B., M. J. Nieuwenhuijsen, N. Best, H. Whitaker, P. Hambly, C. de
Hoogh, J. Fawell, L. Jarup, and P. Elliott. 2005. Relation of trihalomethane
concentrations in public water supplies to stillbirth and birth weight in three water
regions in England. Environmental Health Perspectives 113:225-232.
US EPA (United States Environmental Protection Agency). 1998. Stage 1
Disinfectants and Disinfection byproducts rule. Federal Register 63:69389-69476.
US EPA (United States Environmental Protection Agency). 2006. Stage 2
Disinfectants and Disinfection byproducts rule. Federal Register 71:387-493.
US EPA (United States Environmental Protection Agency). 2007. National Water
Program Fiscal Year 2007 Guidance, Appendix A: Program Activity Measures
Supporting the National Water Program for FY 2007. US EPA Web site at
http://www.epa.gov/ocfo/npmguidance/ owater/2007/2007_appendixa.ppt
(accessed 18/15/07).
Villanueva, C. M., K. P. Cantor, S. Cordier, J. J. K. Jaakkola, W. D. King, C. F.
Lynch, S. Porru, and M. Kogevinas. 2004. Disinfection byproducts and bladder

NCDM Recommendations Version 3.0
Page 132 3/19/2013

cancer: A pooled analysis. Epidemiology 15:357-367.
43. Villanueva, C. M., K. P. Cantor, J. O. Grimalt, N. Malats, D. Silverman, A.
Tardon, R. Garcia-Closas, C. Serra, A. Carrato, G. Castano-Vinyals, R. Marcos, N.
Rothman, F. X. Real, M. Dosemeci, and M. Kogevinas. 2007. Bladder cancer and
exposure to water disinfection by-products through ingestion, bathing, showering,
and swimming in pools. American Journal of Epidemiology 135:148-156.
44. von Gunten, U. 2003. Ozonation of drinking water: Part II. Disinfection and byproduct formation in presence of bromide, iodide, or chlorine. Water Research
37:1469-1487.
45. Waller, K., S. H. Swan, G. DeLorenze, and B. Hopkins. 1998. Trihalomethanes in
drinking water and spontaneous abortion. Epidemiology 9:134-140.
46. Weinberg H. S., S. W. Krasner, S. D. Richardson, and A. D. Thruston, Jr. 2002.
The occurrence of disinfection by-products (DBPs) of health concern in drinking
water: results of a nationwide DBP occurrence study; EPA/600/R-02/068. US
EPA, Athens, Georgia.
47. Whitaker, H., M. J. Nieuwenhuijsen, N. Best, J. Fawell, A. Gowers, and P. Elliot.
2003. Description of trihalomethane levels in three UK water suppliers. Journal of
Exposure Analysis and Environmental Epidemiology 13:17-23.
48. Woo, Y.-T., D. Lai, J. L. McClain, M. K. Manibusam, and V. Dellarco. 2002. Use
of mechanism-based structure-activity relationships analysis in carcinogenic
potential ranking for drinking water disinfection by-products. Environmental
Health Perspectives 110 (suppl 1):75-87.
49. Wright, J. M., J. Schwartz, and D. W. Dockery. 2004. The effect of disinfection
by-products and mutagenic activity on birth weight and gestational duration.
Environmental Health Perspectives 112:920-925.
50. Zwiener, C., S. D. Richardson, D. M. De Marini, T. Grummt, T. Glauner, and F.
H. Frimmel. 2007. Drowning in disinfection byproducts? Assessing swimming
pool water. Environmental Science & Technology 41:363-372.

NCDM Recommendations Version 3.0
Page 133 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: NITRATE
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT
Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
15. Quarterly distribution of number of Community Water Systems (CWS) by
mean nitrate concentration (cut-points: (0-3), (>3-5), (>5-10), (>10-20), (>20)
mg/L nitrate).
16. Yearly distribution of number of CWS by maximum nitrate concentration
(cut-points: (0-3), (>3-5), (>5-10), (>10-20), (>20) mg/L nitrate).
17. Yearly distribution of number of CWS by mean nitrate concentration (cutpoints: (0-3), (>3-5), (>5-10), (>10-20), (>20) mg/L nitrate).
18. Mean concentration of nitrate at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
19. Quarterly distribution of number of people served by CWS by mean nitrate
concentration (cut-points: (0-3), (>3-5), (>5-10), (>10-20), (>20) mg/L
nitrate).
20. Yearly distribution of number of people served by CWS by maximum nitrate
concentration (cut-points: (0-3), (>3-5), (>5-10), (>10-20), (>20) mg/L
nitrate).
21. Yearly distribution of number of people served by CWS by mean nitrate
concentration (cut-points: (0-3), (>3-5), (>5-10), (>10-20), (>20) mg/L
nitrate).
Nitrate measures will be developed from water system attribute and water quality data
stored in state Safe Drinking Water Act (SDWA) databases such as the Safe Drinking
Water Information System (SDWIS/State). Data will be cleaned and transformed to a
standard format. Analytical results of drinking water samples (usually taken at entry
points to the distribution system or representative sampling points after treatment)
will be used in conjunction with information about each CWS (such as service
population and latitude and longitude of representative location of the CWS service
area) to generate the measures.
Concentration of nitrate, mg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution extents,
principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Nitrates and Public Health

Derivation of
Measures

NCDM Recommendations Version 3.0
Page 134 3/19/2013

Nitrate was first identified as a public health threat in drinking water in 1945 when
high nitrate levels from private wells were shown to cause methemoglobinemia or
“blue baby syndrome” in infants who received formula made from well water. When
an individual is exposed to nitrate it can be converted to nitrite (NO2-) in the body
and then oxidize the ferrous iron (Fe+2) in deoxyhemoglobin in the blood to form
methemoglobin containing ferric iron (Fe+3). Methemoglobin cannot transfer oxygen
to tissues; thus nitrate or nitrite can starve the body of oxygen and produce a clinical
condition known as cyanosis, where the lips and extremities turn gray or blue. Infants
younger than four months of age are more sensitive than adults, and can develop
“blue baby” syndrome from intake of nitrate higher than 10 mg/L nitrate or 45 mg/L
nitrate–nitrogen. Blue baby syndrome is fatal in about ten percent of the cases
(ATSDR, 2007). Usually there are no outward signs of cyanosis at methemoglobin
levels below 20 percent (Dabney et al, 1990).
In addition, there is some evidence to suggest that exposure to nitrate in drinking
water is also associated with adverse reproductive outcomes such as spontaneous
abortions, intrauterine growth retardation, and various birth defects such as
anencephaly, related to fetal exposures to nitrate. However, the evidence is
inconsistent (Manassaram et al, 2006).
Similarly, long term exposure to higher nitrate levels in drinking water has been
suggested as a risk factor for cancer. Cancer at several sites (i.e. gastric, colorectal,
bladder, urothelial, brain, esophagus, ovarian and non-Hodgkins lymphoma have
been shown to be associated with nitrate in drinking water in some studies (Sandor et
al, 2001; Weyer et al, 2001; Gulis et al, 2002; De Roos et al, 2003; Volkmer et al,
2005; Ward et al, 2005b; Chiu et al, 2007; ). Other studies have not found any
association (Ward et al, 2003; Ward et al, 2005, 2005c; Ward et al, 2006; Zeegers et
al, 2006). Significant regional differences in cancer risk may occur (Mueller et al,
2001). Occupational exposures are also of concern as nitrate fertilizer workers have
shown increased risk for stomach cancer (Zandjani et al. 1994).
Sources of Nitrate
Nitrate is the most commonly found contaminant in groundwater aquifers worldwide
(Ward, 2005 from: Spalding and Exner 1993). Nitrate (NO3-) originates in drinking
water from nitrate-containing fertilizers, sewage and septic tanks, and decaying
natural material such as animal waste. Nitrate is very soluble in water, can easily
migrate, and does not evaporate (EPA Consumer Fact Sheet). Anthropogenic sources
of nitrates are increasing resulting in increased nitrate levels in water resources.
Surface water and shallow wells in both rural and urban areas can be affected.
Consequently, private wells are especially vulnerable to excess levels of nitrates.
Excess levels of nitrate and nitrite can occur in community water supplies. A U.S.
Geological Survey (USGS) study found nitrate levels exceeded regulatory monitoring
standards in 2% of a sample of 242 public drinking water wells between 1992 and
1999 (Squillace et al, 2002). Levels of nitrates in private wells are less well known;
private wells are not regularly monitored and are often more vulnerable to higher
levels of nitrates because they draw water from shallower groundwater aquifers. The
NCDM Recommendations Version 3.0
Page 135 3/19/2013

USGS estimates approximately 22% of domestic wells in agricultural areas of the
U.S. exceed the MCL (Ward, 2007).
Nitrate Regulation and Monitoring
Congress established the Safe Drinking Water Act in 1974, which set enforceable
Maximum Contaminant Levels (MCLs) and non-enforceable Maximum Contaminant
Level Goals (MCLGs) for certain specified contaminants. In the case of nitrate in
drinking water, the MCLG of 10 mg/L (ppm) was established from human data from
studies of methemoglobinemia in young children. (Johnson and Kross 1990; Walton,
1950). The MCL is also set at 10 ppm, and any exceedance of the MCL is potentially
serious as there is no additional margin of safety between the MCLG and the MCL.
2002). The MCLG and MCL for nitrite are 1 mg/L. While evidence to suggest MCL
exposures for chronic health endpoints remains inconclusive, there is some evidence
to suggest that chronic exposure to nitrate levels below the MCL may be of concern
(Ward, 2005).
Use of Measure

These measures assist by providing data that can be used for surveillance purposes.




Limitations of The
Measure

Data Sources
Limitations of Data
Sources

Distribution measures provide information on the number of CWS and the
number of people potentially exposed to nitrate at different concentrations.
Maximum concentrations provide information on the peak potential exposure
to nitrate at the state level.
Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.

The current measures are derived for CWS only. Private wells are another important
source of population exposure to nitrate. Transient non-community water systems,
which are regulated by EPA, may also be an important source of nitrate exposure.
Measures do not account for the variability in sampling, numbers of sampling repeats,
and variability within systems. Concentrations in drinking water cannot be directly
converted to exposure, because water consumption varies by climate, level of
physical activity, and between people (EPA 2004). Due to errors in estimating
populations, the measures may overestimate or underestimate the number of affected
people.
State grantee
Nitrate levels can vary substantially in groundwater; thus high levels may not be
captured by even quarterly sampling. Estimates of the number of people potentially
exposed may be unreliable as they are based on estimates made by the water system
operator. Concentrations in drinking water cannot be directly converted to exposure
because overall water consumption, and the proportion of water consumed that comes
from the tap is quite variable (EPA 2004). In systems that have more than one Entry
point to the Distribution system, the actual nitrate level at any given house is a
mixture of the levels from all contributing sources. Compliance samples are taken at
each entry point to the distribution system. In systems with separate wells serving
some branches or sections of the distribution system, the system mean would tend to

NCDM Recommendations Version 3.0
Page 136 3/19/2013

underestimate the nitrate concentration of people served by wells with higher nitrate
concentrations.

Related Indicators
References

Exposure may be higher or lower than estimated if data from multiple entry points for
water with different nitrate levels are averaged to estimate levels for the PWS.
Public Water Use
51. ATSDR Case Studies in Environmental Medicine: Nitrate/Nitrite Toxicity.
http://www.atsdr.cdc.gov/HEC/CSEM/nitrate/index.html Downloaded 08/07/07
52. Bosch, H. M., A. B. Rosenfield, R. Huston, H. R. Shipman, and F. L. Woodward. 1950.
Methemoglobinemia and Minnesota well supplies. Am. Water Works Assoc J 42:161170.
53. Chiu HF, Tsai SS, Yang CY. 2007. Nitrate in drinking water and risk of death from
bladder cancer: an ecological case-control study in Taiwan. J Toxicol Environ Health A
70(12):1000-1004.
54. Coss A, Cantor KP, Reif JS, Lynch CF, Ward MH. 2004. Pancreatic cancer and drinking
water and dietary sources of nitrate and nitrite. Am J Epidemiol 159(7):693-701.
55. Dabney BJ, Zelarney PT, Hall AH. 1990. Evaluation and treatment of patients exposed to
systemic asphyxiants. Emerg Care Q 6(3):65-80
56. De Roos AJ, Ward MH, Lynch CF, Cantor KP. 2003. Nitrate in public water supplies
and the risk of colon and rectum cancers. Epidemiology 14(6):640-649.
57. Gulis G, Czompolyova M, Cerhan JR. 2002. An ecologic study of nitrate in municipal
drinking water and cancer incidence in Trnava District, Slovakia. Environ Res 88(3):182187.
58. Johnson CJ and Kross BC. 1990. Continuing importance of nitrate contamination of
groundwater and wells in rural areas. Am J Ind Med 18(4):449-456.
59. Mueller BA, Newton K, Holly EA, Preston-Martin S. 2001. Residential water source and
the risk of childhood brain tumors. Environ Health Perspect 109(6):551-556.
60. Ruckart PZ, Henderson AK, Black ML, Flanders WD. 2007. Are nitrate levels in
groundwater stable over time? J Expo Sci Environ Epidemiol Apr 11; [Epub ahead of
print]
61. Sandor J, Kiss I, Farkas O, Ember I. 2001. Association between gastric cancer mortality
and nitrate content of drinking water: ecological study on small area inequalities. Eur J
Epidemiol 17(5):443-447.
62. U.S. Environmental Protection Agency Office of Water: Candidate Contaminants List.
http://www.epa.gov/safewater/ccl/index.html Downloaded 08/02/07
63. U.S. Environmental Protection Agency. Office of Water (4606) Occurrence Estimation
Methodology and Occurrence Findings Report for the Six-Year Review of Existing
National
Primary Drinking Water Regulations. EPA-815-R-03-006 www.epa.gov June 2003.
http://www.epa.gov/safewater/standard/review/pdfs/support_6yr_occurancemethods_final.pdf
Downloaded 08/02/07
64. U.S. Environmental Protection Agency (2007b): Technical Factsheet on: Nitrate/Nitrite.
http://www.epa.gov/safewater/dwh/t-ioc/nitrates.html Downloaded 08/07/07
65. Volkmer BG, Ernst B, Simon J, Kuefer R, Bartsch G Jr, Bach D, Gschwend JE. 2005.
Influence of nitrate levels in drinking water on urological malignancies: a communitybased cohort study. BJU Int 95(7):972-976.
66. Walton, G. 1951. Survey of literature relating to infant methemoglobinemia due to
nitrate-contaminated water. Am J Public Health 41:986-996.
67. Ward MH, Cantor KP, Riley D, Merkle S, Lynch CF. 2003. Nitrate in public water

NCDM Recommendations Version 3.0
Page 137 3/19/2013

supplies and risk of bladder cancer. Epidemiology 14(2):183-190.
68. Ward MH, Heineman EF, McComb RD, Weisenburger DD. 2005. Drinking water and
dietary sources of nitrate and nitrite and risk of glioma. J Occup Environ Med
47(12):1260-1267.
69. Ward MH, deKok TM, Levallois P, Brender J, Gulis G, Nolan BT, VanDerslice J. 2005b.
Workgroup report: Drinking-water nitrate and health--recent findings and research needs.
Environ Health Perspect 113(11):1607-1614.
70. Ward MH, Heineman EF, McComb RD, Weisenburger DD. 2005c. Drinking water and
dietary sources of nitrate and nitrite and risk of glioma. J Occup Environ Med
47(12):1260-1267.
71. Ward MH, Cerhan JR, Colt JS, Hartge P. 2006. Risk of non-Hodgkin lymphoma and
nitrate and nitrite from drinking water and diet. Epidemiology 17(4):375-382.
72. Weyer PJ, Cerhan JR, Kross BC, Hallberg GR, Kantamneni J, Breuer G, Jones MP,
Zheng W, Lynch CF. 2001. Municipal drinking water nitrate level and cancer risk in
older women: the Iowa Women's Health Study. Epidemiology 12(3):327-338.
73. Zandjani F, Hogsaet B, Andersen A, Langard S. 1994. Incidence of cancer among nitrate
fertilizer workers. Int Arch Occup Environ Health 66:189-93.
74. Zeegers MP, Selen RF, Kleinjans JC, Goldbohm RA, van den Brandt PA. 2006. Nitrate
intake does not influence bladder cancer risk: the Netherlands cohort study. Environ
Health Perspect 114(10):1527-1531.

NCDM Recommendations Version 3.0
Page 138 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: PUBLIC WATER USE
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Exposure
22. Number of people receiving water from community water systems.

Derivation of Measures

Units

This measure will be developed from water system attribute and water quality
data stored in state Safe Drinking Water Act (SDWA) databases such as the
Safe Drinking Water Information System (SDWIS/State). Data will be cleaned
and transformed to a standard format.
1. Number of people

Geographic Scope

State

Geographic Scale

State

Time Period

2009 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Public Water Use and Public Health
The public water use index provides some data to explore the relative
importance of community water supplies as sources of drinking water and to
provide context for subsequent community drinking water system (CWS)
indicators. SDWA collects data for a number of different types of public water
systems of which community water systems (CWS) are a sub-set. The
community water systems represent non-transient public water systems that
serve year round community residents and are the focus of the initial indicators.
The range of state populations served by CWS as their primary residential
drinking water source varies from 95% to as low as 40% within the United
States. Understanding the relative population coverage of these indicators by
state helps to understand representativeness of these data for prioritization and
evaluation across the United States and within individual states and
communities.

Use of Measure

This measure can be useful in providing data for surveillance purposes.
• Estimated population potentially exposed to contaminants in CWS.

Limitations of The
Measure

Data Sources
Limitations of Data
Sources

The current measure is derived for CWS only. Private wells are another
important source of population exposure to water contaminants. Transient noncommunity water systems, which are regulated by EPA, may also be an
important source of potential exposure.
State grantee
Population estimates are rough and may overestimate or underestimate the
number of affected people.

NCDM Recommendations Version 3.0
Page 139 3/19/2013

Related Indicators
Additional Information

All other community water indicators.
1. U.S. Environmental Protection Agency, Water On Tap, Office of Water (4601)
EPA 816-K-09-002, December 2009.
http://water.epa.gov/drink/guide/upload/book_waterontap_full.pdf
2. U.S. Environmental Protection Agency, Public Drinking Water Systems: Facts
and Figures
http://water.epa.gov/infrastructure/drinkingwater/pws/factoids.cfm
3. U.S. Environmental Protection Agency, Public Drinking Water Systems
Programs. http://water.epa.gov/infrastructure/drinkingwater/pws/index.cfm

NCDM Recommendations Version 3.0
Page 140 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: COMBINED RADIUM-226 AND -228
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT
Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
1. Yearly distribution of number of Community Water Systems (CWS) by
maximum Radium concentration (cut-points: 0-3, >3-5, >5-10, >10 pCi/L
Radium).
2. Yearly distribution of number of CWS by mean Radium concentration
(cut-points: cut-points: 0-3, >3-5, >5-10, >10 pCi/L Radium).
3. Mean concentration of Radium at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
4. Yearly distribution of number of people served by CWS by maximum
Radium concentration (cut-points: 0-3, >3-5, >5-10, >10 pCi/L Radium).
5. Yearly distribution of number of people served by CWS by mean Radium
concentration (cut-points: 0-3, >3-5, >5-10, >10 pCi/L Radium).
Combined Radium-226 and -228 measures will be developed from water system
attribute and water quality data stored in state Safe Drinking Water Act (SDWA)
databases such as the Safe Drinking Water Information System (SDWIS/State).
Data will be cleaned and transformed to a standard format. Analytical results of
drinking water samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
pCi/L combined Radium-226 & -228

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Radium-226 and -228 and Public Health
Radium is a naturally occurring silvery-white radioactive metal that can exist in
several forms called isotopes. Radium is produced constantly by the radioactive
decay of uranium and thorium. Uranium and thorium are found in small amounts
in most rocks and soil. Some of the radiation from radium is being released
constantly into the environment. It is this radioactive decay that causes concern

Derivation of Measures

NCDM Recommendations Version 3.0
Page 141 3/19/2013

about the safety of radium and all other radioactive substances. Two of the main
radium isotopes found in the environment are radium-226 and radium-228. The
decay of radium-226 results in the formation of radon which exists as a gas and is
mobile in environmental media. Radium has been used as a radiation source for
treating cancer, in radiography of metals, and combined with other metals as a
neutron source for research and radiation instrument calibration. Until the 1960s,
radium was a component of the luminous paints used for watch and clock dials,
instrument panels in airplanes, military instruments, and compasses (ATSDR,
2010).
Everyone is exposed to low levels of radium in the air, water, and food. Higher
levels may be found in the air near industries that burn coal or other fuels or near
sites that mine or mill uranium. It also may be found at higher levels in drinking
water from groundwater wells. Miners, particularly miners of uranium and hard
rock, are exposed to higher levels of radium. It may also be found at radioactive
waste disposal sites (ATSDR, 1990).
It is not known whether long-term exposure to radium at the levels that are
normally present in the environment (for example, 1 pCi of radium per gram of
soil) is likely to result in harmful health effects. However, exposure to higher
levels of radium over a long period of time may result in harmful effects
including anemia, cataracts, fractured teeth, cancer (especially bone cancer), and
death. Patients who were injected with radium in Germany, from 1946 to 1950,
for the treatment of certain diseases including tuberculosis were significantly
shorter as adults than people who were not treated. Some of these health effects
may take years to develop and mostly are due to gamma radiation. Radium gives
off gamma radiation, which can travel fairly long distances through air.
Therefore, just being near radium at the high levels that may be found at some
hazardous waste sites may be dangerous to your health.
Exposure to high levels of radium results in an increased incidence of bone, liver,
and breast cancer. The EPA and the National Academy of Sciences, Committee
on Biological Effects of Ionizing Radiation, has stated that radium is a known
human carcinogen.

Biomonitoring Information
Urine tests can determine if you have been exposed to radium. Another test
measures the amount of radon (a breakdown product of radium) in exhaled air.
Both types of tests require special equipment and cannot be done in a doctor's
office. These tests cannot tell how much radium you were exposed to, nor can
they be used to predict whether you will develop harmful health effects (ATSDR,
1990). Levels of radium in the U.S. population are unknown.
Sources of Radium
Radium forms from the decay of uranium or thorium in the environment.
NCDM Recommendations Version 3.0
Page 142 3/19/2013

Radium -226 is formed from the decay of uranium-238; Radium-228 is formed
from the decay of thorium. Radium is abundant in low levels everywhere
because it originates from uranium which is commonly found in all rocks, soil
and water. (EPA, 2010)
Radium Regulation and Monitoring
The EPA has set a drinking water limit of 5 picocuries per liter (5 pCi/L) for
radium-226 and radium-228 (combined) (EPA, 2009). A gross alpha particle
activity measurement may be substituted for the required radium-226
measurement provided that the measured gross alpha particle activity does not
exceed 5 pCi/L. The EPA lifetime exposure cancer risk estimate for radium at
the MCL, is approximately 1-2 cases per 10,000 people.
Monitoring frequency
Once a CWS has satisfied initial monitoring requirements (4 quarterly samples at
every entry point to the distribution system within the first quarter after initiating
the source); the required frequency for Combined Radium-226 and -228
monitoring is once every three years if the average of the initial monitoring
results for the contaminant is greater than one-half the MCL but at or below the
MCL. States may allow CWS to reduce the frequency of monitoring from once
every three years to once every six or nine years at each sampling point, if the
average of the initial monitoring results for each contaminant is below the
detection limit. If a system has a monitoring result that exceeds the MCL while
on reduced monitoring, the system must collect and analyze quarterly samples at
that sampling point until the system has results from four consecutive quarters
that are below the MCL, unless the system enters into another schedule as part of
a formal compliance agreement with the State (CFR, 2002).
Use of Measure

Limitations of The
Measure

These measures assist by providing data that can be used for surveillance
purposes.
• Distribution measures provide information on the number of CWS and the
number of people potentially exposed to combined Radium-226 and -228 at
different concentrations.
• Maximum concentrations provide information on the peak potential
exposure to combined Radium-226 and -228 at the state level.
• Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.
The current measures are derived for CWS only. Private wells may be another
source of population exposure to combined Radium-226 and -228. Transient
non-community water systems, which are regulated by EPA, may also be an
important source of combined Radium-226 and -228 exposure. Measures do not
account for the variability in sampling, numbers of sampling repeats, and
variability within systems. Concentrations in drinking water cannot be directly
converted to exposure, because water consumption varies by climate, level of
physical activity, and between people (EPA 2004). Due to errors in estimating

NCDM Recommendations Version 3.0
Page 143 3/19/2013

Data Sources
Limitations of Data
Sources

Related Indicators
References

populations, the measures may overestimate or underestimate the number of
affected people.
State grantee
The required monitoring frequency for combined Radium-226 and -228 is
infrequent and may be as intermittent as every nine years; therefore most states
will have very little data on this contaminant.
Ground water systems may have multiple wells with different combined Radium226 and -228 concentrations that serve different parts of the population.
Compliance samples are taken at each entry point to the distribution system. In
systems with separate wells serving some branches or sections of the distribution
system, the system mean would tend to underestimate the combined Radium-226
and -228 concentrations of people served by wells with higher combined
Radium-226 and -228 concentrations. Exposure may be higher or lower than
estimated if data from multiple entry points for water with different combined
Radium-226 and -228 levels are averaged to estimate levels for the PWS.
Public Water Use; Uranium
1. Agency for Toxic Substances and Disease Registry (ATSDR). Toxic Substances
Portal. Radium. 2010. Available at:
http://www.atsdr.cdc.gov/substances/toxsubstance.asp?toxid=154
2. Agency for Toxic Substances and Disease Registry (ATSDR). 1990. Toxicological
Profile for Radium. Atlanta, GA: U.S. Department of Health and Human Services,
Public Health Service Available at:
http://www.atsdr.cdc.gov/toxfaqs/TF.asp?id=790&tid=154
3. Code of Federal Regulations (CFR), 2002. Title 40 Protection of the Environment
Chapter I--Environmental Protection Agency Part 141--National Primary Drinking
Water Regulations 141.26 Monitoring frequency and compliance requirements for
radionuclides in community water systems. Available at: URL:
http://www.access.gpo.gov/nara/cfr/waisidx_02/40cfr141_02.html
4. U.S. Environmental Protection Agency (U.S. EPA). Radiation Protection, Radium,
2010. Available at:
http://www.epa.gov/radiation/radionuclides/radium.html
5. U.S. Environmental Protection Agency (U.S. EPA). The Analysis of Regulated
Contaminant Occurrence Data from public Water Systems in Support of the Second Sixyear Review of National Primary Drinking Water Regulations. EPA-815-B-09-006,
October 2009.

NCDM Recommendations Version 3.0
Page 144 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: TETRACHLOROETHENE (TETRACHLOROETHYLENE) (PCE)
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
6. Yearly distribution of number of Community Water Systems (CWS) by
maximum PCE concentration (cut-points: 0-1, >1-2, >2-5, >5 µg/L
PCE).
7. Yearly distribution of number of CWS by mean PCE concentration
(cut-points: 0-1, >1-2, >2-5, >5 µg/L PCE).
8. Mean concentration of PCE at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
9. Yearly distribution of number of people served by CWS by maximum
PCE concentration (cut-points: 0-1, >1-2, >2-5, >5 µg/L PCE).
10. Yearly distribution of number of people served by CWS by mean PCE
concentration (cut-points: 0-1, >1-2, >2-5, >5 µg/L PCE).
PCE measures will be developed from water system attribute and water quality
data stored in state Safe Drinking Water Act (SDWA) databases such as the
Safe Drinking Water Information System (SDWIS/State). Data will be cleaned
and transformed to a standard format. Analytical results of drinking water
samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
PCE, µg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be the approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Derivation of Measures

Rationale

Tetrachloroethene (PCE) and Public Health
Tetrachloroethene (PCE) is a volatile halogenated short-chain hydrocarbon.
Tetrachloroethene is used in dry cleaning, metal cleaning, the synthesis of other
chemicals, and household products such as water repellants, silicone lubricants,
and spot removers. PCE is produced and used in high volumes in the U.S. and
has been detected in urban and ambient air and occasionally in soils and
drinking water most likely contaminated by industrial discharge (Moran et al.,
2007; Rowe et al., 2007). Because of its volatility, this solvent does not persist
in the soil or water following the discontinuation of contamination.

NCDM Recommendations Version 3.0
Page 145 3/19/2013

Inhalation is the most common exposure route for the general population
including indoor sources from paints, adhesives, and cleaning solutions.
Volatilization from contaminated water (e.g., shower water) as well as the use
of household products containing this solvent can result in higher indoor than
outdoor air concentrations (ATSDR, 1997; Martin et al., 2005). Nearby dry
cleaning establishments, industries producing PCE, and contaminated waste
disposal sites can also contribute to human exposure (Armstrong and Green,
2004; ATSDR, 1997 and 2000; Schreiber et al., 1993; Wallace et al., 1991).
Drinking water may contribute to exposure when underground drinking water
supplies have been contaminated. Workers in industries such as dry cleaning,
aircraft maintenance, electronics manufacturing, and chemical production may
be exposed by inhalation or by dermal contact with PCE. The EPA has
established drinking water standards and other environmental standards for
PCE, and the FDA regulates PCE and trichloroethene as indirect food additives.
Workplace standards have been established by OSHA, and ACGIH has
recommended occupational guidelines and biological exposure indices for
monitoring workers. Human health effects from PCE at low environmental
doses or at biomonitored levels from low environmental exposures are
unknown. PCE is well absorbed by ingestion and inhalation, and animal studies
have demonstrated that liquid forms can be dermally absorbed. Following
absorption, part of the solvent dose is excreted into expired air; for PCE, about
97-99% of the dose is eliminated unmetabolized into expired air, though it has
an elimination half-life of several days (ATSDR 1997; Monster, 1986). The
retained solvent can undergo hepatic metabolism. PCE is metabolized to
trichloroacetic acid and trichloroethanol, which are eliminated in the urine.
Accidental or intentional high dose acute exposure by ingestion or inhalation
can result in loss of motor coordination, somnolence, and unconsciousness.
Inhaling high doses of PCE may also produce cardiac arrhythmias attributed to
enhanced sensitivity to catecholamines. High dose acute exposure to PCE has
resulted in reversible kidney impairment, and prolonged, low level PCE
exposure has been associated with altered renal enzyme excretion and liver
enlargement (ATSDR, 1997). Chronic occupational exposure to PCE may be
associated with mild degrees of neurological impairments, including reaction
times, verbal skills, cognitive ability, and motor function (Armstrong and
Green, 2004). Various epidemiologic studies of chronic PCE exposure in dry
cleaning workers found increased incidences of esophageal and cervical
cancers and non-Hodgkins lymphoma, but confounding exposures (e.g., other
solvents and trichloroethene) were likely (IPCS, 2006). In animal studies, PCEinduced kidney and liver tumors and caused leukemia (IARC, 1995). IARC
classifies PCE as a probable human carcinogen, and NTP classifies it as
reasonably anticipated to be a human carcinogen (IARC, 1995; NTP, 2004).
Additional information about these solvents is available from ATSDR at:
http://www.atsdr.cdc.gov/toxpro2.html.
In an analysis of occurrence data from the EPA 6 Year Review of National
NCDM Recommendations Version 3.0
Page 146 3/19/2013

Primary Drinking Water Regulations, PCE was detected in 1,262 systems
serving close to 32 million people (EPA, 2009). Concentrations of PCE were
greater than the MCL in 241 systems serving close to 15 million people. PCE
was the fifth highest occurring regulated volatile organic chemical found based
on the percent of detections found from the 6 Year Review data (EPA, 2009).
Biomonitoring Information
Levels of halogenated solvents in blood reflect recent exposure. In the
NHANES 2003-2004 subsample, the level of blood PCE for adults at the 75th
percentile of the U.S. population appear similar to the levels at the 75th
percentile reported for non-smoking adults in a subsample of NHANES 19992000 participants (CDC, 2009; Lin et al., 2008) and were similar or slightly less
than levels reported in a nonrepresentative subsample of the earlier NHANES
III (1988-1994) (Ashley et al., 1994; Churchill et al., 2001). A recent study of
low income, urban children in the Midwest reported slightly lower median PCE
levels (Sexton et al., 2005; Sexton et al., 2006) than the NHANES III levels
(Ashley et al., 1994; Churchill et al., 2001).
Comparatively higher blood levels of PCE and trichloroethene have been noted
for urban and industrial residential settings than for rural settings (Barkley et
al., 1980; Begerow et al., 1996; Brugnone et al., 1994). Residing near drycleaning facilities or storing recently dry-cleaned clothes at home can
contribute to increased blood PCE levels (Begerow et al., 1996; Popp et al.,
1992). In contrast, PCE blood levels in occupationally exposed workers have
been reported to be many thousand times higher than the general population
(Begerow et al., 1996; Furuki et al., 2000; Monster et al., 1983). The
occupational biological exposure index associated with an 8-hour exposure of
25 ppm is 500 μg/L PCE in blood (ACGIH, 2007). Non-occupational exposures
are usually well below this level. Finding a measurable amount of any of these
solvents in blood does not mean that the level of the solvent causes an adverse
health effect. Biomonitoring studies of blood halogenated solvents can provide
physicians and public health officials with reference values so that they can
determine whether or not people have been exposed to higher levels of
halogenated solvents than levels found in the general population.
Biomonitoring data can also help scientists plan and conduct research on
exposure and health effects.
Sources of PCE
The major source of PCE in drinking water is discharge from factories and dry
cleaners. A federal law called the Emergency Planning and Community Right
to Know Act requires facilities in certain industries, which manufacture,
process, or use significant amounts of toxic chemicals, to report annually on
their releases of these chemicals. For more information on the uses and releases
of chemicals in your state, contact the Community Right-to-Know Hotline:
(800) 424-9346 (EPA, 2010).

NCDM Recommendations Version 3.0
Page 147 3/19/2013

Use of Measure

Limitations of The
Measure

Data Sources
Limitations of Data
Sources

PCE Regulation and Monitoring
The EPA limits the amount of PCE that may be present in drinking water to 5
parts of PCE per billion parts of water (5 ppb), or 5 ug/L.
These measures assist by providing data that can be used for surveillance
purposes.
• Distribution measures provide information on the number of CWS and the
number of people potentially exposed to PCE at different
concentrations.
• Maximum concentrations provide information on the peak potential
exposure to PCE at the state level.
• Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.
The current measures are derived for CWS only. Private wells may be another
source of population exposure to PCE. Transient non-community water
systems, which are regulated by EPA, also may be an important source of PCE
exposure. Measures do not account for the variability in sampling, numbers of
sampling repeats, and variability within systems. Concentrations in drinking
water cannot be directly converted to exposure, because water consumption
varies by climate, level of physical activity, and between people (EPA 2004).
Due to errors in estimating populations, the measures may overestimate or
underestimate the number of affected people.
State grantee

Related Indicators

Ground water systems may have multiple wells with different PCE
concentrations that serve different parts of the population. Compliance samples
are taken at each entry point to the distribution system. In systems with
separate wells serving some branches or sections of the distribution system, the
system mean would tend to underestimate the PCE concentration of people
served by wells with higher PCE concentrations. Exposure may be higher or
lower than estimated if data from multiple entry points for water with different
PCE levels are averaged to estimate levels for the PWS.
Public Water Use

References

1.

2.

3.

4.
5.

ACGIH. TLVs and BEIs Based on the documentation of the threshold limit values
for chemical substances and physical agents and biological exposure indices. 2007.
Signature Publications. Cincinnati OH. p.104.
Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological
profile for tetrachloroethylene update. 1997 [online]. Available at URL:
http://www.atsdr.cdc.gov/toxprofiles/ tp18.html. 4/22/09
Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological
profile for Tetrachloroethylene update. 2000 [online]. Available at URL:
http://www.atsdr.cdc.gov/toxprofiles/ tp14.html. 4/22/09
Armstrong SR, Green LC. Chlorinated hydrocarbon solvents. Clin Occup Environ
Med 2004;4(3):481-496.
Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Wooten JV. Blood
concentrations of volatile organic compounds in a nonoccupationally exposed US

NCDM Recommendations Version 3.0
Page 148 3/19/2013

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.
17.

18.
19.

20.

21.

population and in groups with suspected exposure. Clin Chem 1994;40(7 Pt
2):1401-1404.
Barkley J, Bunch J, Bursey JT, Castillo N, Cooper SD, Davis JM, et al. Gas
chromatography mass spectrometry computer analysis of volatile halogenated
hydrocarbons in man and his environment—a multimedia environmental study.
Biomed Mass Spectrom 1980;7(4):139-147.
Begerow J, Jermann E, Keles T, Freier I, Ranft U, Dunemann L. Internal and
external tetrachloroethene exposure of persons living in differently polluted areas
of Northrhine-Westphalia (Germany). Zentralbl Hyg Umweltmed.
1996;198(5):394-406.
Brugnone F, Perbellini L, Guiliari C, Cerpelloni M, Soave M. Blood and urine
concentrations of chemical pollutants in the general population. Med Lav 1994;
8(5):370-389.
Centers for Disease Control and Prevention (CDC). Fourth National Report on
Human Exposure to Environmental Chemicals, 2009. Available at:
http://www.cdc.gov/exposurereport/pdf/FourthReport.pdf
Churchill JA, Ashley DL, Kaye WE. Recent chemical exposures and blood
volatile organic compound levels in a large population- based sample. Arch
Environ Health 2001; 56(2):157-166.
Furuki K, Ukai H, Okamoto S, Takada S, Kawai T, Miyama Y, Mitsuyoshi K, et
al. Monitoring of occupational exposure to tetrachloroethene by analysis for
unmetabolized tetrachloroethene in blood and urine in comparison with urinalysis
for trichloroacetic acid. Int Arch Occup Environ Health. 2000; 73(4):221-227.
International Agency for Research in Cancer (IARC). IARC Monographs on the
Evaluation of Carcinogenic Risks to Humans, Dry Cleaning, Some Chlorinated
Solvents and Other Industrial Chemicals Vol. 63, 1995. Available at:
http://monographs.iarc.fr/ENG/Monographs/vol63/mono63.pdf
International Programme on Chemical Safety (IPCS). Concise International
Chemical Assessment Document 68-Tetrachloroethene. 2006 [online]. Available
at URL: http:// www.inchem.org/documents/cicads/cicads/cicad68.htm. 4/22/09
Lin YS, Egeghy PP, Rappaport SM. Relationships between levels of volatile
organic compounds in air and blood from the general population. J Exp Sci
Environ Epidemiol 2008; 18:421-429.
Martin SA, Simmons MB, Ortiz-Serrano M, Kendrick C, Gallo A, Campbell J, et
al. Environmental exposure of a community to airborne trichloroethylene. Arch
Environ Occup Health 2005; 60(6):341-316.
Monster AC. Biological monitoring of chlorinated hydrocarbon solvents. J Occup
Med 1986; 28:583-588.
Monster AC, Regouin-Peeters W, Van Schijndel A, van der Tuin J. Biological
monitoring of occupational exposure to tetrachloroethene. Scand J Work Environ
Health 1983; 9:273-281.
Moran MJ, Zogorski JS, Squillace PJ. Chlorinated solvents in groundwater of the
United States. Environ Sci Technol 2007; 41:74-81.
National Toxicology Program (NTP). Report on Carcinogens, 11th ed. 2004.
[online]. Available at URL: http://ntp.niehs.nih.
gov/ntp/roc/eleventh/profiles/s066dich.pdf. 4/22/09
Popp W, Muller G, Baltes-Schmitz B, Wehner B, Vahrenholz C, Schmieding W,
et al. concentrations of tetrachloroethene in blood and trichloroacetic acid in urine
in workers and neighbours of dry-cleaning shops. Int Arch Occup Environ Health
1992; 63:393- 395.
Rowe BL, Toccalino PL, Moran MJ, Zogorski JS, Price CV. Occurrence and

NCDM Recommendations Version 3.0
Page 149 3/19/2013

potential human-health relevance of volatile organic compounds in drinking water
from domestic wells in the United States. Environ Health Perspect 2007;
115(11):1539-1546.
22. Schreiber JS, House S, Prohonic E, Smead G, Hudson C, Styk M, et al. An
investigation of indoor air contamination in residences above dry cleaners. Risk
Anal 1993; 13(3):335-344. Sexton K, Adgate JL, Church TR, Ashley DL,
Needham LL, Ramachandran, et al. Children’s exposure to volatile organic
compounds as determined by longitudinal measurements in blood. Environ Health
Perspect 2005; 113(3):342-349.
23. Sexton K, Adgate JL, Fredrickson AL, Ryan AD, Needham LL, Ashley DL. Using
biologic markers in blood to assess exposure to multiple environmental chemicals
for inner-city children 3-6 years of age. Environ Health Perspect 2006;
114(3):453-459.
24. U.S. Environmental Protection Agency (U.S. EPA). Basic Information about
Tetrachloroethylene in Drinking Water, 2010. Available at:
25. http://water.epa.gov/drink/contaminants/basicinformation/tetrachloroethylene.cfm
26. U.S. Environmental Protection Agency (U.S. EPA). The Analysis of Regulated
Contaminant Occurrence Data from public Water Systems in Support of the
Second Six-year Review of National Primary Drinking Water Regulations. EPA815-B-09-006, October 2009.
27. Wallace L, Nelson W, Ziegenfus R, Pellizzari E, Michael L, Whitmore R, et al.
The Los
Angeles TEAM Study: Personal exposures, indoor-outdoor air concentrations, and
breath concentrations of 25 volatile organic compounds. J Exp Anal Environ
Epidemiol
1991; 1(2):157-192.

NCDM Recommendations Version 3.0
Page 150 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: TRICHLOROETHENE (TRICHLOROETHYLENE) (TCE)
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
3. Yearly distribution of number of CWS by maximum TCE concentration
(cut-points: 0-1, >1-2, >2-5, >5 µg/L TCE).
4. Yearly distribution of number of CWS by mean TCE concentration (cutpoints: 0-1, >1-2, >2-5, >5 µg/L TCE).
5. Mean concentration of TCE at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
6. Yearly distribution of number of people served by CWS by maximum
TCE concentration (cut-points: 0-1, >1-2, >2-5, >5 µg/L TCE).
7. Yearly distribution of number of people served by CWS by mean TCE
concentration (cut-points: 0-1, >1-2, >2-5, >5 µg/L TCE).
TCE measures will be developed from water system attribute and water quality
data stored in state Safe Drinking Water Act (SDWA) databases such as the
Safe Drinking Water Information System (SDWIS/State). Data will be cleaned
and transformed to a standard format. Analytical results of drinking water
samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
TCE, µg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be the approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Derivation of Measures

Rationale

Trichloroethene (TCE) and Public Health
Trichloroethene (TCE) is a volatile halogenated short-chain hydrocarbon. TCE
is used primarily as an industrial degreaser, solvent, and in the synthesis of
other chemicals. In the past, it was used in dry cleaning, food processing,
household cleaners, and as a general anesthetic. TCE is produced and used in
high volumes in the U.S. and has been detected in urban and ambient air and
occasionally soils and drinking water most likely contaminated by industrial
discharge (Moran et al., 2007; Rowe et al., 2007). Because of its volatility, this
solvent does not persist in the soil or water following the discontinuation of
contamination.

NCDM Recommendations Version 3.0
Page 151 3/19/2013

Drinking or breathing high levels of TCE may cause nervous system effects,
liver and lung damage, abnormal heartbeat, coma, and possibly death (ATSDR,
2003). Inhalation is the most common exposure route for the general population
including indoor sources from paints, adhesives, and cleaning solutions.
Volatilization from contaminated water (e.g., shower water) as well as the use
of household products containing this solvent can result in higher indoor than
outdoor air concentrations (ATSDR, 1997b; Martin et al., 2005). Nearby dry
cleaning establishments, industries producing this solvent, and contaminated
waste disposal sites can also contribute to human exposure (Armstrong and
Green, 2004; ATSDR, 1997a, 1997b, and 2000; Schreiber et al., 1993; Wallace
et al., 1991). Drinking water may contribute to exposure when underground
drinking water supplies have been contaminated. Workers in industries such as
dry cleaning, aircraft maintenance, electronics manufacturing, and chemical
production may be exposed by inhalation or dermal contact. The EPA has
established drinking water standards and other environmental standards for
TCE, and the FDA regulates TCE as an indirect food additive. OSHA has
established workplace standards , and ACGIH has recommended occupational
guidelines and biological exposure indices for monitoring workers (ACGIH,
2007). Human health effects from TCE at low environmental doses or at
biomonitored levels from low environmental exposures are unknown. TCE is
well absorbed by ingestion and inhalation, and animal studies have
demonstrated that liquid forms can be dermally absorbed. Following
absorption, part of the solvent dose is excreted into expired air (ATSDR1997a;
Monster, 1986). The retained solvent can undergo hepatic metabolism. TCE is
metabolized to trichloroacetic acid and tricholoroethanol, which are eliminated
in the urine. Accidental or intentional high dose acute exposure by ingestion or
inhalation can result in loss of motor coordination, somnolence, and
unconsciousness. Inhaling high doses of TCE may also produce cardiac
arrhythmias attributed to enhanced sensitivity to catecholamines. Prolonged,
low level exposure to TCE has been associated with altered renal enzyme
excretion and liver enlargement (ATSDR, 1997a, b). Chronic occupational
exposure to TCE may be associated with mild degrees of neurological
impairments, including reaction times, verbal skills, cognitive ability and motor
function (Armstrong and Green, 2004). In animal studies, TCE induced kidney
and liver tumors; and caused lung and testicular tumors (IARC, 1995). A recent
EPA toxicological review (EPA/635/R-09/011F) characterized TCE as
carcinogenic in humans by all routes of exposure (EPA, 2011). For cancer, the
inhalation unit risk is 2 × 10-2 per ppm [4 × 10-6 per μg/m3], based on human
kidney cancer risks (Charbotel et al.; 2006) and adjusted, using human
epidemiologic data, for potential risk for non-Hodgkin lymphoma (NHL) and
liver cancer. The oral unit risk for cancer is 5 × 10-2 per mg/kg/day, resulting
from physiologically based pharmacokinetic model-based route-to-route
extrapolation of the inhalation unit risk based on the human kidney cancer risks
(Charbotel et al. 2006) and adjusted, using human epidemiologic data, for
potential risk for NHL and liver cancer. There is high confidence in these unit
NCDM Recommendations Version 3.0
Page 152 3/19/2013

risks for cancer, as they are based on good quality human data, as well as being
similar to unit risk estimates based on multiple rodent bioassays. Evidence is
sufficient to conclude that TCE operates through a mutagenic mode of action
for kidney tumors. Evidence is insufficient and TCE-specific quantitative data
are lacking on early-life susceptibility.
Additional information about TCE is available from ATSDR at:
http://www.atsdr.cdc.gov/toxpro2.html.
In an analysis of occurrence data from the EPA 6 Year Review of National
Primary Drinking Water Regulations, TCE was detected in 1,013 systems
serving 29.5 million people (EPA, 2009). Concentrations of TCE were greater
than the MCL in 195 systems serving close to 12 million people. TCE was the
fifth highest occurring regulated volatile organic chemical found based on the
percent of population served by systems with at least one sample detection
found from the 6 Year Review data (EPA, 2009).
Biomonitoring Information
Levels of halogenated solvents in blood reflect recent exposure. Blood levels of
TCE were generally not detected in the NHANES 2003-2004 subsample and
were detected infrequently in previous U.S. surveys (CDC, 2009).
Comparatively higher blood levels of tetrachloroethene and TCE have been
noted for urban and industrial residential settings than for rural settings
(Barkley et al., 1980; Begerow et al., 1996; Brugnone et al., 1994). Finding a
measurable amount of any of these solvents in blood does not mean that the
level of the solvent causes an adverse health effect. Biomonitoring studies of
blood halogenated solvents can provide physicians and public health officials
with reference values so that they can determine whether people have been
exposed to higher levels of halogenated solvents than levels found in the
general population. Biomonitoring data can also help scientists plan and
conduct research on exposure and health effects.

Use of Measure

Sources of TCE
TCE does not occur naturally in the environment. However, it has been found
in underground water sources and many surface waters as a result of the
manufacture, use, and disposal of the chemical (ATSDR, 2003).
TCE Regulation and Monitoring
The EPA has set a maximum contaminant level for TCE in drinking water of
0.005 milligrams per liter (0.005 mg/L) or 5 parts of TCE per billion parts
water. The EPA has also developed regulations for the handling and disposal of
trichloroethylene.
OSHA has set an exposure limit of 100 parts of TCE per million parts of air
(100 ppm) for an 8-hour workday, 40-hour work week (ATSDR, 2003).
These measures assist by providing data that can be used for surveillance
purposes.

NCDM Recommendations Version 3.0
Page 153 3/19/2013

Limitations of The
Measure

Data Sources
Limitations of Data
Sources

Related Indicators
References

• Distribution measures provide information on the number of CWS and the
number of people potentially exposed to TCE at different
concentrations.
• Maximum concentrations provide information on the peak potential
exposure to TCE at the state level.
• Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.
The current measures are derived for CWS only. Private wells may be another
source of population exposure to TCE. Transient non-community water
systems, which are regulated by EPA, also may be an important source of TCE
exposure. Measures do not account for the variability in sampling, numbers of
sampling repeats, and variability within systems. Concentrations in drinking
water cannot be directly converted to exposure because water consumption
varies by climate, level of physical activity, and between people (EPA 2004).
Due to errors in estimating populations, the measures may overestimate or
underestimate the number of affected people.
State grantee
Ground water systems may have multiple wells with different TCE
concentrations that serve different parts of the population. Compliance samples
are taken at each entry point to the distribution system. In systems with
separate wells serving some branches or sections of the distribution system, the
system mean would tend to underestimate the TCE concentration of people
served by wells with higher TCE concentrations. Exposure may be higher or
lower than estimated if data from multiple entry points for water with different
TCE levels are averaged to estimate levels for the PWS.
Public Water Use
1. ACGIH. TLVs and BEIs Based on the documentation of the threshold limit values
for chemical substances and physical agents and biological exposure indices. 2007.
Signature Publications. Cincinnati OH. p.104.
2. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological
profile for tetrachloroethylene update. 1997a [online]. Available at URL:
http://www.atsdr.cdc.gov/toxprofiles/ tp18.html. 4/22/09
3. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological
profile for trichloroethylene update. 1997b [online]. Available at URL:
http://www.atsdr.cdc.gov/toxprofiles/ tp19.html. 4/22/09
4. Agency for Toxic Substances and Disease Registry (ATSDR). ToxFAQs™ for
Trichloroethylene (TCE), July 2003. Available at:
http://www.atsdr.cdc.gov/toxfaqs/tf.asp?id=172&tid=30
5. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological
profile for Methylene chloride update. 2000 [online]. Available at URL:
http://www.atsdr.cdc.gov/toxprofiles/ tp14.html. 4/22/09
6. Armstrong SR, Green LC. Chlorinated hydrocarbon solvents. Clin Occup Environ
Med 2004;4(3):481-496.
7. Barkley J, Bunch J, Bursey JT, Castillo N, Cooper SD, Davis JM, et al. Gas
chromatography mass spectrometry computer analysis of volatilie halogenated
hydrocarbons in man and his environment—a multimedia environmental study.

NCDM Recommendations Version 3.0
Page 154 3/19/2013

Biomed Mass Spectrom 1980;7(4):139-147.
8. Begerow J, Jermann E, Keles T, Freier I, Ranft U, Dunemann L. Internal and
external tetrachloroethene exposure of persons living in differently polluted areas
of Northrhine-Westphalia (Germany). Zentralbl Hyg Umweltmed.
1996;198(5):394-406.
9. Brugnone F, Perbellini L, Guiliari C, Cerpelloni M, Soave M. Blood and urine
concentrations of chemical pollutants in the general population. Med Lav
1994;8(5):370-389.
10. Centers for Disease Control and Prevention (CDC). Fourth National Report on
Human Exposure to Environmental Chemicals, 2009. Available at:
http://www.cdc.gov/exposurereport/pdf/FourthReport.pdf
11. Charbotel B, Fevotte J, Hoirs M, Martin JL, Bergeret A. Case-control study on
renal cell cancer and occupational exposure to trichloroethylene. Part II:
Epidemiological aspects. Ann. Occup. Hyg. Vol. 50, No. 8, pp. 777-787, 2006.
12. International Agency for Research in Cancer (IARC). IARC Monographs on the
Evaluation of Carcinogenic Risks to Humans, Dry Cleaning, Some Chlorinated
Solvents and Other Industrial Chemicals Vol. 63, 1995. Available at:
http://monographs.iarc.fr/ENG/Monographs/vol63/mono63.pdf
13. Martin SA, Simmons MB, Ortiz-Serrano M, Kendrick C, Gallo A, Campbell J, et
al. Environmental exposure of a community to airborne trichloroethylene. Arch
Environ Occup Health 2005;60(6):341-316.
14. Monster AC. Biological monitoring of chlorinated hydrocarbon solvents. J Occup
Med 1986;28:583-588.
15. Monster AC, Regouin-Peeters W, Van Schijndel A, van der Tuin J. Biological
monitoring of occupational exposure to tetrachloroethene. Scand J Work Environ
Health 1983;9:273-281.
16. Moran MJ, Zogorski JS, Squillace PJ. Chlorinated solvents in groundwater of the
United States. Environ Sci Technol 2007;41:74-81.
17. National Toxicology Program (NTP). Report on Carcinogens, 11th ed. 2004.
[online]. Available at URL: http://ntp.niehs.nih.
gov/ntp/roc/eleventh/profiles/s066dich.pdf. 4/22/09
18. Rowe BL, Toccalino PL, Moran MJ, Zogorski JS, Price CV. Occurrence and
potential human-health relevance of volatile organic compounds in drinking water
from domestic wells in the United States. Environ Health Perspect
2007;115(11):1539-1546.
19. Schreiber JS, House S, Prohonic E, Smead G, Hudson C, Styk M, et al. An
investigation of indoor air contamination in residences above dry cleaners. Risk
Anal 1993;13(3):335-344.
20. U.S. Environmental Protection Agency (U.S. EPA). The Analysis of Regulated
Contaminant Occurrence Data from public Water Systems in Support of the
Second Six-year Review of National Primary Drinking Water Regulations. EPA815-B-09-006, October 2009.
21. U.S. Environmental Protection Agency (U.S. EPA). Toxicological Review of
Trichloroethylene (CAS No. 79-01-76) In Support of Summary Information on the
Integrated Risk Information System (IRIS), September 2011. EPA/635/R09/011F
http://www.epa.gov/IRIS/toxreviews/0199tr/0199tr.pdf.
22. Wallace L, Nelson W, Ziegenfus R, Pellizzari E, Michael L, Whitmore R, et al.
The Los Angeles TEAM Study: Personal exposures, indoor-outdoor air
concentrations, and breath concentrations of 25 volatile organic compounds. J Exp
Anal Environ Epidemiol 1991;1(2):157-192.

NCDM Recommendations Version 3.0
Page 155 3/19/2013

CONTENT DOMAIN: COMMUNITY WATER
INDICATOR: URANIUM (U)
ENVIRONMENTAL PUBLIC HEALTH TRACKING
Type of EPHT Indicator
Measures

Hazard, Exposure
Level of Contaminant in Finished Water
1. Yearly distribution of number of Community Water Systems (CWS) by
maximum Uranium concentration (cut-points: 0-5, >5-15, >15-30, >30
µg/L Uranium).
2. Yearly distribution of number of CWS by mean Uranium concentration
(cut-points: cut-points: 0-5, >5-15, >15-30, >30 µg/L Uranium).
3. Mean concentration of Uranium at CWS-level, by year.

Units

Potential Population Exposure to Contaminants in Finished Water
4. Yearly distribution of number of people served by CWS by maximum
Uranium concentration (cut-points: 0-5, >5-15, >15-30, >30 µg/L
Uranium).
5. Yearly distribution of number of people served by CWS by mean
Uranium concentration (cut-points: 0-5, >5-15, >15-30, >30 µg/L
Uranium).
Uranium measures will be developed from water system attribute and water
quality data stored in state Safe Drinking Water Act (SDWA) databases such as
the Safe Drinking Water Information System (SDWIS/State). Data will be
cleaned and transformed to a standard format. Analytical results of drinking
water samples (usually taken at entry points to the distribution system or
representative sampling points after treatment) will be used in conjunction with
information about each CWS (such as service population and latitude and
longitude of representative location of the CWS service area) to generate the
measures.
Uranium, µg/L

Geographic Scope

State and Community Water System by County

Geographic Scale

The finest detail will be approximate point location of the community water
distribution system represented by water withdrawal point, water distribution
extents, principal county served, or principal city served.

Time Period

1999 or earliest year available to most current year of data abstraction.

Time Scale

Calendar year

Rationale

Uranium (U) and Public Health
Uranium is a silver-white metal that is extremely dense and weakly radioactive.
It usually occurs as an oxide and is extracted from ores containing less than 1%
natural uranium. Natural uranium is a mixture of three isotopes: 238U (greater
than 99%), 235U (about 0.72%), and 234U (about 0.01%). Uranium has many
commercial uses, including nuclear weapons, nuclear fuel, in some ceramics,

Derivation of Measures

NCDM Recommendations Version 3.0
Page 156 3/19/2013

and as an aid in electron microscopy and photography. Depleted uranium (DU)
refers to uranium in which the proportions of 235U and 234U isotopes have
been reduced compared with the proportion in natural uranium. Since the
1990's, DU has been used by the military in armor-piercing ammunition and as
a component of protective armor for tanks. Natural and depleted uranium are
primarily chemical toxicants, with radiation playing a minor role or no role at
all (ATSDR, 2009).
Everyone is exposed to uranium in food, air, and water as part of the natural
environment. (ATSDR, 2009). Variable concentrations of uranium occur
naturally in drinking water sources. In some locations the natural
concentrations may have increased due to mining and milling of uranium. Thus,
the primary exposure sources for non-occupationally exposed persons are likely
dietary and drinking water. Populations most heavily exposed to uranium are
those employed in mining and milling operations, or in uranium enrichment
and processing activities (ATSDR, 2009). In workplaces that involve uranium
mining, milling, or processing, human exposure occurs primarily by inhaling
dust and other small particles. Exposure to DU may occur in military personnel
from retention of internal shrapnel that contains DU or exposure to dust
generated from ammunition impact.
Absorption of uranium compounds is low by all routes of exposure (i.e.,
ingestion, inhalation, and skin contact). Depending upon the specific compound
and solubility, 0.1%-6% of an ingested dose may be absorbed. Inhaled
uranium-containing particles are retained in the lungs, where limited absorption
occurs (less than 5%). After long term or repeated exposure, kidneys, liver, and
bones can accumulate uranium with the largest amounts being stored in bones
(Li et al., 2005). Uranium is eliminated in feces and urine; about 50% of the
absorbed dose is eliminated in the urine within the first 24 hours. After
exposure to soluble uranium salts, the initial half-life of uranium is about 15
days (Bhattacharyya et al., 1992), which represents distribution and excretion,
with much slower elimination from bone. After inhalation, the half-life of
insoluble uranium in the lungs is several years (Durakovic et al., 2003).
Human health effects from uranium at low environmental doses or at
biomonitored levels from low environmental exposures are unknown. Health
outcomes that may occur with uranium overexposure, based on both observed
human effects and animal studies, include non-malignant respiratory disease
(fibrosis, emphysema) and nephrotoxicity. Studies of persons with chronic
exposure to elevated uranium salts in drinking water have shown changes in
urinary biomarkers potentially associated with impaired kidney function
(Kurttio et al., 2006). IARC and NTP have no ratings for uranium human
carcinogenicity. Radiation risks from exposure to natural uranium are very low.
Alpha radiation (such as that from uranium) is classified as a human
carcinogen. However, human studies have not found elevated rates of cancer
from uranium exposure, and high-dose animal studies have not found cancer
NCDM Recommendations Version 3.0
Page 157 3/19/2013

following inhalation, oral, or dermal exposure to uranium.
Workplace air standards and guidelines for external exposure to soluble and
insoluble uranium compounds have been established by OSHA and ACGIH,
respectively. Drinking water and other environmental standards have been
established by U.S. EPA. Information about external exposure (i.e.,
environmental levels) and health effects is available from ATSDR at:
http://www.atsdr.cdc.gov/toxpro2.html.
In an analysis of occurrence data from the EPA 6 Year Review of National
Primary Drinking Water Regulations, uranium was detected in 4,101 systems
serving close to 55 million people (EPA, 2009). Concentrations of uranium
were greater than the MCL in 448 systems serving close to 8.4 million people
(EPA, 2009).
Biomonitoring Information
Levels of urinary uranium reflect recent and ongoing or accumulated exposure.
A previous nonrandom subsample from NHANES III (n = 499) (Ting et al.,
1999) and other small populations have shown urinary concentrations that are
similar to those in NHANES 1999-2000, 2001-2002, and 2003-2004 (Dang et
al.,1992; Galletti, 2003; Karpas et al.,1996; Tolmachev et al., 2006). Older
studies have demonstrated urinary uranium concentrations that are consistent
with levels in the U.S. population, in that the levels were below their respective
detection limits (Byrne et al., 1991; Hamilton et al., 1994; Komaromy-Hiller et
al., 2000). In a study of 105 persons exposed to natural uranium in well water,
urinary levels of uranium were as high as 9.55 μg/L (median 0.162 μg/L)
(Orloff et al., 2004). Eighty-five percent of those levels were above the 95th
percentile of the NHANES 1999-2000 population. The U.S. Nuclear
Regulatory Commission (NRC) has set an action level of 15 μg/L urinary
uranium to protect people who are occupationally exposed (NRC, 1978).
Finding a measurable amount of uranium in urine does not mean that the level
of uranium causes an adverse health effect. Biomonitoring studies on levels of
uranium provide physicians and public health officials with reference values so
that they can determine whether people have been exposed to higher levels of
uranium than are found in the general population. Biomonitoring data can also
help scientists plan and conduct research on exposure and health effects.

Sources of Uranium
Uranium is a naturally-occurring element found in the earth’s crust. It is
naturally abundant in rocks, soil and water. Significant concentrations of
uranium can occur in phosphate rock deposits, and in minerals such as
pitchblende and uraninite. The total amount of Uranium on earth stays virtually
the same because it has such a long half-life (4.47x109 years for U-238) (EPA,
2010).

NCDM Recommendations Version 3.0
Page 158 3/19/2013

Uranium Regulation and Monitoring
The EPA limits the amount of uranium that may be present in drinking water to
30 ug/L (EPA, 2009). A gross alpha particle activity measurement may be
substituted for the required uranium measurement provided that the measured
gross alpha particle activity does not exceed 15 pCi/l.

Use of Measure

Limitations of The
Measure

Data Sources
Limitations of Data
Sources

Monitoring frequency
Once a CWS has satisfied initial monitoring requirements (4 quarterly samples
at every entry point to the distribution system within the first quarter after
initiating the source); the required frequency for Uranium monitoring is once
every three years if the average of the initial monitoring results for the
contaminant is greater than one-half the MCL but at or below the MCL. States
may allow CWS to reduce the frequency of monitoring from once every three
years to once every six or nine years at each sampling point, if the average of
the initial monitoring results for each contaminant is below the detection limit.
If a system has a monitoring result that exceeds the MCL while on reduced
monitoring, the system must collect and analyze quarterly samples at that
sampling point until the system has results from four consecutive quarters that
are below the MCL, unless the system enters into another schedule as part of a
formal compliance agreement with the State (CFR, 2002).
These measures assist by providing data that can be used for surveillance
purposes.
• Distribution measures provide information on the number of CWS and the
number of people potentially exposed to Uranium at different
concentrations.
• Maximum concentrations provide information on the peak potential
exposure to Uranium at the state level.
• Mean concentrations at the CWS level provide information on potential
exposure at a smaller geographic scale.
The current measures are derived for CWS only. Private wells may be another
source of population exposure to Uranium. Transient non-community water
systems, which are regulated by EPA, may also be an important source of
Uranium exposure. Measures do not account for the variability in sampling,
numbers of sampling repeats, and variability within systems. Concentrations in
drinking water cannot be directly converted to exposure, because water
consumption varies by climate, level of physical activity, and between people
(EPA 2004). Due to errors in estimating populations, the measures may
overestimate or underestimate the number of affected people.
State grantee
The required monitoring frequency for Uranium is infrequent (every 3 to 6
years) and may be as intermittent as every nine years; therefore most states will
have very little data on this contaminant.
Ground water systems may have multiple wells with different Uranium
concentrations that serve different parts of the population. Compliance samples

NCDM Recommendations Version 3.0
Page 159 3/19/2013

Related Indicators
References

are taken at each entry point to the distribution system. In systems with
separate wells serving some branches or sections of the distribution system, the
system mean would tend to underestimate the Uranium concentrations of
people served by wells with higher Uranium concentrations. Exposure may be
higher or lower than estimated if data from multiple entry points for water with
different Uranium levels are averaged to estimate levels for the PWS.
Public Water Use; combined Radium-226 and -228
1. Agency for Toxic Substances and Disease Registry (ATSDR). 1999. Toxicological
Profile for uranium. Atlanta, GA: U.S. Department of Health and Human Services,
Public Health Service.
2. Bhattacharyya MH, Breitenstein BD, Metivier H, Muggenburg BA, Stradling GN,
Volf V. Guidebook for the treatment of accidental internal radionuclide contamination
of workers. In: Gerber GB, Thomas RG, eds. Radiation protection dosimetry. Vol. 41
(1). Kent (England): Nuclear Technology Publishing; 1992. pp. 1-49.
3. Byrne AR, Benedik L. Uranium content of blood, urine and hair of exposed and
non-exposed persons determined by radiochemical neutron activation analysis, with
emphasis on quality control. Sci Total Environ 1991;107:143-157.
4. Centers for Disease Control and Prevention (CDC). Third National Report on
Human Exposure to Environmental Chemicals. Atlanta (GA). 2005. 4/20/09
5. Code of Federal Regulations (CFR), 2002. Title 40 Protection of the Environment
Chapter I--Environmental Protection Agency Part 141--National Primary Drinking
Water Regulations 141.26 Monitoring frequency and compliance requirements for
radionuclides in community water systems. Available at: URL:
http://www.access.gpo.gov/nara/cfr/waisidx_02/40cfr141_02.html
6. Dang HS, Pullat VR, Pillai KC. Determining the normal concentration of uranium in
urine and application of the data to its biokinetics. Health Phys 1992;62:562-566.
7. Durakovic A, Horan P, Dietz LA, Zimmerman I. Estimate of the time zero lung
burden of depleted uranium in Persian Gulf War veterans by the 24-hour urinary
excretion and exponential decay analysis. Mil Med 2003;168(8):600-605.
8. Ejnik JW, Carmichael AJ, Hamilton MM, McDiarmid M, Squibb K, Boyd P, et al.
Determination of the isotopic composition of uranium in urine by inductively coupled
plasma mass spectrometry. Health Phys 2000;78:143-146.
9. Galletti M, D'Annibale L, Pinto V, Cremisini C. Uranium daily intake and urinary
excretion: a preliminary study in Italy. Health Phys 2003;85:228-235.
10. Gwiazda RH, Squibb K, McDiarmid M, Smith D. Detection of depleted uranium in
urine of veterans from the 1991 Gulf War. Health Phys 2004;86:12-18.
11. Hamilton EI, Sabbioni E, Van der Venne MT. Element reference values in tissues
from inhabitants of the European community. VI. Review of elements in blood, plasma
and urine and a critical evaluation of reference values for the United Kingdom
population. Sci Total Environ 1994;158:165-190.
12. Karpas Z, Halicz L, Roiz J, Marko R, Katorza E, Lorber A, et al. Inductively
coupled plasma mass spectrometry as a simple, rapid, and inexpensive method for
determination of uranium in urine and fresh water: comparison with LIF. Health Phys
1996;71(6):879-885.
13. Komaromy-Hiller G, Ash KO, Costa R, Howerton K. Comparison of
representative ranges based on U.S. patient population and literature reference
intervals for urinary trace elements. Clin Chim Acta 2000;296(1-2):71-90.
14. Kurttio P, Auvinen A, Salonen L, Saha H, Pekkanen J, Makelainen I, et al. Renal
effects of uranium in drinking water. Environ Health Perspect 2002;110(4):337-342.

NCDM Recommendations Version 3.0
Page 160 3/19/2013

15. Kurttio P, Harmionen A, Saha H, Salonen L, Karpas Z, Komulainen H, Auvinen
A. Kidney toxicity of ingested uranium from drinking water. Am J Kidney Dis
2006;47(6):972-982.
16. Li WB, Roth P, Wahl W, Oeh U, Hollriegl V, Paretzke HG. Biokinetic modeling
of uranium in man after injection and ingestion. Radiat Environ Biophys 2005;44:2940.
17. May LM, Heller J, Kalinsky V, Ejnik J, Cordero S, Oberbroekling KJ, et al.
Military deployment human exposure assessment: urine total and isotopic uranium
sampling results. J Toxicol Environ Health A 2004;67(8-10):697-714.
18. Orloff KG, Mistry K, Charp P, Metcalf S, Marino R, Shelly T, et al. Human
exposure to uranium in groundwater. Environ Res 2004;94:319-326.
19. Ough EA, Lewis BM, Andrews WS, Bennett LG, Hancock RG, Scott K. An
examination of uranium levels in Canadian forces personnel who served in the Gulf
War and Kosovo. Health Phys 2002;82(4): 527-532.
20. Ting BG, Paschal DC, Jarrett JM, Pirkle JL, Jackson RJ, Sampson EJ, et al.
Uranium and thorium in urine of United States residents: reference range
concentrations. Environ Res 1999;81:45-51.
21. Tolmachev S, Kuwabara J, Noguchi H. concentration and daily excretion of
uranium in urine of Japanese. Health Phys 2006;91(2):144-153.
22. U.S. Nuclear Regulatory Commission ( NRC). Regulatory Guide 8.22-Bioassay at
uranium mills. Washington (DC): NRC; July 1978.
23. U.S. Environmental Protection Agency . Radiation Protection, Uranium, 2010.
Available at: http://www.epa.gov/radiation/radionuclides/uranium.html
24. U.S. Environmental Protection Agency. The Analysis of Regulated Contaminant
Occurrence Data from public Water Systems in Support of the Second Six-year
Review of National Primary Drinking Water Regulations. EPA-815-B-09-006,
October 2009.

NCDM Recommendations Version 3.0
Page 161 3/19/2013

CONTENT DOMAIN: REPRODUCTIVE HEALTH OUTCOMES
INDICATOR: PREMATURITY
Type Of
EPHT
Indicator
Measure

Health Outcome

1. Percent of preterm (less than 37 weeks gestation) live singleton births
2. Percent of very preterm (less than 32 weeks gestation) live singleton births
Derivation 1. Number of live singleton births before 37 weeks of gestation to resident mothers,
divided by total number of live singleton births to resident mothers
of Measure
2. Number of live singleton births before 32 weeks of gestation to resident mothers,
divided by total number of live singleton births to resident mothers
1. Preterm live singleton births
Unit
2. Very preterm live singleton births
Geographic State and national
Scope
Geographic State and County
Scale
2000-current
Time
Period
Time Scale Preterm: Annual
Very Preterm: 5 yr annual average

NCDM Recommendations Version 3.0
Page 162 3/19/2013

Rationale

Preterm birth (at less than 37 completed weeks of gestation and among all births regardless
of plurality) affects more than 500,000, or 12.5%, of live births in the United States and is a
leading cause of infant mortality and morbidity (8, 9, 13). Of those births, the majority
(about 84%) of premature babies are born moderately preterm (between 32 and 36
completed weeks of gestation). The remaining 16% of those are born very preterm (at less
than 32 weeks of gestation), representing more than 80,000, or 2%, of live births in the
United States. Of those infants born very preterm, about 63% are born between 28–31
weeks of gestation, and about 37% are born at less than 28 weeks of gestation.
The preterm birth rate rose 18% between 1990 and 2004 (from 10.6% in 1990 to 12.5% in
2004) and more than 30% since 1981 (from 9.4%) (9). For 2003–2004, increases were seen
among both moderately preterm and very preterm births. The percentage of infants born
very preterm increased from 1.92% to 2.01% between 1990 and 2004 (9); it also increased
between 2003 and 2004 from 1.97% to 2.01%, respectively.
Preterm birth rates are higher among black mothers compared to Hispanic and white
mothers. Between 2002 and 2003, the rates increased for the three largest race and ethnic
groups: non-Hispanic white (11.0 to 11.3%), non-Hispanic black (17.7 to 17.8%), and
Hispanic (11.6 to 11.9 %) (9). Since 1990, preterm birth rates have risen by one-third
(about 33%) for non-Hispanic white births (from 8.5%) and by 8% for Hispanic births
(11.0%). In contrast, preterm rates among non-Hispanic black infants have declined
slightly over this period (from 11.9%). However, the preterm birth risk of non-Hispanic
blacks continues to be substantially higher that the risk of other race and ethnic groups. Of
particular concern is the very preterm rate, about twice as high among non-Hispanic black
infants compared to non-Hispanic white and Hispanic births (3.99% compared to 1.6% and
1.73%, respectively).
Preterm birth is a leading cause of infant mortality, morbidity, and long-term disability (8,
9, 13, 14). All infants born preterm are at risk for serious health problems; however, those
born earliest are at greater risk of medical complications, long-term disabilities, and death.
Studies have shown that infants born prematurely, especially those with VLBW, have an
increased risk for neurological problems ranging from attention deficit hyperactivity
disorder to cerebral palsy or mental retardation compared with infants born at term
gestation (1, 6, 8, 14). Preterm birth is associated with nearly half of all congenital
neurological defects such as cerebral palsy (9); it is also associated with congenital
gastrointestinal defects such as gastroschisis.
Preterm infants are at greater risk for serious health problems for several reasons: the earlier
an infant is born, the less it will weigh, the less developed its organs will be, and the more
medical complications it will likely face later in life. Very preterm infants have the greatest
risk of death and lasting disabilities, including mental retardation, cerebral palsy,
respiratory (premature lung) and gastrointestinal problems (including birth defects such as
gastroschisis), and vision and hearing loss. Preterm births account for health care
expenditure of more than $3 billion per year (14).

NCDM Recommendations Version 3.0
Page 163 3/19/2013

Studies have shown that major risk factors associated with preterm birth include (2, 4, 7, 8,
10, 14):
1. Plural births
2. Previous preterm birth
3. Certain uterine or cervical abnormalities of the mother
4. Mother’s age, race, poverty (for example, black women, women younger than 17 and
older than 35 years, and poor women are at greater risk than other women)
5. Male fetal gender (associated with singleton preterm birth)
6. Certain lifestyles and environmental factors, including:
o
Late or no prenatal care,
o
Maternal smoking, alcohol consumption (especially in early pregnancy), illegal
drug use, exposure to the medication diethylstilbestrol (DES), domestic violence,
lack of social support, stress, long working hours with long periods of standing,
being underweight before pregnancy, obesity, marital status, and spacing (less
than 6–9 months between giving birth and the beginning of the next pregnancy),
o
Neighborhood-level characteristics,
o
Environmental contaminants (e.g., exposure to air pollution and drinking water
contaminated with chemical DBP or lead).
Certain medical conditions during pregnancy (e.g., infections, diabetes, hypertension, blood
clotting disorders/thrombophilia, vaginal bleeding, certain birth defects of the fetus) may
also increase the risk of preterm birth.
The strength of the association of each of these risk factors with preterm birth varies, and
remains a subject of significant debate in the literature (14).
The rise in the occurrence of multiple/plural births, which are much more likely than
singleton births to be preterm, influenced the overall preterm birth rate over the past two
decades. However, preterm rates for singleton births have also increased, up to 11% since
1990 (9). This increase in singleton preterm births was only in infants born moderately
preterm; the singleton very preterm birth rate declined slightly, from 1.69% in 1990 to
1.61% in 2004.
Preterm births are associated with many modifiable risk factors, and prevention of preterm
births may greatly contribute to the overall reduction in infant illness, disability, and death.
Several studies are being conducted to improve our understanding of the precise causes of
preterm births, especially those with VLBW, and to learn how to prevent them. These
studies look at how genes, maternal stress, race, occupational and environmental factors,
and infections may contribute to preterm birth (8). Better understanding of the specific
causes of preterm births is needed before tailored interventions can be developed.
Neighborhood-level characteristics have proven to be useful predictors of preterm birth
risks (10). Neighborhoods are the geographic units where interventions can be targeted, and
those interventions can be an effective way to reduce preterm birth rates and other adverse
birth outcomes. Neighborhood-level characteristics contributing to prematurity include the
social, economic, and environmental risk factors such as certain aspects of the built
NCDM Recommendations Version 3.0
Page 164 3/19/2013

environment.
Preterm births data are readily available in all state health departments and can be used to
examine trends. These trends may reflect the contributions of environmental exposures and
other modifiable risks to preterm births. These trends can also be used to evaluate the
effectiveness of existing and new prevention programs.
“Live birth means the complete expulsion or extraction from its mother of a product of
human conception, irrespective of the duration of pregnancy, which, after such expulsion or
extraction, breathes, or shows any other evidence of life, such as beating of the heart,
pulsation of the umbilical cord, or definite movement of voluntary muscles, whether or not
the umbilical cord has been cut or the placenta is attached. Heartbeats are to be
distinguished from transient cardiac contractions; respirations are to be distinguished from
fleeting respiratory efforts or gasps.” All states require the reporting of live births
regardless of length of gestation or birth weight (3).

These measures can be utilized to enhance public health prevention actions and
interventions, and inform policy makers and the public regarding risk factors management
and mitigation.
Limitations Uncertainties associated with gestational age estimates:
The interval between the first day of the mother’s last normal menstrual period (LMP) and
Of The
the day of birth is one method used to determine the gestational age of the newborn.
Measure
However, this measurement is subject to error for many reasons, including imperfect
maternal recall or misidentification of the LMP due to postconception bleeding, delayed
ovulation, or intervening early miscarriage (9). Thus, for the purpose of calculating national
statistics of preterm births, these data are being edited for gestational ages that are clearly
inconsistent with the infant’s plurality and birth weight, but substantial inconsistencies in
the data still persist (9).
Use Of The
Measure

The National Center for Health Statistics (NCHS) and most state vital records offices report
gestational age based on an algorithm that uses both the mother’s reported last normal
menses and the clinician’s estimate of gestational age. The LMP indicator is used unless its
value appears to be inconsistent with birthweight, falls outside likely parameters, or was not
reported. If any of these circumstances exist, the clinical estimate is used. Nationwide in
2004, approximately 5.9% of gestational age values were based on the clinical estimate (9).
Changes in reporting of the gestational age over time may affect trends in preterm birth
rates, especially by race (9). These reporting problems may occur more frequently among
some subpopulations and among births with shorter gestations.
NCDM Recommendations Version 3.0
Page 165 3/19/2013

Difficulties of interpreting preterm and very preterm birth rates:
The preterm birth rates might be an indicator of pregnancy outcome that does not
necessarily predict the true health risk associated with early birth. Preterm rates based on
live singleton births may be affected by maternal characteristics; a low preterm birth rate
might indicate a low-risk population, and a high preterm birth rate might indicate maternal
characteristics that predispose to preterm birth.
Data
Sources

Birth certificate data from Vital Statistics state systems (both numerator and denominator);
National Vital Statistics System (NVSS), CDC, NCHS
http://www.cdc.gov/nchs/VitalStats.htm;
CDC Wonder: Natality Data Request, CDC http://wonder.cdc.gov/natality.html
CDC GIS Reproductive Health Atlas: http://cdc.gov/reproductivehealth/gisatlas/index.htm

Limitations Vital statistics data are readily available, of high quality, and useful for various purposes,
including public health surveillance; however, they cannot be correctly interpreted unless
Of Data
various qualifying factors and classification methods are considered (see “Limitations of
Sources
the Measure”). The factors to be considered will vary depending on the intended use of the
data; however, most of the limiting factors result from imperfections in the original records,
and they should not be ignored. Yet, their existence does not lessen the value of the data
for calculating/estimating this measure.
One important limitation of the national data is the timeliness of when the data are
available. The national file cannot be compiled until all states have submitted their data.
Often times there is delay of 2‐3 years before national statistics are available. There are also
some differences between national data and state data handling of unknowns, imputation
rules, and close out dates. There may be differences or delays in processing resident births
that occur out of state. These process issues, along with the need to close off national
statistics at specified intervals following a reporting period, may lead to small discrepancies
between national data compiled by NCHS and data maintained by state vital statistics
registries.
Related
Indicators
References

Low birthweight
1. Ananth C. W., Joseph K. S., Oyelese Y., Demissie K., Vintzileos A. M. Trends in
Preterm Birth and Perinatal Mortality Among Singletons: United States, 1989
through 2000. Obstet Gynecol,2005, Vo. 105, No. 5, 1084-1091
2. Blackmore C. A. and Rowley D. L. 1994. Preterm Birth. Editors: Wilcox L.S. and
Marks J S. In: From Data to Action CDC’s Public Health Surveillance for Women,
Infants, and Children. CDC’s Maternal & Child Health Monograph 1994. Centers
for Disease Control and Prevention, Atlanta Georgia

NCDM Recommendations Version 3.0
Page 166 3/19/2013

3. Centers for Disease Control and Prevention, National Center for Health Statistics
(NCHS), NCHS Definitions. Available from:
http://www.cdc.gov/nchs/datawh/nchsdefs/list.htm Last accessed: June 19, 2007
4. Cooperstock M. and Campbell J. Excess Males in Preterm Birth: Interactions with
Gestational Age, Race, and Multiple Birth. Obstet Gynecol 1996, 88: 189-193
5. Hamilton B.E., Martin J.A., and Ventura S.J.: Births: Preliminary Data for 2005.
Health E-Stats. Released November 21, 2006. Available from:
http://www.cdc.gov/nchs/products/pubs/pubd/hestats/prelimbirths05/prelimbirths05.
htm
6. Healthy People 2010, Volume 2, Objective 16: Maternal, Infant, and Child Health.
http://www.healthypeople.gov/Document/HTML/Volume2/16MICH.htm
7. Joseph K. S., Allen A. C., Dodds L., Vincer M. J., and Armson B. A. Causes and
Consequences of Recent Increases in Preterm Birth Among Twins. Obstet Gynecol
2001, Vol. 98, No. 1, 57-64
8. March of Dimes, 2007. Available from: http://www.marchofdimes.com
9. Martin J.A., Hamilton B. E., Sutton P., Ventura S. J., Menacker F., and Kimeyer S.
Births: Final Data for 2004. National Vital Statistics Reports, Vol. 55, No 1.
Hyattsville, MD: National Center for Health Statistics. 2006. Available from:
http://www.cdc.gov/nchs/data/nvsr/nvsr55/nvsr55_01.pdf
10. Messer L. C., Buescher P. A., Laraia B. A., and Kaufman J. S. Neighborhood-Level
Characteristics as Predictors of Preterm Birth: Examples from Wake County, North
Carolina. SCHS Studies, No. 148, November 2005. North Carolina Public Health,
State Center for Health Statistics, Raleigh, NC. Available from:
http://www.schs.state.nc.us/SCHS/
11. Monaghan S. C., Little R. E., Hulchiy O., Strassner H., and Gladen B. C. Preterm
Birth in Two Urban Areas of Ukraine. Obstet Gynecol 2000, Vol. 95, No. 5, 752755et al., 2000
12. Petrini J. R., Callagham W. M., Klebanoff M., Green N. S., Lackritz E. M., Howse
J. L., Schwartz R. H., and Damus K. Estimated Effect of 17 AlphaHydroxyprogesterone Caproate on Preterm Birth in the United States. Obstet
Gynecol 2005, 105: 267-272
13. Tan H, Wen S. W., Mark W., Fung K. F. K., Demissie K., and Rhoads G. G. The
Association Between Fetal Sex and Preterm Birth in Twin Pregnancies. Obstet
Gynecol 2004, Vol. 103, No. 2, 327-332
14. Weismiller D. G. Preterm Labor. American Family Physician, 1999, Vol. 59, No. 3
NCDM Recommendations Version 3.0
Page 167 3/19/2013

CONTENT DOMAIN: REPRODUCTIVE HEALTH OUTCOMES
INDICATOR: LOW BIRTHWEIGHT
Type Of
EPHT
Indicator
Measure

Health Outcome

Percent of low birthweight (less than 2500 grams) live term singleton births
2. Percent of very low birthweight (less than 1500 grams) live singleton births
Derivation Number of singleton infants live born at term (at or above 37 completed weeks of
of Measure gestation) with a birthweight of less than 2,500 grams, divided by the total number of
singleton infants live born at term to resident mothers
Number of live singleton births with a birthweight of less than 1,500 grams, divided by
total number of live singleton births to resident mothers
LBW: live singleton term births
Unit
VLBW: live singleton births
Geographic State and national
Scope
Geographic State and County
Scale
2000-current
Time
Period
Time Scale Low birthweight: Annual
Very low birthweight: 5 yr annual average
LBW, a weight of less than 2,500 grams, or 5 pounds, 8 ounces, at birth (regardless of
Rationale
gestational age and plurality), affects about 1 of every 13 babies born each year in the
United States (7). Studies have shown that LBW is an important predictor of future
morbidity and mortality. Note however, that the percent of LWB babies among all births (a
percentage that is confounded by gestational age and plurality) is not recommended as a
population-level measure of perinatal morbidity and mortality (1, 11). It is not
recommended as a measure because preterm delivery, decreased fetal growth, and
genetically determined small body size commonly occur in LBW infants (1). Compared to
infants of normal weight, LBW infants may be at increased risk of perinatal morbidity,
infections, and the longer-term consequences of impaired development such as delayed
motor and social development or learning disabilities. Mortality risk is lowest for infants
born weighing 3,500–4,500 grams (8).
1.

Nationally, the percentage of LBW infants (regardless of gestational age and plurality) has
been increasing steadily; it reached 8.2% of all births in 2005, the highest level reported
since 1968 (4). The 2005 rate was 17% higher than the 1970 (7%) rate, which was 22%
higher than the 1984 low (6.7%). In addition, this rate is 64% higher than the Healthy
People 2010 goal of 5% (5). The percentage of LBW births also increased among singleton
births, from 5.9% in 1990 to 6.31% in 2004 (7% increase).
Increases in the multiple birth rate, obstetric interventions (e.g., induction of labor and
NCDM Recommendations Version 3.0
Page 168 3/19/2013

cesarean delivery), older maternal age at childbearing, and increased use of infertility
therapies likely have affected the trends toward lower birthweights (8). Environmental
exposures have also been implicated as possible risk factors for LBW, but the magnitude of
the contribution to these increased rates remains relatively uncertain. The percentage of
LBW increased among each of the largest racial and ethnic groups: non-Hispanic whites
(from 7.0% in 2003 to 7.2% in 2004), non-Hispanic blacks (from 13.6% in 2003 to 13.7%
in 2004), and Hispanics (from 6.7% in 2003 to 6.8% in 2004) (8).
LBW in singleton births rose between 2003 and 2004 among non-Hispanic white and
Hispanic infants; the increase for non-Hispanic black infants was not statistically significant
(8). Since 1990, singleton LBW rates have risen 8% and 14% for Hispanic and nonHispanic white infants, respectively; the rates have declined 2% among non-Hispanic black
infants.
The youngest and oldest mothers are the most likely to deliver LBW infants. In 2004, the
lowest LBW levels were reported for women aged 25–34 years (7.3% for women aged 25–
29 years and 7.5% for women 30–34 year old); the highest LBW levels were for teenagers
younger than 15 years (13.6%) and women aged 45–54 years (21.2%) (8). However, much
of the elevated LBW risk among older mothers can be attributed to their higher multiple
birth rates; in fact, the LBW rate declined from 21% to 10% for the oldest mothers of
singleton births.
LBW rates also vary widely between states or reporting areas (8). In 2004, more than 10%
of all infants born in Alabama, Louisiana, Mississippi, South Carolina, and the District of
Columbia were LBW., This compares with less than 6.5% of newborns in Alaska, Maine,
Oregon, Vermont, and Washington that were LBW. Different demographic characteristics
of these populations, including maternal age, race, or ethnicity, may explain some of these
differences.
Infants weighing less than 1,500 grams, or 3 pounds, 4 ounces, at birth are considered
VLBW (3); most of them are also premature (born before 37 weeks gestation). (Note that
the percent of VLBW babies among all births is also confounded by plurality; therefore, the
percent of VLBW births among singleton births is recommended as a population-level
measure of prematurity.) Studies have shown that the infant’s birthweight is a predictor of
future morbidity and mortality (8), especially for VLBW infants. VLBW infants have about
a 25% chance of dying in the first year of life; this risk is estimated to be about 100 times
higher for VLBW infants than for normal-weight infants (≥2,500grams) (8). VLBW infants
have an increased risk for developing neurological and intellectual problems (including
attention deficit hyperactivity disorder, cerebral palsy, developmental delay and mental
retardation), visual problems (including blindness), hearing loss, infections, and chronic
lung diseases compared with infants of normal weight or infants born at term gestation (2,
5, 6, 7).
Nationally, the percentage of VLBW infants (regardless of plurality) increased slightly
from 1.45% in 2003 to 1.49% in 2005, and has increased from 1.27% in 1990 (5). The
2005 rate is 66% higher than the Healthy People 2010 goal of 0.9% (5). The VLBW has
NCDM Recommendations Version 3.0
Page 169 3/19/2013

increased since 1990 among whites, blacks, Puerto Ricans, American Indians, and other
population groups (5). For 2004–2005, increases in VLBW rates were statistically
significant for non-Hispanic black infants but not for non-Hispanic white infants (8).
The increase in the rate of multiple births, in which the infants tend to be much smaller than
in singleton births, has likely affected the upward trend in the VLBW rate (8). However, the
VLBW rate among singleton births also increased slightly from 1.12% in 2004 to 1.14% in
2005 (8).
Increases in obstetric interventions (e.g., induction of labor and cesarean delivery), teenage
pregnancy, and older maternal age at childbearing likely contributed to the increased
VLBW rates. Teen mothers, especially those younger than aged 15 years, have a higher
chance of giving birth to a VLBW infant. Environmental exposures, including exposure to
air pollution, drinking water contaminated with chemical DBP, and exposure to pesticides,
have also been implicated as possible risk factors for VLBW, but the exact magnitude of
the contribution to the increased VLBW rates remains relatively uncertain
Birthweight is a multifactorial and heterogeneous birth outcome. Birthweight of an infant
is directly related to its gestational age. As noted above, multiple births are usually LBW,
even those delivered at term. Therefore, the focus of the measure is restricted to singleton
term births. As such, the measure distinguishes between preterm and multiple birth
categories and decreased fetal growth that may be affected by other risk factors, including
environmental factors.
LBW rate is associated with many modifiable risk factors, and preventing LBW may
contribute to the overall reduction in infant illness, disability, and death. Several studies are
being conducted that may help understand the biological, social, and environmental factors
that contribute to LBW births and learn how to prevent them. These studies look at how
genes, hormonal changes, maternal stress, race, occupational and environmental factors,
and infections may contribute to prematurity and LBW (7). Specific causes of LBW births
must be better understood before tailored interventions can be developed.
Neighborhood-level characteristics have proven to be useful predictors of LBW risks (9).
Neighborhoods are the geographic units where interventions can be targeted, and those
interventions can be an effective ways to reduce LBW rates, infant mortality, and other
adverse birth outcomes. Neighborhood-level characteristics contributing to LBW include
social, economic, and environmental risk factors, such as certain aspects of the built
environment.
The percentage of LBW among term singleton births is a useful and feasible measure of
perinatal health. LBW, gestational age, and plurality data are readily available in all state
health departments, and can be used to examine trends that occur over time and space.
These trends may reflect the contributions of environmental exposures and other modifiable
risk factors for LBW.
Exposure to air pollution (both indoor and outdoor) and drinking water contaminated with
NCDM Recommendations Version 3.0
Page 170 3/19/2013

chemical DBPs or lead may serve as examples of environmental risk factors. Maternal
smoking, alcohol consumption, or inadequate weight gain are associated with an increased
risk of intrauterine growth retardation and LBW. Socioeconomic factors, including low
income and lack of education, are reported as risk factors for LBW (10).
Women younger than 15 years or older than 35 years, unmarried mothers, and women who
have had previous preterm birth are at increased risk of having LBW babies. Women who
experience excessive stress, domestic violence, or other abuse also may be at increased risk
of having a LBW baby (7).
“Live birth means the complete expulsion or extraction from its mother of a product of
human conception, irrespective of the duration of pregnancy, which, after such expulsion or
extraction, breathes, or shows any other evidence of life, such as beating of the heart,
pulsation of the umbilical cord, or definite movement of voluntary muscles, whether or not
the umbilical cord has been cut or the placenta is attached. Heartbeats are to be
distinguished from transient cardiac contractions; respirations are to be distinguished from
fleeting respiratory efforts or gasps.” All states require the reporting of live births,
regardless of length of gestation or birth weight (3).
Birthweight is the first weight of the newborn obtained after birth (3).
Low birthweight is defined as less than 2,500 grams or 5 pounds, 8 ounces (3). Before
1979, low birthweight was defined as 2,500 grams or less.
Very low birthweight is defined as less than 1,500 grams or 3 pounds, 4 ounces (3). Before
1979, very low birthweight was defined as 1,500 grams or less.
Term birth is defined here as the birth at or above 37 completed weeks of gestation.
Use Of The
Measure

This indicator can be used to influence public health prevention actions and interventions
and policy makers and inform the public regarding risk factors management and mitigation.
The LBW measure can be used to track the perinatal health in states, regions, counties, and
smaller geographic areas or communities, as needed. Baseline data can be used to monitor
changes or trends.

This measure can also be used to evaluate the effectiveness of existing and new prevention
programs.
Limitations Difficulties of interpreting LBW birth rates among term singleton births:
Using LBW rates alone as a pregnancy outcome measure might not inform the user about
Of The
the true health risk associated with LBW.
Measure
Difficulties of interpreting VLBW birth rates:
Although the percentage of VLBW births has increased during the past 20 years, in large
part this could be due to improvements in fetal health. Conditions that may have resulted in
a fetal death decades ago might today result in fetal survival and a live VLBW birth (6).
NCDM Recommendations Version 3.0
Page 171 3/19/2013

Data
Sources

Recommendations:
LBW rates should be interpreted with caution. The LBW rate should be only one of the
reproductive outcome measures being tracked, and it should be accompanied by the infant
mortality rate (neonatal and postneonatal), fetal death rate if reliable, and morbidity
measures. If feasible, an infant’s anthropometric parameters should also be monitored; this
could include a reduced head circumference measure because smaller head size may predict
lower IQ and cognitive abilities and may be associated with ADD/ADHD.
Birth certificate data from Vital Statistics state systems (both numerator and denominator)
National Vital Statistics System (NVSS), CDC, NCHS;
CDC Wonder: Natality Data Request, CDC http://wonder.cdc.gov/natality.html

CDC GIS Reproductive Health Atlas: http://cdc.gov/reproductivehealth/gisatlas/index.htm
Limitations Although vital statistics data are readily available, of high quality, and otherwise useful for
various purposes, including public health surveillance, they cannot be correctly interpreted
Of Data
unless various qualifying factors and classification methods are considered (see also
Sources
“Limitations of the Measure”). The factors to be considered will vary, depending of the
intended use of the data; however, most of the limiting factors result from imperfections in
the original records, and they should not be ignored. Yet, their existence does not lessen
the value of the data for the purpose of calculating this measure. At the minimum, the
following data quality attributes should be evaluated: completeness of registration,
reporting and quality control procedures, and records geocoding procedures and quality.
One important limitation of the national data is the timeliness of when the data are
available. The national file cannot be compiled until all states have submitted their data.
Often times there is delay of 2‐3 years before national statistics are available. There are also
some differences between national data and state data handling of unknowns, imputation
rules, and close out dates. There may be differences or delays in processing resident births
that occur out of state. These process issues, along with the need to close off national
statistics at specified intervals following a reporting period, may lead to small discrepancies
between national data compiled by NCHS and data maintained by state vital statistics
registries.
Related
Indicators
References

Prematurity
1. Adams M., Andersen A-M. N., Andersen P. K., Haig D., Henriksen T. B., HertzPicciotto I., Lie R. T., Olsen J., Skjerven R., and Wilcox A. Sostrup Statement on
Low Birthweight. Int J Epidemiol 2003, 32: 884-885
2. Ananth C. W., Joseph K. S., Oyelese Y., Demissie K., Vintzileos A. M. Trends in
Preterm Birth and Perinatal Mortality Among Singletons: United States, 1989
through 2000. Obstet Gynecol,2005, Vo. 105, No. 5, 1084-1091
3. Centers for Disease Control and Prevention, National Center for Health Statistics
(NCHS), NCHS Definitions. Available from:

NCDM Recommendations Version 3.0
Page 172 3/19/2013

http://www.cdc.gov/nchs/datawh/nchsdefs/list.htm Last accessed: June 19, 2007
4. Hamilton B.E., Martin J.A., and Ventura S.J.: Births: Preliminary Data for 2005.
Health E-Stats. Released November 21, 2006. Available from:
http://www.cdc.gov/nchs/products/pubs/pubd/hestats/prelimbirths05/prelimbirths05.
htm
5. Healthy People 2010, Volume 2, Objective 16: Maternal, Infant, and Child Health.
http://www.healthypeople.gov/Document/HTML/Volume2/16MICH.htm
6. Kitchen WH, Permezel MJ, Doyle LW, Ford GW, Rickards AL, and Kelly EA.
Changing obstetric practice and 2-year outcome of the fetus of birth weight under
1000 g. Obstet Gynecol. 1992 Feb;79(2):268-75.
7. March of Dimes, 2007. Available from: http://www.marchofdimes.com
8. Martin J.A., Hamilton B. E., Sutton P., Ventura S. J., Menacker F., and Kimeyer S.
Births: Final Data for 2004. National Vital Statistics Reports, Vol. 55, No 1.
Hyattsville, MD: National Center for Health Statistics. 2006. Available from:
http://www.cdc.gov/nchs/data/nvsr/nvsr55/nvsr55_01.pdf
9. Messer L. C., Buescher P. A., Laraia B. A., and Kaufman J. S. Neighborhood-Level
Characteristics as Predictors of Preterm Birth: Examples from Wake County, North
Carolina. SCHS Studies, No. 148, November 2005. North Carolina Public Health,
State Center for Health Statistics, Raleigh, NC. Available from:
http://www.schs.state.nc.us/SCHS/
10. Preliminary Results from the Central Valley/South Coast Children’s Environmental
Health Demonstration Project – Supplement to July 2006 Report, California
Environmental Health Tracking Program, 2006. Available from:
http://www.catracking.com/cvsc
11. Wilcox A. J. On the Importance – and the Unimportance – of Birthweight. Int J
Epidemiol 2001, 30: 1233-1241

NCDM Recommendations Version 3.0
Page 173 3/19/2013

CONTENT DOMAIN: REPRODUCTIVE HEALTH OUTCOMES
INDICATOR: MORTALITY (USING PERIOD LINKED
BIRTH/INFANT DEATH APPROACH)
Type of EPHT Indicator
Measures

Health Outcome

Derivation of Measures

1. Infants: Number of deaths occurring in infant residents under 1
year of age (under 366 days during a leap year) in a given year
divided by the number of live births in the same year.
2. Neonates: Number of deaths occurring in infant residents less than
28 days of age in a given year divided by the number of live births
in the same year
3. Perinates: Number of fetal deaths in infant residents greater than
or equal to 28 weeks gestation plus infant deaths less than 7 days
old in a given year divided by the number of live births plus fetal
deaths at greater than or equal to 28 weeks gestation in the same
year
4. Postneonates: Number of deaths occurring in infant residents at 28
days to less than1 year of age (under 366 days during a leap year)
in a given year divided by the number of live births in the same
year

1. Average Infant (less than 1 year of age) Mortality Rate per 1000 live
births
2. Average Neonatal (less than 28 days of age) Mortality Rate per 1000 live
births
3. Average Perinatal (equal to or greater than 28 weeks gestation to less
than 7 days of age) Mortality Rate per 1000 live births (plus fetal deaths
equal to or greater than 28 weeks gestation)
4. Average Postneonatal (equal to or greater than 28 days to less than 1 year
of age) Mortality Rate per 1000 live births

Both birth and death counts are geographically classified based on
maternal residence at the time of birth.

Geographic Scope
Geographic Scale
Time Period

1. Deaths per 1,000 live births
2. Deaths per 1,000 live births
3. Deaths per 1,000 live births plus fetal deaths at 28 or greater
weeks gestation
4. Deaths per 1,000 live births
State and national
State and County
2000-current

Time Scale

Five year

Units

NCDM Recommendations Version 3.0
Page 174 3/19/2013

Rationale

Fetuses and young children may be particularly susceptible to harmful
effects of environmental contaminants. Many environmental
contaminants have been proposed to be particularly toxic in utero;
many cross the placenta and make their way into the circulatory
system of the developing fetus. However, specific health effects are
often not well understood for years Therefore, gross indicators of
childhood health—such as mortality—should be tracked as part of an
EPHT system. Furthermore, data on births and deaths in a region may
be far more complete than data on other health-related events.
Overall, congenital malformations, deformations, and chromosomal
abnormalities are the leading cause of infant deaths (20.1% of deaths)
(1). Disorders related to short gestation and LBW are second, making
up 16.6% of deaths. However, importantly, cause of death varies over
the first year of life, and combining all causes obscures the fact that
sudden infant death syndrome is the leading cause of death in the
postneonatal period.
Disorders related to short gestation and LBW are the leading cause of
neonatal death (24.3% of deaths) (1). This is in contrast to the leading
cause of postneonatal death, which is sudden infant death syndrome
(21.8%). Congenital malformations, deformations, and chromosomal
abnormalities are the second-leading cause of neonatal deaths (21.4%)
and postneonatal deaths (17.5%) (1).
Restricting infant mortality to deaths during the perinatal, neonatal, or
postneonatal period may limit the etiologic heterogeneity inherent in a
gross measure such as overall infant mortality. Also, it may be more
likely that infants who died within 7 or 28 days, respectively, were
living in reasonable proximity to where they were born, making
ecological associations with environmental exposures potentially more
meaningful. Specifically, exclusion of infants who died within 28 days
might reduce etiologic heterogeneity due to differences in early
prenatal care and other non-environmental factors likely to influence
neonatal survival.
When a fetus or an infant dies around the time of labor and delivery, it
is not always clear whether to classify this event as a live birth and
infant death, or a fetal death. Diagnostic ability for detecting signs of
life, such as breathing or beating of the heart, pulsation of the
umbilical cord, or definite movement of voluntary muscles after
expulsion or extraction from the mother may vary across obstetric
clinics.
Unexplained fetal death and death related to growth restriction are the
leading causes of fetal loss (2). Fetal death is an important contribution

NCDM Recommendations Version 3.0
Page 175 3/19/2013

Use of the Measure

Limitations of the Measure

to reproductive loss, with the rate being many times higher than the
rate of sudden infant death syndrome among infants (1). Although the
rate of late fetal loss (greater than or equal to28 weeks gestation) has
been decreasing in past decades, the rate of intermediate fetal loss (20–
27 weeks gestation) has remained relatively constant (3). Markers of
increased risk for fetal loss include pre-pregnancy obesity, lower
socioeconomic status, non-Hispanic black race, and advanced maternal
age.
Identifying populations with higher infant, neonatal, perinatal, and
postneonatal mortality rates may indicate where potential
environmental problems are. It will assist in targeting outreach
intervention activities and improve our understanding of geographic
variation, time trends, and demographic patterns of infant death.
An important limitation of this health outcome measure is the
heterogeneity in its etiology. Environmental exposure-related causes of
infant death are only one piece of a puzzle that includes many other
factors, such as access to and quality of health care, competency in
childcare, and understanding of injury prevention.
The maternal residence during pregnancy and the infant’s residence
during the first year of life are critical data for linking deaths to
environmental hazards/exposures; these residences may differ from
maternal residence at birth or infant residence at death. The mother
may have lived far from the place at which she gave birth during part
or all of the pregnancy. The infant who died may have been born and
lived for a major portion of its life far from the place of death; it may
be less likely that neonates and perinates who died were born and lived
far from the place of death.

Data Sources
Limitations of Data
Sources

NCHS currently uses a period linkage approach that links death
certificates to birth certificates. This approach would allow
stratification of deaths according to place of birth. However, it does
not address the possibility that migration across states or other
geographies occurred during pregnancy or infancy.
Local, state, or national vital statistics systems (birth, death, and fetal
death records)
It may be reasonable to assume universal reporting of live births and
infant deaths in the United States; however, some births/deaths may be
excluded because of the difficulty in distinguishing a death shortly
after birth as a live birth; a death soon after birth might be reported as
a fetal death rather than as a live birth and infant death. In addition,
some fetal deaths may be missed in some regions, although those
occurring at greater than or equal to28 weeks are less likely to be
missing.
Data on fetal death certificates may not provide all the information that

NCDM Recommendations Version 3.0
Page 176 3/19/2013

can be collected from birth certificates linked to infant deaths within 7
days; however, many variables used for environmental health tracking
(maternal race/ethnicity and age, place of residence) have relatively
complete reporting on the fetal death certificate.

References

Births and deaths will be tabulated according to maternal
race/ethnicity, using linked data from birth certificates.
1. Heron M. Deaths: Leading Causes for 2004. National Vital
Statistics Reports; vol. 56, no. 5. Hyattsville, Maryland:
National Center for Health Statistics. 2007. Available from:
http://www.cdc.gov/nchs/data/nvsr/nvsr56/nvsr56_05.pdf
2. Fretts, RC. Etiology and prevention of stillbirth. Am J Obstet
Gynecol. 193(6): 1923-35. 2005.
3. MacDorman MF, Hoyert DL, Martin JA, Munson ML,
Hamilton BE. Fetal and perinatal mortality, United States,
2003. Natl Vital Stat Rep. 2007 Feb 21;55(6):1-17.

NCDM Recommendations Version 3.0
Page 177 3/19/2013

CONTENT DOMAIN: REPRODUCTIVE HEALTH OUTCOMES
INDICATOR: FERTILITY
Type of EPHT Indicator
Measure
Derivation of Measure(s)
Unit
Geographic Scope
Geographic Scale
Time Period

Health outcome
Total Fertility Rate per 1000 women of reproductive age
TFR = sum of age-specific fertility rates * 5
Rate per 1,000 women of reproductive age
State and national
State and County
2000-current

Time Scale

Year

Rationale

The cause of approximately 10% of fertility problems is unknown, and
environmental contaminants, including endocrine disruptors, have
been considered major contributors. The case of diethylstilbestrol
revealed that environmental contamination can have multigenerational effects on reproduction that should be studied and tracked
long-term. Several indicators have been used to track fertility on a
global, national, state, and local level. Indicators most commonly used
are the general fertility rate (GFR), which is defined as the number of
live births divided by the total number of women of reproductive age
(aged 15–44 years), and the total fertility rate (TFR).
The TFR differs from the GFR in that it adjusts for age-specific
differences in fertility. It also shows the potential impact of current
fertility patterns on reproduction, allowing for more valid comparisons
of rates across time and space.
Fecundity: The physical ability of a woman or couple to conceive and
carry a child to term birth.
Fertility: The ability to conceive a child.

Use of the Measure

The TFR indicates the average number of births to a hypothetical
cohort of 1,000 women if they experienced the age-specific birth rates
observed in a given year. Understanding the geographic distribution
and trends in fertility will provide basic descriptive clues to changes
that may be influenced by environmental risk factors. As more is
learned regarding the link between adverse exposures and fertility,
these rates will provide important background information about how
fertility varies geographically in relation to changes in potentially
related environmental risk factors and how it has varied over time
within the United States. Similar to the GFR, the TFR may not be

NCDM Recommendations Version 3.0
Page 178 3/19/2013

Limitations of the Measure

Data Sources

Limitations of Data
Sources

specific enough to permit tracking of specific changes related to
environmental risk factors. However, if the estimate of 10% is correct,
this measure can be used with other measures, including ambient
concentrations of pollutants, to examine potential associations with
population-level changes in fertility and generate some well- informed
hypotheses or areas for future investigations.
The fertility measure is influenced by social/demographic choices for
reproduction, maternal age, parity, and social class measures, as well
as the use of contraception and infertility treatments leading to
multiple births. These factors all may determine variations in overall
fertility across populations and geographic locations; therefore social
and demographic factors would need to be controlled for to examine
any environmental effects on total fertility.
Numerator:
U.S. National Center for Health Statistics—Vital Statistics Reports
and/or state-specific vital statistics (for more recent years of data)
Denominator:
U.S. Census Bureau
National-level data sources may differ slightly from state-level vital
statistics data sources

NCDM Recommendations Version 3.0
Page 179 3/19/2013

CONTENT DOMAIN: REPRODUCTIVE HEALTH OUTCOMES
INDICATOR: SEX RATIO AT BIRTH AMONG SINGLETON BIRTHS
Type of EPHT Indicator
Measure
Derivation of Measure(s)
Unit
Geographic Scope
Geographic Scale
Time Period

Health outcome
Male to Female sex ratio at birth (term singletons only)
Sex ratio=total males/total females at birth among term singleton
births only
Ratio
State and national
State and county
2000-current

Time Scale

Year

Rationale

Population growth is, in part, related to the number of live male
children (1). Numerous studies have reported changes in the ratio of
males to females at birth; many of the studies have found a reduction
in male relative to female births in different countries throughout the
world (2-5). Although the mechanism that determines the sex of the
infant is not completely understood, some (6-12), but not all (3-4),
have suggested that environmental hazards can affect the number of
males. Biological parent(s) and/or the fetus can come in contact with
and become exposed to different hazards referred to as endocrine
disruptors (7-8, 10, 12). Fewer males are conceived when exposure to
endocrine disruptors results in a decrease in testosterone. Because
states have accurate Vital Statistics (VS) records on the sex of live
births, changes over time in the sex ratio of infants can be measured as
the ratio of males to females. This ratio of total males/total females
born in a pre-defined polygon (e.g., state, county, ZIP code, census
tract, block group) at a certain time (one birth year or multiple years) is
referred to as the Sex Ratio (SR).
The SR can be used to monitor the proportion of males to females in
states, counties, or smaller-resolution polygons, when data are
available and such analyses are justified. Baseline data can be used to
determine if the proportion of males is changing over time. When the
number of male births is the same as the number of female births, the
SR is equal to 1.000. Many studies have observed baseline SR values
that are usually higher than 1.000, and closer to 1.050(1, 3, 13). In
2002, the U.S. SR was 1.048 (1). If the SR is decreasing over time, the
implication is that fewer males than females are born for that period of
time. If consistent decreases in the SR occur, this outcome could be
used to determine if such changes are the result of environmental
hazards that can disrupt the endocrine system or some other

Use of the Measure

NCDM Recommendations Version 3.0
Page 180 3/19/2013

Limitations of the Measure

Data Sources
Limitations of Data
Sources

physiological system related directly or indirectly to the expression of
the neonates’ sex at birth.
Unfortunately, other factors besides endocrine disruptors can affect the
expression of sex (6, 13-15). Decreases in male births inversely
related to parental smoking, gestation length, parental age, and birth
order. Reproductive practices and social morays regarding sex
preferences—males over females, for example, can affect the observed
SR (3, 4, 7). Case-control studies have to be carried out to determine
if decreases in the SR over time are due to contact with and exposure
to endocrine disruptors; but effect modifiers have to be controlled in
order to understand this relationship, factors that modify it need to be
better accounted for. (8).
State’s VS data, CDC Wonder, CDC VS data, and U.S. Census 2000
data in Summary File (SF) 1.
There may be discrepancies between national and state data as noted in
the templates for measures of prematurity and growth retardation
above.

References
1. Mathews TJ, Brady E, Hamilton, E. Trend analysis of the sex
ratio at birth in the United States. National Vital Statistics
Reports; volume 53, number 20. Hyattsville, Maryland:
National Center for Health Statistics. 2005.
2. Grech V, Vassallo-Agius P, Savona-Ventura C. Secular trends
in sex ratios at birth in North America and Europe over the
second half of the 20th century. J Epidemiol Community
Health 2003;57:612-5.
3. Marcus M, Kiely J, Xu F, et al. Changing sex ratio in the
United States, 1969-1995. Fertil Steril 1998;70:270-3.
4. Martuzzi M, Di Tanno N, Bertollini R. Declining trends of
male proportion at birth in Europe. Arch Environ Health
2001;56:358-364.
5. Parazzini F, La Vecchia C, Levi F, et al. Trends in male:
female ratio among newborn infants in 29 countries from five
continents. Hum Reprod 1998;13:1394-6.
6. Fukuda M, Fukuda K, Shimizu T, et al. Parental
preconceptional smoking and male: female ratio of newborns.
Lancet 2002;359:1407-8.
7. Garry VF, Holland SE, Erickson LL, et al. Male reproductive
hormones and thyroid function in pesticide applicators in the
Red River Valley of Minnesota. J Toxicol Environ Health Part
NCDM Recommendations Version 3.0
Page 181 3/19/2013

A 2003;66:965-86.
8. Gomez del Rio I, Marshall T, Tsai P, et al. Number of boys
born to men exposed to polychlorinated biphenyls. Lancet
2002;360:143-4.
9. Karmaus W, Huang S, Cameron L. Parental concentration of
dichlorodiphenyl dichloroethene and polychlorinated biphenyls
in Michigan fish eaters and sex ratio in offspring. J Occup
Environ Med 2002;44:8-13.
10. Mackenzie CA, Lockridge A, Keith M. Declining sex ratio in
a first nation community. Environ Health Perspect
2005;113:1295-8.
11. Sakamoto M, Nakano A, Akagi H. Declining Minamata male
birth ratio associated with increased male fetal death due to
heavy methylmercury pollution. Environ Res 2001;87:92-8.
12. Weisskopf MC, Anderson HA, Hanrahan LP, et al. Decreased
sex ratio following maternal exposure to polychlorinated
biphenyls from contaminated Great Lakes sport-caught fish: a
retrospective cohort study. Environ Health 2003;2:2.
13. Vatten LJ, Skjærven R. Offspring sex and pregnancy outcome
by length of gestation. Early Hum Dev 2004;76:47-54.
14. Juntunen KST, Kvist AP, Kauppila AJI. A shift from a male to
a female majority in newborns with increasing age of grand
multiparous women. Hum Reprod 1997;12:2321-3.
15. Nicolich MJ, Huebner WW, Schnatter AR. Influence of
parental and biological factors on the male birth fraction in the
United States: An analysis of birth certificate data from 1964
through 1988. Fertil Steril 2000;73:487-92.

NCDM Recommendations Version 3.0
Page 182 3/19/2013


File Typeapplication/pdf
AuthorCDC User
File Modified2013-06-20
File Created2013-06-20

© 2024 OMB.report | Privacy Policy