Final Report 08 15 2014

Potential Public Health Impacts of Natural
Gas Development and Production in the

Marcellus Shale in Western Maryland
July 2014
Maryland Institute for Applied Environmental Health
School of Public Health
University of Maryland, College Park

Prepared for the Maryland Department of the Environment
and the
Maryland Department of Health and Mental Hygiene

ii
1 ACKNOWLEDGEMENTS
Research Team
Donald Milton, MD, DrPH
Professor and Director
Maryland Institute of Applied Environmental Health

Sacoby Wilson, MS, PhD
Assistant Professor

Chengsheng Jiang, PhD
Research Assistant Professor

Laura Dalemarre, MPH
Program Associate

Amir Sapkota, PhD
Associate Professor

Thurka Sangaramoorthy , MPH, PhD
Assistant Professor
Department of Anthropology
College of Behavioral and Social Sciences

Meleah Boyle
Graduate Research Assistant

Contributors
Keeve Nachman, PhD, MHS
Assistant Scientist
Johns Hopkins School of Public Health

Liz Ducey, MPS
Geographical Information Systems Support

Kelsey Babik
Graduate Student

Amelia Jamison
Graduate Student
Department of Anthropology

Kim Stinchcomb
Graduate Student

Carly Brody
Undergraduate Student

Harihar Batal

Christian Jenkins

Josh Trowell
iii
2 TABLE OF CONTENTS
1 ACKNOWLEDGEMENTS ...................................................................................................... ii
2 TABLE OF CONTENTS ......................................................................................................... iii
3 LIST OF TABLES .................................................................................................................. vii
4 LIST OF FIGURES ................................................................................................................. ix
5 GLOSSARY OF TERMS ....................................................................................................... xii
6 EXECUTIVE SUMMARY .................................................................................................... xv
6.1 Baseline Health Assessment ........................................................................................... xvi
6.1.1 Vulnerable Populations ............................................................................................ xvi
6.1.2 Physical Determinants of Health ............................................................................. xvi
6.1.3 Social Determinants of Health ................................................................................ xvii
6.2 Impact Assessment .......................................................................................................... xix
6.2.1 Hazard Evaluation Methods and Summary ............................................................. xix
6.2.2 Air Quality ................................................................................................................ xx
6.2.3 Flowback and Production Water-Related ................................................................ xxi
6.2.4 Noise ....................................................................................................................... xxii
6.2.5 Earthquakes ............................................................................................................. xxii
6.2.6 Social Determinants of Health ............................................................................... xxiii
6.2.7 Occupational Health ............................................................................................... xxiii
6.2.8 Healthcare Infrastructure ....................................................................................... xxiii
6.2.9 Cumulative Exposures/Risk ................................................................................... xxiv
6.3 Recommendations .......................................................................................................... xxv
6.3.1 Comprehensive Gas Development Plans (CGDP) .................................................. xxv
6.3.2 Disclosure of Well Stimulation Materials ............................................................... xxv
6.3.3 Air Quality ............................................................................................................. xxvi
6.3.4 Flowback and Production Water-Related ............................................................. xxvii
6.3.5 Noise ..................................................................................................................... xxvii
6.3.6 Earthquakes .......................................................................................................... xxviii
6.3.7 Social Determinants of Health ............................................................................. xxviii
6.3.8 Healthcare Infrastructure ....................................................................................... xxix
6.3.9 Cumulative Exposure/Risk .................................................................................... xxix
6.3.10 Occupational Health ................................................................................................ xxx
iv
7 INTRODUCTION .................................................................................................................... 1
7.1 Health Impact Assessment Process and the Public Health Study ...................................... 1
7.2 Natural Gas Development & Production ........................................................................... 2
7.2.1 Conventional vs. Unconventional Natural Gas ........................................................... 2
7.2.2 The Marcellus Shale ................................................................................................... 2
7.2.3 Terminology ................................................................................................................ 3
7.2.4 Unconventional Natural Gas Development (UNGD) ................................................. 3
7.2.5 Unconventional Natural Gas Production (UNGP) ...................................................... 5
8 SCOPING UPDATE ................................................................................................................. 6
9 BASELINE HEALTH ASSESSMENT .................................................................................... 8
9.1 Introduction ........................................................................................................................ 8
9.2 Overview of Allegany and Garrett Counties ..................................................................... 8
9.3 Demographics .................................................................................................................... 8
9.4 Vulnerable Populations ...................................................................................................... 9
9.5 Health Indicators ................................................................................................................ 9
9.5.1 Environmental Health ................................................................................................. 9
9.5.2 Physical Health Indicators ........................................................................................ 10
9.5.3 Major Causes of Morbidity and Mortality ................................................................ 11
9.6 Social Determinants of Health ......................................................................................... 12
9.7 Healthcare Infrastructure ................................................................................................. 13
10 IMPACT ASSESSMENT ....................................................................................................... 15
10.1 Overview of Key Determinants of Human Exposures to UNGDP Related Hazards ...... 15
10.1.1 Overview of Exposure Assessment Methods for UNGDP Related Hazards ............ 16
10.1.2 Linking Exposure to Hazards with Adverse Health Outcomes ................................ 18
10.2 Methods............................................................................................................................ 19
10.2.1 Overview of Data Collection .................................................................................... 19
10.2.2 Identification of Hazards of Concern to Western Maryland Communities .............. 19
10.2.3 Ranking of Hazards ................................................................................................... 20
10.2.4 Identifying Chemicals of Concern ............................................................................ 23
10.3 Community Impacts ......................................................................................................... 26
10.3.1 Air Quality ................................................................................................................ 26
10.3.2 Flowback and Production Water-Related ................................................................. 40
10.3.3 Noise ......................................................................................................................... 49
v
10.3.4 Earthquakes ............................................................................................................... 58
10.3.5 Social Determinants of Health .................................................................................. 62
10.3.6 Healthcare Infrastructure .......................................................................................... 68
10.3.7 Cumulative Exposures/Risk ...................................................................................... 73
10.4 Occupational Impacts ....................................................................................................... 76
10.4.1 Injuries and Fatalities ................................................................................................ 76
10.4.2 Job Hazards Overall .................................................................................................. 77
10.4.3 Assessment ................................................................................................................ 81
11 REGULATORY LANDSCAPE ............................................................................................. 83
11.1 Federal Regulations ......................................................................................................... 83
11.1.1 Water ......................................................................................................................... 83
11.1.2 Air ............................................................................................................................. 84
11.1.3 Waste Disposal and the Right to Know .................................................................... 84
11.2 State and Local Regulations ............................................................................................. 85
11.2.1 Setback Requirements ............................................................................................... 86
11.2.2 Chemical Disclosure ................................................................................................. 86
11.2.3 Other Forms of Well Stimulation ............................................................................. 87
12 RECOMMENDATIONS ........................................................................................................ 88
12.1 Comprehensive Gas Development Plans (CGDP) ........................................................... 88
12.2 Disclosure of Well Stimulation Materials ........................................................................ 89
12.3 Air Quality ....................................................................................................................... 91
12.4 Flowback and Production Water-Related ........................................................................ 92
12.4.1 Water & Soil Quality ................................................................................................ 92
12.4.2 NORM ...................................................................................................................... 93
12.5 Noise ................................................................................................................................ 94
12.6 Earthquakes ...................................................................................................................... 94
12.7 Social Determinants of Health ......................................................................................... 95
12.7.1 Traffic Safety ............................................................................................................ 95
12.7.2 Empower communities ............................................................................................. 96
12.8 Healthcare Infrastructure ................................................................................................. 96
12.9 Cumulative Exposure/Risk .............................................................................................. 98
12.10 Occupational Health ...................................................................................................... 98
13 LIMITATIONS ..................................................................................................................... 100
vi
14 REFERENCES ..................................................................................................................... 101
15 APPENDIX 1: Baseline Health Assessment ........................................................................ 123
15.1 Overview of Allegany and Garrett Counties ................................................................. 123
15.1.1 Geography ............................................................................................................... 123
15.1.2 Schools .................................................................................................................... 124
15.1.3 Hospitals ................................................................................................................. 125
15.1.4 Important Landmarks .............................................................................................. 126
15.2 Demographics ................................................................................................................ 129
15.3 Vulnerable Populations .................................................................................................. 130
15.3.1 Age .......................................................................................................................... 131
15.3.2 Socioeconomic Status ............................................................................................. 133
15.4 Environmental Health .................................................................................................... 134
15.4.1 Drinking Water ....................................................................................................... 136
15.4.2 Air ........................................................................................................................... 138
15.4.3 National Scale Air Toxics Assessment (NATA) .................................................... 139
15.5 Physical Health Indicators ............................................................................................. 141
15.5.1 Life Expectancy ...................................................................................................... 141
15.5.2 Poor Physical Health Days ...................................................................................... 142
15.5.3 Chronic Diseases ..................................................................................................... 143
15.5.4 Major Causes of Morbidity and Mortality .............................................................. 145
15.6 Social Determinants of Health ....................................................................................... 155
15.6.1 Sexually Transmitted Infections (STIs) .................................................................. 155
15.6.2 Crime ....................................................................................................................... 156
15.6.3 Injuries .................................................................................................................... 158
15.6.4 Mental Health .......................................................................................................... 160
15.6.5 Substance Abuse ..................................................................................................... 161
15.7 Health Care Infrastructure .............................................................................................. 162
15.7.1 Providers ................................................................................................................. 162
15.7.2 Insurance Status ...................................................................................................... 165
16 APPENDIX 2 ........................................................................................................................ 166
vii
3 LIST OF TABLES
Table 6-1: Hazard Evaluation Summary ....................................................................................... xx
Table 10-1: Ranking of Exposure Assessment Methods .............................................................. 17
Table 10-2: Description of the evaluation criteria used for hazard ranking ................................. 20
Table 10-3. Chemicals Commonly Used in Shale Fracturing and Consequence of Not Using the
Chemicals, Source: [26] ........................................................................................................ 23
Table 10-4: Summary of selected air pollutants associated with the UNGDP process, as
described in Leidos report [28], with slight modification. .................................................... 27
Table 10-5: RESI scenarios by development year ........................................................................ 31
Table 10-7: Air Quality Evaluation .............................................................................................. 39
Table 10-8: Flowback and Production Water Related Evaluation ................................................ 49
Table 10-9. Proposed Setbacks specific to Occupied Dwellings, Source: Maryland Best
Management Practices [16] ................................................................................................... 50
Table 10-10. Marylands Maximum Allowable Noise Levels for Receiving Land Categories ... 50
Table 10-11. Noise Associated with UNGD ................................................................................. 52
Table 10-12. Summary Statistics, Stratified by Distance, Location, and Time ............................ 55
Table 10-13: Noise Evaluation ..................................................................................................... 58
Table 10-14: National Inventory by Classes of Injection Well [112] ........................................... 61
Table 10-15: Earthquake Evaluation ............................................................................................ 62
Table 10-16. Percent Change in STIs, Disorderly Conduct Arrests, and Substance Abuse Arrests
............................................................................................................................................... 65
Table 10-17: Social Determinants of Health Evaluation .............................................................. 67
Table 10-18: Health Care Infrastructure Evaluation ..................................................................... 72
Table 10-19: Cumulative Exposures/Risk Evaluation .................................................................. 75
Table 10-20: Occupational Health Evaluation .............................................................................. 81
Table 15-1: Demographics, US Census 2012 ............................................................................. 129
Table 15-2: Life Expectancy, 2009 ............................................................................................. 142
Table 15-3: Poor Physical Health Days, 2006-2012 ................................................................... 142
Table 15-4: Cancer Incidence Rates, 2006-2010 ........................................................................ 145
Table 15-5. Sexually Transmitted Infections (STIs), 2011 ......................................................... 156
Table 15-6. Total Crime, 2010 .................................................................................................... 157
Table 15-7. Violent and Property Crime, 2010 ........................................................................... 157
Table 15-8. Unintentional Injuries, 2006-2010 ........................................................................... 158
Table 15-9: Alcohol-Impaired Driving Deaths, 2008-2012 ....................................................... 159
viii
Table 15-10. Suicide, 2000-2010 ................................................................................................ 160
Table 15-11. Mental Health, 2006-2012 ..................................................................................... 160
Table 15-12. Substance Abuse, 2006-2012 ................................................................................ 161
Table 15-13. Health Care Infrastructure ..................................................................................... 163
Table 16-1: Health Effects Associated with Chemicals Used During UNGDP ......................... 166

ix
4 LIST OF FIGURES
Figure 7-1: Major Activities Associated with UNGDP Process [15] ............................................. 3
Figure 7-2: UNG Compressor Station ............................................................................................ 5
Figure 10-1: Source to Effect Continuum for a Typical Environmental Hazard .......................... 15
Figure 10-2: Carcinogenicity Classification for Chemicals used During UNGDP ...................... 25
Figure 10-3: Target organ systems for chemicals used during UNGDP, from Cal EPA's OEHHA
Toxicity Criteria database and ATSDR's Toxic Substance Portal ........................................ 25
Figure 10-4: Seasonal comparisons of air quality for Garrett County and the State of Maryland
for selected criteria air pollutants, 2013 ................................................................................ 28
Figure 10-5: Ambient concentrations for selected VOCs near well pads in WV. Data taken from
University of WV study by McCawley et al. [34] ................................................................ 30
Figure 10-6: Variability in ambient concentrations of Acetone and Heptane across different well
pads in WV. Data taken from University of WV study by McCawley et al. [34] ................ 31
Figure 10-7: Estimated Marcellus Shale well production curve for Maryland during the first five
years. Source: Regional Economic Studies Institute 2014. [10] ........................................... 34
Figure 10-8: Estimated yearly emissions for PM
2.5
in Western Maryland under 25% and 75%
extraction scenarios ............................................................................................................... 35
Figure 10-9: Estimated yearly emissions for NO
x
Figure 10-10: Estimated yearly emissions for VOCs in Western Maryland under 25% and 75%
Figure 10-11. Conceptual model of water contamination pathways, from Rozell and Reaven
2012....................................................................................................................................... 42
Figure 10-12: Well Pad, West Virginia ........................................................................................ 51
Figure 10-13: Time Series, Indoor L
eq
by Distance from Compressor Station ............................ 54
Figure 10-14: Time Series, Outdoor Leq by Distance from Compressor Station ........................ 55
Figure 10-15: Boxplots, Leq by Distance from Compressor Station ............................................ 57
Figure 10-16. Cumulative counts of earthquakes with a magnitude ! 3.0 in the central and
eastern United States, 1970-2013, [105] ............................................................................... 59
Figure 10-17: UNGDP-related Traffic, West Virginia ................................................................. 63
Figure 10-18: Silica Dust from a Well Pad, West Virginia .......................................................... 78
Figure 10-19: Comparisons of arithmetic means of TWAs (mg/m3) for job titles with five or
more samples in relation to a calculated OSHA PEL (based on 53% silica) and NIOSH REL
for respirable silica. Maximum values for each job title shown by diamonds at the end of
dashed lines, Source: [30] ..................................................................................................... 79
Figure 10-20: Natural Gas Flaring ................................................................................................ 79
x
Figure 15-1: Major Cities and Towns in Allegany and Garrett Counties ................................... 124
Figure 15-2: Location of Community Assets and Sensitive Human Receptors ......................... 128
Figure 15-3: Map of Zip Codes in Allegany and Garrett Counties ............................................ 129
Figure 15-4: Age Distribution for Allegany and Garrett Counties, Maryland, the Region, and the
U.S., Source: U.S. Census 2012 ......................................................................................... 131
Figure 15-5: Children Less than Age 5 and Adults Greater than 65 in Allegany and Garrett
Counties, Source: U.S. Census 2012 .................................................................................. 132
Figure 15-6: Comparison of Percent Poverty and Percent Less than High School Education for
Allegany and Garrett Counties, Source: U.S. Census 2012 ................................................ 133
Figure 15-7: Percent Poverty and Unemployment for Allegany and Garrett Counties, Maryland,
the Region, and the U.S., Source: U.S. Census 2012 ......................................................... 133
Figure 15-8: Spatial Distribution of Conventional Gas Wells, NPDES-Permitted Facilities,
Superfund Sites, Brownfields, LUSTs, and TRI Facilities in Allegany and Garrett Counties
............................................................................................................................................. 135
Figure 15-9: Location of Private Wells in Garrett County ......................................................... 137
Figure 15-10: Average Daily PM2.5 Concentrations, 2011 ....................................................... 137
Figure 15-11: Total TRI Releases for 2000, 2005, and 2010 ..................................................... 138
Figure 15-12: NATA Cancer Risk, 2002 and 2005 .................................................................... 140
Figure 15-13: Respiratory Hazard Index, 2002 and 2005 ........................................................... 141
Figure 15-14: Preventable Hospital Stays, 2011 ......................................................................... 142
Figure 15-15: Percent of Adults with High Blood Pressure, 2006-2012 .................................... 143
Figure 15-16: Percent of Obese Adults and Percent of Adults with Diabetes, 2006-2012 ........ 144
Figure 15-17: Percent of Adult Smokers, 2006-2012 ................................................................. 144
Figure 15-18: Number of Deaths from Various Cancers per 100,000 (Age-Adjusted) in Allegany
and Garrett Counties Compared to Maryland and the Region (2000-2010), Source: National
Cancer Institute ................................................................................................................... 147
Figure 15-19: Total Cancer Deaths per 100,000, 2000-2010 ..................................................... 148
Figure 15-20: Total Chronic Respiratory Deaths per 100,000, 2000-2010 ................................ 149
Figure 15-21: Total Flu and Pneumonia Deaths per 100,000, 2000-2010 .................................. 150
Figure 15-22: Cardiovascular Disease Deaths per 100,000, 2000-2010 .................................... 151
Figure 15-23: Cerebrovascular Disease Deaths per 100,000, 2000-2010 .................................. 151
Figure 15-24: Septicemia Deaths per 100,000, 2000-2010 ........................................................ 152
Figure 15-25: All-Cause Mortality, 2000-2010 .......................................................................... 153
Figure 15-26: Percent Low Birth Weight, 2006-2012 ................................................................ 154
Figure 15-27: Percent Premature Births and Low Birth Weight, 2006-2012 ............................. 154
xi
Figure 15-28: Infant Mortality, 2006-2010 ................................................................................. 155
Figure 15-29: Chlamydia Rate, 2011 and HIV Rate, 2010 per 100,000 .................................... 156
Figure 15-30: Total Accidental Deaths and Motor Vehicle Deaths per 100,000, 2006-2010 .... 159
Figure 15-31: Percent Adult Excessive Drinking ....................................................................... 162
Figure 15-32: Number of Dentists, 2012 and Primary Care Physicians, 2011 per 100,000 ....... 163
Figure 15-33: HPSA Designations in Allegany and Garrett Counties, 2013 ............................. 164
Figure 15-34: Uninsured Populations, 2011 ............................................................................... 165

xii
5 GLOSSARY OF TERMS
ATSDR: Agency for Toxic Substances and Disease Registry, a part of CDC
AAA: American Automobile Association
BTU: British Thermal Units
CH
4
: Methane
CO: Carbon monoxide
CO
2
: Carbon dioxide
CAL EPA: California Environmental Protection Agency
CAA: Clean Air Act
CWA: Clean Water Act
CDC: Center for Disease Control and Prevention
CERCLA: Comprehensive Environmental Response, Compensation, and Liability Act
CERCLIS: Comprehensive Environmental Response, Compensation, and Liability
Information System
CDGP: Comprehensive Gas Development Plan
CNGDP: Conventional Natural Gas Production
CRA Cumulative Risk Assessment
dB: Decibel
dBA: A-weighted decibel level used for measuring loudness to the human ear
DPM: Diesel Particulate Matter
DEP: Department of Environmental Protection
DHMH: Maryland Department of Mental Health and Hygiene
EPA: Environmental Protection Agency
EPA IRIS: Environmental Protection Agencys Integrated Risk Information System
EPCRA: Emergency Planning and Community Right-to-Know Act
EIA: Environmental Impact Assessment
FTE: Full-time employees
GCDH: Garrett County Department of Health
GCMH: Garrett County Memorial Hospital
H
2
S: Hydrogen Sulfide
HC: Hydrocarbons
HH: Household
HS: High School
xiii
HIA: Health Impact Assessment
HPSA: Health Professional Shortage Areas
HVHF: High Volume Hydraulic Fracturing
IARC: International Agency for Research on Cancer
LULU: Locally Unwanted Land Use
LUST: Leaking Underground Storage Tank
"g/L: Micrograms per liter
"g/m
3
: Micrograms per cubic meter
MD: Maryland
MDE: Maryland Department of the Environment
MDNR: Maryland Department of Natural Resources
MIAEH: Maryland Institute of Applied Environmental Health
MOU: Memorandum of Understanding
MLMC: Mountain Laurel Medical Center
MOSH: Maryland Occupational Safety and Health
MSDS: Material Data and Safety Sheets
MUA: Medically Underserved Areas
NATA: National-Scale Air Toxics Assessment
NIOSH: National Institute for Occupational Safety and Health
NGDP: Natural Gas Development And Production
NO
x
: Nitrogen Oxides
NORM: Naturally Occurring Radioactive Material
NPDES: National Pollution Elimination Discharge System
NYSDEC: New York State Department of Environmental Conservation
ODNR: Ohio Department of Natural Resources
OEHHA: Office of Environmental Health Hazard Assessment
OSHA: Occupational Safety and Health Administration
O
3
: Ozone
PA: Pennsylvania
PM: Particulate Matter
PM
2.5
: Particulate Matter with aerodynamic diameter 2.5 microns or less
PM
10
: Particulate Matter with aerodynamic diameter 10 microns or less
ppb: Parts per billion
xiv
ppm: Parts per million
PEL: Permissible Exposure Limits
PSE: Physicians Scientists and Engineers for Healthy Energy
pCi/L: Picocurie, amount of ionizing radiation per liter
PAH: Polycyclic Aromatic Hydrocarbon
POTW: Publicly Operated Treatment Work
RCRA: Resource Conservation and Recovery Act
RESI: Regional Economic Studies Institute, Towson University
REL: Respiratory Exposure Limits
SDWA: Safe Drinking Water Act
SDH: Social Determinants Of Health
SES: Socio-Economic Status
STIs: Sexually-Transmitted Infections
SO
x
: Sulfur Oxides
SP-OMS: SoundPro Outdoor Measuring System
TRI: Toxic Release Inventory
TWA: Time Weighted Average
UNGD: Unconventional Natural Gas Development
UNGDP: Unconventional Natural Gas Development and Production
USGS: United States Geological Survey
UST: Underground Storage Tank
VOCs: Volatile Organic Compounds
WMHS: Western Maryland Health System
WHO: World Health Organization
WV: West Virginia

xv
6 EXECUTIVE SUMMARY
On October 18, 2013, the Maryland Department of Health and Mental Hygiene (DHMH) signed
a memorandum of understanding (MOU) with the Maryland Institute for Applied Environmental
Health (MIAEH), School of Public Health, University of Maryland, College Park to conduct an
assessment of the potential public health impacts associated with drilling in the Marcellus Shale
in Maryland and to provide a Marcellus Shale Public Health Report. This document is the final
report.
The MOU specified that the project is designed to provide a baseline assessment of current
regional population health, an assessment of potential public health impacts, and possible
adaptive and public health mitigation strategies in the event that natural gas extraction takes
place within Marylands Marcellus Shale resource. In particular, the project is not designed to
make recommendations about whether or when to allow unconventional natural gas development
and production (UNGDP) in Maryland. Rather this study is designed to inform decisions by
clearly describing the risks and potential public health responses.
This public health study draws upon several methods of a rapid Health Impact Assessment (HIA)
including: scoping, assessment of baseline health and potential health impacts of shale gas
development, and this final report with recommendations for public health responses. The
scoping process sought input from a wide range of stakeholders through public meetings and
publication of a draft detailed scoping document. Comments on the scoping document were used
to make modifications to the scope and are reflected in this final report. Due to time constraints,
we will not publish a revised scoping report. Rather, we describe the revisions to the scope in
section 8 of this final report. Although global climate change is a major concern and some
stakeholders wanted it included, it remains beyond the scope of this study. Our focus is on public
health impacts that would be concentrated in and unique to the Garrett and Allegany County
populations living and working near the sites of shale gas development.
The baseline health assessment examined demographics, potential vulnerable populations, a wide
range of health indicators, social determinants of health, and healthcare infrastructure in Garrett
and Allegany counties. The impact assessment is based on available data from other states with
ongoing UNGDP regarding exposure and health outcomes and on epidemiologic and toxicologic
data from other contexts that are relevant to potential UNGDP related exposures. Our
assessments of potential health impacts are not predictions that these effects will necessarily
occur in Maryland, where regulation is likely to be stricter than in some states where UNGDP is
already underway. Rather, we provide assessments of the impacts that could occur and that need
to be addressed by preventive public health measures if and when drilling is allowed. Thus, the
focus of our recommendations is on answering this question: Given the baseline population
health, vulnerabilities, and potential impacts of UNGDP, how can Maryland best protect public
health if and when UNGDP goes forward?
We presented a draft of this report and its recommendations in a final progress report at a public
meeting June 28, 2014. This final report reflects stakeholder input received at and subsequent to
that meeting. Several of the recommendations have been significantly revised. The final
recommendations in this report supersede the recommendations contained in the slides
posted from the June 28, 2014 meeting.
xvi
6.1 Baseline Health Assessment
The first step in the Health Impact Assessment process is identifying the health trends and issues
currently impacting the population. Therefore, to assess the baseline health of Allegany and
Garrett County residents, we considered demographics, potential vulnerable populations, a wide
range of health indicators, environmental health, social determinants of health, and healthcare
infrastructure. Ideally, a baseline assessment would, and stakeholder input urged us to, collect
primary, representative, individual health and exposure data. However, conducting a new survey
was beyond the time and funding limitations of this study, as is often true of health impact
assessments. In this study, most health data was only available at the county-level. Thus analysis
at the town or census tract level was not possible. Many of the physical and social determinants
of health covered in this baseline assessment were raised as concerns at scoping meetings by
stakeholders including residents of Allegany and Garrett Counties, health practitioners,
policymakers, environmental non-governmental organizations, and health advocates. While this
baseline assessment only focuses on residents of Allegany and Garrett Counties, it may have
relevance for individuals in surrounding counties or with other shale deposits in the State of
Maryland. A brief summary of the baseline assessment is provided here, a more detailed
summary is contained in the body of this report and the full baseline health assessment is
available in the Appendix.
6.1.1 Vulnerable Populations
It is important to recognize underlying social, economic, geographic, and individual level
vulnerabilities that may increase risk of disease and premature mortality for populations in
Garrett and Allegany counties. For example, residential proximity is an important factor in
geographic vulnerability; there are many conventional gas wells in Western Garrett County and
individuals who live near them could have higher exposure compared to individuals who live
farther away. There are many existing conventional gas wells and EPA-regulated facilities and
land uses including Superfund sites, brownfields, leaking underground storage tanks (LUSTs),
and National Pollution Discharge Elimination System (NPDES) sites in Allegany and Garret
Counties. The particular subpopulations living near these facilities may have environmental
exposure burden disparities and cumulative impacts that increase exposure and health risks.
Additionally, almost 40% of the population (children under age 18 and adults over 65) may be
considered more vulnerable to certain exposures including chemical and physical agents and
social stressors. Other factors including genetics, pre-existing disease, exposure to psychosocial
stress, socioeconomic status, educational attainment, pregnancy, and behaviors such as alcohol
consumption, smoking history, nutrition, and lifestyle can also influence vulnerability to disease.
Occupational exposures and lack of access to health care infrastructure may also contribute to
risk of disease. One group that is particularly at risk is surface owners who do not have mineral
rights; they are subject to stress associated with the lack of control as well as any negative
impacts associated with UNGDP activities. Unfortunately, we do not have accurate data on
mineral rights ownership.
6.1.2 Physical Determinants of Health
To assess the baseline physical health of the Allegany and Garrett counties and compare to the
surrounding region (counties in Pennsylvania and West Virginia) and the State of Maryland, the
xvii
HIA team obtained and analyzed environmental health data, health status data, chronic disease
data, cancer incidence and cancer mortality data, other mortality data, and information on birth
outcomes including low birth weight and premature births and infant mortality. Currently, daily
PM
2.5
levels average around 13 "g/m
3
for Allegany and Garrett Counties, which is slightly
higher than daily average for the state of Maryland. We observe that TRI releases are
significantly higher in Allegany County compared to Garrett County, but there has been a
decrease in total TRI releases in the state from 2000 to 2010. From the NATA dataset, we found
that the estimated lifetime cancer and respiratory risks from air toxics were higher for individuals
in Allegany County compared to Garrett County. For both, we observe that overall risk has
decreased in the state of Maryland which could be due to reduction in air pollution levels or
changes in the risk calculation. The rate of poor physical health days and preventable hospital
stays was higher in Allegany County compared to Garrett, the region, and the state of Maryland.
The percent of adults with hypertension, diabetes, obese, or smokers was higher in Allegany
County compared to Garrett County and the overall trend in the state of Maryland for 2006-2012.
Only for hypertension, was the percentage noticeably higher in Allegany County compared to the
region. We obtained cancer incidence data from NCI for 2006-2010 and found that Allegany had
a higher cancer incidence rate for non-Hodgkins lymphoma, leukemia, and colorectal cancer
compared to both Garrett County and the state of Maryland. While, Garrett only had a bladder
cancer rate higher than both Allegany and the state of Maryland. Over the 10-year period
between 2000 and 2010, the cancers with the highest mortality rates in Allegany and Garrett
counties were colorectal, breast, and prostate cancers. Deaths from these cancers were higher in
these counties compared to the region and State of Maryland overall. Furthermore, compared to
the leading causes of death from cancer nationwide, these counties rates of colorectal cancer
deaths were higher. Over the ten-year period from 2000 and 2010, rates of chronic respiratory
disease mortality, cardiovascular disease mortality, cerebrovascular disease mortality, and
septicemia mortality were higher in Allegany County compared to the rates of Garrett, Maryland,
and the region. From the 2006-201 time period, percentage of babies born with low birth weight
(LBW) in Allegany (9.1%) was higher than % low birth weight for Garrett (7.5%), MD (9%),
region (8%), and the United States (8.2%); while, infant mortality rates of 8.4 deaths/1000 births
(Allegany) and 10.8 deaths/1000 births (Garrett) were higher than the rates for the MD (7.2
deaths/1000 births), and US (6.9 births/1000 deaths).
6.1.3 Social Determinants of Health
To evaluate the baseline social determinants of health in Allegany and Garrett counties, we
obtained available information regarding sexually transmitted infections (STI), crime, injuries,
mental health, and substance abuse. In 2011, STI rates including HIV remain low in both
counties when compared to the State of Maryland. In Garrett County, crime rates across all
categories remain steady and lower than the Maryland State averages, fluctuating slightly over
the 10-year period between 2000 and 2010. In Allegany County, there was a slow but steady
increase in most crime categories in this same period. This increase runs counter to statewide
trends, which demonstrate major decreases in crime rates across all categories in the last decade.
Total mortality rates from unintentional injury, motor vehicle traffic accidents, and intentional
self-harm (suicide) are much higher in Allegany and Garrett counties than the average for the
State of Maryland. Data gathered from County Health Rankings, the Health Indicators
Warehouse, and the Behavioral Risk Factor Surveillance System (BRFSS) indicate that poor
xviii
mental health, insufficient social and emotional support, and alcohol abuse appear to be the top
indicators of the burden of mental health and substance abuse in Allegany and Garrett counties.
To assess the healthcare infrastructure of Allegany and Garrett counties, we obtained information
from the US Health Resources and Services Administration (HRSA) regarding rates and ratios of
primary care physicians, dentists, and mental health providers to the population. These rates are,
on average, much lower for both counties than the statewide averages, especially for mental
health providers, indicating a critical shortage of providers in both Allegany and Garrett
counties. In addition, Allegany County is a designated Health Professional Shortage Area
(HPSA) for primary care for low-income populations, mental health care for Medical Assistance
populations, and dental care for Medical Assistance populations. Allegany County has a critical
need for specialty providers including vascular surgery, urology, as well as dentists willing to
provide care for adults with no insurance or Medical Assistance. Garrett County is a designated
HPSA for primary and mental health care, and dental care for Medical Assistance populations.
All of Garrett County is considered a Medically Underserved Area (MUA), while substantial
portions of Allegany County (Orleans, Lonaconing, Oldtown, and Cumberland) also qualify as
MUAs. Finally, the percentage of uninsured residents in Allegany County was similar to the
statewide average of 12%, while the percentage was slightly higher (14%) for Garrett County.
The body of this report contains a summary of the baseline health assessment and the Appendix
contains a more detailed baseline health profile and assessment for Allegany and Garrett coun-
ties.
xix
6.2 Impact Assessment
6.2.1 Hazard Evaluation Methods and Summary
To evaluate the potential public health impact of UNGDP process in Allegany and Garrett
Counties, we conducted an extensive review of the literature. We identified the hazards that most
concern community members through a detailed scoping process. We also conducted a site visit
to a community with active UNGDP in Doddridge County, WV to directly observe the impacts,
both positive and negative. During that visit, we gained firsthand knowledge about the hidden
sufferings experienced by residents as well as benefits experienced by the local businesses. We
also met by teleconference with the American Petroleum Institute to obtain their views and
reviewed comments and literature that they submitted as part of their stakeholder input to the
public health study.
Based on the scoping process and existing literature, we categorized the UNGDP associated
hazards into eight broad categories: i) Air quality, ii) Water-related (water quality, soil quality
and naturally occurring radioactive materials), iii) Noise, iv) Earthquakes, v) Social determinant
of health (e.g., sexually transmitted infections (STIs), traffic, crime), vi) Healthcare
infrastructure, vii) Occupational health, and viii) Cumulative exposure/risk. We then ranked each
of these hazards using seven criteria. The scores were summed across the evaluation criteria to
obtain an overall score for the hazards. Based on this overall score, we classified each hazard as:
H: High likelihood that UNGDP related changes will have negative impact on public
health
M: Moderately high likelihood that UNGDP related changes will have negative
impact on public health.
L: Low likelihood that UNGDP related changes will have negative impact on public
health.
After much internal discussion on the study team, we agreed on Moderate as our middle
hazard classification in time for the final progress report presentation but were never truly
comfortable with it. However, following the June 28 meeting, we found that our original
classification term Moderate was frequently misunderstood and resulted in repeated requests
for clarification. Apparently some see moderate as similar to moderation is good and we
were asked if we meant it was not significant. An alternative classification that might be clearer
would be not significant, moderately significant, and highly significant likelihood of
negative impact. But, that seemed unwieldy. High and low are clear and succinct. To retain their
brevity and to be clear that our middle category does not consist of hazards that should be
dismissed, we have chosen Moderately High as the middle category for this final report.
A summary of the hazard classification for each of the eight broad categories of UNGDP
associated hazards is shown in Table 6-1. Three categories were classified as having high
likelihood, three were classified as having moderately high, and one as having low likelihood of
negative impacts on public health. The following sections 6.2.2 through 6.2.9 give summary
explanations of these classifications for each category of hazards. The detailed hazard evaluation
is contained in section 10 Impact Assessment.
xx
Table 6-1: Hazard Evaluation Summary
Topic
Likelihood of Negative
Public Health Impact
Air Quality High
Healthcare Infrastructure High
Occupational Health High
Social Determinants of Health High
Cumulative Exposures/Risks Moderately High
Flowback and Production Water-Related Moderately High
Noise Moderately High
Earthquakes Low
High = high likelihood of negative health impacts, Moderately High =
moderately high likelihood of negative health impacts, Low = low
likelihood of negative health impacts
6.2.2 Air Quality
Epidemiological studies over the past 50 years have documented the relationships between
exposure to selected air pollutants and various adverse health outcomes. Recent data suggests
these air pollutants are associated with UNGDP - some are produced as a part of the process (site
preparation, production), while others are present in the natural gas. At present, linking exposure
to air pollution associated with UNGDP - a new phenomenon- with adverse health outcome is
challenging because: 1) discrepancy in temporal scale between onset of exposure (dating only
few years back) and manifestation of outcomes that are known to have a notable lag time,
particularly for chronic diseases, 2) epidemiological studies designed to investigate such
association are often 3-5 years in duration with additional 1-2 years for data to be published in a
peer-reviewed journals. Despite these challenges, findings have started to emerge in peer-re-
viewed journals linking exposure to air pollution associated with UNGDP increased risk of sub-
chronic health effects, adverse birth outcomes including congenital heart defects and neural tube
defects, as well as higher prevalence of symptoms such as throat & nasal irritation, sinus
problems, eye burning, severe headaches, persistent cough, skin rashes, and frequent nose bleeds
among respondents living within 1500 feet of UNGDP facilities compared to those who lived
>1500 feet [13]. Major determinants of these relationships include the concentration of the
pollutants in the environment, frequency and duration of exposures encountered by individuals
xxi
as well as potency of these pollutants. At present, there is a dearth of information that allows
public health professionals to critically evaluate these aspects. While no information is available
on the concentration profile of air pollutants as a function of distance from the well pads and
compression stations, increasing body of literature on traffic related air pollution show that the
concentrations of traffic related air pollutants reach to background level beyond 500-1000m
(1640-3280 feet) distance from the roads.
Based on our evaluations of the limited but emerging epidemiological evidence from UNGDP
impacted areas and air quality measurements as well as epidemiological evidence from other
fields, we conclude that there is a High Likelihood UNGDP related changes in air quality will
have a negative impact on public health in Garrett and Allegany Counties. The extent of the
impact will be based on population vulnerability, proximity to the sites, and the success of public
health prevention strategies implemented by the State and local communities and control
measures taken by the industry to minimize exposures.
6.2.3 Flowback and Production Water-Related
The scientific literature has documented many plausible pathways by which natural and anthrop-
ogenic contamination may become available for human exposure as a result of unconventional
natural gas development. The evidence base to date suggests that gases, chemical compounds,
and to a lesser extent naturally occurring radioactive materials (NORM), are mobilized during
the drilling and wastewater recovery phases of the fracturing process and may result in
contamination of ground waters used for drinking water. Concerns also exist regarding the
surface impoundment of wastewater in ponds or pits, in regards to both accumulation of
radiological material and the concurrent potential for spills or leaks due to overfilling or ruptures
in impoundment liners. While challenges (some peer-reviewed) exist to assertions that fracturing
activities are impacting drinking water sources, there appears to be scientific consensus that
high-quality baseline and periodic monitoring data are largely absent in states that currently
permit fracturing. This lack of data complicates assessment of the potential impacts of fracturing
activities and may preclude determination of best practices or other interventions aimed at
minimizing exposures. Despite these gaps, there is consistency in the literature that wells within
shorter distances (typically <1 km) of drill sites are likely to be impaired, potentially by fractur-
ing activities. The most commonly documented contamination in these wells is methane gas.
Since horizontal drilling with hydraulic fracturing has not yet occurred in the state, Maryland has
an opportunity to conduct a thorough baseline characterization of ground water conditions prior
to allowing UNGDP. That way, if the state were to proceed with permitting the practice, it would
have comparison data to revisit its decision in the future. Questions remain, however, whether
feasible technologies exist to reverse groundwater impacts that may later be determined to have
arisen as a result of fracturing activities.
Studies of health effects of drinking water exposures to fracturing contaminants do not yet exist,
though many anecdotal reports would suggest that high-quality, rigorous studies should be
conducted to better understand the health consequences of exposure. Evidence exists to show
recovered wastewater can be contaminated with NORM and heavy metals. The composition of
the NORM appears to depend on the geologic composition of bedrock in which drilling is
occurring. It is common for radium isotopes to be used as indices of radiological contamination,
but emerging thought would suggest that radium alone might be an inadequate surrogate for
xxii
monitoring radiological activity. Beneficial reuses of fracking brines, especially those that
involve land application of brines or wastewater, are inadvisable as a result of concerns related to
potential human exposures to radionuclides and heavy metals.
After carefully reviewing the limited evidence from UNGDP impacted areas and current
scientific understanding form non-UNGDP related fields, we conclude that there is a
Moderately High Likelihood that UNGDPs impact on water quality, soil quality and naturally
occurring radioactive materials will have a negative impact on public health in Garrett and
Allegany Counties. The overall score for Flowback and Production Water related concerns are
primarily driven by concerns related to water quality and the large fraction of population relying
on well water..
6.2.4 Noise
Environmental noise associated with UNGDP was identified as a top concern among residents of
Western Maryland. The literature on UNGDP noise is very limited, however a few studies have
shown that at 1,000 to 2,000 feet from a well pad noise levels can be expected to range from 44
dBA to 76 dBA, depending on the phase. Due to a lack of information regarding compressor
stations, we conducted a small pilot study in Doddridge County, WV to understand the noise
levels associated with living near a compressor station. We found at 1,000-2,000 feet from the
compressor station noise levels were 55.78 dBA over a 24-hour period, 52.75 dBA during
daytime hours and 51.75 dBA during nighttime hours. While there are not any epidemiologic
studies on UNGDP noise, we know from other industries that long-term exposure to
environmental noise has been associated with a myriad of health outcomes, including stress and
annoyance, sleep disturbances, hypertension, and cardiovascular disease. Noise levels can be
reduced by distance, enforcement of regulatory standards, and use of sound reduction
technologies.
Based on prior evidence regarding negative impact of noise exposures and noise monitoring
results from UNGDP sites that included our own monitoring results from WV, we conclude that
there is a Moderately High Likelihood that UNGDP related changes in noise exposure will
have negative impacts on public health in Garrett and Allegany Counties.
6.2.5 Earthquakes
Recent studies suggest that an increasing frequency of earthquakes, particularly in the Central
and Eastern US may be associated with UNGDP, primarily linked with deep well injection of
wastewater. The actual process of hydraulic fracturing used for initiation of new wells produces
thousands of micro earthquakes (most too small to feel). The potential public health effects of
earthquakes related to deep well injection is a concern. However, the potential public health
effects associated with micro earthquakes resulting from hydraulic fracturing appears to be
negligible, based on current literature. There is considerable evidence that suggest earthquakes
can persist years after the start/stop of well activities. At present, it remain unclear if the
underground stress produced by horizontal drilling and hydraulic fracturing can cumulate over
space (high well density) and time to produce much more significant earthquakes in the future
years/decades, that could have a much more significant impact on public health.
xxiii
Provided that Maryland does not allow deep well injection of wastewater, we conclude that there
is a Low Likelihood that UNGDP related earthquakes will have a negative impact on public
health in Garrett and Allegany Counties.
Many of the rural communities that will be potentially impacted by unconventional natural gas
development (UNGD) operations are not fully equipped to handle the influx of industrial traffic.
Evidence from UNGDP impacted area suggest that increased truck traffic associated with
UNGDP related activities exposes residents to greater risk of motor vehicle crashes involving
injury, or even death. These communities also experience increases in violent crime, sexually
transmitted diseases, mental health problems and substance abuse. Crime statistics, disease rates,
and police accounts all suggest the introduction of UNGD operations to a community places the
local residents safety as well as the safety of the workers at risk. Most of the research

conducted
on these issues suggest there are solutions available to these problems [47].
Based on data from UNGDP impacted communities as well as previous knowledge related to
boom town, we conclude that there is a High Likelihood UNGDP related activities will have a
negative impact on the social determinants of health.
6.2.7 Occupational Health
The promise of UNGDP operations brings the promise of jobs. Yet the men and women who
work these jobs are at greater risk of harmful occupational exposures than many other industries
in Maryland. Of particular concern are exposures to crystalline silica, hydrogen sulfide, and
diesel particulate matter, as well as fatalities from truck accidents, which accounted for 49% of
oil and gas extraction fatalities in 2012. Recently reported unusually high level of UNGDP
workers exposure to crystalline silica, which is known to cause silicosis and lung cancer, is of
particular concern [30]. Evidence shows that numerous social hazards, such as mental distress,
suicide, stress, and substance abuse, have been associated with working on a UNGDP operation
due to the transient nature of the work. These social hazards also put a strain on communities, as
evidenced by increased incidence in violent crime arrests, drug violations, and sexually
transmitted infections. Based on these, we conclude that there is a High Likelihood of adverse
outcomes among UNGDP workers in Garrett and Allegany Counties.

6.2.8 Healthcare Infrastructure
Our assessment of the impact to healthcare infrastructure in Allegany and Garrett counties is
based upon RESIs estimate of an average of 1327-2825 migrant workers during the first 10
years of drilling and 151-189 migrant workers on average during the 10-year period after
drilling.
Impacts to the healthcare infrastructure are expected to be high due to a substantial increase in
the migrant workforce and population and the associated potential increase in health care
utilization in Allegany and Garrett counties. Research indicates that healthcare infrastructure
impacts will be observed when the influx of workers is the highest, during the initial years of the
project in the development phase, and that this impact will be uneven during the lifecycle of the
xxiv
project. If UNGDP workers are insured, local primary care and public health services will be
supported. However, it is unclear whether an increase in the insured population and UNGDP
revenues will lead to healthcare infrastructure development in Allegany and Garrett counties. If
UNGDP workers are uninsured, they would stress an already under-resourced healthcare
infrastructure. In addition, utilization rates for primary and public health care systems, especially
in the areas of emergency, urgent care, and trauma care, is likely to rise as a result of an increase
in the UNGDP workforce regardless of their insurance status. Because Allegany and Garrett
counties healthcare infrastructure needs are substantial (e.g. HPSA and MUA areas) and a high
number of their populations are vulnerable and because of the large number of expected long-
term migrant workers relative to population size, we predict that UNGDP would have a high
likelihood of negatively impacting healthcare infrastructure.
6.2.9 Cumulative Exposures/Risk
Exposure does not happen in vacuum. Community members impacted by UNGDP will be
exposed to multiple chemical hazards (VOCs, PM, PAHs), physical hazards (noise, radiation),
and a host of psychosocial stressors including those related to public safety, potential loss of
property values, disruption of existing social fabric, crime, among others. In addition, such
developments also disproportionately impact underserved communities such as those with low
SES, and without a strong political voice. The question of combined effect of these cumulative
exposures, as well as the interactions between chemical and non-chemical stressors needs to be
considered. While there is strong agreement in scientific community that the traditional single
chemical centric risk assessment methods are inadequate in dealing with such issues, the
emerging field of cumulative risk assessment is still in its infancy. Epidemiological and clinical
evidence from other disciplines document: 1) interactions between chemical hazards, 2)
interaction between chemical and physical hazards, and that 3) psychological stress increases
susceptibility to respiratory infections that are known to be major drivers of asthma morbidity.
Furthermore, significant evidence suggests that disadvantaged communities are
disproportionately exposed and are more vulnerable to the effect of these hazards. Based on this,
it is reasonable to assume that the combined effect of UNGDP related hazards described in this
report may be higher than the simple sum, and that the impact will be more pronounced in
disadvantaged communities and will be disproportionately felt by vulnerable subpopulations
such as property owners without mineral rights, elderly, children, and individuals with
preexisting diseases. Therefore, we conclude that there is a Moderately High Likelihood that
the UNGDP related activities will have a net negative impact in the cumulative exposure/risk.
xxv
6.3 Recommendations
6.3.1 Comprehensive Gas Development Plans (CGDP)
Potential public health impacts and prevention and mitigation strategies should be included in the
CGDP so that the required and routine public hearings on the plan can include an informed
discussion of health as well as environmental impacts.

R1. Require assessment of air quality and other potential health impacts and
propose strategies to protect the community and workers from exposure to
hazardous air pollutants.
R2. Require assessment of whether application of standard setback distances will be
adequate to protect public health, including consideration of prevailing winds
and topography.
R3. Require disclosure of planned well stimulation methods and classes and amounts
of chemicals to be used.
R4. Require a quality assurance plan.
R5. Require an air, water, and soil-monitoring plan.
R6. Require assessment of impact on and a monitoring plan for potential fugitive
emissions from existing and historic gas wells within the horizontal extent of the
fractured area.
R7. Require that all UNGDP materials and wastes be stored in closed tanks; open
pits shall only be used for storage of fresh water.
6.3.2 Disclosure of Well Stimulation Materials
Recommendations concerning disclosure were revised and moved to a separate section based on
feedback received at and following the public progress report on June 28, 2014. The final
recommendations are now in line with the proposed legislation H.B. 1030 [8]. Three phases of
disclosure are included a preliminary more general disclosure with the CGDP, a specific
detailed disclosure with the well permit application, and a specific detailed disclosure after well
stimulation is finished.
R8. Require preliminary disclosure at time of CGDP submission (see CGDP
recommendations), detailed disclosure at time of well permit application, and
detailed reporting of actual materials used within 30 days of finishing well
stimulation activities. Require notification of MDE, local emergency responders
and public notice of significant variances from materials and concentrations pro-
posed in the permit within 24-hours of occurrence.
R9. Require detailed disclosures to include CAS numbers, volume and concentration
of every chemical or distinct material including proppants, their physical form,
xxvi
and identification of engineered nanomaterials including drilling muds and
hydraulic fracturing and other fluids used in well stimulation. Do not allow
claims of trade secrets for identities and concentrations of specific chemicals or
nanomaterials used in well stimulation.
R10. Require detailed disclosures to include base fluid volume and sources including
percentages that are recycled fracturing fluid, production water, and fresh
water.
R11. Require simultaneous submission to state regulators and FracFocus.
R12. Collaborate with California to develop a State controlled and archived Internet
Web site consistent with the provisions of California SB 4.
R13. Implement the provisions of H.B. 1030 for timely access to disclosed information
by medical professionals, emergency responders, poison control centers, local
officials, scientists, and the public.
6.3.3 Air Quality
fields, we conclude that there is a High Likelihood UNGDP related changes in air quality will
have a negative impact on public health in Garrett and Allegany Counties. Should Maryland
move forward with UNGDP, the following recommendations should be implemented to prevent
or minimize potential negative impacts on public health.
R14. Require a minimal setback distance of 2000 feet from well pads and from
compressor stations not using electric motors.
R15. Require electrically powered motors wherever possible; do not permit use of
unprocessed natural gas to power equipment. This recommendation is designed
to reduce VOCs and PAHs emissions from drilling equipment and compressors.
R16. Require all trucks transporting dirt, drilling cuttings to be covered.
R17. Require storage tanks for all materials other than fresh water and other UNGDP
equipment to meet EPA emission standards to minimize VOC emissions.
R18. Establish a panel consisting of community residents and industry personnel to
actively address complaints regarding odor.
R19. Conduct Air Quality Monitoring
a. Initiate air monitoring to evaluate impact of all phases of UNGDP on
local air quality (baseline, development and production).
b. Conduct source apportionment that allows UNGDP signal to be
separated from the local and regional sources.
c. Conduct air monitoring with active input from community members in
planning, execution, and evaluation of results.
xxvii
d. Conduct air monitoring in a manner to capture both acute and chronic
exposures, particularly short-term peak exposures.
e. Clearly communicate to community members expectations about what is
achievable through air monitoring.
Based on our evaluations of the limited data available from UNGDP impacted areas, we
conclude that there is a Moderately High Likelihood that UNGDPs impact on water quality,
soil quality and naturally occurring radioactive materials will have a negative impact on public
health in Garrett and Allegany Counties. The overall score for Flowback and Production Water
related concerns are primarily driven by concerns related to water quality. Should Maryland
move forward with UNGDP, the following recommendation should be implemented to prevent
6.3.4.1 Water & Soil Quality
R20. Prohibit well pads within watersheds of drinking water reservoirs and protect
public and private drinking water wells with appropriate setbacks.
R21. Implement UMCES-AL/MDE water monitoring plan. Require monitoring of
water quality during initial gas production and at regular intervals thereafter.
R22. Implement the UMCES-AL recommendations for management and recycling of
flowback and production fluids.
R23. Require identification and monitoring of signature chemicals in fracturing
fluids to allow for future identification of ground water
infiltration/contamination.
R24. Conduct soil monitoring in areas potentially impacted by UNGD upset
conditions.
R25. Prohibit flowback and production wastewater or brine use to suppress road
dust, de-ice roads, or other land/surface applications.
6.3.4.2 NORM
R26. Conduct research to identify the appropriate suite of priority radionuclides for
assessment of radiological activity.
In the meantime, metrics such as total alpha activity, or total gamma activity should be used to
assess radiological contamination and support decision-making.
6.3.5 Noise
Based on our monitoring results from Doddridge County, WV as well as other noise monitoring
reports, we conclude that there is a Moderately High Likelihood that UNGDP related changes
in noise exposure will have negative impacts on public health in Garrett and Allegany Counties.
Should Maryland move forward with UNGDP, the following recommendation should be
implemented to prevent or minimize potential negative impacts on public health.
xxviii
R27. Implement noise reduction strategies recommended by UMCES-AL in the MD
Best Management Practices, including requiring electric motors wherever power
supplies are available and construction of artificial sound barriers.
R28. Require a setback of 2,000 feet for natural gas compressor stations using diesel
engines, 1000 feet for stations using electric motors and sound barriers.
R29. Establish a system to actively address noise complaints.
6.3.6 Earthquakes
Based on our review of literature, there is clear evidence that deep well injection of wastewater is
related to earthquakes that are greater than magnitude 3. However, earthquakes related to
hydraulic fracturing itself are very small (less than magnitude 3). Provided that Maryland does
not allow deep well injection of wastewater, there is a Low Likelihood UNGDP related
earthquakes will have a negative impact on public health in Garrett and Allegany Counties.
Should Maryland move forward with UNGDP, following recommendation should be taken into
consideration to minimize potential negative impact on public health.
R30. Collect baseline data on seismic activities using methods that can record
earthquakes smaller than magnitude 3.
R31. Restrict issuing UIC Class II permits for disposal of UNGDP fluids until
licensing requirements adequately addresses earthquake risk.
R32. Implement use of sensitive seismic monitoring technology to better detect small
earthquake activity that could presage larger seismic events as well as using a
traffic-light system that sets thresholds for seismic activity notification.
Based on our review of social determinants of health, we conclude that there is a High
Likelihood UNGDP related activities will have a negative impact on the social determinants of
health. Should Maryland move forward with UNGDP, the following recommendation should be
6.3.7.1 Traffic Safety
R33. Increase state and local highway patrols to closely monitor truck traffic subject
to the Oilfield Exemption from highway safety rules.
R34. Empower local communities to control truck speed and traffic patterns.
R35. Route truck traffic to maintain separation between UNGDP activities and the
public (such as, avoid trucking during school bus transport).
R36. Consider use of pipelines to move UNGDP fluids between sites.
6.3.7.2 Empower communities
R37. Enact a Surface Owners Protection Act as recommended in the MDE Part I
report.
xxix
R38. Engage local communities in monitoring and ensuring that setback distances are
properly implemented.
R39. Create maps using buffer zones (setback distance) to identify specific areas were
fracking should be restricted (homes, churches, schools, hospitals, daycare
centers, parks, recreational water bodies) and make these available for
community members.
Based on our evaluations of the current healthcare infrastructure in Garret and Allegany Counties
as well as expected number of migrant workers that will come to these areas, we conclude that
there is a High Likelihood UNGDP related activities will have a negative impact on public
healthcare infrastructure in Garrett and Allegany Counties. Should Maryland move forward with
UNGDP, the following recommendations should be implemented to prevent or minimize
potential negative impact on public health.
R40. Closely monitor whether prospective UNGDP companies provide adequate
health insurance coverage for all employees.
R41. Organize a local health care forum with key stakeholders to assess health care
services and anticipated needs related to UNGDP.
R42. Inform and train emergency and medical personnel on specific medical needs of
UNGDP workforce.
R43. Review and monitor county-level tax revenues and assess improvements
necessary to meet increased services need.
R44. Establish a committee of state and local stakeholders (including UNGDP officials
and local providers and residents) for early identification of impacts to
healthcare infrastructure.
R45. Initiate monitoring of UNGDP healthcare-related costs.
6.3.9 Cumulative Exposure/Risk
The combination of chemical, physical, and psychosocial stressors can lead to effects that are
cumulative involving potentially additive or multiplicative interactions among the exposures.
Observed health impacts, if any, will result from these cumulative impacts. We anticipate the
cumulative risk from the physical, chemical and psychosocial stressors will be greater than the
simple sum of individual risks. We further anticipate that the impact will be disproportionately
felt by vulnerable subgroups such as children, elderly, individuals with existing diseases, poor
residents, and individuals without mineral rights. We conclude that there is a Moderately High
Likelihood that the UNGDP related activities will have a net negative impact in the cumulative
exposure/risk.
Most of the recommendations in this report are targeted at primary prevention (i.e. to prevent the
occurrence of adverse health effects). However, a monitoring method is needed to verify the
effectiveness of primary prevention activities and to improve them as necessary. Furthermore,
secondary and tertiary prevention should not be neglected. Thus, disease surveillance and
xxx
targeted longitudinal epidemiologic studies are needed for both evaluation of primary prevention
effectiveness and as a means of providing continuing improvement of regulations. Surveillance
and epidemiologic studies will need to incorporate appropriate exposure assessment programs,
and to be most useful, need to be started immediately so as to provide comparable baseline data
should Maryland decide to move forward with UNGDP.
R46. Initiate a birth outcomes surveillance system
R47. Initiate a longitudinal epidemiologic study of dermal, mucosal, and respiratory
irritation
R48. Develop a funding mechanism for public health studies
6.3.10 Occupational Health
Based on our evaluations of the limited but emerging studies of UNGDP workers exposures to
respirable crystalline silica (frack sand) and what is known from epidemiologic and toxicologic
studies of crystalline silica (silicosis, lung cancer), we conclude that there is a High Likelihood
of adverse outcomes among UNGDP workers in Garrett and Allegany Counties. Should
Maryland move forward with UNGDP, the following recommendations are made to prevent
most and minimize residual potential negative impacts on occupational health.
R49. Require implementation of NIOSH and OSHA recommended controls for silica
exposure in UNGDP operations.
R50. Provide MOSH with resources to regularly inspect UNGDP workplaces and
monitor worker exposures.
R51. Establish community outreach programs to help transient workers feel more
welcome in the community as a means of reducing rates of depression, suicide,
and drug use.
R52. Require employers to provide employee assistance programs including
counseling and substance abuse treatment.
1
7 INTRODUCTION
On June 6, 2011, Governor Martin O'Malley issued Executive Order 01.01.2011.11, establishing
the Marcellus Shale Safe Drilling Initiative (Initiative). The Initiatives purpose is to assist state
policymakers and regulators in determining whether and how gas production from the Marcellus
Shale and other shale formations in Maryland can be accomplished without unacceptable risks to
public health, safety, the environment, and natural resources. On October 18, 2013, the Maryland
Department of Health and Mental Hygiene (DHMH) signed a MOU with the Maryland Institute
of Applied Environmental Health (MIAEH) at the University of Maryland, College Park to
evaluate the potential public health impacts associated with drilling in the Marcellus Shale in
Maryland. The study, as outlined in the MOU, will include:
Detailed Scoping, including timetable for remaining deliverables, methods, and public
input to determine study objectives.
Baseline Assessment of current regional population health, including demographics,
causes of morbidity and mortality, local health priorities, vulnerable populations, local
healthcare and social service infrastructure.
Impact Assessment of the potential exposures, including hazards and known health
impacts both directly and indirectly associated with hydraulic fracturing, assessment of
current exposures and data gaps prior to onset of hydraulic fracturing.
Final Report, which will include the study findings, monitoring and assessment
recommendations, and public health response and mitigation strategies.
7.1 Health Impact Assessment Process and the Public Health Study
Health Impact Assessment (HIA) is a tool that is designed to support decision and policymaking.
HIA combines array of data sources, analytic methods and input from stakeholders including
community members to determine if a proposed policy, plan, program, and/or decision has the
potential to impact the health of the community, and how these effects are distributed within
population subgroups that differ by geography, SES, and demographic characteristics [9]. This
information is then fed back to the policymakers to help them make an informed decision on the
pending policy, plan, program and/or decision.
HIA is not a quantitative risk assessment, rather it provides information that is qualitative in
nature that can be used to assess whether and how community wellbeing may be impacted, both
directly and indirectly. It consists of 6 steps:
1. Screening: Initial step to determine the need for HIA.
2. Scoping: With community input, identify the most important hazard and health impact to
focus on.
3. Assessment: Analyze the baseline characteristics of the population and provide
anticipated potential effects.
4. Recommendations: Based on the assessment, develop recommendations for minimizing
health effects, and approaches for monitoring.
5. Reporting: Prepare a report for the decision makers, disseminate the findings and
recommendations to all the stakeholders including community members.
6. Monitoring and Evaluation of the HIA Process: Evaluate if the HIA process helped the
decision making process.
2
This public health study, which draws upon several elements of the HIA, including scoping,
baseline assessment, impact assessment as well as reporting, was conducted to inform the
Marcellus Shale Safe Drilling Initiative Advisory Commission, State legislators and the
Governor about potential health impacts associated with UNGDP related activities so they make
an informed decision that takes into account the health and well-being of Marylanders. Should
Maryland decided to move forward with UNGDP, this report provides set of recommendations
that will minimize negative impact on public health. This public health study does not address
economic benefits associated with UNGDP as these issues are addressed in a separate Economic
Report prepared for the commission by RESI [10]. As outlined in the National Academy of
Science Report, quantitative risk assessment is beyond the scope of HIA. As such, this study did
not conduct a formal quantitative risk assessment [9].
7.2 Natural Gas Development & Production
7.2.1 Conventional vs. Unconventional Natural Gas
While increasing domestic production of natural gas provides economic growth and jobs, there is
concern that new extraction technologies could negatively impact public health, safety, the
environment, and natural resources. There are several key differences between conventional
natural gas development and production (CNGDP) and unconventional natural gas development
and production (UNGDP). CNGDP requires vertical drilling and hydraulic fracturing, while
UNGDP uses new horizontal drilling techniques along with hydraulic fracturing. There are also
other, less common alternative well stimulation technologies sometimes combined with
horizontal drilling, including acid well stimulation. This report is focused on horizontal drilling
and hydraulic fracturing. The direction and length of the lateral section of the well can range
from 4,000 to 5,000 feet [11]. The additional horizontal drilling leads to more cuttings a
mixture of coarse chips and finer particles of rock that are produced as the well is drilled that
have to be removed. The use of horizontal drilling requires a larger well pad to accommodate
increased on-site storage of equipment and fluid [11]. CNGDP requires up to three acres per well
pad, while UNGDP requires up to six acres [11]. UNGDP also requires significantly more time,
water, and fluid CNGDP lasts approximately one month and uses up to 80,000 gallons of
water, while UNGDP lasts approximately three months and uses up to four million gallons of
water [11]. The water is mixed with a mixture of chemicals and natural or manufactured sand
grains used to hold open the fractures created during hydraulic fracturing called proppants [11].
This mixture often referred to as fracking fluid, is forced into the gas-bearing rock under
intense pressure to fracture the rock proximate to the wall. These fractures form pathways so that
the natural gas can be released and captured.
7.2.2 The Marcellus Shale
The Marcellus Shale gas formation is abundant in natural gas resources. It is one of the largest
shale regions in the United States; covering over 95,000 square miles and 4,000-8,000 miles feet
depth [12]. This 400-year-old rock contains more than 410 trillion cubic feet of natural gas and is
found beneath the surface of Pennsylvania, Ohio, West Virginia, New York and Western
Maryland [12]. The Marcellus Shale contains both dry and wet gas. Marylands area of
Marcellus Shale is composed of mostly dry gas, which is composed almost entirely of methane,
while wet gas contains not only methane, but also natural gas liquids (NGLs) including ethane,
3
butane and propane. There is greater demand for wet gas due its versatility as a fuel and use as a
feedstock for plastics and other petrochemical production [11]. The revenue generated from
these NGLs sales counterbalance the low price of natural gas, leaving dry gas drilling less
popular and profitable. Thus, it seems likely that the demand for natural gas from Marylands
portion of the Marcellus Shale will not be sufficient to attract significant investment until and
unless the price of natural gas increases significantly. [13]
7.2.3 Terminology
As discussed by Shonkoff and colleagues, there is some confusion regarding terminology [14].
We will be using the term UNGDP to refer to the entire process, from well pad construction to
pipeline development. High-volume hydraulic fracturing (HVHF) refers to the well completion
stage, when a mixture of water, chemicals, and proppant are injected into the well at high
pressure. Unconventional natural gas development (UNGD) includes well pad preparation,
vertical and horizontal drilling, and well completion. Well completion includes completion
transition, hydraulic fracturing, and flowback. Unconventional natural gas production (UNGP)
includes pipeline development and operation of compressor stations. Detailed activities involved
in this process are depicted in Figure 7-1. [15]

Figure 7-1: Major Activities Associated with UNGDP Process [15]
7.2.4 Unconventional Natural Gas Development (UNGD)
Once an applicant is granted the right to extract natural gas, a preliminary Comprehensive Gas
Development Plan (CGDP) has to be prepared including all exploration and production
activities. According to Marylands Best Management Practice, a CGDP is designed to address
the larger, landscape-level issues and cumulative effects, which offers significant benefits to both
the industry and the public [16]. A CGDP will be mandatory in Maryland, once the adopted best
4
management plan is implemented in regulations, and will serve as a prerequisite to an application
for a well permit. The CGDP must contain locations for well pads, roads, pipelines, and other
ancillary equipment, which precedes an individual well permit of more detail and explanation of
each activity, step and process. Engineering, design, and environmental plans must meet or
exceed the standards set by the departments in order for a company to develop the CGDP in a
process that allows for public participation. A completed CGDP, approved by the State is
effective for ten years and enable provisions for individual wells to be made.
The development phase of a well requires pad preparation, drilling, and well completion. UNGD
in the Marcellus Shale uses high volume, slickwater, horizontal hydraulic fracturing. A vertical
well is drilled along with a horizontal or lateral extension. These horizontal, gas-bearing layers
require a higher volume of water to fracture the shale. In order to increase the speed of fluid
inside the well, a combination of chemicals, also known as slickwater, are added to the water to
adjust viscosity or thickness [11]. Marcellus wells are drilled and cased in multiple stages.
Before the process begins, there must be an established well pad, large enough to support all
equipment needed for drilling. In the interest of horizontal drilling, a special drill bit is used to
turn the drill at a predetermined depth, known as the kickoff point. This point is reached after
1000 feet of drilling in order to fully turn a well horizontal. Multiple barriers of steel casing and
cement are installed to protect the hole from collapsing as well as the escape of drilling fluids
and gas from the side of the well. As the hole is drilled deeper and cemented in place, guiding
shoes on the ends of the casing assist in the lengthening of casing down the well safely until it
reaches the annulus, or space between the casing and the drilled well. A wiper plug forces
cement out of the well bore, cleans the inside of casing walls and separates the cement from
additional drilling muds, or lubrication for easier and faster drilling [11].
There are several layers of casing conductor pipe, surface, intermediate, and production that
are designed to protect against environmental contamination. A conductor pipe has the largest
diameter and prevents the top of the well from collapsing and exchange of fluids like water and
gas. This provides a path for drilling muds. Conductor pipe casing also has blowout preventers,
which regulate erratic pressure changes that can be found while drilling. The next three levels of
casing are surface, intermediate and production casing. Surface casing, similar to the conductor
pipe prevents contamination of groundwater water by drilling muds and also preventing sediment
from caving into the well [11]. Intermediate casing regulates potential problems like abnormal
pressure from shallow gas pockets. Production casing is the thinnest in diameter and runs
through the length of the well to isolate the zone containing natural gas from subsurface
formations [17]. Once the casing is in place, perforation guns positioned in the lateral part of the
well create punctured holes for natural gas to flow into the well. The guns small projectile shots
are steel piercing bullets that punch through the steel casing and cement-filled annulus. The
fracking process last three to five days, while allowing up to ten fractures per well and between
six to ten drillable wells per well pad. After the well has been completed, the gas company
performs a series of shut-ins or pressure tests on the well. These tests take another three days,
in order to assess the proper functioning of the drilling and casing. A device referred to as the
Christmas Tree is placed on top of the well surface to allow gas to be pumped into production
pipelines, seal gas in the well in an emergency and monitor production [11]. After the surface
facilities are installed, the well is placed on production for 20 to 30 years, during which time the
well may be refractured several times in an effort to increase production.
5

Figure 7-2: UNG Compressor Station
Following completion, various procedures are conducted to measure the effectiveness of the
hydraulic fracturing process. A combination of micro-seismic mapping and measurements are
taken to identify temperature, production and video imaging logging [11]. Such techniques are
costly, however, the information collected is used to evaluate fractures near the wellbore. Since
horizontal hydraulic fracturing requires large quantities of water, it also needs chemicals, ranging
from benign to toxic. The various chemicals mixed with water create a fracturing fluid, of which
a portion returns to the surface. The flowback water at the surface contains fracturing fluids as
well as various amounts of heavy metals, salts, and naturally occurring radioactive material
found in the gas-bearing unit [11]. This fluid is a concern for the Marcellus Shale due to the
handling, treating and disposal of the contaminated water. Current options for disposal include
deep fluid injection wells or onsite water treatment and recycling. Deep fluid injection consists
of injecting contaminated water into a deep, impermeable formation, where it is stored
permanently; recycling water restores and prolongs its use for future hydraulic fracturing jobs
[11].
7.2.5 Unconventional Natural Gas Production (UNGP)
During the production phase, [18]natural gas travels from the well, where liquids and gases are
separated, through a network of pipes and field compressor stations that serve as a gathering
system. A processing facility is frequently required to remove impurities such as hydrogen
sulfide, helium, carbon dioxide, hydrocarbons, and water vapor that were not removed at the well
head. Once these impurities are removed, the gas is pumped into large high pressure interstate
pipelines.
6
8 SCOPING UPDATE
Our Draft Scoping Report (http://www.marcellushealth.org/detailed-scoping-report.html) was
released for public comment on December 23, 2013. We received 46 comments from concerned
residents, environmental advocacy organizations, and the industry and 2 reviews from external
experts recruited by DHMH to provide input. After carefully considering all of the input, we
made changes to our projects timeline and to the baseline health and health impact assessments.
We altered the timeline for the study as follows: First, we incorporated the baseline health
assessment, impact assessment, and recommendations into a single final report, rather than
issuing a separate baseline health assessment earlier in the process. This provided additional time
to develop the baseline health assessment. Second, we presented a progress report with a
summary of our findings and recommendations at a community meeting in Western Maryland on
June 28, 2014. Third, this final report will be released July 2014 to allow for a public comment
period. All comments on this report will go directly to the Marcellus Shale Safe Drilling
Advisory Commission for consideration, along with comments from external reviewers arranged
by DHMH. This report will not be revised.
We revised the terminology of the report to be more inclusive of the entire development and
production process. Thus, we now refer to unconventional natural gas development and
production (UNGDP) as described in section 7.2.3. We also recognize that well stimulation can
involve other technologies besides hydraulic fracturing and make recommendations accordingly.
We also developed a regional group of counties including the surrounding counties in
Pennsylvania and West Virginia to provide relevant comparisons for data for Garrett and
Allegany counties. We also added several indicators to our baseline health assessment. These
include:
Violent and non-violent crime
Public water service areas
Sources of public water
Water quality of public and private water
Suicide rates
Drug and alcohol addiction
Smoking
Overall mental health status
The impact assessment plan was expanded to include:
Use of brine or flowback on roads
Impact of compressor stations, excluding Cove Point LNG plant associated
Soil contamination
Radon and naturally occurring radioactive materials
Proppants with airborne qualities,
Specific literature recommended by commenters was reviewed. A complete list of literature can
be found at www.marcellushealth.org/resources.
Through our scoping process, community members clearly identified climate change as one of
the issues of concern to them. Fugitive emission of methane, which can occur throughout the
production and distribution process, can significantly contribute to climate change and climate
7
change is considerable threat to public health. However, a different study team would be required
to assess the climate tradeoffs inherent in using shale gas as a transition fuel. This report is
focused on health impacts that are primarily restricted to the local area where UNGDP takes
place.
8
9 BASELINE HEALTH ASSESSMENT
9.1 Introduction
A robust understanding of the health trends and issues currently affecting a community is an
important step in the HIA process. This public health study provides the baseline community
health information needed to fully evaluate potential impacts to human health from UNGDP.
Baseline health of a community can be estimated by examining a wide range of health indicators
including vulnerable populations, chronic and non-chronic disease, major causes of morbidity
and mortality, environmental health, social determinants of health, and healthcare infrastructure.
Factors such as age, genetics, behavior, educational attainment, family income, poverty status,
access to quality healthcare, proximity to hazards, and environmental exposures can influence
individual health status. The Marcellus Shale Commission requested the baseline health
assessment of Garrett and Allegany Counties prior to UNGDP activity in the region.
In order to assess the baseline health of Allegany and Garrett County residents, we considered
demographics, potential vulnerable populations, a wide range of health indicators, environmental
health, social determinants of health, and healthcare infrastructure. We used county level and
census tract level statistics for the baseline health assessment. When possible, data for Allegany
and Garrett counties was compared to the health data of the region (Allegany and Garrett
Counties in Maryland; Bedford, Fayette, and Somerset Counties in Pennsylvania, and Grant,
Hampshire, Mineral, Preston, and Tucker Counties in West Virginia), and the State of Maryland
for an overall baseline health profile.
The full baseline health assessment profile for Allegany and Garrett counties is available in the
Appendix.
9.2 Overview of Allegany and Garrett Counties
Allegany County with a population of 75,087 individuals is located in the northwestern part of
Maryland and is 424.16 square miles. Positioned in the Ridge-and-Valley Country of
the Appalachian Mountains, it is bordered to the north by the Mason-Dixon Line along
with Pennsylvania. To the south, it is surrounded by the Potomac River and West Virginia. To
the west is the Allegheny Front, and to the east is Frostburg, MD.
Garrett County is the western-most county in Maryland, and its bordered to the north by
the Mason-Dixon Line with Pennsylvania, to the south by the Potomac River and West Virginia.
Garrett County with a population of 30,097 individuals is 647.10 square miles of incorporated
and unincorporated jurisdictions. Garrett County has over 76,000 acres of parks, lakes, and
publicly accessible forestland. Nicknamed Marylands Mountaintop Playground," the county
has the states highest elevation at 3,360 feet, as well as its largest inland body of water (Deep
Creek Lake). Garrett County is home to the state's only sub-arctic wetlands and is the only
county in the state to produce natural gas. There are approximately 153 churches, 87 schools, and
3 hospitals in both counties.
9.3 Demographics
As of 2012, 50.4% of the population in Garrett County were female and 49.6% were male;
27.1% of the population were under the age of 18, while 17.7% of adults were 65 years and
9
older; 97.2% of the population identified themselves White, 1% as African-American, 0.8 % as
Hispanic and 1% as other; 3.7% of the population were unemployed and 13% of the residents
were living below the federal poverty line. The median income for Garrett County residents is
approximately $45,354, which is higher than the regional average of $39,026 but much lower
than the Maryland state average of $68,559. In Allegany County, 48% of county residents were
female, and 52% were male. In addition, 18% of the population were under the age of 18, while
18.1% were 65 years and older. For those who reported their race, 88.3% identified themselves
as white, 7.6% as African-American, 1.5% as Hispanic and 2.6% as other. Approximately 16.1%
of Allegany residents live at or below poverty. The median income in Allegany County is
$39,087, which like Garrett County, is slightly higher than the regional average but much lower
compared to the Maryland state average. When comparing Allegany and Garrett Counties,
Garrett County had the highest number of residents with less than a high school education (15%).
9.4 Vulnerable Populations
Garrett and Allegany counties. Vulnerability is commonly defined as how individuals or groups
of individuals or organisms respond to and recover from stressors inadequately or not as well as
the average [19, 20]. Factors that contribute to vulnerability include characteristics at the
individual and/or community levels that moderate the effect of environmental hazards on
community health and well-being, and can be demographic, biological, social, and behavioral.
Demographic factors of interest when assessing vulnerability include race, ethnicity, age, and sex
[7]. Biologic factors include genetic make-up and pre-existing medical conditions; pre-existing
conditions have been associated with reduced response to stressors. Other individual level
vulnerability factors such as low socio-economic status, low educational attainment, and
psychosocial stress have also been associated with negative health outcomes. Health behaviors
play a role in increasing or decreasing an individuals vulnerability. In this study, we are limited
to assessing vulnerability using sociodemographic data and county-level health data. We were
unable to obtain individual health data including family history of disease for populations in both
counties.
9.5 Health Indicators
9.5.1 Environmental Health
A large proportion of Marylanders currently rely on unregulated private wells as sources of
drinking water. An estimated 1 million Maryland residents draw drinking water from private
wells [21]. Elevated levels of nitrates and other chemicals have been noted in Marylands
groundwater [22, 23]. In Garrett County, private wells are concentrated most heavily around
McHenry, Grantsville and Oakland. Over 14,200 well location records are currently available for
the county. Approximately, 8,250 or 58% of well records occur in grid cells that contain
Marcellus Shale gas leases. Annual average PM
2.5
concentrations were ~13 "g/m
3
in both
Allegany and Garrett counties. These mean levels were higher than the mean concentrations for
the state of Maryland as a whole. Scientific literature has shown relationships between PM
exposure (e.g., coarse or fine particles, acute or chronic) and increased respiratory and
cardiovascular health end points including increased mortality, hospital admissions, and
emergency department visits.
10
9.5.2 Physical Health Indicators
The health profile of the residents of this region was compiled by using data collected on overall
life expectancy, poor physical days, preventable hospital stays, chronic diseases, major causes of
morbidity and mortality, and birth outcomes. Data for Allegany and Garrett counties was
compared to the health data of the region (Allegany and Garrett Counties in Maryland; Bedford,
Fayette, and Somerset Counties in Pennsylvania, and Grant, Hampshire, Mineral, Preston, and
Tucker Counties in West Virginia), and the State of Maryland for an overall health profile.
9.5.2.1 Life Expectancy
Data on life expectancy was obtained from the CDCs Community Health Status Indicators
website. Garrett County has the highest average life expectancy (78.2), compared to Allegany
County (77.4), the state of Maryland (67.8), and the region (76.7).
9.5.2.2 Poor physical health days
Data on poor physical health days was obtained from the Behavioral Risk Factor Surveillance
System (BRFSS) for the years of 2006-2012. Allegany County residents had a higher number of
poor physical days (4.8) than those in Garrett County (3.7). Both counties had higher numbers
than those for the State of Maryland (3.1).
9.5.2.3 Preventable hospital stays
We obtained data from the University of Wisconsin County Health Indicators Project for 2011.
The Ambulatory Care Sensitive Conditions (ACSC) rate for preventable hospital stays in
Allegany (88.0) and Garrett (67.6) counties was higher than the overall state rate (60.2). The
ACSC rate for Allegany was higher than the rate for the both the region (85.6) and Garrett
County.
9.5.2.4 Chronic Diseases
9.5.2.4.1 Adult Hypertension
We obtained data on adults with high blood pressure for 2006-2012 from the Behavioral Risk
Factor Surveillance System (BRFSS). Both Allegany and Garrett counties had higher
percentages of adults with high blood pressure (37% and 31% respectively) when compared to
the State of Maryland (30%). Compared to the region (34.3%), Allegany County has a higher
percentage of adults with high blood pressure while Garrett Countys percentage was lower.
9.5.2.4.2 Adult Obesity and Diabetes
Adult obesity and diabetes data were obtained for years 2006-2012 from BRFSS. The
percentages of obese adults in Allegany and Garrett counties were 21% and 30%, respectively,
while, the percentages with diabetes in Allegany and Garrett counties were 12% and 11%,
respectively. These trends mirror each other since obesity has been linked to the development of
Type 2 diabetes [24]. While both counties have lower percentages of obese adults and either
equal or lower percentages of diabetic adults compared to the region (12%), they are both higher
compared to the state of the Maryland (9.7%).
11
9.5.2.4.3 Adult Smoking
We obtained adult smoking data for the years 2006-2012 from BRFSS. In both Allegany (23%)
and Garrett (19.5%) counties, the percent of adults who smoke was much higher than the percent
of adults who do the same across the state (15.4%). However, only the smoking rate for Allegany
was higher than the smoking rate for the region (22.7%).
9.5.3 Major Causes of Morbidity and Mortality
9.5.3.1 Cancer
We obtained cancer incidence data from the National Cancer Institutes (NCI) State Cancer
Profile site (2006-2010) and cancer mortality data from CDC Wonder (2000-2010) on non-
Hodgkin's lymphoma, multiple myeloma, leukemia, malignant melanoma of skin, malignant
neoplasm of breast, malignant neoplasm of prostate, malignant neoplasm of bladder, and
malignant neoplasms of colon, rectum and anus. Prostate cancer, breast cancer, and colorectal
cancer were the cancers with the highest incidence rates. The top three cancers in Allegany and
Garrett counties that result in the highest rates of deaths were colorectal, breast, and prostate
cancers. Death rates from these cancers were higher in these counties compared to the region and
the state of Maryland. Overall, the total cancer death rate in Allegany County (196.1 per
100,000) were higher than that of Garrett County (174.9 per 100,000), the region (191.1 per
100,000), and the State of Maryland (194 per 100,000).
9.5.3.2 Other Mortality Data
9.5.3.2.1 Chronic respiratory disease deaths
We obtained data on chronic respiratory deaths from CDC Wonder. The death rates in Allegany
(54.5 per 100,000) and Garrett counties (51.4 per 100,000) due to chronic respiratory disease
were higher than those for the region (47.8 per 100,000) and the state (37.1 per 100,000).
9.5.3.2.2 Flu deaths
We obtained data on influenza and pneumonia mortality from CDC Wonder. The death rates
attributed to flu in Allegany County (17.2 per 100,000) was higher than those for Garrett County
(14.5 per 100,000), yet both were lower than the death rate from flu for the state (20.1 per
100,000).
9.5.3.2.3 Cardiovascular disease deaths
We obtained heart disease mortality data using CDC Wonder. Cardiovascular disease mortality
rates for Allegany (275.6 per 100,000) and Garrett counties (253.9 per 100,000) were much
higher than the rates for the region (249.5 per 100,000) and the state (216.5 per 100,000).
9.5.3.2.4 Cerebrovascular disease deaths
We obtained data on cerebrovascular disease mortality from CDC Wonder. The rate of stroke-
related mortality for Allegany County (59 per 100,000) was higher than the mortality rates for
Garrett (49.6 per 100,000), the region (53 per 100,000), and the state (46.7 per 100,000).
12
9.5.3.2.5 Septicemia deaths
Data on Sepsis (septicemia) mortality was obtained through CDC Wonder. Septicemia is an
illness that affects all parts of the body that can happen in response to an infection and can
quickly become life-threatening. People with weakened immune systems, infants and children,
elderly citizens, and people with chronic diseases are at risk from this condition. We found that
the septicemia mortality rate for Allegany County was 20.8 per 100,000. This rate is twice as
high as the rate of Garrett County (10 per 100,000) and also higher than the rates of the region
(12.9 per 100,000) and the state (18.8 per 100,000).
9.5.3.2.6 All-Cause mortality
All-Cause mortality rates for Allegany (853 per 100,000) and Garrett (808 per 100,000) were
higher than the rate for Maryland (768 per 100,000).
9.5.3.3 Birth Outcomes
9.5.3.3.1 Low birth weight and premature births
We obtained data on percent low birth weight (< 2800 grams) and infant mortality for Allegany,
Garrett, the region, the state of Maryland, and the US from the Health Indicators Warehouse and
National Vital Statistics System (2006-2012). Percentage of babies born with low birth weight
(LBW) in Allegany (9.1%) was higher than % low birth weight for Garrett (7.5%), MD (9.0%),
region (8%), and the United States (8.2%). The percentages of premature births for Allegany
County (13%) were higher than the rates for Garrett (12%), the region (11.6%), the State of
Maryland (12.9%), and the United States (12.2%).
9.5.3.3.2 Infant mortality
We obtained data on infant mortality for Allegany and Garrett counties, the region, the state of
Maryland, and the US from the Health Indicators Warehouse and National Vital Statistics
System. Infant mortality rates of 8.4 deaths/1000 births (Allegany) and 10.8 deaths/1000 births
(Garrett) were higher than the rates for MD (7.2 deaths/1000 births), and the US (6.9 deaths/1000
births).
9.6 Social Determinants of Health
To evaluate the baseline social determinants of health in Allegany and Garrett counties, we
obtained available information regarding sexually transmitted infections (STI), crime, injuries,
mental health, and substance abuse from a variety of sources, as summarized in the Appendix.
Data regarding STIs was obtained from the Health Indicators Warehouse for years 2010 (HIV)
and 2011 (gonorrhea) and from the 2011 County Health Rankings (chlamydia). STI rates,
specifically chlamydia, gonorrhea, and HIV, in this area are much lower compared to the rest of
the state.
Information regarding violent and property crime was obtained from the Maryland Governors
Office on Crime Control and Prevention Crime Statistics Report for 2000, 2005, and 2010. Data
regarding homicides was obtained from County Health Rankings and the National Center for
Health Statistics for 2010. In Garrett County, crime rates across all categories remain steady and
lower than the Maryland State averages, fluctuating slightly over the 10-year period between
2000 and 2010. In Allegany County, there is a slow but steady increase in most crime categories
13
in this same period. This increase is in contrast to statewide trends, which demonstrate major
decreases in crime rates across all categories in the last decade, from 2000 to 2010. Homicide
rates, as reported in the County Rankings Data shows that rates in both counties are quite low,
much lower than the Maryland State average of 9.3 homicides per 100,000.
Data for deaths resulting from unintentional injuries were obtained from Health Indicators
Warehouse, National Vital Statistics System for the years 2006-2010. Both Allegany and Garrett
counties have much higher total mortality rates from unintentional injury than the Maryland
State average and both are slightly higher than the national average. Mortality from motor
vehicle traffic deaths was also higher for both counties than the State of Maryland average.
Information on alcohol impaired driving deaths was obtained from the 2014 County Health
Rankings Information and the Fatality Analysis Reporting System. The percentage of driving
deaths that were a result of alcohol impairment is lower for Allegany County (29%) and higher
for Garrett County (41%) than the State of Maryland average (33%).
Data on suicide including intentional self-harm by discharge of firearms and intentional self-
harm by other and unspecified means and their sequelae were obtained from CDC Wonder
Mortality from 2000-2010. The total mortality rate from intentional self-harm (suicide) for
Allegany and Garrett counties were significantly higher than the State average.
Data on mental health specific to residents of Allegany and Garrett counties were obtained
through the County Health Rankings Database and the Health Indicators Warehouse from 2006-
2012. Mental health was measured by the number of reported mentally unhealthy days per month
among adults over age 18. In the period 2006-2012, adults in Allegany and Garrett counties
reported slightly higher mentally unhealthy days per month than the Maryland state average. A
related measure on the perceived availability of social-emotional support was obtained through
the Behavioral Risk Factor Surveillance System for 2006-2012. Adults in Allegany County had
lower rates of perceived social and emotional support (18.7%) while adults in Garrett County
had slightly higher rates 20.0% than those for Maryland as a whole (19.8%).
Substance abuse data were extracted from the Health Indicators Warehouse, with measures for
adult binge drinking and excessive drinking, collected from the period 2006-2012. The
Behavioral Risk Factor Surveillance System was used for self-reported data on binge drinking
1

and excessive drinking
2
. Both counties report slightly higher rates when compared to Maryland
State averages (14.4% binge drinking and 15.7% for excessive drinking), wide margins of error
could account for these differences. Information on other types of substance abuse were more
difficult to obtain.
9.7 Healthcare Infrastructure
To assess the healthcare infrastructure of Allegany and Garrett counties, the team obtained
information regarding rates and ratios of primary care physicians, dentists, and mental health

1
Sample respondents age 18+ who drank 5 or more drinks for men, 4 or more drinks for women, at one or more
occasions in the past 30 days [286].
2 Sample respondents age 18+ who drank more than two drinks per day on average (for men) or more than one
drink per day on average (for women) or who drank 5 or more drinks during a single occasion (for men) or 4 or
more drinks (for women) during a single occasion [286].
14
providers to the population from the 2014 County Health Indicators and the Health Resources
and Services Administration (HRSA) Area Resource Files. Rates and ratios of these service
providers are, on average, much lower than the statewide averages, especially for mental health
providers, indicating a critical shortage of providers in both Allegany and Garrett counties.
According to HRSA, Allegany County is a designated Health Professional Shortage Areas
(HPSA) for primary care for low-income populations, mental health care for Medical Assistance
populations, and dental care for Medical Assistance populations. Allegany County has a critical
need for specialty providers including vascular surgery, urology, as well as dentists willing to
provide care for adults with no insurance or Medical Assistance. Garrett County is a designated
HPSA for primary and mental health care, and dental care for Medical Assistance populations.
Furthermore, all of Garrett County is considered a Medically Underserved Area (MUA), while
substantial portions of Allegany County (Orleans, Lonaconing, Oldtown, and Cumberland) also
qualify as MUA.
The team also obtained information on insurance status of individuals living in Garrett and
Allegany counties from the County Health Rankings Database. As of 2011, 11.9% of the total
population of Allegany County and 14% of Garrett County were uninsured; these are similar to
and higher than statewide averages (12%).
15
10 IMPACT ASSESSMENT
10.1 Overview of Key Determinants of Human Exposures to UNGDP
Related Hazards
This section provides a brief overview on key determinants of human exposures to UNGDP
related hazards, which is integral to understanding the relationship between exposures and
adverse health outcomes.
The fate and transport of hazards emitted from a source depend upon several factors including
chemical and physical properties of the hazard as well as meteorological conditions, local
topography, and source characteristics. Depending upon these factors, hazards released from the
UNGDP process may end up in several media (air, dust, water, food and/or soil). For example,
hazards that are volatile or very small in size may end up in air, while those that are non-volatile
and/or larger in particle size end up in the soil. The presence of these hazards in the environment
also depends upon their half-life as well as the extent to which the hazard under consideration
interacts with other hazards.

Figure 10-1: Source to Effect Continuum for a Typical Environmental Hazard
Residents from impacted communities will be exposed to hazards through inhalation if the
hazard is present in the air or in dust. If the hazard is present in dust, food or water, individuals
will be exposed through ingestion. While less common among adults, ingestion of contaminated
dust can be a major driver of exposure among young children as they explore their world through
P
o
t
e
n
t
i
a
l

D
o
s
e
/
P
o
t
e
n
t
i
a
l

E
x
p
o
s
u
r
e
O
b
s
e
r
v
e
d

D
o
s
e
Activity / Contact / Physiology
Pathways
Of Exposure
Routes
Of Exposure
!
"
#
$
%
$
&
'
(
"
)*+&
,'-
.((/
0$&1-
2('%
2
(
*
-
3
1
!
"
4
1
+
&
'
(
"
)
1
-
5
$
%
Critical information
currently lacking
16
hand to mouth activities. Individuals may get exposed to hazards present in the soil or water
through dermal routes. Dermal exposure can be substantial if the hazard under consideration is
lipophilic (having an affinity for fat or oil). Exposure can be further enhanced if the integrity of
the skin surface that comes into contact with the hazard is compromised.
When contact is established between a hazard and external body surface, there is a potential for
human exposure. The magnitude of this potential exposure/dose depends upon:
1. Concentration of the hazard in the environment,
2. Frequency of exposure to the hazard,
3. Duration of exposure to the hazard.
Concentration of the hazard in the environment is directly related to the source activity,
meteorological conditions, topography, atmospheric chemistry, and the half-life of the hazard in
the environment. Frequency of exposure is related to the contact rate between the hazard and the
external body surface, and depends upon the exposure pathways. For example, if the hazard is
present in the air, then exposure is continuous, but if the hazard is in the water and the only route
of exposure is dermal, then frequency of exposure may be once a day or every other day
depending upon how often the individual showers/takes a bath. Duration of exposure reflects the
length of contact between the hazard and the external body surface. For example, if the hazard
under consideration is related to well development, then the duration of exposure may last a few
months while the wells are being drilled. But if the hazard is related to compressor stations
(production phase), then the duration of exposure can be decades, as the compressor stations are
in service for a long period of time. Duration of exposure may also differ by population
subgroups. For instance, workers from out of town may be exposed to the air pollutants for 8
hr/day during the workday, while community residents may be exposed 24hr/day if they work
and reside in the area.
There are additional factors that may modify an individuals exposure. These include individual
level activities, lifestyles and physiological factors. For instance, an individual who leads a very
active life may breathe a higher volume of air compared to someone who leads a sedentary
lifestyle, and in doing so, may be exposed to a higher level of hazards present in the air.
Likewise, overall dermal uptake of a hazard may be considerably higher among individuals
whose skin integrity is compromised due to old age, open wounds or dry skin. In addition, the
potential to detoxify hazards varies across individuals based on their genetic makeup.
Information regarding potential dose and/or observed dose is desirable while evaluating the
impact of UNGDP on public health. Currently such individual level measures of exposure are
lacking Figure 10-1. Available information is restricted to selected media within exposure
pathways (air, dust and water).
10.1.1 Overview of Exposure Assessment Methods for UNGDP Related Hazards
There may be considerable variability in the concentration of UNGDP related hazards in
different microenvironments within one location. Similarly, there likely is variability in the
concentration of hazards related to UNGDP processes across different geographic locations.
Finally, this variability is not constant (i.e., they change from day to day). Since individuals
spend their time moving from one microenvironment to another, it is important to capture the
spatial and temporal variability in concentrations while conducting exposure measurements for
17
epidemiological studies. There are several approaches available for quantifying an individuals
true exposure, with varying accuracy.
Table 10-1: Ranking of Exposure Assessment Methods

Table 10-1 provides a summary of exposure assessment methods that can be used for quantifying
residents exposures to UNGDP related hazards. These methods are ranked in decreasing order
based on their ability to approximate an individuals true exposure. For example, biomarker
(biological samples) is ranked highest because it provides a good estimate of individuals total
exposure that may be coming from inhalation, ingestion or dermal routes of exposure. Personal
air samples, on the other hand, capture inhalation exposures only, and thus may miss exposures
taking place through the dermal and ingestion route of exposure. In addition, a biomarker
indicates that the toxicant has already gotten into the human body (internal dose) while personal
air samples indicate potential exposure/potential dose. Area level samples are less desirable than
individual level measurements as they do not account for the variability that exists between
individuals (between person variability). Likewise, surrogates (e.g. distance of home from well
!"#$ &' ()#&*+,$ -./.
0##,&)12./1&3 &'
!,+$ ()#&*+,$
4351615+.7 7$6$7 2$.*+,$2$3/*
81&2.,9$, 8$*/
:$,*&3.7 01, ;.2#7$*
-$,2.7 ;.2#7$*
0,$. <$6$7 ;.2#7$*
435&&, .1,=5+*/ *.2#7$* ',&2 ,$*15$3>$
435&&, .1,=5+*/ *.2#7$* ',&2 ?&,9#7.>$
01, *.2#7$* ',&2 3$1@AB&,A&&5=>$3/,.7 *1/$ 2&31/&,*
()#&*+,$ ;+,,&@./$*
;&+,>$* &' ?./$, C2+31>1#.7 6* #,16./$ ?$77D
-1*/.3>$ B$/?$$3 ,$*15$3>$ .35 /A$ *&+,>$
E$*15$3>$ &, $2#7&"2$3/ 13 /A$ 3$1@AB&,A&&5
12#.>/$5 B" ',.>913@
E$*15$3>$ &, $2#7&"2$3/ 13 /A$ >&+3/" 12#.>/$5 B"
/A$ ',.>913@
F&,*/
18
pad) are less desirable than area level samples, as they are simple proxies of exposure, and as
such do not provide any quantitative information on the individual hazard themselves.
The cost associated with implementing these exposure assessment methods also varies, with the
individual level measurements being the most expensive to the exposure surrogates being the
least expensive. Individual level measurements entail contacting each participant, collecting
samples (biological, personal air, dermal), and performing detailed laboratory analysis of those
samples. Thus, it requires large field study teams, sampling equipment as well as extensive
laboratory testing; each component requiring numerous resources. Surrogates on the other hand,
do not require contacting individual participants or laboratory analysis. In terms of feasibility,
implementing biomarker-based exposure assessment methods in a large epidemiological study is
less feasible because of the cost and time requirements. Exposure surrogates, including
questionnaire based methods on the other hand can be implemented in large studies. Both cost
and feasibility need to be taken into consideration while deciding which sampling approach to
use for exposure assessment.
10.1.2 Linking Exposure to Hazards with Adverse Health Outcomes
The linkage between hazards, exposures, and adverse health outcomes is established using
epidemiological studies. The causality of these associations is evaluated using a set of criteria,
often referred to as Hills Criteria for Causality. They include:
Strength of Association: Stronger the association, less likely it is due to an extraneous
variable.
Temporality: Exposure precedes the disease on a temporal scale.
Consistency: Multiple, independently conducted studies report the same findings.
Dose-response relationship: As the exposure increases, disease risk increases as well.
Theoretical Plausibility: Current understanding provides theoretical basis for the
observed association.
Specificity in the cause: Ideally, the effect has one primary cause.
Experimental Evidence: Experimental studies support the findings.
Removal of exposure alleviates the risk.
However, it is important to note that this type of information is currently not available in the
context of UNGDP for several reasons:
Recent exposures - UNGDP is a relatively new process; so, the residents of the impacted
communities have been exposed for a relatively short period of time.
Issue of lag time - some of the chronic health outcomes take a long time to manifest after the
onset of exposure (long lag time). For certain chronic diseases such as cancer, prior evidence
suggests that the lag time can be substantial, often several decades. Since the UNGDP-related
exposures are relatively recent, this issue of lag time needs to be considered in
epidemiological studies.
Duration of Epidemiological Studies - Epidemiological studies used for studying the link
between potential exposure to a hazard and adverse health outcome often take 3-5 years to
complete. In addition, the peer-review process that investigators rely on to disseminate their
findings may take an additional 1-2 years. Thus, even if epidemiological studies were
19
initiated at the onset of UNGDP (which is unlikely), the findings from these studies may not
be available in the peer-reviewed literature. Results from epidemiological studies related to
UNGDP are just appearing in the peer-reviewed literature.
These factors are of particular relevance to UNGDP and should be taken into consideration while
evaluating the impact of UNGDP on human health. Simply put, the absence of investigation or
peer-reviewed data does not imply the absence of harm.
10.2 Methods
10.2.1 Overview of Data Collection
10.2.1.1 Literature Search
A literature search was performed using ISI Web of Knowledge (www.isiknowledge.com) and
PubMed (http://www.ncbi.nlm.nih.gov/pubmed) between October 2013 and May 2014.
Additional publications were identified based on communication with experts, references cited
within the published articles, and the citation track feature available from the ISI Web of
Knowledge. Search terms included fracking OR hydraulic fracturing OR natural gas OR
unconventional natural gas OR Marcellus shale AND air quality OR air pollution OR
water quality OR water pollution OR radiation OR health effects OR adverse health
outcomes OR public health.
Additional searches were conducted using Physicians Scientists and Engineers for Healthy
Energy (PSE) Citation Database on Shale Gas and Tight Oil Development
(http://www.psehealthyenergy.org/site/view/1180). We also used Google and Google Scholar to
search for government reports, and reports from non-governmental organizations. Finally, we
considered additional reports and articles submitted to us from the community, industry groups,
and environmental advocacy organizations.
All articles were screened for titles and abstracts. Articles that were not related to UNGDP were
eliminated from the list. Articles that were not related to human health, such as those related to
drilling technology, exploration, and development were excluded as well. Reports and white
papers from governmental agencies (local/state/federal), academic institutions, non-profit
groups, industry, and activists were considered, provided they were related to the environment
and/or human health.
10.2.1.2 Monitoring Data
Baseline and post-UNGDP monitoring data on air and water quality were gleaned from peer-
reviewed literature and reports from state, local, and non-governmental organizations. Whenever
available, raw data were used to come to a conclusion instead of relying on the authors
interpretation of the data. Criteria air pollution data for Maryland was obtained from U.S. EPA.
10.2.2 Identification of Hazards of Concern to Western Maryland Communities
As described in detail in the Scoping Report, we used a detailed scoping process to identify
UNGDP-related hazards that were of most concern to the community members in Western
20
Maryland. Additional hazards were identified based on the literature review. We grouped these
UNGDP-related hazards and stressors into 8 broad categories as shown below:
1) Air quality
2) Flowback and Production Water Related
a) Water quality
b) Soil quality
c) Naturally Occurring Radioactive Materials (NORM)
3) Noise
4) Earthquakes
5) Social determinants of health
a) Sexually transmitted infections
b) Traffic
c) Crime
6) Healthcare infrastructure
7) Occupational health
8) Cumulative exposure/risk

We combined water quality, soil quality and NORM under the Flowback and Production Water
related concerns because they are all related to the wastewater. Similarly we combined sexually
transmitted infections, traffic and crime into Social Determinants of Health. The traffic-related
issues discussed within the framework of public safety pertains to traffic accidents, not air
quality. The fugitive emission of methane throughout the production and distribution process and
the issue of climate change was brought up during the scoping process. Community members
were particularly concerned about the contribution of shale gas development to the impending
threat of climate change on their community. However this report is focused on health impacts of
UNGDP that are restricted to the area where gas production occurs. We did not consider
secondary effects that may manifest due to climate change.
10.2.3 Ranking of Hazards
Based on our review of the literature, we scored each hazard using a set of seven criteria that was
adapted from Witter and colleagues, who previously used them in the Battlement Mesa Health
Impact Assessment [7]. The modified metrics included in our evaluation are: 1) vulnerable
populations, 2) geographic extent, 3) duration of exposure, 4) frequency of exposure, 5)
likelihood of health effects, 6) magnitude of health effects, 7) effectiveness of the setback, and 8)
public health impact. The detailed description of these ranking criteria are provided in Table
10-2.
Table 10-2: Description of the evaluation criteria used for hazard ranking
Evaluation
Criteria
Result Score Description
Presence of
vulnerable
No 1 Affects all populations equally
21
populations
Yes 2
Disproportionately affects
vulnerable population
Duration of
exposure
Short 1 Lasts less than 1 month
Medium 2
Lasts at least one month but less
than one year
Long 3 Lasts one year or more
Frequency of
exposure
Infrequent 1 Occurs sporadically or rarely
Frequent 2
Occurs constantly, recurrently,
and/or numerously
Likelihood of
health effects
Unlikely 0
Prior evidence suggests exposure is
not related to adverse health
outcomes
Unknown 1
Evidence inconclusive/insufficient
data
Possible 2
Prior evidence suggests exposures
may be associated with adverse
health outcomes
Likely 3
Prior evidence suggests similar
exposures to be associated with
adverse health outcomes
Magnitude/severity
of health effects
None 0 No adverse health effects
Unknown 1
Evidence inconclusive/insufficient
data
Low 1
Causes health effects that can be
quickly and easily managed, do not
require medical treatment
22
Medium 2
Causes health effects that necessitate
treatment of medical management
and are reversible
High 3
Causes health effects that are
chronic, irreversible or fatal
Geographic extent
Localized 1
Effects occur in close proximity to
UNG-Development and/or
Production
Community-wide 2
Effects occur across most of the
community
Effectiveness of
setback
Positive 1
Setback is anticipated to minimize
health effects
Negative 2
Setback is not anticipated to
minimize health effects
Public health
impact
No-low impact Green Hazard received a score of 6-9
Moderately-high
impact
Yellow Hazard received a score of 10-14
High impact High Hazard received a score of 15-17

We summed the score for each hazard across the seven evaluation criteria to obtain an overall
score. These overall scores were then used to rank each hazard into three broad categories using
color-coded scheme (Table 10-2). They include:
H: High likelihood that UNGDP related changes will have negative impact on
public health
M: Moderately high likelihood that UNGDP related changes will have negative
impact on public health.
L: Low likelihood that UNGDP related changes will have negative impact on
public health.
This approach enabled us to rank each of the eight hazards, identified with community input,
using a consistent approach. We set the bar for High impact to include only the three highest
possible scores (15,16 & 17) so as to clearly distinguish those hazards that should be of the
greatest concern.

23
10.2.4 Identifying Chemicals of Concern
The fluid composition used to hydraulically fracture a well is a mixture of 99.2% water and
0.79% additives consisting of acids, corrosion inhibitors, friction reducers, clay control,
crosslinkers, scale inhibitors, breakers, iron control, and biocides [25]. These chemicals are an
important part of the process and play an important role in natural gas extraction. Table 10-3
outlines why the additives are used and the consequences of not using them. The industry has
argued that these necessary additives account for a minute fraction of the fracking fluid,
therefore their impact will be negligible. While the statement regarding a small fraction is true, it
needs to be discussed in the right context:
1. USGS estimates suggest that 3 to 7 million gallons of water are used per well.
Furthermore, 5 to 12 wells are located in a single well pad. Taking these two figures
under consideration, even if only 0.8% of the total volumes are additives, this amounts to
340,000 gallons (range 120,000-672,000 gallons) of chemicals used per well pad, a
single point source (1.29 million liters, range 0.45 -2.5 million liters). So the argument
that more than 99% of fluids used are water, while correct, is misleading because it does
not tell the whole story.
2. The less than 1% is chemical argument also overlooks basic principles of toxicity.
While discussing UNGDP related additives and chemicals, their toxicity also needs to be
taken into consideration. If a chemical is highly toxic, even exposures to small amounts
can be detrimental to human health.
These misleading statements, combined with opacity surrounding the nature of
individual chemicals present in the fracking fluid serves to drive public mistrust of the
overall fracking process.
Table 10-3. Chemicals Commonly Used in Shale Fracturing and Consequence of Not Using the
Chemicals, Source: [26]
Chemical Use Consequence of not using chemical
Acid Removes near well damage Higher treating pressure, slightly more
engine emissions
Biocides Controls bacterial growth Increased risk of souring and increased
erosion
Corrosion
inhibitor
Prevent corrosion in the pipe Increased risk of pipe corrosion from
acid
Friction reducer Decreases pumping friction Increased surface pressure and engine
emissions
Gelling agents Improves proppant placement Increased water use and decreased gas
recovery
Oxygen scavenger Prevents corrosion of well
tubulars by oxygen
Increased corrosion and compromised
well integrity

24
The U.S. House of Representatives Committee on Energy and Commerce published a report in
April 2011 outlining a comprehensive list of the chemicals used by 14 oil and gas companies
during hydraulic fracturing between 2005-2009 [27]. A total of 2,500 products containing 750
chemicals were reported. We cross referenced these chemicals with four databases to identify
their carcinogenic potential and specific organ toxicity. These databases included International
Agency for Research on Cancer (IARC), the US Environmental Protection Agencys (EPA)
Integrated Risk Information System (IRIS), Cal EPAs Office of Environmental Health Hazard
Assessment (OEHHA) Toxicity Criteria database, and the Agency for Toxic Substances and
Disease Registry (ATSDR) Toxic Substances portal. We grouped the chemicals according to
carcinogenicity and target organ system. The four categories for carcinogenicity:
Known human carcinogen
Probable carcinogen
Possible carcinogen
Not a likely to be carcinogen.
Each of these categories includes classifications from all four databases. Carcinogen includes
IARC Group 1, EPA IRIS Group A, chemicals classified as carcinogens by CalEPA, and
chemicals classified as known to be a human carcinogen by ATSDR. Probable carcinogen
includes IARC Group 2A, EPA IRIS Group B1 and B2, and chemicals classified as reasonably
anticipated to be a human carcinogen by ATSDR. CalEPA does not have a carcinogenicity
classification other than carcinogen. Possible carcinogens include IARC Group 2B and EPA
IRIS Group C. Neither CalEPA nor ATSDR have a classification for possible carcinogens. Not a
likely carcinogen includes IARC Group 3 and EPA IRIS Group E. We identified 11 target organ
systems: nervous, endocrine, circulatory, lymphatic (immune), digestive, respiratory, urinary,
reproductive system, skeletal, integumentary (skin), and muscular systems. As shown in Figure
10-2, six chemicals used in UNGDP were identified by IARC as known human carcinogens, an
additional two were identified as probable human carcinogens and eight were identified as
possible carcinogens.

25

Figure 10-2: Carcinogenicity Classification for Chemicals used During UNGDP

Figure 10-3: Target organ systems for chemicals used during UNGDP, from Cal EPA's OEHHA
Toxicity Criteria database and ATSDR's Toxic Substance Portal
"
#
$
%
&
'"
'#
Carcinogen Probable Carcinogen Possible Carcinogen
N
u
m
b
e
r

o
f

C
h
e
m
i
c
a
l
s

IARC
EPA IRIS
CAL EPA
ATSDR
0
2
4
6
8
10
12
14
16
18
N
u
m
b
e
r

O
f

C
h
e
m
i
c
a
l
s

CAL EPA
ATSDR
26
10.3 Community Impacts
The major stressors we identified and evaluated were determined through our charge from
DHMH, the scoping process, and our review of the literature. The eight stressors that were
addressed were: air quality, flowback and production water-related concerns (water quality, soil
quality, and naturally-occurring radiological materials), noise, earthquakes, social determinants
of health, healthcare infrastructure, occupational health, and cumulative exposures/risk. In this
section, we evaluate each stressor and assess if UNGDP related changes in the stressor are likely
to negatively impact public health (High Likelihood, Moderately High Likelihood, and Low
Likelihood).
10.3.1 Air Quality
10.3.1.1 Air Pollutants Associated with UNGDP Activities
Details regarding broad categories of air pollutants associated with different stages of the
UNGDP process are described in the Ambient Air Monitoring report prepared by Leidos
Incorporated for MDE [28]. Table 10-4 shows a list of selected pollutants associated with
different UNGDP activities. The information in Table 10-4 was taken from the Leidos report
with slight modifications, particularly related to traffic. In addition, it should be noted that Table
10-4 does not provide separate descriptions for two important pollutants (ozone and polycyclic
aromatic hydrocarbons (PAHs)).
27
Table 10-4: Summary of selected air pollutants associated with the UNGDP process, as
described in Leidos report [28], with slight modification.

In reviewing 353 chemicals associated with the UNGDP process, one study estimated that up to
75% of the chemicals have a potential to adversely affect eyes, skin and other sensory organs as
well as respiratory and gastrointestinal systems; an additional 40-50% have the potential to affect
nervous, immune and cardiovascular systems; 37% have the potential to affect the endocrine
system; and 25% may have carcinogenic potential [29].
Currently, MDE is collecting baseline air quality data for criteria air pollutants as well as
selected VOCs at the Piney Run Reservoir. Additional monitoring data is available for Garrett
County from the EPA Air Quality Data Mart. In general, monitoring data from 2013 suggest that
air quality in Garrett County is better than Maryland as a whole, with noted exceptions for SO
2

concentrations (Figure 10-4).
!"#$%&#'() +&#','#- .(/$&0 (1 !2'33'()3
4(25/3#'()
6$(7/&#3 849:
;(" %)7 .9<=
;()>20#?%)0
?-7$(&%$5()3
@0#?%)0:
9$A%)'& B+C3:
B<.
C%$#'&/D%#0
@%##0$
E$%11'& F F F F
.'#0 6$06%$%#'() 0)A')03 F F
G/3#3 3/360)3'() F
G'030D 0)A')03 F F F F
H%3 03&%60 1$(2
I0DD5($0
F F
.#($%A0 (1 7$'DD')A 1D/'73:
2/73 %)7 &/##')A3
F F
J%$A0 6/263 F F F F
C$(66%)# ?%)7D')A F
KD(I5%&L %)7 1D%$')A F F F F
KD(I5%&L D'M/'73 F F
E$%11'& F F F F
J0%L%A0 1$(2 ,%D,03:
30%D3: %)7 A%3L0#3
F F
N0)#')A %)7 1D%$')A F F F F
4(26$033($ 0)A')0
0"?%/3#
F F F
C)0/2%#'& 6/263 %)7
70,'&03
F F
O0DD
P0&(26D0#'()3
.%20 %3 1$%&L')A %)7
&(26D0#'()3
F F F F
.'#0 G0,0D(620)#
%)7 G$'DD')A
C$06%$%#'()
G$'DD')A
K$%&#/$')A %)7
4(26D0#'()
O0DD C$(7/&#'()
%)7 4(26$033($
.#%#'()3
28

Figure 10-4: Seasonal comparisons of air quality for Garrett County and the State of Maryland
for selected criteria air pollutants, 2013
10.3.1.2 Overview of Studies Related to Air Pollution
Studies Based on Individual Level of Measurements: As stated earlier, exposure data collected
on individual respondents are ideal as they provide a good approximation of an individuals true
exposure. Currently such data are not available for residents impacted by UNGDP. A study
conducted by investigators from the National Institute for Occupational Safety and Health
(NIOSH) measured workers exposures to respirable crystalline silica (frac sand) at 11 sites
across five states [30]. The authors collected 111 samples from the breathing zone of workers
that showed unusually high levels of exposure to respirable crystalline silica among workers. In
multiple instances, these exposures were > 10 times higher than the occupational health
29
standards such as OSHAs permissible exposure limit or NIOSHs recommended exposure limit.
This is of significant concern to public health because:
a) Crystalline silica is known to cause silicosis
b) Crystalline silica is a known human carcinogen
c) The respirators used by workers to protect themselves are not recommended at such high
exposure levels (i.e., they do not provide adequate protection)
d) People living, working, or attending school near and downwind of a well pad would be at
high risk of exposure. Because respirable crystalline silica particles are very small and
remain airborne indefinitely in outdoor air, they can travel from well pads to nearby
communities where they may disproportionately affect vulnerable populations such as
children, the elderly, asthmatics and individuals living with chronic obstructive
pulmonary diseases (COPDs).
Studies Based on Area Level Measurements: Most of the studies on air quality available to date
have relied on ambient monitoring near UNGDP facilities. One study collected 163 background
air samples at locations >0.5 miles from well pads and compared them to area samples collected
within <0.5 miles of well pads during the well completion phase [1]. Results showed that
concentrations of VOCs were significantly higher within 0.5 miles from the well pad (median
benzene 2.6 g/m
3
, range 0.9-69 g/m
3
) compared to >0.5 miles from well pads (median
benzene 0.9 g/m
3
, range 0.1-14 g/m
3
). The corresponding values for hexane were 7.7 g/m
3

(range 1.7-255 g/m
3
and 4.0 g/m
3
(range 0.23-62 g/m
3
) and). Based on a twelve month field
study, Colborn et al. 2014 reported the highest levels of non-methane hydrocarbons NMHC
concentrations during the initial drilling phase. The methane concentrations reported were
particularly high ranging from 1600 to 5500 ppb (mean 2473 ppb), while methylene chloride
ranged from 2.7 to 1730 ppb (mean 206 ppb). The authors reported that the levels of PAHs
detected in this particular study were higher than the ones that produced lower developmental
and IQ scores in children in a separate study [31].
Results from extensive air monitoring performed near UNGDP sites in Fort Worth, Texas
showed elevated levels of methane, ethane, propane and butane. In some cases, the methane
concentrations exceeded 5000 ppb. However, benzene concentrations were reported to be
consistently below 0.7 ppb. Some of the high UNGDP activity sites had average benzene
concentrations less than 0.2 ppb. This is an interesting observation for a high activity UNGDP
site located in Fort Worth Texas; given that background benzene concentrations at urban
locations routinely exceed 0.2 ppb level. A separate report on air quality monitoring by the
Pennsylvania Department of Environmental Protection conducted in Southwest and Northeast
PA showed VOC concentrations significantly lower than the ones reported by McCawley in a
West Virginia study [3234]. For example, the benzene levels in the PA study ranged from 0.29
to 1.7 g/m
3
across monitoring stations [32, 33]. It is important to note that the measurements
made using Open Path Fourier Transform Infrared (OP-FTIR) consistently showed
concentrations that were an order of magnitude higher than the ones obtained using canister
samplers. However, the validity of OP-FTIR measurements is questionable as they are likely
influenced by other factors including humidity levels. So a direct comparison of FTIR and
canister results is not recommended.
More relevant air pollution data for MD comes from a recent University of West Virginia study
that collected various air quality and noise data associated with UNGDP processes in WV. The
30
air quality measurements were taken at location 625 feet away from the well pads. Results
suggest that the concentrations of selected VOCs were considerably higher than the ones
reported for Colorado, including benzene (mean 32.2 g/m
3
median 9.35 g/m
3
, 95
th
percentile
160 g/m
3
), hexane (mean 10.4 g/m
3
, median 8.1 g/m
3
, 95
th
percentile 22 g/m
3
), acetone
(mean 99.3 g/m
3
median 90 g/m
3
, 95
th
percentile 210 g/m
3
). The overall distribution of
concentrations for selected VOCs is presented in Figure 10-5. The concentrations of these VOCs
in the West Virginia study varied considerably across different well pads. An example of this
variability is provided in Figure 10-6. The WV study also collected air samples from control sites
(Morgantown, WV) using an identical method. Although the sample size at the control site was
limited (3), none of the control samples had detectable levels of VOCs.

Figure 10-5: Ambient concentrations for selected VOCs near well pads in WV. Data taken from
University of WV study by McCawley et al. [34]
1
101
201
301
C
o
n
c
e
n
t
r
a
t
i
o
n

(
u
g
/
m
3
)
2
-
B
u
t
a
n
o
n
e

2
-
P
r
o
p
a
n
o
l
4
-
M
e
t
h
y
l
-
2
-
p
e
n
t
a
n
o
n
e
A
c
e
t
o
n
e
B
e
n
z
e
n
e
C
u
m
e
n
e
E
t
h
a
n
o
l
H
e
p
t
a
n
e
H
e
x
a
n
e
T
o
l
u
e
n
e
V
i
n
y
l

C
h
l
o
r
i
d
e
31

Figure 10-6: Variability in ambient concentrations of Acetone and Heptane across different well
pads in WV. Data taken from University of WV study by McCawley et al. [34]
10.3.1.3 Estimated Emissions for Pollutants from UNGDP Activities
10.3.1.3.1 Drilling Scenarios
The Regional Economic Studies Institute (RESI) has developed Drilling Scenarios for Western
Maryland, as reported in the Impact Analysis of the Marcellus Shale Safe Drilling Initiative [10].
According to this scenario, the estimated UNG-Development phase will last for 10 years (2017-
2026) with peak development from 2018-2021 (RESI Drilling Scenarios). A well can produce
for 25-30 years, therefore, we estimate UNG-Production to end 30 years after the last well is
drilled in 2026 (2017-2056). Table 10-5 provides the number of new and existing wells in
Western MD under 25% and 75% extraction scenarios.
Table 10-5: RESI scenarios by development year
Scenario 1: 25% Scenario 2: 75%
Year Number
of New
Wells
Drilled
Number
of New
Well
Pads
Total
Number
of Wells
Total
Number
of Well
Pads
Number
of New
Wells
Drilled
Number
of New
Well
Pads
Total
Number
of Wells
Total
Number
of Well
Pads
2017 8 4 8 4 36 12 36 12
2018 16 4 24 8 72 12 108 24
2019 29 3 53 11 63 9 171 33
2020 22 3 75 14 54 9 225 42
2021 18 3 93 17 63 9 288 51
2022 15 2 108 19 42 6 330 57
2023 12 2 120 21 36 6 366 63
2024 12 2 132 23 36 6 402 69
2025 12 2 144 25 36 6 438 75
0
50
100
150
200
A
c
e
t
o
n
e

(
u
g
/
m
3
)
B
R
O
O
K
E
D
O
N
N
A
L
E
M
O
N
L
e
w
is

W
e
t
z
e
l

T
h
r
e
e
L
e
w
is

W
e
t
z
e
l

T
w
o
M
A
U
R
Y
W
E
E
K
L
E
Y
0
50
100
150
H
e
p
t
a
n
e

(
u
g
/
m
3
)
B
R
O
O
K
E
D
O
N
N
A
L
E
M
O
N
L
e
w
is

W
e
t
z
e
l

T
h
r
e
e
L
e
w
is

W
e
t
z
e
l

T
w
o
M
A
U
R
Y
W
E
E
K
L
E
Y
32
2026 6 0 150 25 12 0 450 75
10.3.1.3.2 Emission Estimates
The City of Fort Worth natural gas air quality study [35] estimated that the total yearly emissions
of organic compound from 375 well pads, 8 compressor stations, 1 gas processing plant, a
saltwater treatment facility, a drilling operation, a fracking operation and a completion operation
would sum up to be 20,818 tons per year. The report suggested that the majority of these yearly
emissions are attributable to well pads, accounting for 75% of the total emissions.
In a recent study, Roy and colleagues [36] provided process-level emission estimates along with
uncertainty for each sources in UNGDP related activities. The authors assumed that there will be
significant decreases in the emissions from each source by 2020, compared to 2009, because of
stricter emission controls. These process-level estimates from Roy and colleagues are provided
in Table 10-6 for NO
x
, PM
2.5
and VOCs. We used these process-level emission estimates and the
well development scenarios derived by RESI for Allegany and Garrett Counties to derive total
yearly emissions for NO
x
, PM
2.5
and VOCs using the following assumptions:
The source emissions for compressor stations in Table 10-6 are based on the volume of
gas processed (billion cubic feet). To derive the estimated production volumes, we used
predicted number of wells and the estimated Marcellus Shale well production curve from
the RESI Report Figure 10-7.
We further assumed that 85% of the estimated ultimate recovery (EUR) will be extracted
by the end of year three as described in the Impact Analysis of the Marcellus Shale Safe
Drilling Initiative [10].
We took the 2009 process level to calculate overall emissions. This was done because the
likelihood of implementing stricter emission control policies (as described in Roy et al.
2014) in the next 6 years (2020) remains unclear.
The total yearly emissions were derived by summing up all process-level emissions for a
given year.
33

Drill rigs (tons/well drilled) 4.4 (0.811.5) 2.9 (0.58.1) 0.3 (0.031) 0.1 (0.010.4) 0.5 (0.11.8) 0.1 (0.020.5)
Frac (tons/well drilled) 2.2 (0.74.3) 1.8 (0.63.4) 0.16 (0.030.4) 0.1 (0.010.3) 0.25 (0.070.7) 0.14 (0.030.5)
Trucks (tons/well drilled) 6.9 (1.420) 1.5 (0.24.5) 0.07 (4x10
-4
-0.3) 0.02 (2x10
-4
-0.09) 0.4 (0.022.2) 0.2 (0.011.2)
Flowback
Dry well n/a n/a n/a n/a 3.8 (2x10
-3
-29) 1.01 (5x10
-4
-8.3)
Wet well n/a n/a n/a n/a 21 (0.09-145) 5.5 (0.02-37.5)
Dry gas n/a n/a n/a n/a 0.5 (0.08-0.8) 0.1 (0.02-0.2)
Wet gas n/a n/a n/a n/a 3.3 (2.4-4.4) 0.8 (0.5-1.0)
Compressor Stations (tons/BCF) 3.3 (1.0-5.2) 1.5 (0.3-3.0) 0.3
(4x10
-4
- 0.1)
0.3
(4x10
-4
-0.1)
1 (0.3-3.0) 0.4 (0.06-1.0)
Pneumatics (tons/producing well)
Sources
NOx PM2.5 VOCs
2009 2020 2009 2020 2009 2020
Table 10-6: Process level emission estimates for selected pollutants based on 2009 and 2020 emission levels, [36]
34

Figure 10-7: Estimated Marcellus Shale well production curve (million cubic foot) for Maryland
during the first five years. Source: Regional Economic Studies Institute 2014. [10]
Based on these assumptions, we calculated yearly emissions (Figure 10-8, Figure 10-9, & Figure
10-10) associated with UNGDP-related activities in Garrett and Allegany Counties for both 25%
and 75% extraction scenarios described in detail in the Impact Analysis of the Marcellus Shale
Safe Drilling Initiative. Results based on the 25% extraction scenario suggest that during peak
production years, approximately 22 tons of PM
2.5
will be emitted per year (range 1.76-51.62
tons/year). In addition, 468 tons of NO
x
per year (range 107-1159 tons/year) and 517 tons of
VOCs per year (range 80-2867 tons/year) will be produced during the peak years (Figure 10-9
and Figure 10-10). When we considered the 75% extraction scenario, yearly emissions for PM
2.5

were estimated to be 52 tons/year (range 3.8-113.3 tons/year). The corresponding estimates for
NO
x
and VOCs were 1,151 tons/year (range 263-2,860 tons/year) and 1,462 tons/year (range
390-6,708 tons/year), respectively (Figure 10-9 and Figure 10-10).
35

Figure 10-8: Estimated yearly emissions for PM
2.5
extraction scenarios

Figure 10-9: Estimated yearly emissions for NO
x
u
1u
2u
Su
4u
Su
6u
2u16 2u17 2u18 2u19 2u2u 2u21 2u22 2u2S 2u24 2u2S 2u26 2u27
E
m
i
s
s
i
o
n

(
t
o
n
s
)

Yeai
Yeaily PN
2.S
Emission
7S% Extiaction Scenaiio
u
2uu
4uu
6uu
8uu
1uuu
12uu
14uu
E
m
i
s
s
i
o
n

(
t
o
n
s
)

Yeai
Yeaily N0x Emission
36

Figure 10-10: Estimated yearly emissions for VOCs in Western Maryland under 25% and 75%
10.3.1.4 Studies on UNGDP Related Exposures and Adverse Health Outcomes
McKenzie and colleagues evaluated the health risks associated with UNGDP air emissions. [1].
They estimated the chronic and sub chronic non-cancer hazard indices, and the cancer risks for
residents living within a ! mile radius of UNGDP facilities and compared them with that of
residents living greater than ! mile away. The results suggest that residents who lived closer to
the wells were at greater risk of adverse health outcomes related to UNGDP-related air emissions
compared to those who lived more than ! mile away. The subchronic hazard quotient (HQ) of 5
observed for residents <1/2 mile away from wells was considerably higher than the subchronic
HQ of 0.2 observed for those living >1/2 mile away.
In a separate study, [2] investigated the relationship between maternal residence near UNGDP
wells and risk of adverse birth outcomes in rural Colorado. The authors calculated maternal
exposure during pregnancy using inverse distance weighted well count data within a specified
radius. This inverse distance weighting approach assigned higher weights to wells that are closer
to the mothers residence compared to those that were located further away. The index was then
divided into tertiles (low, medium and high). Mothers at the highest tertile of exposure were
more likely to give birth to children with congenital heart defects (CHDs) compared to mothers
at the lowest tertile of exposure (Odds Ratios (OR) 1.3, 95% Confidence Interval (CI): 1.2-1.5).
The authors observed similar associations for neural tube defects (NTDs) as well (OR 2.0, CI:
1.0-3.9). In a similar study, Hill [37] investigated maternal residency in areas heavily impacted
by UNGDP and risk of adverse birth outcome including low birth weight (LBW) and preterm
birth (PTB). The study included 22,000 live births in Colorado and 2,500 live births in
u
2uu
4uu
6uu
8uu
1uuu
12uu
14uu
16uu
E
m
i
s
s
i
o
n

(
t
o
n
s
)

Yeai
Yeaily v0Cs Emission
37
Pennsylvania. In Colorado, mothers who lived within 1 km of well were more likely to have
LBW babies as well were at increased risk of delivering prematurely (PTB) compared to mothers
who lived 2-5 km of well. Similarly in PA, the prevalence of LBW and PTB increased in 2.5 km
radius of the well after the well development [37].
A survey of PA residents living in counties impacted by UNGDP [3] found an increased
prevalence of symptoms such as throat & nasal irritation, sinus problems, eye burning, severe
headaches, persistent cough, skin rashes, and frequent nose bleeds among respondents living
within 1500 feet of UNGDP facilities compared to those who lived >1500 feet away. Some
noted limitations of the study include a small sample size (108 respondents) and non-random
samples. Furthermore, the analysis was not adjusted for potential confounders. But the study did
have some highly exposed individuals, and a large exposure gradient (distance to the facility
ranging from 350 feet to 5 miles). Findings of this particular study related to headaches,
throat/nose irritation, severe headaches, skin rashes and nose bleeds are consistent with the
common symptoms reported to us by residents of West Virginia during our site visit (November
16, 2013). The findings of Steinzor et al. [3] serve as an important hypothesis generating step
that needs to be further confirmed with detailed epidemiologic investigations that take into
account potential confounders.
In a separate study funded by The American Natural Gas Alliance, Fryzek and colleagues [38]
investigated the association between childhood cancer incidence in Pennsylvania and UNGDP
by linking childhood cancer data from 1990 through 2009 with 29,000 wells drilled during the
same time period. The authors reported no association between UNGDP and childhood cancer.
This particular study suffers from two serious flaws in study design: 1) the first UNGDP well
was dug in PA in 2006 with production starting in 2008, so the vast majority of cancer cases in
the study predated the exposure of interest; and 2) the study overlooked the issue of lag time that
is known to exist for chronic outcomes such as cancer. Thus, the design of this study was such
that it could not possibly have found an effect. This study highlights the need for high quality
epidemiological investigations with robust exposure assessments that enable investigators to
carefully match the temporal scale of exposure and outcome of interest.
10.3.1.5 External Evidence for the Health Effects of Air Pollution
10.3.1.5.1 Air Pollution and Cardiovascular Disease Deaths
Previous research has shown an association between exposure to air pollution and cardiovascular
disease morbidity and mortality. Morris and colleagues [39] found a positive association between
ambient carbon monoxide levels and hospital admissions for congestive heart failure. Venners
and colleagues [40] found that risk of cardiovascular mortality was associated with an increase in
peak SO
2
levels. Dockery et al (2001) reported that particulate matter was associated with
increased heart rate, decreased heart rate variability, and cardiac arrhythmias. In addition,
exposure to black carbon is associated with cardiovascular disease including emergency
department visits and hospitalizations [41]. Research conducted by Adar and Kaufman [42]
reviewed several studies that examined the impact of traffic related pollution on cardiovascular
disease, and despite the variations in techniques used to assess this relationship, there was
consistent evidence that confirmed the association between cardiovascular disease and traffic
exposure. In addition, another study by Hoffman and colleagues [43] found that, long-term
exposure to heavily trafficked residential areas was associated with coronary heart disease.
Moreover, there was a 1.95 fold increase in cardiopulmonary mortality in residents that lived
38
close (" 150 m) to major roadways which means that traffic exposure should be considered as a
risk factor for cardiovascular disease in addition to more traditional factors [43]. Gan and
colleagues [44] found that individuals living near a roadway were 29% more likely to die from
heart disease. Jerrett and colleagues [45] spatially examined the relationship between air
pollution and mortality in Los Angeles and found that PM
2.5
was more associated with ischemic
heart disease mortality than with cardiopulmonary or all-cause mortality. Additional air pollution
from UNGDP activities in Western Maryland could lead to an increase in heart disease morbidity
and mortality in areas and expand health disparities in areas with Marcellus shale deposits.
10.3.1.5.2 Air Pollution and Cerebrovascular Disease Deaths
Previous research has shown that exposure to particulate air pollution may similarly increase the
risk of stroke. We suspect that PM
2.5
levels will increase in Garrett and Allegany counties due to
UNGDP activities including emissions from diesel truck traffic, gas flaring, compressor stations
and other sources of air pollution. Studies of small-area variation have found a positive
association between stroke mortality rates and living in areas of high-ambient pollution.
Residents who live near UNGDP facilities particularly those who have previously had a stroke,
who are elderly, have diabetes, have heart disease, smoke, are overweight, or are in poor health
may have higher risk of strokes compared to residents with similar conditions who live farther
way from activities. Wellenius and colleagues [46] found that an increase in PM
10
, CO, NO
2
, and
SO
2
was associated with an increase risk for stroke admissions. Hong and colleagues [47] found
exposure to PM
2.5
and PM
10
was associated with an increased risk of ischemic stroke attack. In
addition, Wellenius and colleagues [46] found that exposure to PM
2.5
levels considered safe by
the U.S. EPA increased the risk of stroke onset within hours of exposure. While, Kettunen and
colleagues [48] found that levels of PM
2.5
, ultrafine particles, and CO were associated with
increased risk of stroke mortality, but only in the warmer season. Additionally, Franklin, Zeka,
and Schwartz [49] found a 1.03% increase in stroke related mortality with a 10 #g/m
3
increase in
the previous days PM
2.5
level. Time series studies using hospital discharge data reveal a
statistically significant positive association between daily measures of PM
10
and cerebrovascular
hospitalizations, but the results have been inconsistent [50]. Overall, local health departments
and clinics should monitor for increase in stroke morbidity and mortality in areas with UNGDP
activities due to a decrease in local air quality because of PM
2.5
and PM
10
.
10.3.1.5.3 Air Pollution and All-cause Mortality
This rate could increase due to exposure to air pollution from UNGDP activities in the counties
that could lead to more heart disease and respiratory problems. Previous studies have shown an
association between air pollution and all-cause mortality. Franklin, Zeka, and Schwartz [49]
observed a 1.21% increase in all-cause mortality with a 10 #g/m
3
increase in fine particulate
matter from the previous day. Ostro and colleagues [51] studied the relationship between fine
particulate matter and all-cause mortality in California and found that a 10 #g/m
3
change in the
2-day PM
2.5
concentrations corresponded to a .6% increase in all-cause mortality.
10.3.1.5.4 Air Pollution and Low Birth Weight
Studies have shown exposure to high levels of particulate matter is associated with premature
births [52] and others have shown that UNGDP operations increase local particulate matter
concentrations [6, 7]. Brauer and colleagues [53] found that residing within 50 m of highways
was associated with an 11% increase in low birthweight. Other combustion-related pollution
39
such as carbon monoxide has been associated with a significantly increased risk for low birth
weight among women living near heavily trafficked areas [54]. One study conducted in Los
Angeles, found that women who lived within 1 mile from high levels of particulate matter from
air emissions had at least a 27% increased risk for preterm delivery [55]. Black carbon a
constituent of fine particulate matter that comes from diesel exhaust has been shown to also
contribute to low birth weight among infants [56]. Therefore, it is likely that the number of
premature births and low birth weight babies will increase in this area.
10.3.1.5.5 Air Pollution and Infant Mortality
Previous research has shown a relationship between exposure to air pollution including traffic-
related pollution and infant mortality. Woodruff and colleagues [57] found an odds ratio of 1.07
for overall postneonatal mortality and 2.13 for respiratory-related postneonatal mortality. While,
Woodruff and colleagues [58] found an odds ratio of 1.16 for a 10 #g/m
3
increase in PM
10
for
respiratory causes of neonatal mortality and 1.20 odds ratio for a 10 ppb increase in ozone and
death from SIDS. Exposure to air pollution related to UNGDP activities and increases in social
stressors could have an impact on maternal stress and infant mortality rates.
10.3.1.6 Assessment
fields (external evidence), we conclude that there is a High Likelihood that UNGDP related
changes in air quality will have a negative impact on public health in Garrett and Allegany
Counties. Table 10-7 describes the scoring system we used to arrive at this conclusion.
Table 10-7: Air Quality Evaluation
Evaluation Criteria Score
Vulnerable populations 2
Duration of exposure 3
Frequency of exposure 2
Likelihood of health effects 3
Magnitude/severity of health effects 3
Geographic extent 1
Effectiveness of Setback 1
Overall Score 15
Hazard Rank H

The rationale used for scoring:
1. Vulnerable population received a score of 2 as exposure to air pollution is not equal for
all members of a population. Concentrations of air pollution will decrease as the distance
from the UNGDP facility increases. Therefore individuals living closer to the UNGDP
40
facilities will experience higher exposures. These individuals may be property owners
who do not have mineral rights.
2. Duration of exposure received a score of 3. While the exposure to air pollution resulting
from site development may decrease once the site preparation is completed, exposures
related to production, such as those associated with compressor stations will continue to
persist for years/decades.
3. Frequency of exposure received a score of 2 as exposure to air pollution occurs
continuously, 24 hrs/day, 7 days/week.
4. Likelihood of health effects was assigned a score of 3 because emerging epidemiological
evidence shows that exposure to UNGDP related changes in air quality may be associated
with adverse birth outcomes including NTD and CHD. There is also strong
epidemiologic evidence from studies outside of UNGDP settings that show exposures to
air pollutants associated with UNGDP related activities, including crystalline silica,
VOCs, and PM have negative effects on human health.
5. Magnitude/severity of health effects was assigned a score of 3 because exposure to air
pollutants that are present in UNGDP processes are known to cause human health effects
that can be irreversible, chronic, and at times fatal.
6. Geographic extent received a score of 1 because as outlined in the first bullet, the impact
will be more pronounced in the immediate vicinity of the UNGDP facilities.
7. Effectiveness of setback was assigned a score of 1 because evidence from traffic-related
air pollution studies indicated that the concentrations of traffic-related pollutants drop to
the background level beyond 500-700m (1640-2296 feet). Likewise, a study from
Colorado reported air pollution levels significantly higher within 0.5 miles (2640 feet) of
UNGDP facilities compared to >0.5 miles. Based on this, we concluded that an adequate
setback from the corner of a UNGDP facility to the corner of a residential property (2000
feet) can minimize exposure.
This section details concerns related to human contact with natural and anthropogenic
compounds made available for exposure through activities related to UNGDP. Specifically, it
will focus on exposure to chemical and radiological hazards present in water and soil impacted
by hydraulic fracturing activities.
This section will rely primarily on evidence from the peer-reviewed scientific literature. To a
lesser extent, information from the grey literature, including governmental and independent
consulting firm reports, will be included.
10.3.2.1 Water-related exposure pathways resulting from hydraulic fracturing
Humans can be exposed to fracking-related chemicals through a number of environmental
pathways. These exposure pathways can be grouped by water source.
1. Ground water. Some ground water aquifers are used by people for drinking, cooking,
bathing, and other household purposes. In the state of Maryland, more than one third of
residents rely on ground water for their water supply [21]. Over 1.1 million Maryland
41
residents rely on individual, domestic wells, accounting for 31% of ground water usage in
the state [21]. Ground waters that are developed for private wells may be vulnerable to
both naturally-occurring and anthropogenic contamination resulting from fracking
activities.
Private wells, as compared to community water systems, are uniquely vulnerable, in that
they are not protected by the Safe Drinking Water Act and are thus unregulated by the
EPA [59]. In addition, hydraulic fracturing activities are also exempted from
consideration under SDWA [60].
2. Production water and flowback. Large volumes of water are used in the process of
UNGDP; according to the USGS, stimulation of the shale formation to prompt gas
recovery can require between 3 and 7 million gallons of water per well [61, 62]. While
UNGDP has been shown to generate less wastewater per unit of natural gas recovered
than conventional drilling, the dramatic increase in drilling over the past decade has
resulted in an overall increase in the amount of wastewater produced. Lutz et al. have
estimated that since 2004, the generation of wastewater has increased by 570%, an
amount that exceeds the wastewater disposal infrastructure capacity [63]. Production
water refers to water that comes to the surface with the gas that originates from the
subsurface, whereas flowback water refers to the water injected into the well during the
fracking process [64].
To recover natural gas from production wells, the injected water must first be removed
and brought to the surface. Much of the injected water and fluids may be unrecovered
recent estimates place the loss rate of injected water at 47 91 % [65]. Additionally, an
analysis of Marcellus Shale well logs reveals that the low permeability shale retains little
free water, and thus fracturing fluids may be absorbed into the shale [65].
Once water is removed from production wells, numerous methods are employed to manage the
water; many of these can create opportunities for human exposures.
1. Storage of production waters. After recovery, flowback water is often temporarily
stored at the surface at impoundment ponds or pits prior to reuse or disposal. Concerns
exist that surface leaks and spills are possible at impoundment ponds [60]
2. Treatment of production waters. While less common today, production waters from
fracking operations in Pennsylvania were often sent to commercial or municipal
wastewater treatment plants (WWTPs). After treatment at these plants, treated production
waters were typically discharged into surface waters like rivers and streams. In 2011, the
Pennsylvania Department of Environmental Protection (PADEP) requested that
production waters from fracking operations no longer be sent to commercial and
municipal WWTPs. Some production waters from wells in Pennsylvania and other states
have been sent to industrial wastewater treatment plants [66]. In addition, a number of
alternative strategies have been pursued for disposal of fracking production water.
3. Water Reuse. Recycling/reuse of fracking water has become more common since 2011
[67], though it requires a pre-treatment before reuse, and some well operators are not
willing to pay the cost related to separation and filtration [68]. Prior to 2011, it has been
estimated that only 13% of wastewater was recycled [63]. It was estimated that 70% of
production water was reused in the state of Pennsylvania [66].
42
4. Deep Injection Wells. To a limited extent in Pennsylvania, and a greater extent in Ohio
and Texas, deep injection wells (Class II) are being used as a method for disposal of
fracking wastewater at depths considered to be below aquifers that would be used as
sources of drinking water.
5. Road and Land Application. Production water has also been periodically used as a road
de-icer (due to its high salinity) and as a dust suppressant for road maintenance. In some
instances, production water and sludge have been used as an agricultural land amendment
[67].
6. Transport of Production Water. Transport of production water for treatment, re-use, or
other purposes, creates the opportunity for spillage or leakage, which may lead to
unwanted exposures [69].
10.3.2.2 Origins of Chemical and Radiological Hazards in Water Sources
Existing research suggests that contamination of shallow aquifers used for drinking water may
occur through a variety of mechanisms, though there is a lack of high quality baseline data to
challenge the certainty of such processes [64]. Ground water resources appear to be at increased
risk of contaminant infiltration (by naturally occurring chemical hazards and radiological
materials that may exist in the subsurface) as a result of fracturing activities [66]. Poorly
constructed or faulty well casings may allow for chemicals present in production waters to leak
from production wells into the surrounding geology, a scenario that can introduce contamination
into shallow aquifers used for drinking water. Furthermore, chemicals associated with fracturing
use may be abandoned or improperly sealed in oil and gas wells [65].
Rozell and Reaven [68] present a conceptual model of pathways by which ground waters and
production waters can become contaminated (Figure 10-11).

Figure 10-11. Conceptual model of water contamination pathways, from Rozell and Reaven
2012
The process of fracking requires the use of liquid mixtures of numerous chemicals to prop open
subsurface fractures to allow the movement of natural gas through drilled wells for recovery at
the surface. The composition of these mixtures is typically considered to be a trade secret, and
thus in many cases it is not disclosed to the public [70]. Chemicals comprise between 0.5 2 %
of fracking fluids; while they may constitute a small fraction of the total fluids used, the very
large volume of fluid used (3-7 million gallons per well) combined with 5-12 wells per well pad
means the overall volume of chemicals used at a single location can be substantial [71]. A list of
43
chemicals suspected to be used in fracking has been compiled and made available to the public
[72]. The fluid solutions used to prop open fractures likely varies considerably by drilling
company, and poses challenges in terms of characterization of water quality and hazard
assessment. Based on this composition, recommendations have been made that candidate
chemicals for monitoring the impact of fracking flowback on water quality should include
sodium hydroxide, 4,4-dimethyloxazolidine, and hydrochloric acid [60, 73].
In addition to the chemical constituents of flowback, production water also brings large
quantities of brine. The characteristics of recovered brines from produced waters have been
described, and treatment and disposal of these brines has been seen as a burden to well operators.
The concentrations of metals in recovered fluids increase over time once recovery of flowback
water from the well begins, and the concentrations of inorganic elements such as barium,
strontium and radioactive radium appear to increase with salinity [60, 64]. Brines are typically
rich in chloride and sodium bicarbonate, among other constituents. At elevated concentrations,
constituents of these brines have been demonstrated to have adverse effects on ecological
receptors [69].
10.3.2.3 Gases/Chemicals
Multiple mechanisms can allow for the migration of gases (including methane, ethane and
propane) into shallow aquifers used for drinking water, raising the risk of gas accumulation to
concentrations that pose a risk of explosion [60]. This gas may originate from the target
formation and may migrate from the well annulus and through the cement sheath into the
surrounding geology. Alternatively, the fracking process may create pathways for stray gas (that
originates outside of the borehole, but has been released as a result of the fracturing) to migrate
into groundwater resources [65]. Leaky casings, abandoned oil and gas wells, and existing or
even newly-formed faults resulting from fracturing activities can serve as potential opportunities
for migration of gases [60, 66, 74]. Bacteriogenic gases may play a role in contamination [74],
but studies on ground waters within a kilometer of shale gas production sites have shown
relatively enriched thermogenic carbon isotope fingerprints. Other investigations have found that
wells where stray gas was evident had gas composition profiles of production gases consistent
with Marcellus and Upper Devonian formations [60].
A USGS-led investigation of isotopic signatures of gases in groundwater in northern Tioga
County in Pennsylvania noted complicating factors (arising from multiple potential gas sources)
in pinpointing the precise origins of the identified hydrocarbons [75]. However, the authors
noted that ground waters had evidence of both thermogenic and biogenic methane, and wells had
evidence of thermogenic methane [75]. The isotopic signatures of gases detected in ground
waters and wells suggest the same source as gases found in storage field observation wells [75].
Pathways for stray gases into shallow aquifers may also facilitate the flow of fluids from the
fracturing site to the surface and shallow aquifers [74]. Other researchers (including those funded
by the energy industry) have contested this claim, asserting that pre-existing hydraulic gradients
and factors related to bedrock permeability limit the upward flow of fracking fluid and brines
that may result from fracking [76]. Contamination of groundwater sources with fracking fluids
has not been studied extensively; the existing literature is summarized in a following section.
After hydraulic fracturing has been performed, the fraction of wastewater that has been
recovered from the well may be temporarily stored in surface impoundments. Some criteria have
been established in certain states for this practice; for example, in Pennsylvania, the
44
impoundments must have a plastic liner 30 mm in thickness with seams sealed to prevent leaks
[67]. Despite these criteria, concerns remain over ruptures in liner materials and overflows of
impoundments with fracturing water.
The Gradient consulting firm (whose client list includes members of the energy industry)
evaluated concerns related to the potential for constituents of flowback to impair processes at
publicly-owned treatment works, and concluded that flowback chemical constituents were
unlikely to impact Publicly-Owned Treatment Works (POTWs) [77].
Limited investigation has examined the impact of industrial wastewater treatment on the
contaminant profile of treated flowback water. Warner and colleagues [66] studied effluent from
the Josephine Brine Treatment facility in Pennsylvania, and reported that the treatment process
resulted in elevated concentrations of chloride and bromide relative to background levels [66].
Whereas concentrations of barium and radium in treated flowback were significantly reduced
compared to that of untreated flowback [66].
Concerns have been raised that improper treatment, resulting in either enrichment with (or
improper removal of) halides could result in the formation of trihalomethanes, some of which are
recognized carcinogens [60]. A study of effluent from a commercial wastewater treatment plant
(CWT) that accepted flowback water showed decreased diversity of disinfection byproducts,
while the actual concentration of two disinfection byproducts, dibromochloronitromethane and
chloroform, at the CWT were far higher than those in the effluent of typical POTWs [78]. The
authors also reported finding elevated concentrations of bromide and chloride, which are
precursors to disinfection byproducts in the CWT that accepted flowback water [78].
10.3.2.4 Radiological materials
Wastewater from UNGDP operations has been shown to carry residual levels of radionuclides,
often referred to as naturally occurring radioactive materials (NORM). The presence of NORM,
and the nature of NORM, is highly dependent on the shale formation in which fracturing is
occurring [64]. According to scientists from the USGS, the Marcellus Shale is recognized to
have elevated uranium content, whose daughter product Ra
226
can be present in shale brine at
levels exceeding 10,000 pCi/L [67]. Prominent NORM found in production water from the
Marcellus Shale includes radioactive radium (often Ra
226
and Ra
228
) with activities ranging from
185 to 592 Bq/L [66]. When radium is present under circumstances of high salinity and reducing
conditions, it can be dissolved in and mobilized by water [67].
Examinations of effluent from a Pennsylvania facility treating flowback demonstrated significant
reductions in radium and barium content, lowering activity of residual radium to less than 2
Bq/L, the industrial discharge limit [66]. Despite these reductions, the authors described
accumulation of radium in point-of-discharge stream sediments to levels approximately 200
times higher than what was observed in background and upstream samples at levels in excess of
standards for radioactive waste disposal [66].
The chemical composition of flowback brine derived from fracking wells in the Marcellus Shale
region was recently examined for similarity with brines from oilfields and other processes.
Among other findings, the authors reported that flowback brines from fracking wells had
concentrations of Ra
226,
Ra
228
, and Ba at levels that far exceed radiologically-based drinking
water standards [79]. The authors cautioned that flowback water must be managed carefully, to
avoid human exposures to relatively high levels of these radionuclides [79]. The study team also
45
reported that levels of other constituents of the brine, including total dissolved solids, chlorine,
bromine, sodium, calcium, and strontium, were elevated above typical seawater concentrations
by factors of 5 10 [79].
A study of soils and sediment samples near roads where brines from conventional oil and gas
wells were spread as a de-icing agent found increases in Ra
226
[80]. As compared to background
roads (where brines were not used), sediments recovered near roads (where brines from
conventional oil and gas wells were used for de-icing) were found to contain elevated
concentrations of elemental contaminants such as Ra
226
[80]. Ra
226
was 20% above background.
No significant increases in Ra
226
were observed in effluent from POTWs that received recovered
water from fracking wells [80].
While much of the research surrounding radiological hazards focuses on the activity of Ra
226
, a
recent examination of pit sludge from fracking operations characterized the frequency and
activity of a wide array of radionuclides beyond Ra
226
and Ra
228
including beryllium, potassium,
scandium, cobalt, cesium, thallium, lead
210
, lead
214
, bismuth
212
, bismuth
214
, thorium, uranium,
Sr
89
and Sr
90
[81]. While the results did not exceeded regulatory guidelines for any one particular
radionuclides, the total beta activity in one sludge sample (1329 pCi/g) exceeded regulatory
guidelines by more than 8 times (eg Texas Administrative Code, Title 16, Part 1, Chapter4,
Subchapter F, Rule 4.614) which lead authors to question the adequacy of solely using radium
as an indicator of NORM contamination and as a basis for a complete risk assessment [81].
In summary, concerns have been raised regarding the potential for human exposure to
radionuclides present in NORM from unconventional gas recovery. The majority of attention has
been focused around radioactive radium (and to a lesser extent radon, which is largely assumed
to be released at well heads [59]) as an indicator for NORM, though other radionuclides may
also be present and pose cumulative risks. Despite this, regulatory oversight aimed at exposure
mitigation appears to be minimal, and the likelihood of human exposures and disease resulting
from potential exposures are largely uncharacterized [67]. Studies of exposures to radiological
material from fracking are underway at PADEP and EPA, though the results of those studies
have yet to be released.
10.3.2.5 Evidence of well water contamination
There have been a limited number of studies examining the potential impacts of fracking on
groundwater wells used for drinking. Studies have examined contamination with gases, brine,
various chemical contaminants (including those thought to be constituents in fracking fluids),
and radiological hazards. To date, studies reporting the infiltration of gases, chemicals and other
process wastes into groundwater sources have been mixed [64].
Osborn and colleagues [82] examined drinking water wells in New York and northeastern
Pennsylvania and found that methane concentrations in drinking water wells located in active
drilling areas (within 1 km of unconventional gas wells) were higher than those in areas >1 km
away, with concentrations 17 times higher on average. The authors reported that the ratios of
methane to higher chain hydrocarbons like ethane, propane and butane suggest a thermogenic
origin in active drilling areas and primarily biogenic methane in areas where drilling was not
occurring. An investigation of the geochemical and isotopic features of the water recovered from
shallow wells did not suggest mixing with brine or fracturing fluids from drilling, and the authors
concluded that there was no evidence of contamination with these compounds [82].
46
A separate investigation examined relationships between geographic proximity to natural gas
wells and methane and ethane concentrations for 141 drinking water wells in the Appalachian
Plateaus of Pennsylvania [83]. Significant spatial relationships were observed, where drinking
water wells less than 1 km from gas wells had average methane concentrations six times higher
than those further away. Geographic distance from gas wells was also found to be significant for
both methane and ethane concentrations. The isotopic signatures of gases examined in the study
were consistent with a hypothesis of thermogenic origin and were unlikely to be of biogenic
origin. The authors concluded that living within 1 km of gas wells likely predicts exposures to
drinking water contamination with stray gases [83].
Fontenot and colleagues [84] reported an investigation of 100 drinking water wells in the Barnett
Shale Formation, including 91 wells in areas of active extraction, 4 wells from areas of non-
active extraction, and 5 reference sites. Comparisons were made between concentrations of
selected contaminants in water from the active area wells compared to those at inactive,
reference, and historic sites (measured between 1989 and 1999). The authors found significantly
elevated concentrations of arsenic, selenium, strontium and barium at active area wells compared
to historic sites. Arsenic and barium levels were also found to be significantly higher in
measured non-active and reference area wells. Despite this high background level, the maximum
detected arsenic and barium concentrations in the active area were nearly 18 and 3 times higher
than those in the inactive/reference area, respectively. The authors also reported proximity to the
nearest gas well as an important factor in predicting the contaminant concentrations. Methanol
and ethanol concentrations were also examined in active and inactive/reference areas. While the
compounds were found in wells from both areas, methanol concentrations were highest in the
active area. The authors were not able to speak definitively to the sources of the contamination,
citing the need for sampling data collected, before, during and after extraction activities to
pinpoint with certainty drilling as the source of contamination. However, they suggest that this
scenario is plausible, and that private wells closer to natural gas extraction may be at increased
risk of contamination as compared to those further away. [84]
A study of 1,701 water wells conducted in Northeastern Pennsylvania examined historic and
background surveys of methane content and other water quality measures in groundwater to
characterize potential sources of methane in drinking water wells [85]. The study, which
included an author who had an affiliation with the oil and gas industry, concluded that methane
contamination of water is related primarily to topography and groundwater geochemistry, and
that activities related to shale gas recovery have not contributed to gas impacts on drinking water
sources. The authors also assert that fracturing activities have not created or accentuated
fractures that could allow gas migration [85].
Investigators in Colorado reported results of a study examining 176 groundwater wells in the
Wattenberg field in northern Colorado where the occurrence of drilling and fracturing is
increasing in frequency [86]. The authors found that three quarters of sampled wells contained
measurable concentrations of methane, and that the majority of methane detected in sampled
wells was of biogenic origin (only two sampled wells had thermogenic methane). They
concluded that while fracturing is a possible pathway for thermogenic gas migration into
groundwater, the majority of methane present in their study was from microbial sources [86].
Overall, new UNGDP activities could lead to exposure and health risks for populations on well
water due to potential contamination of ground water and well water from fracking fluids,
47
recharge, or spills including radionuclides, heavy metals, methane, and benzene among other
contaminants.

10.3.2.6 Evidence of Soil Contamination
Soil can be contaminated with drilling fluids, flowback, produced waters, and other wastes. As
outlined above, these fluids and wastes may contain numerous contaminants including
radionuclides. Soil contamination is likely to occur through: 1) unintentional spills and leaks of
waste or chemicals used during UNGDP, 2) the spread of waste onto fields, and 3) the use of
wastewater or brine on roads.
Limited evidence in the literature suggests that land application of wastewater and flowback is a
practice that could lead to severe vegetation damage and mortality [8789]. Land application
of waste is a common waste disposal method in several states [87, 88]. In a land application
study in West Virginias Fernow Experimental Forest, Adams and colleagues observed visible
changes to ground vegetation, including browning and wilting leaves, leaf scorch, curling, and
drop following land application of drill pit fluids (Adams et al. 2011). The fluids met the
regulatory requirements for land application chemicals: chlorides below 12,500 mg
L-1
and pH
between 6 and 10. A few days after land application, nearly all the ground vegetation died. After
7-10 days, overstory trees began showing similar damage (Adams et al. 2011). Two-years after
the application, 56% of the trees in the area were dead. Damage was attributed to direct contact
with the fluids, as well as root uptake from the soil. When they evaluated the soil chemistry, they
found statistically significant differences in Ca, Mg, Al, Mn, Zn, and the C/N ratio between the
test and control sites [88, 89].
Aminto and Olson [73] used a four-compartment model (including soil, water, air, and biota) to
evaluate 12 hazardous components (sodium hydroxide, ethylene glycol, 4,4-dimethyl
oxazolidine, 3,4,4-trimethyl oxazolodine, 2-amino-2-methyl-1-propanol, formamide,
glutaraldehyde, benzalkonium chloride, ethanol, hydrochloric acid, methanol, and propargyl
alcohol) used in hydraulic fracturing fluid. They found that sodium hydroxide, hydrochloric acid,
and 4,4-dimethyl oxazolidine were the highest mass concentrations found in the soil
compartment [73]. Sang and colleagues (2014) found in controlled laboratory experiments, that
flowback fluid has the potential to activate colloid mobilization. Mobilization is dependent on
certain chemical constituents, several of which are found in hydraulic fracturing fluid, such as
inorganic salts and organic compounds, such as surfactants [87].
Overall, there is little information on the impact of UNGDP activities on soil quality. As
discussed above, there are three exposure pathways that could contribute to contamination of soil
and groundwater. Accidental spills and leaks due to storage of flowback and production waters
can be minimized. According to the Maryland Best Management Practices, the State would
require use of enclosed tanks, constructed of metal with liners instead of impoundment ponds. In
addition, a barrier that can hold the total volume of the largest storage container or tank located
in the enclosed area would surround the tanks [16].

These practices may minimize the potential
for contamination.
10.3.2.7 Characterization of water-related human health burden
Despite evidence suggesting that human exposures to contaminants originating from fracking are
likely, to date, there is a dearth of studies that have examined relationships between exposed
48
persons and health outcomes [64]. While these studies are critical for decision-making efforts
that aim to consider public health concerns, it is critical to recognize that the absence of
investigation does not constitute an absence of risk or harm. We were unable to locate any
studies of fracking-impaired waters on human health outcomes; this is consistent with the lack of
identification of studies noted in a recent review [64].
A study of 39 unique ground water samples collected in Garfield County, Colorado (a region
with highly concentrated drilling activity) examined the propensity for flowback water to elicit
endocrine activity on estrogen and androgen receptors. As compared to samples collected from a
reference region, ground water samples from drilling areas were far more likely to exhibit
endocrine activity; 89%, 41%, 12% and 46% exhibited estrogenic, antiestrogenic, androgenic
and antiandrogenic activity, respectively. The authors concluded that natural gas drilling
operations may contribute to elevated level of endocrine disrupting compounds in ground and
surface water [90]. While this study did not characterize likely exposures or associated human
health burdens, the findings point toward future directions for epidemiologic investigations.
While there are not any epidemiological studies that have evaluated associations between soil
quality and health, Bamberger and Oswald [91] published a study documenting 24 cases of
livestock, domesticated animals, and humans that have been adversely impacted by exposure to
contaminated water and soil. In case study three, a cattle pasture had been contaminated by
wastewater due to a tear in an impoundment pond, and soil tests detected high levels of chloride,
sulfate, sodium, and strontium. As a result of the contamination, the cattle experienced
reproductive issues, including spontaneous abortion and stillbirth [91].
10.3.2.8 Limitations of existing database Critical data gaps
1. Baseline water quality data are largely unavailable for states that already allow
unconventional natural gas production. This lack of data precludes high quality
characterization of the impacts of fracturing activities on drinking water sources, and thus
complicates efforts to conduct epidemiologic investigations of potential associations
between fracking-related contamination and human health outcomes.
2. The majority of studies examining NORM in fracking brines or recovered waters are
typically limited to examination of radioactive radium, and do not include other
radionuclides that may co-occur and create additional radiation exposures.
3. Given the proprietary nature of unconventional natural gas development, data are largely
unavailable regarding the composition of fluids used for fracturing.
4. We were unable to locate a comprehensive database of best practices aimed at
minimizing leakages, storage problems, and other failures that could lead to human
exposures; this, too, may be related to the proprietary nature of fracturing.
5. The utility of radium isotopes as indices of contamination with NORM is unclear, but
emerging data suggest that other radionuclides may also contribute significantly to
cumulative radiological activity.
10.3.2.9 Assessment
health in Garrett and Allegany Counties. The overall score for the Flowback and Production
49
Water Related hazard category is primarily driven by concerns related to water quality. Table 10-
7 provides an overview of the scoring for each evaluation criteria.
Table 10-8: Flowback and Production Water Related Evaluation
Geographic extent 2
Overall Score 13
Hazard Rank M

1. Vulnerable population received a score of 2 as exposure to contaminated water
disproportionately affects residents near the UNGDP facilities, particularly those who
rely on well water.
2. Duration of exposure received a score of 3 because exposure will persist for longer than 1
year.
3. Frequency of exposure received a score of 2 as exposure to contaminated water is
frequent.
4. Likelihood of health effects was assigned a score of 1 because despite evidence of
exposure, evidence regarding adverse health outcomes could not be determined because
of insufficient data.
5. Magnitude/severity of health effects was assigned score of 1 because despite evidence of
exposure, evidence regarding adverse health outcomes could not be determined because
of insufficient data.
6. Geographic extent received score of 2 because exposure can be widespread if the
drinking water aquifer is contaminated.
7. Effectiveness of setback was assigned score of 2 because setback will not mitigate
exposure.
10.3.3 Noise
Environmental noise associated with UNGDP was identified as a top concern by residents in
Allegany and Garrett Counties during the Scoping process. Increased noise levels are expected
during all phases of development and production. Setback regulations (Table 10-9) and
adherence to the state or local noise standards (Table 10-10) are two methods being proposed to
minimize noise during development and production [16]. Local governments will be responsible
for enforcing the noise standards; however, if the counties do not have the capacity for
50
monitoring and enforcement, the permittee may be required to hire an independent contractor to
conduct periodic noise monitoring and to respond to noise complaints [16]. The current noise
standards adopted by MDE are outlined in Table 10-10. The residential noise standards for both
day and night are relatively high considering the literature on health effects associated with noise
exposure and may not adequately protect public health.
Table 10-9. Proposed Setbacks specific to Occupied Dwellings, Source: Maryland Best
Management Practices [16]
Distance From To
1,000 Borehole Any occupied dwelling
1,000 Compressor stations Any occupied dwelling

Table 10-10. Marylands Maximum Allowable Noise Levels for Receiving Land Categories
Day/Night
3
Industrial Commercial Residential
Day 75 67 65
Night 75 62 55
10.3.3.1 Hazards associated with noise
Major sources of environmental noise are transportation, including vehicular traffic, aircrafts,
and railroads, as well as industrial operations. Urban areas typically have higher noise levels
compared to rural areas. Most of the increased noise in urban areas is due to traffic-related noise.
Noise is considered a major stressor because of its ability to lead to a number of adverse health
effects.
Most of the literature on noise and health effects has focused on transportation (traffic, airplanes,
and trains) sources. Adverse health effects from noise are dependent on the duration of exposure
and the intensity of the noise. Long-term exposure to A-weighted decibels ranging from 35-75
have been associated with a myriad of health effects, from disruption of sleep and school
performance to hypertension

[92]. Children, elderly, chronically ill, and hearing impaired
individuals have been found to be more susceptible to environmental noise [93]. While increased
noise levels are associated with both the UNG-Development and the UNG-Production phase,
exposures associated with the UNG-Development phase are temporary as the development
activity ceases to exist once the wells are constructed. The noise associated with the production
phase, on the other hand, is permanent. Only a few studies have evaluated noise associated with
UNGDP activities.

3
Daytime hours are 7 a.m. to 10 p.m. and Nighttime hours are 10 p.m. to 7 a.m., COMAR 26.02.03.01
51

Figure 10-12: Well Pad, West Virginia (photo: Brigid Kenney)
10.3.3.2 Noise Associated with UNG-Development
McCawley [34] monitored and recorded the average A-weighted decibel levels (dBA) in West
Virginia at 9 sites located around 5 well pads at different stages of natural gas development,
including site preparation, vertical drilling, horizontal drilling, hydraulic fracturing, and
flowback. He found the average noise levels across the sites were lower than 70 dBA, but the
levels were frequently over 55 dBA [34]. The Colorado School of Public Health conducted a
HIA to assess the potential health impacts associated with natural gas drilling in Battlement
Mesa. They determined that significant sources of noise would be heavy truck traffic,
construction equipment, diesel engines used throughout drilling and hydraulic fracturing, and
drill rig brakes [7]. Based on these sources and the estimated baseline noise levels in the
community, they determined that noise associated with natural gas extraction would produce
negative health effects [7]. New York evaluated the noise impact associated with UNGDP in
their draft supplemental Environmental Impact Assessment (EIA) using a model to estimate the
noise levels at varying distances associated with each stage of well pad construction and drilling.
Noise levels were estimated based on data obtained from the industry for the construction
equipment. They found that noise levels at a distance of 250-2,000 feet would range from 52-75
dBA during well pad construction, 44-68 dBA during drilling, and 72-90 dBA during high-
52
volume hydraulic fracturing [94]. Noise associated with construction, drilling, and hydraulic
fracturing would last approximately 60 days per well pad.
Table 10-11. Noise Associated with UNGD
Phase/Activity Distance (feet) Average dBA Source
Well Development
Access road construction 50-500 69-89 NYSDEC, 2011
Access road construction 1,000-2,000 57-63 NYSDEC, 2011
Truck traffic, construction 625 56-73 McCawley M,
2013
Truck traffic
4
< 500 65-85 Witter et al, 2010
Site preparation 625 58-69 McCawley M,
2013
Well pad preparation 50-500 64-84 NYSDEC, 2011
Well pad preparation 1,000-2,000 52-58 NYSDEC, 2011
Drilling
Vertical drilling 625 54 McCawley M,
2013
Rotary air well drilling 50-500 58-79 NYSDEC, 2011
Rotary air well drilling 1,000-2,000 45-52 NYSDEC, 2011
Horizontal drilling 50-500 56-76 NYSDEC, 2011
Horizontal drilling 1,000-2,000 44-50 NYSDEC, 2011
Well Completion
Hydraulic fracturing 625 47-60 McCawley M,
2013
Hydraulic fracturing
5
50-500 82-102 NYSDEC, 2011
Hydraulic fracturing 1,000-2,000 70-76 NYSDEC, 2011
Hydraulic fracturing & flowback 625 55-61 McCawley M,
2013

4
This is an estimate based on anticipated noise associated with diesel truck traffic and residential proximity to truck
routes
9
.
5
Average dBA for pumper truckers with a sound pressure level of 110 and 115.
53
10.3.3.3 Noise associated with UNGP
Current literature on noise impacts associated with UNGDP focuses on well construction and
hydraulic fracturing. There have not been any studies to evaluate noise levels associated with
production, including noise originating from compressor stations. Natural gas compressor
stations are a more permanent source of noise in the community. To better understand noise
exposure levels associated with compressor stations, we conducted a pilot study to monitor and
evaluate residential exposure to noise associated with natural gas compressor stations in West
Virginia.
Methods
All noise monitoring was conducted around compressor stations in Doddridge County, West
Virginia between April 11-17, 2014, using 3M Quest SoundPro noise monitors (3M Personal
Safety Division, St. Paul, MN). All monitors were set to collect slow, A-weighted decibel levels
(dBA) L
eq
, L
min
, L
max
, L
peak
, L
5
, and L
95
and C-weighted decibel levels L
eq
, L
min
, L
max
, L
peak
in 1-
minute intervals.
Short-term Measurements: Short-term measurements (20 min) were collected at increasing
distance from compressor stations in Doddridge County, WV. The monitors were placed in a
safe outdoor location using a tripod. The exact geographical coordinate of the monitor location
was recorded.
Medium-term (24 hr) Measurements: 24-hour noise measurements were collected inside and
outside homes that were near compressor stations in Doddridge County, WV. A total of three
homes were located less than 1,000 feet from the compressor stations, three homes were located
between 1,000 and 2,000 feet, and two homes were located between 2,000 and 2,500 feet. An
additional 3 homes were recruited as control homes, located beyond 3,500 feet from the
compressor stations. Noise monitors (Quest SoundPro SE/DL Series) were placed inside and
outside each home for 24 hours. Indoor monitors were typically placed in a bedroom and outdoor
monitors were placed in the yard facing the natural gas compressor stations (NGCS). Outdoor
monitors were encased in an environmental protection kit (3M SoundPro Outdoor Measuring
System (SP-OMS)). Outdoor measurements for two homes located 2,000 to 2,500 feet were not
for a full 24-hours, due to battery failure. Following the method used by Murphy and King
(2014), we evaluated the difference between the C-weighted dB and the A-weighted dB to
determine the presence of low-frequency noise. A difference greater than 15 dB indicates the
potential for low frequency noise and would require further spectral analysis. Monitors were
factory calibrated prior to use and then were pre-calibrated using a Quest QC-10/QC-20
Calibrator onsite prior to each measurement. Following each measurement, the monitor was
post-calibrated and the data were downloaded using Quest Suite Professional. The average sound
equivalent was calculated using logarithmic averages and was stratified by distance from
compressor station, time of day (daytime 7:00 am-10:00 pm and nighttime 10:00 pm-7:00 am),
and location (indoor and outdoor).
Results
Noise levels associated with compressor stations were dependent on the distance from the
compressor station, location (indoor vs. outdoor), and time of day. Overall the average L
eq
for the
combined compressor stations was 60.20 dBA (range 35.3 to 94.8 dBA), and the average short-
term L
eq
for the combined compressor stations was 61.43 (range 45.3 to 76.1 dBA) (Table
54
10-12). Average outdoor noise levels were 58.33 (35.3 to 85.0 dBA) compared to 61.27 (35.3 to
95.8 dBA) indoors. Both the short-term and 24-hour measurements decreased with distance from
the compressor stations, 63.15 dBA at less than 1,000 feet to 54.09 dBA at 2,000 to 2,500 feet
for 24-hour measurements and 63.34 dBA at less than 1,000 feet to 54.10 at 2,000-2,500 feet for
short-term measurements (Table 10-12). Noise levels were generally higher during daytime
hours compared to nighttime hours, 61.44 dBA and 56.38 dBA, respectively. Noise levels were
higher indoors compared to outdoors for homes located within 2,500 feet of a compressor
station, 61.27 and 58.33, respectively (Table 10-12). The contribution of outdoor noise to indoor
noise varies depending on the type of home and whether the windows are opened or closed. A 17
dB reduction in noise levels would be expected in a cold-climate home with windows open and a
27 dB reduction with windows closed [95]. We observed a 3-7 dB difference in indoor versus
outdoor noise levels, much lower than would be expected. There is little indication of low-
frequency noise at varying distances from natural gas compressor stations. We observed a
difference greater than 15dB at sites located less than 500 feet from the compressor stations.

Figure 10-13: Time Series, Indoor L
eq
by Distance from Compressor Station
55

Figure 10-14: Time Series, Outdoor Leq by Distance from Compressor Station
The control homes in West Virginia were set in a semi-rural/rural community, located more than
3,500 feet from a compressor station. It is anticipated that current noise levels in Western
Maryland are comparable to the noise levels at the control homes located more than 3,500 feet
from a compressor station. Overall, the average L
eq
at the control homes was 51.40 dBA, with
45.02 dBA indoor and 54.03 dBA outdoors (Table 10-12). Noise levels at homes within 2,500
feet of the compressor station were on average 8.7 dBA higher, with a 16.25 dBA difference
indoor and a 4.3 dBA difference outdoor than the levels observed at the control homes.
Table 10-12. Summary Statistics, Stratified by Distance, Location, and Time
Distance
(feet)
Location Time of Day N
1
Mean L
eq

(dBA)
Range L
eq

(dBA)
All distances All locations All times 21205 60.20 35.3-94.8
Indoor All times 11520 61.27 35.3-94.8
Outdoor All times 9388 58.33 35.3-85
Short All times 297 61.43 45.3-76.1
All locations Daytime 13575 61.44 35.3-94.8
All locations Nighttime 7630 56.39 35.3-73.3
<1000 All locations All times 8818 63.15 35.3-94.8
Shoit All times 178 6S.S4 Su-76.1
Indoor All times 4320 64.59 35.3-94.8
Daytime 2700 66.49 35.3-94.8
56
Distance
(feet)
Location Time of Day N
1
Mean L
eq

(dBA)
Range L
eq

(dBA)
Nighttime 1620 53.85 35.3-70.1
Outdoor All times 4320 60.97 55.3-85
Daytime 2700 61.25 55.3-85
Nighttime 1620 60.46 55.3-73.3
1000-2000 All locations All times 8963 55.48 35.3-77.6
Shoit All times SS SS.4u 46.2-67.8
Indoor All times 4320 57.28 35.3-75.7
Daytime 2700 57.86 35.3-75.7
Nighttime 1620 56.12 35.3-65.3
Outdoor All times 4320 52.36 35.3-77.6
Daytime 2700 52.75 35.3-77.6
Nighttime 1620 51.62 36.9-57.9
2000-2500 All locations All times 3694 54.09 35.3-80.3
Shoit All times 66 S2.1u 4S.S-S7.1
Indoor All times 2880 53.75 35.3-80.3
Daytime 1800 54.31 35.3-80.3
Nighttime 1080 52.61 35.3-72.6
Daytime 678 55.32 35.3-76.5
Nighttime 70 55.41 50.9-69.6
>3500 All locations All times 8704 51.50 35.3-74.1
Indoor All times 4384 45.02 35.3-69.3
Daytime 2764 45.95 35.3-69.3
Nighttime 1620 42.72 35.3-65.1
Daytime 2700 54.23 35.3-74.1
Nighttime 1620 53.66 35.3-58.4
1
N refers to the number of 1-minute intervals

57

Figure 10-15: Boxplots, Leq by Distance from Compressor Station
10.3.3.4 UNGDP Noise Evaluation
Both daytime and nighttime noise levels associated with natural gas compressor stations
routinely exceed the Marylands maximum allowable noise level of 65 dBA for residential areas,
the nighttime noise level is just above the maximum allowable noise level of 55 dBA, as
depicted in Figure 10-14. The exceedance was less common at control homes located >3500 feet
from the compressor stations. This shows that residents living more than 3,500 feet away from
natural gas activity are not expected to experience high levels of noise. This finding should be
taken into consideration while deciding on setback distances. Furthermore, the findings
presented here are from compressor stations and are not related to development activities. As
such, they represent chronic noise exposure that community members will have to encounter for
years/decades, not transient exposures that go away after the completion of a well.
There have not been any epidemiologic studies to evaluate health outcomes associated with
UNGDP noise; however, numerous studies have evaluated the health impact of long-term
exposure to environmental noise from other industries. The most common health effects
associated with environmental noise are annoyance, stress, sleeping disturbances, headaches,
hypertension, and cardiovascular problems [96100]. Nighttime noise levels as low as 35 dBA
have been found to cause sleep disruption [92]. Children, elderly, and hearing impaired
individuals are more susceptible to environmental noise [93].
In addition to noise-related health outcomes, there may be synergistic effects between noise and
air pollution associated with UNGDP. Several studies have evaluated the relationship between
air quality and noise on health [101104].
10.3.3.5 Assessment
in noise exposure will have negative impacts on public health in Garrett and Allegany Counties.
Table 10-12 provides an overview of scoring for each evaluation criteria we used to arrive at this
conclusion.

58
Table 10-13: Noise Evaluation
Geographic extent 1
Overall Score 12
Hazard Rank M

1. Vulnerable population received a score of 2 as exposure to noise disproportionately
affects residents near the UNGDP facilities. Property owners without mineral rights are
disproportionally burdened as they do not have a voice, and may not be able to sale their
property even if they want to move away.
2. Duration of exposure received a score of 3 because exposure to noise will persist for
longer than 1 year (fracturing of well, compressor stations).
3. Frequency of exposure received a score of 2 as exposure to noise is frequent.
4. Likelihood of health effects was assigned a score of 2 because noise exposure is known
to elicit hearing loss, and increase stress levels.
5. Magnitude/severity of health effects was assigned a score of 1 because the adverse health
effects are reversible.
6. Geographic extent received a score of 1 because noise exposure is localized.
7. Effectiveness of setback was assigned a score of 1 because adequate setbacks will
mitigate noise exposures.
10.3.4 Earthquakes
In four years, from 2010-2013, roughly 450 earthquakes, with magnitudes of 3.0 or larger,
occurred across the central and eastern United States at an average rate of 100 per year. That is a
five-fold increase in earthquake occurrence recorded over a 30 year period from 1970 to 2000
[105]. Limited research has pointed to anthropogenic activities such as UNGDP

as a potential
reason for the increase, while others have pointed that these small events are nothing to be
concerned about [105109]. To date, none of the earthquakes recorded in Maryland, including
the August 2011 event, have been linked to NGDP. Yet earthquake events in the Marcellus Shale
area have been attributed to fracking activities [110].
59

Figure 10-16. Cumulative counts of earthquakes with a magnitude $ 3.0 in the central and
eastern United States, 1970-2013, [105]
In Figure 10-16, the dashed line corresponds to the long-term rate of 20.2 earthquakes per year,
with an increase in the rate of earthquake events starting around 2009.
Earthquakes associated with overall NGDP process can broadly be grouped into 2 categories:
those associated with well development/production and those that are associated with the
disposal of wastewater through injection.
10.3.4.1 Earthquakes associated with development/production
During the developmental process, micro-earthquakes, ones with a magnitude of 2 or lower, are
produced during the hydraulic fracturing (or fracking) stage [106]. So far, none of the
thousands that have been recorded by seismographic networks in the Marcellus Shale area have
been large enough to pose a serious risk [106]. Yet, a modeling simulation conducted to assess
the potential for fault reactivation and large seismic events associated with shale-gas hydraulic
fracturing operations showed when hydraulic fracturing is conducted in areas with existing
faults, it may lead to micro-seismic events [109]. The magnitude of these particular micro-
seismic events is somewhat larger than that associated with micro-seismic events originating
from regular hydraulic fracturing because of the availability of larger surface area that can
rupture [109].
10.3.4.2 Earthquakes associated with injection of wastewater
Disposal of wastewater from NGDP activities is done by injecting the water deep underground,
hundreds to thousands of meters below the water table and drinking water aquifers. This is done
so as not to contaminate drinking water [105]. However, according to the US Geological Service
(USGS), wastewater injection increases the underground pore pressure, which may, in effect,
lubricate nearby faults thereby weakening them. If the pore pressure increases enough, the
weakened fault will slip, releasing stored tectonic stress in the form of an earthquake. Even faults
that have not moved in millions of years can be made to slip and cause an earthquake if
conditions underground are appropriate [105]. This situation is what is believed to have caused
60
a series of more than 100 earthquakes in Youngstown, Ohio during the fall and winter of 2011 to
2012, the largest of which, recorded in December 2011, had a magnitude of 3.9 [110]. All the
events were recorded by the Division of Geological Survey of the Ohio Department of Natural
Resources (ODNR) and analyzed with velocity models (1D Earth models) for precise epicenter
locations. Each quake was also assessed to determine the effect of vertical velocity
heterogeneities on focal depth, which is the depth at which the earthquakes rupture began. The
study looked in-depth into the December 2011 quake, and the associated Northstar 1 injection
well, since this quake was the largest [105]. The first earthquake experienced by Youngstown
occurred 13 days after the well became active and ended when ODNR shut the well down. More
importantly, from the modeling, it was shown that the earthquakes rupture centers were in an
ancient fault near the Northstar 1 well, suggesting wastewater injections caused the existing,
dormant faults to slip, a theory that is also supported by the USGS [105].
Furthermore, the USGS report suggests that at some NGDP sites, increases in seismicity
coincides with the injection of wastewater in deep disposal wells [105]. Others have found a
relationship between the magnitude of an earthquake and the total volume of fluid injected into
the ground, with 1 million cubic meters of fluid linked to quakes of magnitude 5 or less [111]. In
these instances, rate of injection seemed to influence the frequency of quakes [111]. Another
study examined the source of three earthquake events near a NGDP site in November of 2011
and indicated that even small- to moderately-sized injection-induced events could release
additional tectonic stress and induce an even larger earthquake event [107].

The earthquakes, of
magnitudes 5.0, 5.7, and 5.0, sequentially, were felt in 17 states, with the epicentral region the
Wilzetta North field. Production of oil from the Wilzetta Northfield occurred primarily in the
1950s and 1960s, with NGDP production continuing into the present. The active wastewater
fluid injection wells were located within the Wilzetta North field or just over a kilometer outside
it, and use of these injection wells began after 1993 and continues to occur. The group measured
the aftershocks of each earthquake event to identify the faults that ruptured in the sequence.
Their results show that the tip of the initial rupture plane was within 200 m of active injection
wells and within 1 km of the surface. Additionally 30% of early aftershocks occur within the
sedimentary section [107]. Using Coulomb stress calculations, they concluded the first event
(magnitude of 5.0) was induced by increased fluid pressure from the injection wells.
Additionally, the aftershocks of this first event deepen away from the well, suggesting stress was
transferred and added to the increased magnitude experienced in the second event (magnitude of
5.7). This study suggests that decades-long timeframes between the beginning of fluid injections
and the induction of earthquakes are possible. Furthermore, the sequential rupture of three faults
suggests that stress changes from the initial rupture triggered the successive earthquakes,
including one larger than the first [107].
However, Rutqvist and colleagues suggests that these small earthquake events are not something
to be highly concerned about [109]. They conducted numerical simulation studies to evaluate the
potential for injection-induced fault reactivation and notable seismic events associated with
shale-gas hydraulic fracturing operations [109]. Specifically, they focused on the Marcellus
Shale, which has an approximate depth of 1500 m (~4,500 feet). Their repeated, modeled
injection-events and fault slips resulted in a total rupture length of 50 m, with an offset
displacement of less than 0.01 m [109]. They concluded that any fractures to the earth caused by
waste injection would occur at great depths below the ground, too low to activate faults or
impact drinking water supplies [109]. Yet, this study only used mathematical models to assess
61
the impact of massive physical events that have been previously shown in Oklahoma and Ohio to
be of significant concern [109].
Most of the deep injection wells used for disposal of wastewater are classified as Underground
Injection Control (UIC) program Class II wells. License and Operational requirements for the
UIC Class II wells are regulated under the Safe Drinking Water Act or designated State
Authority. Therefore the licensing and operational requirements only address the issues related to
potable aquifers to ensure that they are not contaminated for drinking water purposes. These
wells are required to report average injection pressure, flow rate and cumulative volume. It does
not take into consideration the diffusion of pore pressure into the basement faults, seismic
monitoring as well as injection pressure that may cause the critically stressed faults to fail [106].
The national and Maryland inventory of injection wells is provided in Table 10-14.
Table 10-14: National Inventory by Classes of Injection Well [112]

10.3.4.3 Assessment
not allow deep well injection of wastewater, there is a Low Likelihood that UNGDP related
Classes of
Well
uses lnvenLory
Mu
lnvenLory
Class l
ln[ecL hazardous wasLes, lndusLrlal non-hazardous llqulds,
or munlclpal wasLewaLer beneaLh Lhe lowermosL uSuW
680 wells
0
Class ll
ln[ecL brlnes and oLher flulds assoclaLed wlLh oll and gas
producLlon, and hydrocarbons for sLorage.
172, 068 wells
0
Class lll
ln[ecL flulds assoclaLed wlLh soluLlon mlnlng of mlnerals
beneaLh Lhe lowermosL uSuW.
22,131 wells
0
Class lv
ln[ecL hazardous or radloacLlve wasLes lnLo or above
uSuWs. 1hese wells are banned unless auLhorlzed under a
federal or sLaLe ground waLer remedlaLlon pro[ecL.
33 slLes
0
Class v
All ln[ecLlon wells noL lncluded ln Classes l-lv. ln general,
Class v wells ln[ecL non-hazardous flulds lnLo or above
uSuWs and are Lyplcally shallow, on-slLe dlsposal sysLems.
Powever, Lhere are some deep Class v wells LhaL ln[ecL
below uSuWs.
400,000 Lo 630,000 wells. noLe: an
lnvenLory range ls presenLed
because a compleLe lnvenLory ls
noL avallable.
13701
Class vl
ln[ecL Carbon uloxlde (CC2) for long Lerm sLorage, also
known as Ceologlc SequesLraLlon of CC2
6-10 commerclal wells expecLed
Lo come onllne by 2016.
(lnLeragency 1ask lorce on
Carbon CapLure and SLorage)
unknown
62
Table 10-15 provides the scoring for the evaluation criteria that we used to arrive at this
conclusion.
Table 10-15: Earthquake Evaluation
Geographic extent 2
Overall Score 7
Hazard Rank L

1. Vulnerable population received a score of 1 as it does not affect populations
disproportionately.
2. Duration of exposure received a score of 1 because exposure to an earthquake is short.
3. Frequency of exposure received a score of 1 as exposure to earthquakes is infrequent.
4. Likelihood of health effects was assigned a score of 0 because the low level of
earthquakes associated with UNGDP are not likely to have a direct impact on public
health. This assumes MDE will not issue permit for deep well injection of waste.
5. Magnitude/severity of health effects was assigned a score of 0.
6. Geographic extent received a score of 2 because earthquakes are not confined to the
immediate vicinity of UNGDP facilities.
7. Effectiveness of setback was assigned a score of 2 because adequate setbacks will not
mitigate exposure.
As UNGDP operations continue to grow, the need to address impacts on social determinants of
health intensifies. Multiple studies [57, 64]

have identified the main issues of concern within
communities including increased rates of industrial traffic, violent crimes, mental health
problems, substance abuse, sexually transmitted infections and the resulting impact on the local
police force as well as local healthcare facilities.
63
10.3.5.1 Traffic Impacts Other Than Air Pollution

Figure 10-17: UNGDP-related Traffic, West Virginia
As UNGDP operations continue to grow, impacted communities will experience significantly
higher levels of traffic. Of particular concern is the increased level of truck traffic. It is estimated
that, on average, a multistage well can require upwards of 1000 truck round trips to deliver
equipment (e.g., bulldozers, graders, pipe), chemicals, sand, and water needed for well
development and fracturing) [5, 6, 64]. Additionally, the wastewater from UNGD operations may
be trucked offsite for disposal [7], which would result in even greater numbers of large trucks on
local roads. In Bradford County, PA, an area rich in UNGDP operations, truck counts were
approximately 40% higher than a comparable 5-year average prior to UNGDP [5, 113].

Local
traffic would also increase since an average of 120 to 150 workers per day [7] would commute
into the community to work on site.
Increased traffic raises issues of air pollution, discussed in section 10.3.1 above. Increased truck
traffic in local and residential areas raises other issues such as increased frequency of collisions,
need for road maintenance. According to Adgate, data from the Pennsylvania Department of
Transportations Crash Reporting System indicates a significant increase in the number of total
accidents and accidents involving heavy trucks between 1997 and 2011 in counties with a
relatively large degree of shale gas development compared to counties with no development [5,
114]. Even though nationally, the number of automobile accidents has been on the decline since
2005 [115, 116],

heavy-truck crashes rose 7.2% in rural Pennsylvania counties heavily impacted
by UNGDP [115].

In fact, Pennsylvania counties with the highest density of UNGDP operations
had the largest increase in large-truck crashes after UNGDP activity began in 2005 [115]. This
trend is seen in UNGDP sites across the country. The Texas Department of Transportation noted
a 40% increase in reported fatal motor vehicle accidents from 2008 to 2011 in 20 counties with
UNGDP operations [4, 5].

All these accidents have not come without a price. The Bureau of Labor Statistics reported that
the fatality rate for oil and gas workers is more than 8 times higher than that of other occupations
[5, 117]. Beyond the obvious human cost, there is the economic burden that local communities
64
are carrying for these incidents. In Pennsylvania, a large-truck accident can cost a local
community over $200,000 related to deaths, injuries and property damage [115]. An explanation
for the increased numbers of accidents experienced by UNGDP truckers is the oil field
exemption from highway safety rules. These exemptions allow truckers in the oil and gas
industry to work longer hours than drivers in other industries [113, 118], placing them at greater
risk for crashes and fatalities. Additionally, much of the truck traffic, and therefore risk of
accidents, is concentrated over the first 50 days following well development [7], suggesting that
either truckers are rushing to meet deadlines and/or truckers new to the area are unprepared for
the mountainous terrain that is typical of shale areas, and have more accidents based upon these
factors.
Accidents may also be attributed to deteriorating road conditions surrounding UNGDP sites.
Since each UNGDP well site requires thousands of truck trips to deliver UNGDP fluid and
materials and to haul away UNGDP wastewater, the local and rural roadways will be strained to
keep up with the wear and tear [115, 119]. Together, truck-driving exemptions, poor road
quality, unfamiliarity with the area, and pressure to complete a run in a timely manner all
contribute to increase risk to UNGDP driver safety.
Because most of these rural communities have few roads that allow access to the UNGDP site,
most of the industrial traffic would use the same roads that children use for walking or bicycling
to school and bus stops [113], placing them at greater risk of emission exposures as well as
placing them in harms way. Indirectly, increases in traffic may cause some members of the
community to decrease their time spent doing outdoor-fitness activities (walking, cycling,
running, etc.) [113, 120, 121], thereby lowering their overall physical well-being.
Increased traffic on the roads leads to congestion, congestion that makes it harder for first
responders to do their jobs. The increase in traffic accidents has resulted in a significant increase
in 911 calls and emergency dispatches [115, 122]. In fact, in Bradford County, PA, the increased
traffic has delayed the response times of emergency vehicles [115], placing those who requested
them in great danger.
10.3.5.2 Crime
Increase in crimes rates is a major concern for communities with UNGDP operations. A study
conducted by Haggerty and colleagues [5] focused on counties within the six major oil- and gas-
producing states in the U.S. West. The group conducted statistical analyses to determine whether
or not the level of influence of oil and gas extraction on income had been associated with
increases or decreases in county well-being [5]. They determined that the average number of
violent and property crimes per 1,000 people increases with increased length of specialization in
oil and gas and increases at a faster rate for counties whose UNGDP income was higher [5].
Additionally, the longer a county has been specialized in oil and gas, the higher the countys
crime rate [5].

While, Haggerty and colleagues [5] focused primarily on violent crime, Food and Water Watch,
a non-governmental organization and consumer rights group, conducted an impact assessment on
UNGDP operations in Pennsylvania looking at crimes associated with alcohol abuse. According
to their impact assessment, counties with the highest density of UNGDP wells (at least 15 wells
per square mile) had a greater increase in disorderly conduct, drunk driving, and public
intoxication arrests than counties with no wells, after 2005 when UNGDP began.

In the most
heavily impacted counties in Pennsylvania, average annual number of disorderly conduct arrests
65
rose 17.1 % from 1,336 prior to commercial UNGDP (2000 to 2005) to an average of 1,564 per
year after UNGDP. This was three times higher than the average number of disorderly conduct
arrests in counties in Pennsylvania with fewer UNGDP operations.

Additionally, in these same,
heavily impacted counties, the average annual number of public intoxication arrests rose 11.9%,
along with steep increases in drunk driving, traffic violations and bar fights [113].
10.3.5.3 Illness, Mental Health, and Substance Abuse
Communities with UNGD operations are also experiencing increasing rates of physical and
mental illness. The development of UNGD operations in Pennsylvania has been linked to a rise
in sexually transmitted infections [113, 114].

In heavily fracked, rural counties in Pennsylvania,
the average annual number of gonorrhea and chlamydia cases increased by 32.4% while the
average annual number of the same cases in non-fracking counties only increased 20.1 percent
from the previous year. Comparing the two, the most heavily impacted rural Pennsylvania
counties had a 61% greater increase in STI rates than counties without UNGDP [113]. This
phenomenon is not unique to the Marcellus Shale area. In UNGD regions in North Dakota,
doctors are treating more chlamydia cases. Furthermore, this region reported increased sexual
and domestic assault rates and local women feeling increasingly unsafe [113, 115].

Overall, in
communities with UNGD operations, a trend has emerged with increases in arrests for both
crime and substance abuse and STIs corresponding to periods of increased natural gas
development [4, 7, 64, 116, 117].
The following table compares the changes in the percentages of these issues across a state
(Pennsylvania) verses one rural community (Battlement Mesa, Colorado) [7, 113]. The change is
over the time from baseline assessment of the area for UNGD operations to the peak of
operations. Generally, there is an upward trend in all areas for both statewide UNGD operations
and the operations at the community level. The decrease in substance abuse arrests seen in
Battlement Mesa could be due to collection time of the data. Following the peak production,
UNGD operations taper off, are not as intensive, and require fewer workers. Additionally, the
data for Pennsylvania are statewide, so while some peak UNGD operations are winding down in
one area, in another the peak could just be beginning.
Table 10-16. Percent Change in STIs, Disorderly Conduct Arrests, and Substance Abuse Arrests
Percent Change
Pennsylvania (heavily
fracked)
Battlement Mesa,
Colorado
STIs + 32.4% + 216.7%
Disorderly Conduct Arrests + 17.1% + 31.8%
Substance Abuse Arrests + 11.9% - 33.4%

Along with increasing rates of STIs, communities with UNGD operations have reported
increases in substance abuse. Studies have shown that alcohol and other illicit drug use is highest
among the workers in the natural gas development and production industry [7, 118, 119, 123].
Substance abuse has long been associated with mental health issues, and here with UNGD
operations, the situation is no different. Witter and colleagues reported that the transient nature
66
of the migrant worker along with a high intensity and stressful job make for the perfect
combination for psychosocial stressors [7]. The workers are away from social controls and
comforts of their home community, the difficult employment fosters the desire for release
(through drugs, alcohol, fights, sex, etc.), and high salaries in a predominantly male workforce
put the workers at risk for engaging in risky behavior that negatively impact their mental and
physical health as well as the health of those who live in the community [7, 124].
Residents of UNGD communities have also experienced mental health problems related to the
operations. Ferrar and colleagues [125] noted that individuals who believed their physical health
had been affected tended to report higher stress levels due to loss of trust and perceived lack of
transparency in the UNGD industry and local government. Seventy-nine percent of subjects
reported being denied or receiving false information, while 58% reported that their
concerns/complaints were ignored. Interestingly, residents reported more psychosocial stressors
than physical stressors, suggesting that residents mental well-being was impacted more so than
their physical well-being [64]. An increase in alcohol consumption as a coping mechanism has
been shown in previous research in areas with UNGDP activities and could occur in the two
Maryland counties of concern [126].
10.3.5.4 Impact on Residents, Police, and Healthcare System
Some community members in Garfield County, Colorado reported that the development of such
an intensive industry in a relatively non-industrial area has negatively affected their sense of
community livability and social cohesion. Additionally, land values near UNGD operations are
declining, further affecting the psychosocial health of the community [4]. If residents cannot sell
their land and homes, they may feel trapped and helpless in their situation. Furthermore, studies

have shown that prolonged exposure to stress increases the levels of stress hormones in the body
and places the individual under the stress at greater risk for health and cardiovascular disease [4,
120122].
The impact of UNGDP operations on public safety extends beyond the direct threat of residents
and workers physical well-being. The safety of all living in a community with UNGD
operations could be indirectly impacted by the industry. First, local and state police departments
may be ill-equipped to handle the additional increases in crime. Local forces have a limited
number of officers they can spare and in some cases, the state police act as local law enforcement
when the community is highly rural. When these small-staffed and already stretched departments
see large increases in crime, it keeps them preoccupied and unable to handle all the situations.
For instance, the Pennsylvania State Police have linked an increase in arrests and crimes
involving natural gas workers with community members not receiving help when they need it the
most [113, 127].
The boom of UNGD operations can also overwhelm the local healthcare system that is described
in the Healthcare Infrastructure section. Rural hospitals are not designed to handle large influxes
of triage patients, like those seen in occupational accidents, traffic accidents, or the result of
fights. Furthermore, with an increase in population of transient workers, hospitals and clinics are
experiencing increases in the incidence of patients exhibiting STIs, mental illness, and substance
abuse, issues that small, rural healthcare systems do not have the resources to handle adequately
[113, 128]. For instance, when confronted about their lack of treatment for STIs, workers often
cite lack of access to healthcare facilities due to geographic isolation or lack of facilities with
67
available walk-in testing along with clinic hours overlapping with their own working hours as a
rationale for not seeking treatment [7, 124, 128].
10.3.5.5 Assessment
Based on our review of social determinants of health (section 10.3.5), we conclude that there is a
High Likelihood that UNGDP related activities will have a negative impact on the social
determinants of health. Table 10-16 provides the scoring for the evaluation criteria that we used
to arrive at this conclusion.
Table 10-17: Social Determinants of Health Evaluation
Geographic extent 2
Overall Score 15
Hazard Rank H
1. Vulnerable population received a score of 2 as exposure to public safety issues
disproportionately affects residents near UNGDP facilities. Issues of sexually transmitted
infections, crime and traffic safety all disproportionately affect community members in
the high UNGDP activity areas.
2. Duration of exposure received a score of 3 because exposure to public safety related
issues will last for more than 1 year.
3. Frequency of exposure received a score of 2 as exposure is frequent.
4. Likelihood of adverse effect was assigned a score of 3 because evidence from Colorado
and Pennsylvania show public safety to be negatively impacted in UNGDP impacted
communities.
5. Magnitude/severity of health effects was assigned a score of 2 because the adverse health
effects (STD) require medical treatments.
6. Geographic extent received a score of 2 because the entire community is at risk.
mitigate issues related to public safety.
68
A communitys healthcare infrastructure consists of both healthcare facilities (i.e., private and
public healthcare services, hospitals, and emergency transport services) and trained healthcare
professionals. In rural communities, disparities in infrastructure or professional capacity to
address community health needs may exist due to shortages of primary care physicians and non-
physician providers, and specialists; high rates of uninsured, elderly, or poor patients who often
require additional health care; and limited public resources allocated to health care [129].
Introduction of new industries, including those in the extractive sector, can have mixed impacts
on a communitys health care infrastructure, particularly those in rural, resource-poor settings.
Revenue flows from the extraction of natural resources, when distributed in an effective and
equitable manner, can fund public services such as healthcare infrastructure; the potential
increase in workers with health insurance can also have a positive impact on local health care
industry [130]. At the same time, increases in population, particularly among those engaged in
high-risk occupations may intensify local health care utilization [131, 132]; without adequate
strengthening and expansion of existing healthcare infrastructure, these population changes could
overextend an already fragile system. Citizens living in Allegany and Garrett counties, during
our scoping phase, conveyed their concerns about the negative impact of UNGDP on their
already limited healthcare infrastructure. Because UNGDP could modify the usage rates of the
healthcare infrastructure in Allegany and Garrett counties, a review of potential health impacts is
needed.
10.3.6.1 Current healthcare infrastructure conditions
In Allegany County, Western Maryland Health System (WMHS) provides a continuum of care
ranging from primary care to nursing home services. Services offered by WMHS include acute
and chronic care, community health and wellness, clinical prevention, care coordination, home
care, community health workers, and provider recruitment. WMHS is the only licensed hospice
care facility in Allegany County. WMHS is also a Level III trauma center, the only trauma center
in Western Maryland. The closest Level I trauma center for both Allegany and Garrett Counties
is in Morgantown, WV at West Virginia University Hospital. WMHS operates a regional
medical center in Cumberland (a 275-bed hospital), along with two diagnostic centers, a nursing
and rehabilitation center in Frostburg, a community health and wellness center, two urgent care
centers (one in Short Gap, WV and the other in Frostburg, MD), and three primary care centers
(2 in Frostburg and 1 in LaVale, MD). There are approximately 187 physicians affiliated with
WMHS. The primary care facilities are open Monday from 8:00 a.m. - 4:30 p.m. and Tuesday -
Friday 7:30 a.m. until 6:00 p.m.
In addition to WMHS, other facilities include an inpatient psychiatric hospital, a federally
qualified community health center, and nine nursing homes/assisted living facilities in Allegany
County. The Thomas B. Finan Center is a state owned and operated inpatient psychiatric facility
located in Cumberland, MD with 80 beds. It provides services to those 18 years of age and older
and includes inmates with criminal histories, non-criminals who have been involuntarily
committed, and voluntary patients. Tri-State Community Health Center is a federally qualified
community health center with OB/GYN and primary care services. Tri-State operates five
community health center sites located in Allegany (Cumberland) and Washington (Hancock)
counties in Maryland; Fulton County in Pennsylvania; and Morgan County in West Virginia.
There are a total of nine nursing facilities and assisted living facilities throughout the county
69
including: EGLE Nursing and Rehab, Allegany Health Nursing and Rehab, Devlin Manor Health
Care Center, Golden Living Center, The Lions Center, Frostburg Village Nursing Care Center,
Moran Manor Health Center, Frostburg Nursing and Rehab, and Kensington-Algonquin and
Country House.
Public health services are provided by the Allegany County Health Department, which provides
screening and prevention programs, family planning, WIC, inpatient and outpatient behavioral
health services, mental health care management, dental services, and food and water protection.
The Allegany County Health Department is open five days a week from 9:00 a.m. - 5:00 p.m.
Finally, Allegany College of Maryland and Frostburg State University provide training to local
health care providers in nursing, psychology, dental hygiene, radiologic technology, respiratory
therapy, and other areas and support continuing education for health care professionals. The
Western Maryland Area Health Education Center facilitates continuing education and training
for health professionals and conducts health workforce development.
In Garrett County, Garrett County Memorial Hospital (GCMH) operates a 55-bed, not-for-profit,
acute care hospital facility, including a 10-bed sub-acute rehabilitation unit. GCMH is the only
hospital in the region, serving a population of 31,000, including residents of Garrett County and
communities in the surrounding West Virginia counties. Services at the Hospital include a 24-
hour emergency department; inpatient care; observations services; obstetrics; pediatrics;
medical/surgical intensive care unit; operating room: radiology; lab; cardiopulmonary services;
as well as community and worksite wellness; safe sitter; and CPR programs and other ancillaries.
In a study conducted by GCMH, individuals utilizing GCMH also reported using a second
facility for services, with a majority (72%) traveling to Morgantown, WV, or 23% Cumberland,
Maryland in Allegany County [133].
Garrett County also has a Federally-Qualified Health Center in Oakland, MD: Mountain Laurel
Medical Center (MLMC). MLMC uses a patient centered medical home model for the delivery
of primary health care and offers services such as primary health care services, acute and chronic
illness care, care coordination, and health education. They are open Monday, Wednesday,
Friday: 7:30AM 5:00PM; Tuesday and Thursday: 7:30AM 7:00PM and have a 24 hour on-
call access.
There are three assisted living facilities in Garrett County: Dennett Road Nursing Home and
Oakland Nursing and Rehabilitation Center--each a 100-bed facility and located in Oakland,
MD; and Goodwill Mennonite Nursing Home in Grantsville, MD (89-bed facility). In addition,
there is one licensed hospice facility in the county (Hospice of Garrett County in Oakland, MD).
Public health services in Garrett County are provided through the Garrett County Department of
Health (GCDH) including adult and geriatric services, behavioral health, dental health,
environmental health, WIC, person health including health education and outreach, and home
health.
Allegany County is a designated HPSA for primary care for low-income populations, mental
health care for Medical Assistance populations, and dental care for Medical Assistance
populations. Allegany County has a critical need for specialty providers including vascular
surgery, urology, as well as dentists willing to provide care for adults with no insurance or
Medical Assistance. Garrett County is a designated HPSA for primary and mental health care,
and dental care for Medical Assistance populations. All of Garrett County is considered a
70
medically underserved area (MUA), while substantial portions of Allegany County (Orleans,
Lonaconing, Oldtown, and Cumberland) also qualify as a MUA.
10.3.6.2 Rates of insurance coverage
Insurance status of individuals living in Garrett and Allegany counties were obtained from the
County Health Rankings Database. In 2011, there were an estimated 6,532 uninsured individuals
living in Allegany County, approximately 11.9% of total population, including 4% of children.
In Garrett County, an estimated 3,473 individuals were uninsured, approximately 14% of the
total population. In the State of Maryland, an average of 12% of the total population is
uninsured, with most counties having between 8-16% of the total population uninsured.
10.3.6.3 Migrant workforce and health care usage
UNDGP entails multiple labor intensive phases that could extend several years for larger
projects. Much of the impact to health care infrastructure is related to the influx of workers
during the initial development phase. The RESI final report predicts workforce numbers based
on two possible extraction scenarios which were developed based on conservative and feasible
extraction rates given gas reserves in Maryland and the production curve of a horizontal well
[10]. Under scenario one, 25% of the total shale gas would be extracted, and scenario two, 75%
of the total shale gas would be extracted. RESI estimates that under scenario one, drilling activity
will increase employment over baseline by approximately 1327 jobs (1056 in Garrett County and
271 in Allegany County) on average from 2017 to 2026, and in the period after drilling, from
2027 to 2036, economic activity will change the baseline employment with an increase of 151
jobs (113 in Garrett County and 38 in Allegany County). Under scenario two, drilling activity
will increase employment by approximately 2825 jobs (2093 in Garrett County and 732 in
Allegany County) on average from 2017 to 2026, and by 189 jobs (80 in Garrett County and 109
in Allegany County) from 2027 to 2036.
Although these predictions by RESI project the number of jobs that could be created from
specific levels of expenditures based on the number of wells, they do not distinguish between
jobs to local and out-of-state workers. Literature indicates that shale gas drilling depends heavily
on a migrant workforce residing in Texas and Oklahoma and moves with rig operations to new
extraction sites; local residents are often faced with part-time, short-term, and low-wage
employment prospects found in supportive industries trucking, construction, and retail jobs [134,
135]. Because data on the number of workers estimated to be migrating into Allegany and
Garrett counties to work on the Marcellus Shale Gas Development is unknown, we will use the
RESI numbers to approximate low and high levels of the migrant workforce: approximately from
1327-2825 migrant workers on average during the first 10 years of drilling, and 151-189 migrant
workers on average during the 10-year period after drilling. We also do not have information on
whether this migrant workforce will have insurance. Research literature on this is inconclusive.
For instance, a HIA done for Battlement Mesa in Garfield County, Colorado by the University of
Colorado School of Public Health predicted the impact to health care infrastructure based on
information that all migrant workers would have health insurance [7]. A recent impact
assessment of fracking in Carroll County, Ohio by Policy Matters Ohio uses case studies from
Sublette County, Wyoming, Lycoming County Pennsylvania, and McKenzie County, North
Dakota to conclude that most of these jobs do not offer health insurance. As result, there have
been negative impacts on local healthcare infrastructures due to uncompensated care for
emergency room visits [135].
71
Oil and gas extraction and production workers experience seven times the fatality rate of general
industry; a vast majority are due to motor vehicle incidents, contact with objects and equipment,
and fires and explosions [6, 136]. These workers also experience non-fatal injuries and illnesses
(injuries due to being struck by objects or being caught in objects, equipment or material) at a
higher rate than other industries [137]. Because of the exposure to such safety hazards, UNGDP
workers can increase utilization of emergency, urgent, and trauma care services due to higher
rates of occupational related incidents and injuries. Insured workers using healthcare services
could offer positive support to existing systems as long as their rate of utilization meets available
capacity. If utilization rates surpass current healthcare infrastructure capacity, then this could
have adverse consequences for the availability, access, and quality of services. Uninsured
UNGDP workers, like any other uninsured population, would place stress on healthcare
infrastructure because those who are uninsured are unable to pay for medical care when they do
seek care and often go into medical debt; an influx of additional uninsured populations into
Allegany and Garrett counties may stress these under-resourced health systems [138].
The use of primary and public health care systems, especially in the areas of emergency, urgent
care, and trauma care, may rise as a result of an increase in the UNDGP workforce. These
services may or may not be supported by employers through the provision of insurance. If these
utilization rates are within current capacity, local healthcare infrastructure could potentially
benefit economically from revenues introduced by UNDGP industries. However, if any potential
revenues are not reinvested back into the maintenance of current health care infrastructure and
the development of new infrastructure, an influx of UNDGP workers may exacerbate existing
infrastructure pressures.
10.3.6.4 Characterization of healthcare infrastructure impact
There could be negative impacts to local healthcare infrastructure due to the increase in UNGDP
workforce and their potential health care utilization rates. Impacts to the healthcare infrastructure
are expected to be high given that we can expect 1327-2825 migrant workers on average during
the first 10 years of drilling, and 151-189 migrant workers on average during the 10-year period
after drilling moving into counties with a total population of 29,889 (Garrett) and 73,521
(Allegany). If all or most of these workers are insured, local primary care and public health
services will be supported and this support could potentially expand services to all community
residents. However, the impact of financial support through the increase in the insured
population may not be adequate to foster the development of the existing healthcare
infrastructure because it is unclear whether revenues from UNGDP will be substantial enough to
directly impact health care infrastructure in Allegany and Garrett counties.
Even though there have been popular and social media accounts of the demands placed on rural
and remote health services by extractive industry workers and visitors, there is a critical data gap
of evidence-based research around UNGDP and the health of the broader community. A handful
of studies that have been conducted indicate that extractive industry workers place similar
demands on health care infrastructure as local residents, with an increased demand on emergency
department services [131, 132, 139].

Given this literature and the vast health care infrastructure
needs of Allegany and Garrett counties (i.e., as federally designated HPSA and MUA areas with
high levels of uninsured and medically assisted populations), we predict that an increase in
health care utilization, regardless of whether workers are insured or uninsured, would strain the
existing healthcare infrastructure, likely leading to decreased quality, availability, and access to
services. Even a small stressor to the existing healthcare infrastructure would impact the
72
residents of Allegany and Garrett counties, particularly those who utilize services most often and
vulnerable populations such as older adults, pregnant women, and young children.
Studies of boomtown-related theory and research have documented the cyclical nature of the
natural gas extraction industry [140142]. This temporal aspect of the process of natural gas
extraction leads us to predict that any impact on healthcare infrastructure will be observed during
the initial years of the project in the development phase, which is the most labor-intensive. The
cyclical nature of change also leads us to anticipate that any impact to health care infrastructure
will be uneven throughout the lifecycle of the project. Any impact, positive or negative, on the
healthcare infrastructure will be concentrated during the first phase of development, when labor
needs are high and larger numbers of workers are expected; impacts will decline in the
production and reclamation phases as labor force requirements even out and eventually decline
[6]. Large numbers of workers are expected relative to population size for more than a year, and
therefore, there is an increased likelihood that this would stress local health care infrastructure,
especially those serving emergency, urgent, and trauma care needs. We expect that residents of
Allegany and Garrett counties will experience negative impacts as a result of changes to their
healthcare infrastructure. The long-term exposure to the effects of the project along with the
potential impact to the health of county residents, a high number of who are vulnerable, could be
noticeable. The impact of changes to the healthcare infrastructure is predicted to be negative as a
result of the UNGDP.
10.3.6.5 Assessment
Based on our evaluations of the current healthcare infrastructure in Garret and Allegany Counties
as well as expected number of migrant workers that will come to these areas, we conclude that
there is a High Likelihood that UNGDP related activities will have a negative impact on public
healthcare infrastructure in Garrett and Allegany Counties. Table 10-17 provides an overview of
scoring we used for each evaluation criteria to arrive at this conclusion.
Table 10-18: Health Care Infrastructure Evaluation
Geographic extent 2
Overall Score 15
Hazard Rank H

73
1. Vulnerable population received a score of 2 as healthcare infrastructure impacts
disproportionately those who are more likely to use healthcare services such as the
elderly, the disabled, and children.
2. Duration of exposure received a score of 3 because exposure (the influx of UNGDP
workers) will last for more than 1 year.
3. Frequency of exposure received a score of 2 as UNGDP worker health care utilization
rates over the length of a UNGDP cycle will be constant.
4. Likelihood of adverse effect was assigned a score of 2 because stress on healthcare
infrastructure will preclude individuals from receiving timely treatment.
5. Magnitude/severity of health effects was assigned a score of 2 because health
infrastructure effects are noticeable but with proper management and resources, can be
reversible.
6. Geographic extent received a score of 2 because the entire community is at risk.
mitigate issues related to healthcare infrastructure.
10.3.7 Cumulative Exposures/Risk
Conventional risk assessment methods were designed to assist regulators and risk managers in
addressing threats resulting from a single chemical or source to a hypothetical individual, instead
of a population [143, 144]. This approach fails to account for the fact that exposures do not
happen in a vacuum, and that individuals are simultaneously exposed to multiple chemical,
biological and physical hazards as well as psychosocial stressors.
This shortcoming of traditional risk assessment has given rise to cumulative risk assessment
(CRA) or community-based risk assessment approaches [145]. Cumulative risk is the combined
risk from aggregate exposures from all relevant routes, to multiple hazards or stressors, including
chemical, biological, physical and psychosocial stressors [143, 145, 146]. Under this framework,
the CRA is divided into 3 distinct phases: 1) planning and scoping and problem formulation, 2)
analysis phase, and 3) risk estimation and characterization. In this approach, the impacted
community is the central focus, instead of a specific chemical or the source. CRA is a tool for
organizing and analyzing information to examine, characterize and possibly quantify the
combined adverse human health effects from multiple stressors [20, 143, 145148]. The scoping
process allows engagement of stakeholders, particularly impacted community members, from the
onset. This process helps to identify concerns that are of high priority to the impacted
community. As such, it is a useful tool for a community that is being impacted by new threats
including UNGDP activities.
CRA is often not quantitative like conventional risk assessment [143, 145, 148]. This is because
CRA deals with the combined effects of multiple hazards (chemical, physical, and biological)
and psychosocial stressors, and calculating specific risk, including interactions among various
mixtures/stressors is methodologically complex [20, 143, 147, 149, 150]. Although there has
been some advancement made in terms of aggregate exposure and dealing with hazards that have
common mechanisms of toxicity, similar modes of action, or have common target organs, there
are no clear approaches to deal with interactions between multiple stressors, particularly non-
74
chemical stressors such as psychosocial stress from loss of property value, loss of community
identity, family conflict, poverty, unemployment, lack of access to amenities, unsafe community
conditions and working environments, limited access to healthcare resources, discrimination,
residential crowding, street crime, traffic congestion and other circumstances, on risk [20, 143].
The issue of cumulative exposure/risk is of paramount interest among communities impacted by
UNGDP activities. These communities encounter a multitude of hazards and psychosocial
stressors simultaneously during the development phase (i.e., hydraulic fracturing) and production
phase (i.e., compressor stations). The cumulative impact from the spatial concentration of
environmental hazards, pollution-intensive facilities, and noxious land uses combined with the
potential impacts of future UNGDP activities may lead to negative health outcomes and
community stress and lower quality of life and community sustainability. For example,
individuals who currently live near multiple facilities could see an increase in exposure and
respiratory health risks by new UNGDP activities. For example, McKenzie and colleagues [1]
estimated the chronic and subchronic non-cancer hazard indices for residents living within ! mile
radius of UNGDP facilities and compared it with residents living greater than ! mile away in
Garfield County, Colorado. The subchronic HQ of 5 was observed for residents <1/2 mile of
wells was considerably higher than subchronic HQ of 0.2 observed for those living >1/2 miles
away. In addition, pollution sources tend to concentrate in poor and under-resourced
communities leading to disparities in burden and exposure and higher risk of poor health
outcomes.
Cumulative exposure assessment should also include positive exposures including the economic
benefits of UNGDP activities in a host community. The development and production of shale
resources in Allegany and Garrett counties could improve the economy and provide jobs for
local residents. However, studies on extractive industries have shown loss of jobs and increase in
unemployment rates in boom towns during the bust phase [151, 152]. Other industries that
need clean environments including good air and water quality and healthy ecosystems including
agriculture, tourism, fishing, and recreational industries are incompatible with UNGDP [151,
153, 154]. This could potentially lead to a net loss in jobs, and an increase in the unemployment
and poverty rates [113] in both Allegany and Garrett counties. This suggest the overall impact of
UNGDP on job creation in western MD is more complex than simple estimation of how many
workers do the UNGDP industry need to complete the process.
For reasons mentioned above, public health advocates have long stressed the need to incorporate
cumulative exposure/risk as the true impact of UNGDP activities simply cannot be quantified by
simple measure of criteria air pollutants, VOCs, contaminants in drinking water supplies, or any
other hazards for that matter. What these quantitative measures fail to account for, are the slow
and hidden sufferings encountered on daily basis by impacted community members that simply
cannot be measured. To understand these hidden costs, the study team embarked on a site visit of
a community in Doddridge County of WV where UNGDP activities are already underway.
During this site visit, study team members were given a tour of the UNGDP sites across the
county. Study team members were informed about the hidden sufferings experienced by
individual community members that led to chronic stress, poor quality of life, sense of
helplessness, and mental health issues including depression and anxiety.
Examples included:
75
Chronically stressed property owners who cannot stop the development in his own
property because he/she does not have the mineral rights.
A mother who cannot let her children play outside because of the odor, and the symptoms
her children exhibit if they play in the front yard (throat and eye irritation, skin rash). She
cannot sell her house and move away because no one wants to buy her property next to a
UNGDP facility.
Community members who feel that the social fabric of their community has been
irreversibly destroyed.
Families who cannot sleep in their own house because of the constant noise from the
compressor station next to their property.
A resident with a pre-existing condition who is convinced the worsening of his/her
symptoms coincides with the odor in the air that comes from the nearby UNGDP facility.
A neighbor whose two small children both suffer from frequent nosebleeds.
It is clear that communities currently impacted by UNGDP activities need a place-based
cumulative exposure/risk assessment to capture their cumulative risks from exposures to multiple
chemicals, media, pathways and non-chemical stressors (e.g., psychosocial stressors) or the
stakeholders underlying vulnerabilities, as described in the NRC report [155, 156]. Yet, there
are no studies to date that have applied the framework of CRA to look at the risk experienced by
UNGDP impacted communities.
10.3.7.1 Assessment
We anticipate the cumulative risk from the physical, chemical and psychosocial stressors will be
greater than the simple sum of individual risks. We further anticipate that the impact will be
disproportionately felt by vulnerable subgroups such as children, elderly, individuals with
existing diseases, poor residents, and individuals without mineral rights. We conclude that there
is a Moderately High Likelihood that the UNGDP related activities will have a net negative
impact in the cumulative exposure/risk. Table 10-18 provides overview of scoring we used for
each evaluation criteria to arrive at this conclusion.
Table 10-19: Cumulative Exposures/Risk Evaluation
Geographic extent 2
Overall Score 14
76
Hazard Rank M

1. Vulnerable population received a score of 2 as cumulative risk will not be uniformly
distributed, and that the most vulnerable subgroups will be disproportionately burdened.
2. Duration of exposure received a score of 3 because cumulative exposure will persist
beyond first year.
3. Frequency of exposure received score of 2 as exposure is frequent.
4. Likelihood of adverse effect was assigned score of 2 because previous evidence
document the relationship between exposure to individual hazards and risk.
5. Magnitude/severity of health effects was assigned score of 1 because evidence regarding
the magnitude/severity of health effect could not be determined because of insufficient
data.
6. Geographic extent received score of 2 because the entire community is at risk.
7. Effectiveness of setback was assigned score of 2 because adequate setback will not
mitigate issues related to public safety.
10.4 Occupational Impacts
As the demand for natural gas from UNGDP increases, so does the demand for jobs in the
industry. This is seen as economic life-saver in areas such as New York, Pennsylvania, and
Maryland that have been economically depressed for last decade. According to a report by the
Pennsylvania Labor and Industry Department, from October 2009 through March 2010, 48,000
new jobs were created by the UNGDP industry and its related supply chain, a number expected
to increase as the industry grows [157]. Yet this fast growth poses concerns for the safety for
those who will be filling those job vacancies. This section covers the injuries and fatalities,
overall job hazards (physical, chemical, and social) associated with UNGDP, who is primarily
affected, and how these occupational injuries impact the local community and health care
system.
10.4.1 Injuries and Fatalities
Across the natural resources and mining section, there were 23,280 reported nonfatal injuries
with an average of 11 days away from work, while for the mining industry (which includes oil
and gas extraction), there were 7,060 reported nonfatal injuries with an average of 28 days away
from work [158]. Contact with objects (33.7 per 10,000 full-time employees (FTE)),
overexertion in lifting and lowering (8.8 per 10,000 FTE), and fall on the same level (8.3 per
10,000 FTE) are the most common events leading to nonfatal injuries [158].

These injuries are
due to sprains, strains, and tears (34.9 per 10,000 FTE) and fractures (12.9 per 10,000 FTE)
[158]. Over half of the nonfatal injuries occurred in workers who have been with their employer
for at least one year. This correlates with the age of employees at the time of injury there were
10,060 injuries among workers aged 16-34 and 12,360 injuries among workers 35-64.
77
In the U.S., fatalities in the oil and gas extraction industries reached a high in 2012, with 138
total fatalities, which accounted for 78% of the fatal work injuries in the mining industry [159].
Fatalities are most likely to occur in operations run by small subcontractors (those with less than
19 employees), whether they are engaged in drilling or well servicing [159]. In 2012, the top
three events that led to the fatalities were transportation incidents (49%), contact with objects or
equipment (18%), and fires and explosions (15%) [160]. The increased transportation fatalities
are due, in part, to a fifty year old Department of Transportation exemption that allows drivers in
the oil and gas industry to work longer hours than most truck drivers. [161].
10.4.2 Job Hazards Overall
10.4.2.1 Physical Hazards
The overall job hazards associated with UNGDP can be categorized into physical, chemical, or
social hazards. Physical hazards consist of exposure to high noise levels, slips, trips, and falls,
temperature extremes, fatigue, naturally occurring radioactive material, electrical and other
hazardous energy, working in confined spaces, ergonomic hazards, high pressure lines and
equipment, and machine hazards [162].

According to OSHA, exposure to high noise levels is one
of the most common health hazards throughout the oil and gas extraction industry, and hearing
loss has been characterized by the CDC as the most common work-related illness in the US [162,
163]. Furthermore, excessive noise and/or continuous noise, such as that typically experienced
on a drill site, has documented health impacts such as permanent tinnitus or hearing loss [162].
NIOSH sets occupational noise standards at 85 dBA over 8-hours while OSHAs standards are a
bit higher at 90 dBA over 8 hours [163, 164]. Yet noise sampling in New Yorks Marcellus
Shale UNGD measured sound levels from an air compressor (generates some of the highest noise
on site) at 105 dBA [94]. This loud, continuous noise not falling within the set regulations
creates a dangerous work environment.
10.4.2.2 Chemical Hazards
Chemically, hundreds of chemicals used in hydraulic fracturing. According a 2012 Natural
Resources Defense Council issue brief, there are at least twenty-nine states in which hydraulic
fracturing activities are underway [165]. Only fourteen of the twenty-nine states require some
disclosure and accessibility by the public through FracFocus.org. The requirements of disclosure
vary from state to state; therefore the information stored in FracFocus is incomplete. Table 10-3
lists the types of chemicals, their use in hydraulic fracturing, and the consequences of not using
the chemical.
78

Figure 10-18: Silica Dust from a Well Pad, West Virginia
A 2010 NIOSH study found the specific chemical agents of most concern in UNGD to be silica,
diesel particulate matter (DPM), VOCs, and hydrogen sulfide [30]. The previously mentioned
2010 NIOSH study identified crystalline silica during UNGDP as the most significant health
hazard to workers during UNGD, a finding also supported by the American Public Health
Association [30, 166]. A report by Esswein and colleagues described work crew exposures to
respirable crystalline silica during hydraulic fracturing [30]. According to the report, workers are
exposed to large quantities of silica at multiple points during the UNGD process and currently
deployed engineering controls are not adequately protecting workers. The report documented
116 air samples at 11 fracking sites in five states (AR, CO, ND, PA and TX) taken to evaluate
worker exposure to crystalline silica. The results showed that 47% of the 116 samples collected
exceeded the OSHA permissible exposure limit (PEL) and 79% exceeded the NIOSH
recommended exposure limit (REL). Furthermore, 31% of samples that exceeded the NIOSH
REL value exceed that value by a factor of 10 or more. The report concluded that the use of a
half-face air-purifying respirator does not adequately protect workers because the half-face air-
purifying respirators have a maximum use concentration of 10 times the occupational health
exposure limit.
79

Figure 10-19: Comparisons of arithmetic means of TWAs (mg/m3) for job titles with five or
more samples in relation to a calculated OSHA PEL (based on 53% silica) and NIOSH REL for
respirable silica. Maximum values for each job title shown by diamonds at the end of dashed
lines, Source: [30]

Figure 10-20: Natural Gas Flaring
Given the amount of diesel-based heavy duty vehicles, machineries and generators at the UNGD
sites, worker exposure to DPM at these sites is of significant public health concern [136].
Workers exposed to diesel exhaust are at increased risk of adverse health outcomes ranging from
irritation of the eyes and nose, headaches and nausea, to respiratory disease and lung cancer
[136]. Though diesel exhaust has been classified by International Agency for Cancer Research
(IARC) as a known human carcinogen (Group 1),

there are currently no federal workplace
standard for worker exposure to DPM [167]. The state of California does regulate DPM at
workplace with an 8-hour TWA not to exceed 20 mg/m
3
[136]. Limited monitoring data
80
available showed a quarter of the UNGD sites monitored exceeded the 20 mg/m
3
threshold
[168].
While research on the impact of hydrogen sulfide at UNGDP sites is lacking, its impact on
workers at other natural gas processing plants is raising concern for UNGDP workers. Hydrogen
sulfide is a compound that is released through the venting and flaring of natural gas throughout
the extraction and refining processes as a safety precaution [169].

This is because hydrogen
sulfide is extremely flammable and when mixed with air can be explosive. Additionally, it may
travel to sources of ignition and flash back. When ignited, the burning gas produces toxic vapors
like sulfur dioxide [170].

Short-term exposure to hydrogen sulfide has been linked to nausea,
headache, shortness of breath, sleep disturbance, throat and eye irritation, while long-term
exposure causes olfactory nerves paralysis, respiratory inflammation, chronic bronchitis, and
chronic tearing of the eyes [169].

As previously stated, data on hydrogen sulfide exposure at
UNGDP work sites is limited, but information on its dangers during other natural gas processing
and refining show hydrogen sulfide to be a very dangerous chemical if not handled properly. In
2010, a natural gas well salvage and capping business based in Zanesville, Ohio failed to provide
training, along with eye protection, a written respiratory protection program, a written hazard
communication program, and material safety data sheets on hydrogen sulfide that resulted in the
death of a worker at a natural gas well site in Londonderry, Ohio [171].

Similarly, workers were
injured while vacuuming explosive dust to clean out a natural gas processing unit in a Eustace
Gas Processing Plant in Eustace, Texas in 2011 [172].

These incidences show UNGDP workers
are at increased risk of adverse health outcome, if more adequate safety measures are not put in
place.
10.4.2.3 Social Hazards
Due to the transient nature of employment in the UNGDP sector, workers experience a number
of psychosocial issues, including mental distress, suicide, stress, and substance abuse. A project
conducted by the UC Davis Center for Reducing Health Disparities on migrant Latina/o
agriculture workers and the communities they migrate to focus on the mental burden these
workers face. While not employed by the same industry or even of the same ethnic background,
UNGDP workers have much in common with Latina/o migrant agriculture workers and their
experiences produce similar mental health outcomes. Migrant workers tend to be desperate for
obtaining and maintaining employment in order to provide basic necessities for their families
[173].

There are stressors that lead to depression, anxiety, and drug and alcohol abuse. The UC
Davis project also identified illicit drugs as being sometimes used as a means to cope with or
relax after working long hours. Migrant workers experience social stressors such as avoidance at
best or discrimination at worst from the communities, and do not bring their families along. As
such, they are even more isolated from their support system and more at risk for turning to
substances to cope.
This also puts a strain on communities that host UNGDP activities. In Garfield County, Colorado
the increase in UNGDP activity coincided with increases in violent crime arrests and drug
violations for adults and juveniles alike, along with an increase incidence of STIs [7].

Furthermore, substance abuse information extracted from the Garfield County 's 2006
assessment on community needs indicated depression, anxiety and stress along with tobacco
smoking and alcohol abuse appear to be the top indicators of the burden of mental health and
substance abuse, respectively.
81
Furthermore, a disproportionate number of the workers lack health insurance [174].

Rural
healthcare facilities are not trauma centers; they are designed to be family clinics. The increases
in accidents and other health issues, such as STIs have put a strain on the healthcare system as
local hospitals and governments are not compensated for their healthcare services. A hospital in
a North Dakotan fracking boom town saw its ambulance visits increase four-fold and its debt
increase 2,000% to $1.2 million over the past five years [174].
A study conducted by the U.S. Chamber of Commerces Institute for 21st Century Energy has
shown shale oil and gas extraction has accounted for 4,000 ambulatory health care and hospital
jobs in North Dakota and more than 2,000 healthcare jobs in Louisiana, with 4,000 more
projected by 2035 [175].

Yet the fate of these workers after the UNGDP activity begins to wane
is a critical issue that should not be overlooked, this time around. Socially, UNGDP poses a
threat to the livelihood of the communities the activities take place in as well as those who work
in them.
10.4.3 Assessment
Based on our review of the occupational health hazards associated with UNGDP (section 10.4),
we conclude that there is a High Likelihood that UNGDP related activities will have a negative
impact on occupational health. Table 10-20 provides the scoring for the evaluation criteria that
we used to arrive at this conclusion.
Table 10-20: Occupational Health Evaluation
Geographic extent 2
Overall Score 17
Hazard Rank H

1. Vulnerable population received a score of 2 workers are disproportionately affected.
2. Duration of exposure received a score of 3 because these workers are employed in the
UNGDP industries for > 1 year where their exposures continue to persist.
3. Frequency of exposure received score of 2 as workers exposures are frequent.
4. Likelihood of adverse effect was assigned score of 3 because evidence suggests that these
workers are exposed to very high level of hazards, including crystalline silica.
82
5. Magnitude/severity of health effects was assigned score of 3 because the potential
adverse health effects (silicosis, lung cancer) are irreversible.
6. Geographic extent received score of 2 because the workers are from different areas.
7. Effectiveness of setback was assigned score of 2 because adequate setback will not
mitigate workers exposure.
83
11 REGULATORY LANDSCAPE
Here, we briefly describe the scope and implications of pertinent federal regulations and examine
the approach to regulation of UNGDP taken by selected states.
11.1 Federal Regulations
While Congress has the authority to regulate hydraulic fracturing activities under the Commerce
Clause (Article I, Section 8, Clause 3 of the U.S. Constitution), legislation of the practices
involved in hydraulic fracturing has been primarily put in the hands of the states. Each state is
allowed to regulate hydraulic fracturing activities as it sees fit while still maintaining the
minimum federal regulations. Yet at the federal level there is little power to regulate hydraulic
fracturing through most of the major federal environmental statutes, which include: the Safe
Drinking Water Act (SDWA), the Clean Water Act (CWA), the Clean Air Act (CAA), the
Resource Conservation and Recovery Act (RCRA), the Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA), and the Emergency Planning and
Community Right-To-Know Act (EPCRA) due to the enactment of the 2005 Energy Policy Act.
This act was intended to help formulate a new national energy strategy that would address
energy production in the United States, focusing on areas such as renewable energy, energy
efficiency, climate change technology and domestic extraction of oil and gas. In the process, it
created exemptions for natural gas and oil drilling (which hydraulic fracturing falls under),
known commonly as Halliburton loopholes [176], which allow hydraulic fracturing activities to
bypass the major federal environmental statutes something that most other large energy
industries are not privy to [177].
11.1.1 Water
Our waters are protected by the Safe Drinking Water Act (SDWA) and the Clean Water Act
(CWA). The SDWA mandates regulation of underground injection activities in order to protect
groundwater resources [59].

Under this statute, groundwater is classified as underground water
reserves (e.g., aquifers). Additionally, the SDWA is designed for public municipal drinking
water; therefore, the 15% of Americans on private drinking wells and one-third of Maryland
residents using private wells are not protected under this Act [21, 178]. Under the 2005 Energy
Policy Act, UNGDP is excluded from the SDWAs underground injection terminology unless
diesel fuels are used during the injection process [179].

Additionally, previous sections of this
report have shown groundwater may be affected, altered or contaminated by UNGDP fluids or
mobilization of naturally occurring minerals, gases or radiation. Together, this places all of us
who drink municipal water at risk for exposure to UNGDP chemicals, and places those using
private wells at an even greater risk.
The CWA was enacted to protect and improve water quality in the nations rivers, streams,
creeks, and wetlands [180].

In order to achieve this goal, the CWA requires permits for all
discharges of pollutants to those waters. Under the 2005 Energy Policy Act, term pollutant
does not include water, gas, or other material that is injected into a well to facilitate production
of oil or gas, and UNGDP is exempt from the permit requirements [179].

Furthermore, the 2005
Energy Policy Act broadened the discharge permit exemption to include stormwater discharge
from oil and gas construction activities. Although in a suit brought against the US Environmental
84
Protection Agency (EPA) by the National Resource Defense Council (NRDC), the Court decided
this broadened exemption was unlawful, the EPA has yet to set forth a better measure to regulate
this kind of discharge [181].

Without these protections in place, the natural habitats surrounding
UNGDP sites are in danger of being destroyed. This has both direct and indirect consequences
for public health. Directly, habitat destruction may result in floods, heat waves, water shortages,
landslides, earthquakes; while indirectly, we will see changes in disease risk, reduced crop yields
(malnutrition/stunting), and depletion of natural medicines associated with habitat destruction
[182].

11.1.2 Air
In 1970, the Clean Air Act (CAA) was passed in an effort to protect and enhance the quality of
the Nations air resources so as to promote the public health and welfare [183]. As previously
stated, natural gas production produces toxic air pollution, including volatile organic compounds
(VOCs) (which react with sunlight to form ground level ozone or smog), methane, hydrochloric
acid, and hydrogen sulfide, all contributors to greenhouse gases. Under the CAA, when
numerous small sources of air pollution, such as individual oil and gas wells, are under common
control and in close proximity they are treated as a major source and subject to CAA best
technology requirements, and require an emission permit to ensure their emissions are under a
set threshold [183].

With the passage of the 2005 Energy Policy Act, most oil and gas production
sites are not are treated as a major source and are not required to obtain an emission permit
[179].

Many supports of the natural gas industry argue emissions from natural gas power
generation are half of that of coal and are comprised more of methane than carbon dioxide, and
are therefore not of a great concern in terms of global greenhouse gas emissions. However,
according to the EPA, methane is more efficient at trapping radiation than carbon dioxide.
Pound for pound, the comparative impact of methane on climate change is over 20 times greater
than carbon dioxide over a 100-year period [184]. Furthermore, the natural gas/petrol industry
was the biggest methane emitter from 1990 to 2012 [185].

The impact of greenhouse gases in
terms of climate change on human health ranges from increases in tropical disease incidents such
as malaria and cholera to widespread crop failure to mass population displacement [182].

11.1.3 Waste Disposal and the Right to Know
The health and safety of the land and those who occupy it is protected under the Resource
Conservation and Recovery Act (RCRA) and the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA). RCRA enacted by Congress as a cradle to grave
regulatory framework for managing solid waste, including hazardous waste [186].

Under the
1980 amendments to RCRA known as the Solid Waste Disposal Act Amendments of 1980,
Congress temporarily exempted oil and gas exploration and production wastes from regulation
under RCRA until the completion of an EPA study to determine whether such wastes should be
regulated as hazardous waste under RCRA [187].

EPA finalized this study in 1988, and
concluded that regulation of hazardous oil and gas waste under RCRA was unnecessary. This
means the hydraulic fracturing fluid, a mixture of water and all the hydraulic fracturing additives
(chemicals), used to perform high volume horizontal hydraulic fracturing are not considered
hazardous material or hazardous waste and therefore is not regulated in transportation. Therefore
labeling, shipping, record keeping, training, etc. are not required for transport. This is contrary to
85
the previously mentioned studies in the report that have identified know toxins and carcinogens
in hydraulic fracturing fluid indicating that they should be regulated under RCRA.
Congress enacted CERCLA in 1980, creating a framework for cleanup of toxic materials through
creation of the Superfund Program [188].

The oil and gas industry was taxed in order to pay into
the Superfund and in exchange was exempted from CERCLAs requirements until 1985. But
according to the Environmental Defense Center, the industry continues to be exempt while not
paying the tax [189].
These two acts (RCRA and CERCLA) tie directly into the Emergency Planning and Community
Right to Know Act (EPCRA). EPCRA was enacted in 1984 in the wake of the chemical
explosion and disaster in Bhopal, India, and generally requires companies to disclose information
related to locations and quantities of chemicals stored, released, or transferred [190].

This is done
for the safety and welfare of those who live and work in an area in which toxic or harmful
substances are used, deposited, or transported; they have the right to know what they are being
exposed to in their homes and work. Under the 2005 Energy Policy Act, oil and gas exploration
and production wastes were exempted from this requirement. EPCRA only applies to hydraulic
fracturing when diesel fuels are used [179].

Industry argues revealing chemical compositions of
fracture fluid formulations could reveal valuable corporate trade secrets. Additionally, industry is
allowed to withhold the specific chemical identity from the reports filed under sections 303, 311,
312 and 313 of EPCRA if the facilities submit a claim with substantiation to EPA. Additionally,
the industry argues material safety data sheets (MSDS) are posted on-site at UNGDP sites, as
required by law, and MSDSs are freely accessible to the public online [191].

However, simply
providing MSDSs is not the same as understanding the exposures risks associated with chemical
mixtures, especially if the individual reading the MSDS is not well versed in chemistry or
toxicology. Furthermore, UNGDP companies tend to leave out key details when listing
compounds on their MSDS. "Frac fluid with additives" is a commonly listed compound on
fracking MSDSs, yet nowhere on the sheet does it list what the additives are [191].

11.2 State and Local Regulations
The 2005 Energy Policy Act largely exempted oil and gas development from regulation at the
federal level based on an assertion that the oil and gas industry was adequately managed under
state regulations. It was assumed that one size fits all federal regulations would be
inappropriate given the diversity of geology and environments among states. This has resulted in
a checkerboard of varying regulations at the state and local levels particularly with regards to
setback and disclosure requirements [192, 193]. While most regulation occurs at the state level
rather than local level, some local governments have taken an aggressive stance toward UNGDP
regulation [194, 195]. In Pennsylvania, the Supreme Court recently struck down limits imposed
by the state legislature on local zoning control of oil and gas development [196, 197] providing
an increasing role for local government in that state. The district court is now charged with
determining if the remainder of Pennsylvania Act-13 is consistent with the state constitution after
invalidation of the zoning preemption. In a recent decision in June 2014, the New York State
Court of Appeals ruled that towns can use zoning ordinances to ban hydraulic fracturing. This
ruling supports efforts of towns across New York including Dryden in Tompkins County and
Middlefield in Otsego County who have modified their zoning ordinances to ban the practices on
86
the grounds that it would threaten public health, environment, and the social fabric of local
communities [198, 199].
The status of state regulations has been extensively reviewed (Richardson et al. 2013; Konschnik
& Boling 2014; McFeeley 2012). However, regulations are rapidly evolving with new laws
passed, new regulations proposed or issued, and court decisions rendered on an almost daily
basis.[197, 200202] Hence, any review is rapidly out of date. A comprehensive review of state
statutes and regulations across the U.S. is outside the scope of this document.
11.2.1 Setback Requirements
It is clear that states (and some localities) differ widely in setback requirements. A recent report
found that setback requirements ranged from 100 to 1000 feet [193]. The City of Dallas, Texas,
recently enacted an ordinance requiring 1500 feet setbacks [195]. Maryland should base setback
regulations on best available science. If the State decides to allow UNGDP, continual
improvement of regulation and enforcement will be necessary as new information becomes
available, regardless of whether the state decides to allow development to begin in 2015 or after
several more years of study. Thus, recognition of a need for continued evolution of standards and
regulations, will be important to acknowledge in the initial regulations [192].
11.2.2 Chemical Disclosure
A second area of important variation among states is the requirement for disclosure of
information about chemicals used in UNGDP. Twenty-two states have some requirements for
disclosure of hydraulic fracturing chemicals [203]. Of these states, 15 require disclosure to
FracFocus.org, including 10 that make the website the primary or sole location for reporting.
FracFocus has received much criticism for its relatively primitive technology preventing
searching and aggregation, onerous terms of use, and its lack of date-time stamping and logging,
quality control, verification, and provisions for permanent archiving of data [203205].
Recommendations for critical improvements were made by a Task Force of the Secretary of
Energy Advisory Board [205]. Four states require that records submitted to FracFocus also be
submitted directly to the state. California also requires that a the data be made available on a
state government website that allows the public to easily search and aggregate [submitted data]
using search functions on that Internet Web site [206 Sec 3160].
All current state disclosure laws and regulations make some provision for protecting confidential
trade secret information [203]. Yet, the case for the existence of valid, commercially important
trade secret information about the identity of chemicals used in the UNGDP industry was
undermined recently by a major oil and gas industry service companys commitment to disclose
100% of the chemical ingredients we use in hydraulic fracturing fluids [207]. Part of the
strategy here may be to disclose in an effort to defuse concerns about the chemicals being used.
A letter to the Alaska Oil and Gas Conservation Commission from ten law professors who
specialize in intellectual property made the case against trade secret protection in regulation of
the UNGDP industry without questioning the legitimacy of the secrets [208]. They argue, "the
publics interest in assuring that hydraulic fracturing is managed in a manner that addresses all
significant risks may legitimately outweigh commercial concerns. Furthermore, trade secrecy
should not impede disclosure of information when the information describes public risks that the
87
trade secret claimant is itself creating. Thus, even full-blown property rights do not legitimate
harming third parties or avoiding duties.
Some states have or are developing administrative procedure to review trade secret claims. The
Wyoming Supreme Court recently ruled that, under that states constitution, failure to disclose
chemical constituents of hydraulic fracturing fluids required detailed justification [200].
However, [a]dministrative agencies are poorly positioned to evaluate and monitor trade secrecy
claims and this function is resource intensive[208]. California does not allow [t]he identities of
the chemical constituents of additives, including CAS identification numbers to be claimed as
trade secrets [206 Sec 3160]. But, the states new Well Stimulation law does require
companies to submit extensive information and provide for administrative review of the validity
of any submitted claims of confidential trade secrets. Such an effort may be within the reach of
resources in a large state such as California but may not be feasible in Maryland. Thus,
alternative approaches [192], or disallowing trade secrets may be necessary to strike the proper
balance between public and private interests if this industry is to operate in Maryland.
11.2.3 Other Forms of Well Stimulation
Another area of important variation between states is the extent to which UNGDP regulations
address all forms of well stimulation for oil and gas production including potential new and
emerging technologies. In his recent review of disclosure laws, McFeeley [203] found that only
three states, California, Ohio, and Wyoming require chemical disclosures for all types of well
stimulation while most states address only hydraulic fracturing. California titles its act Well
Stimulation and specifically mentions acid well stimulation as well as hydraulic fracturing
[206]. It seems likely that new extraction technologies will continue to evolve. Thus, a forward
looking approach, rather than focusing on the technology of the moment, seems warranted.
88
12 RECOMMENDATIONS
12.1 Comprehensive Gas Development Plans (CGDP)
Potential public health impacts and prevention and mitigation strategies should be included in the
CGDP so that the required and routine public hearings on the plan can include an informed
discussion of health as well as environmental impacts.
R1. Require assessment of air quality and other potential health impacts and
propose strategies to protect the community and workers from exposure to
hazardous air pollutants.
Air quality is one of the major potential areas of UNGDP health impacts. It is
essential, therefore, that air quality and resulting potential health impacts be
addressed at the earliest stage of development in the Comprehensive Gas
Development Plan.
R2. Require assessment of whether application of standard setback distances will be
adequate to protect public health, including consideration of prevailing winds
and topography.
Each area for proposed development will have unique features not limited to the
geology of the site. Differing vulnerabilities of adjacent populations as well as
physical features of the landscape may impact the likely effectiveness of setback
requirements in preventing health effects from UNGDP. The CGDP, as the first step
in development, needs to address this issue.
R3. Require disclosure of planned well stimulation methods and classes and amounts
of chemicals to be used.
Although test well results as well as passage of time between submission of a CGDP
may result in some changes in the details of planned well stimulation methods, the
general approach, methods (hydraulic fracturing, acid stimulation, or other), and types
and expected amounts of chemicals expected to be used should be available in
advance of the public hearing mandated for consideration of the CGDP. More
specific data will be required at later stages as described below. However, because
individual well permits are not subject to mandatory public hearings unless MDE is
petitioned by individual stakeholders, and because the health impacts and adequacy of
prevention measures cannot be assessed at the CGDP stage without knowledge of
proposed well stimulation methods, disclosure at the CGDP stage is essential.
R4. Require a quality assurance plan.
Simply having proposed prevention plans in place is not sufficient. A method of
ensuring that the planned methods are implemented and monitored is also essential.
R5. Require an air, water, and soil-monitoring plan.
Air, water, and soil monitoring are complex undertakings and adequate monitoring
will require significant planning. The CGDP should provide the plan so that it is
subject to public review at the mandatory public hearing.
89
R6. Require assessment of impact on and a monitoring plan for potential fugitive
emissions from existing and historic gas wells within the horizontal extent of the
fractured area.
Experience in WV suggests that horizontal drilling and hydraulic fracturing can cause
long dormant and abandoned wells to begin leaking. The leaks may be associated
with a variety of air quality problems, as well as atmospheric methane releases. At the
time of the CGDP, the existing wells need to be identified and assessed for potential
impact of new development, and a plan for monitoring these potential emission
sources put in place.
R7. Require that all UNGDP materials and wastes be stored in closed tanks; open
pits shall only be used for storage of fresh water.
This requirement is consistent with the MDE report. The plan for siting of the
required tanks and related infrastructure needs to be included in the CGDP.
12.2 Disclosure of Well Stimulation Materials
Recommendations concerning disclosure were revised and moved to a separate section based on
feedback received at and following the public progress report on June 28, 2014. The final
recommendations are now in line with the proposed legislation H.B. 1030 [8]. Three phases of
disclosure are included a preliminary more general disclosure with the CGDP, a specific
detailed disclosure with the well permit application, and a specific detailed disclosure after well
stimulation is finished.
R8. Require preliminary disclosure at time of CGDP submission (see CGDP
recommendations), detailed disclosure at time of well permit application, and
detailed reporting of actual materials used within 30 days of finishing well
stimulation activities. Require notification of MDE, local emergency responders
and public notice of significant variances from materials and concentrations
proposed in the permit within 24-hours of occurrence.
We recommend three phases to the disclosure process. The first phase occurs as a part
of the CGDP and is necessarily more general and preliminary. As described in the
recommendations for the CGDP (see 12.1 above), this is necessary for informed
public discussion of the merits of the plan. The second phase, close in time to the
actual well stimulation activity, is the appropriate time for detailed disclosures. The
final phase is a record of materials actually used in the well stimulation process. An
additional requirement is made for immediate notification of relevant responders and
the public if variations from the approved detailed submission occur.
R9. Require detailed disclosures to include CAS numbers, volume and concentration
of every chemical or distinct material including proppants, their physical form,
and identification of engineered nanomaterials including drilling muds and
hydraulic fracturing and other fluids used in well stimulation. Do not allow
claims of trade secrets for identities and concentrations of specific chemicals or
nanomaterials used in well stimulation.
90
As described in section 11.2.2 above, the legitimacy of claims of trade secrets in the
hydraulic fracturing industry is questionable, and even if one grants that some
legitimate trade secrets exist, we believe public risk should outweigh commercial
concerns especially where the potential risks are created by the trade secret claimant.
Were trade secrets to be allowed, administrative due process would be required each
time the state declined to release requests made by the public and initial validity of
claims should also be reviewed at the time of submission rather than allow
unnecessary delay in access by medical practitioners, emergency responders, or
public health researchers. This administrative burden is likely beyond the resources of
MDE. What is important for public health is disclosure of the amounts of specific
materials injected into the ground, released into the air, or otherwise potentially
released into the environment. The exact formulation of products used is not needed
rather it is the final concentrations and amounts in the fluids. Structuring the
disclosures in this way may avoid some issues of trade secrets.
R10. Require detailed disclosures to include base fluid volume and sources including
percentages that are recycled fracturing fluid, production water, and fresh
water.
While recycling of fracturing fluids is highly desirable, this also raises the potential
for accumulation of naturally occurring contaminants and reaction products of
fracturing chemicals. Thus, knowing the sources of the fluids is important.
R11. Require simultaneous submission to state regulators and FracFocus.
FracFocus is a somewhat useful web portal for information on UNGDP materials and
methods. However, as described in section 11.2.2 above, it also has numerous well-
known severe limitations that make is unsuitable as the primary or sole repository for
disclosed information.
R12. Collaborate with California to develop a State controlled and archived Internet
Web site consistent with the provisions of California SB 4.
Under its new Well Stimulation legislation and proposed implementing regulations,
California is in the process of developing an accessible and more useable Internet
Web site for disclosure of well stimulation material identities. This web site will
address many of the limitations in FracFocus. It would be more cost effective for
Maryland to partner with California than to try and develop its own web portal.
R13. Implement the provisions of H.B. 1030 for timely access to disclosed information
by medical professionals, emergency responders, poison control centers, local
officials, scientists, and the public.
The provisions of H.B 1030, considered by the House of Delegates during its 2014
session, particularly its requirement for timely disclosure of information, are
supported by this studys analysis.
91
12.3 Air Quality
fields, we conclude that there is a High Likelihood that UNGDP related changes in air quality
will have a negative impact on public health in Garrett and Allegany Counties. Should Maryland
R14. Require a minimal setback distance of 2000 feet from well pads and from
compressor stations not using electric motors.
Evidence from traffic related air pollution studies show that concentration of traffic
related pollutants drop down to background level beyond 500-700m (1640-2296 feet).
Likewise study from Colorado shows concentration of air pollutant significantly
higher within 0.5 miles (2640 feet) of UNGDP facilities compared to >0.5 miles.
Based on this, we concluded that adequate setback from periphery of the UNGDP
facility to the periphery of residential property can minimize exposure. Based on this
data, we recommend minimal setback distance be 2000 feet.
R15. Require electrically powered motors wherever possible; do not permit use of
unprocessed natural gas to power equipment. This recommendation is designed
to reduce VOCs and PAHs emissions from drilling equipment and compressors.
VOC and PAH emission into the local environment can be eliminated by using
electrically powered motors. This is consistent with the recommendation in the
UMCES-AL report.[209]
R16. Require all trucks transporting dirt, drilling cuttings to be covered.
Fugitive dust from trucks transporting dirt, drilling cuttings and other waste materials
is of concern to the community. Spill from these trucks also contribute to the soil and
water contamination issues. To minimize, require all trucks transporting these
materials to be covered.
R17. Require storage tanks for all materials other than fresh water and other UNGDP
equipment to meet EPA emission standards to minimize VOC emissions.
The EPA issued final standards for emissions from storage tanks and other UNGDP
equipment in 2012 with subsequent updates. [210, 211] Maryland should require all
facilities, not merely large ones, to meet these standards.
R18. Establish a panel consisting of community residents and industry personnel to
actively address complaints regarding odor.
Community residents from Doddridge County in WV complained they often
encounter periods of intense odor that is sometimes followed by acute respiratory
ailments. These residents feel powerless as there is no one to help them understand
the causes of such episodes and ways to minimize them. We recommend
establishment of a panel consisting of community and industry representatives that
will work to identify the causes of such episodes and minimize/eliminate them.
92
R19. Conduct Air Quality Monitoring
a. Initiate air monitoring to evaluate impact of all phases of UNGDP on
local air quality (baseline, development and production).
b. Conduct source apportionment that allows UNGDP signal to be
separated from the local and regional sources.
c. Conduct air monitoring with active input from community members in
planning, execution, and evaluation of results.
d. Conduct air monitoring in a manner to capture both acute and chronic
exposures, particularly short-term peak exposures.
e. Clearly communicate to community members expectations about what is
achievable through air monitoring.
Air monitoring should be conducted to determine the impact of UNGDP on
community air quality. This process should incorporate input from community
members regarding the location of the monitor, type of pollutants to be monitored,
and sampling interval to capture peak concentrations. Community members should be
informed regarding the expected outcomes, and results should be disseminated in a
timely manner.
12.4 Flowback and Production Water-Related
health in Garrett and Allegany Counties. The overall score for the Flowback and Production
Water Related hazard category is primarily driven by concerns related to water quality. Should
Maryland move forward with UNGDP, the following recommendation should be implemented to
prevent or minimize potential negative impacts on public health.
12.4.1 Water & Soil Quality
R20. Prohibit well pads within watersheds of drinking water reservoirs and protect
public and private drinking water wells with appropriate setbacks.
The potential for contamination of drinking water is of significant concern to
community residents. Risk of public drinking water reservoir contamination should
be limited by prohibiting well pads in the reservoir watersheds. Because many rely on
public and private wells as their primary source of drinking water, appropriate
safeguards for well water are also important. Appropriate setbacks for private and
public groundwater wells should be established for each well based on hydrogeologic
evaluation as part of the CGDP.
R21. Implement UMCES-AL/MDE water monitoring plan. Require monitoring of
water quality during initial gas production and at regular intervals thereafter.
93
The UMCES-AL and MDE reports provide a reasonable water-monitoring plan.
Maryland should incorporate that recommendation and require monitoring at regular
intervals.
R22. Implement the UMCES-AL recommendations for management and recycling of
flowback and production fluids.
The UMCES-AL report by Eshleman and Elmore [209] provides extensive
recommendations for management and recycling of flowback and production fluids.
In particular, we endorse their recommendation 3-J. UNGDP in Maryland should not
be permitted until an adequate means of disposal of any residual waste, without
extensive trucking, is identified.
R23. Require identification and monitoring of signature chemicals in fracturing
fluids to allow for future identification of ground water
There is a need to identify a panel of signature chemicals that are specifically
associated with UNGDP. The monitoring campaign described in UMCES-AL and
MDE reports should be augmented with these signature chemicals or potential
tracer that can be added in the fracturing fluid to identify water
R24. Conduct soil monitoring in areas potentially impacted by UNGD upset
conditions.
Periodic soil monitoring should be conducted to track potential contamination with
semi-volatiles, heavy metals, and radionuclides. These sampling plans should be
augmented with more intensive campaign if there is evidence of accidental spills
(upset conditions).
R25. Prohibit flowback and production wastewater or brine use to suppress road
dust, de-ice roads, or other land/surface applications.
Flowback, production water or brine contains many chemical agents, heavy metals,
NORMs and other materials used in fracturing fluids. Therefore, their use as road
dust suppressor, deicers and/or other land/surface application should be prohibited
consistent with the recommendations in the UMCES-AL report.
12.4.2 NORM
R26. Conduct research to identify the appropriate suite of priority radionuclides for
assessment of radiological activity.
Studies have relied on radium as a surrogate for overall radioactivity. Emerging
evidence suggest that there may be additional radionuclides that may be of concern to
human health, and may in fact be present at appreciable concentration. There is a
need to characterize a suite of radionuclides that are of concern and use them in the
monitoring studies. In the meantime, metrics such as total alpha activity, or total
gamma activity should be used to assess radiological contamination and support
decision-making.
94
12.5 Noise
in noise exposure will have a negative impact on public health in Garrett and Allegany Counties.
Should Maryland move forward with UNGDP, the following recommendation should be

R27. Implement noise reduction strategies recommended by UMCES-AL in the MD
Best Management Practices, including requiring electric motors wherever power
supplies are available and construction of artificial sound barriers.
Currently technologies do exist to reduce noise levels. In fact such technology is used
in urban locations such as Fort Worth, TX (personal communication, API). But
because of the cost associated with them, such technologies are not used in places
such as Doddridge County, WV. Maryland should require such noise reduction
strategies at all locations.
R28. Require a setback of 2,000 feet for natural gas compressor stations using diesel
engines, 1000 feet for stations using electric motors and sound barriers.
Based on our data from WV, noise hazard can be minimized through setback
distance. Therefore, Maryland should require a setback of 2,000 feet for facilities
using diesel engines.
R29. Establish a system to actively address noise complaints.
Panel established with community and industry representatives to monitor the issues
related with odor should also be tasked with monitoring the noise complaints and
addressing them.
12.6 Earthquakes
not allow deep well injection of wastewater, there is a Low Likelihood UNGDP related
Should Maryland move forward with UNGDP, following recommendation should be taken into
consideration to minimize potential negative impact on public health.
R30. Collect baseline data on seismic activities using methods that can record
earthquakes smaller than magnitude 3.
Earthquakes associated with hydraulic fracturing are of small magnitude. There is a
need to collect baseline data on these small earthquakes so changes in trend over time
can be established.
95
R31. Restrict issuing UIC Class II permits for disposal of UNGDP fluids until
licensing requirements adequately addresses earthquake risk.
Previous studies have established link between deep well injection of wastewater and
increased incidence of earthquakes greater than magnitude 3. Maryland should
restrict issuing UIC Class II permits for disposal of UNGDP fluids. Deep well
injection of UNGDP fluids in existing wells should also be banned.
R32. Implement use of sensitive seismic monitoring technology to better detect small
earthquake activity that could presage larger seismic events as well as using a
traffic-light system that sets thresholds for seismic activity notification.
An advance warning system should be developed to warn citizens on potential future
earthquakes, based on small earthquake activity.
Based on our review of social determinants of health (section 10.3.5), we conclude that there is a
High Likelihood UNGDP related activities will have a negative impact on the social
determinants of health. Should Maryland move forward with UNGDP, the following
recommendation should be implemented to prevent or minimize potential negative impacts on
public health.
12.7.1 Traffic Safety
R33. Increase state and local highway patrols to closely monitor truck traffic subject
to the Oilfield Exemption from highway safety rules.
The Oilfield Exemption from highway safety rules allows truck operators to work
extended hours without sleep. This creates a dangerous situation with sleep deprived
operators driving their vehicles at high speed through rural roads that are not designed
to handle such heavy traffic. To minimize this hazard, additional highway patrols
should be hired to closely monitor truck traffic.
R34. Empower local communities to control truck speed and traffic patterns.
Local communities should be empowered to determine and enforce routes for truck
traffic, as well as installing speed bumps to control speed.
R35. Route truck traffic to maintain separation between UNGDP activities and the
public.
Truck traffic should be routed during off peak hours, such as after morning commute,
school bus transport and before afternoon rush hours.
R36. Consider use of pipelines to move UNGDP fluids between sites.
When possible, consider using pipelines to move UNGDP fluids between sites as it
will minimize the issues related to spill, traffic accidents as well as traffic-related air
pollution.
96
12.7.2 Empower communities
Strong, resilient communities are an important defense against the psychosocial stressors and can
make a major contribution to limiting exposure to chemical and physical hazards. Empowered
local communities improve community resilience. Processes should be implemented to ensure
public participation in decision-making associated with UNGDP activities, particularly actions to
reduce or eliminate negative environmental, social, and public health impacts of UNGDP
activities.
R37. Enact a Surface Owners Protection Act as recommended in the MDE Part I
report.
During scoping process, community members expressed concerns about surface right
owners who lack mineral rights for their property. Since mineral rights trumps surface
rights, this particular subgroup is considered to be among the most vulnerable. They
experience chronic stress that is detrimental to their mental and physical health.
Maryland should enact a Surface Owners Protection Act as recommended in the
MDE Part I report.
R38. Engage local communities in monitoring and ensuring that setback distances are
properly implemented.
We recommend that local communities should be empowered to ensure adequate
setback distances are maintained.
R39. Create a mapping tool for community members using buffer zones (setback
distance) around homes, churches, schools, hospitals, daycare centers, public
parks and recreational water bodies.
A user friendly mapping tool should be created that enables community members to
incorporate buffer zones (setbacks) around sensitive human receptor sites and
ecological assets including homes, churches, schools, hospitals, daycare centers,
parks, recreational water bodies and map specific areas where UNGDP should be
restricted.
We recommend that the user friendly mapping tool should be freely available to
community members. The team began the development of a public participatory GIS
tool for the project. DHMH staff should expand the online mapping tool, host the
tool, and make it available for use by residents, health practitioners, advocates, and
other stakeholders. This tool will aid communities in implementing recommendation
number R38.
12.8 Healthcare Infrastructure
Based on our evaluations of the current healthcare infrastructure in Garrett and Allegany
Counties as well as expected number of migrant workers that will come to these areas, we
conclude that there is a High Likelihood that UNGDP related activities will have a negative
impact on public healthcare infrastructure in Garrett and Allegany Counties. Should Maryland
or minimize potential negative impact on public health.
97
R40. Closely monitor whether prospective UNGDP companies provide adequate
health insurance coverage for all employees.
Insured workers using healthcare services could offer positive support to existing
systems as long as their rate of utilization is within the scope of available capacity.
Uninsured workers, like any other uninsured population, would place stress health
care infrastructure due to their inability to pay for services.
R41. Organize a local health care forum with key stakeholders to assess health care
services and anticipated needs related to UNGDP.
The use of primary and public health care systems, especially in the areas of
emergency, urgent care, and trauma care, may rise as a result of an increase in the
UNDGP workforce. It is important to assess current healthcare infrastructure capacity
to meet these anticipated needs.
R42. Inform and train emergency and medical personnel on specific medical needs of
UNGDP workforce.
UNGDP workers have specific emergency, urgent, and trauma care needs due to
higher rates of occupational related incidents and injuries and providers most likely to
service UNGDP workers (e.g., emergency personnel and trauma specialists) should
be adequately prepared and trained to respond to their needs.
R43. Review and monitor county-level tax revenues and assess improvements
necessary to meet increased services need.
Prioritizing health infrastructure at a high level when appropriating local government
revenues derived from UNGDP and engaging in long-term planning for healthcare
infrastructure development is critical to alleviating existing and anticipated healthcare
infrastructure pressures.
R44. Establish a committee of state and local stakeholders (including UNGDP officials
and local providers and residents) for early identification of impacts to
healthcare infrastructure.
Previous research indicates that healthcare infrastructure impacts will be concentrated
during the first phase of UNGDP, when labor needs are high and larger numbers of
workers are expected. Initiating ongoing monitoring of healthcare infrastructure
utilization rates by collecting information on patients occupational status is strongly
recommended along with close monitoring of healthcare infrastructure access with
attention to emergency and trauma care and vulnerable populations.
R45. Initiate monitoring of UNGDP healthcare-related costs.
There is a critical data gap of evidence-based research and monitoring around
healthcare-related costs of UNGDP. Economic analysis of medical and healthcare
infrastructure costs of increased disease rates and injuries from UNDGP should be
initiated.
98
12.9 Cumulative Exposure/Risk
The combination of chemical, physical, and psychosocial stressors can lead to effects that are
cumulative involving potentially additive or multiplicative interactions among the exposures.
Observed health impacts, if any, will result from these cumulative impacts. Most of the
recommendations in this report are targeted at primary prevention (i.e., to prevent the occurrence
of adverse health effects). However, a monitoring method is needed to verify the effectiveness of
primary prevention activities and to improve them as necessary. Furthermore, secondary and
tertiary prevention should not be neglected. Thus, disease surveillance and targeted longitudinal
epidemiologic studies are needed for both evaluation of primary prevention effectiveness and as
a means of providing continuing improvement of regulations. Surveillance and epidemiologic
studies will need to incorporate appropriate exposure assessment programs, and to be most
useful, need to be started immediately so as to provide comparable baseline data in the event that
Maryland decides to move forward with UNGDP at some point in the future.
R46. Initiate a birth outcomes surveillance system
Birth outcomes are strongly influenced by exposures occurring during pregnancy, and thus are
potentially one of the earliest health effects that might occur as a result of exposures generated
by new development. Recent studies (see Section 10.3.1.4) have suggested an association of
adverse birth outcomes with UNGDP close to the mothers residence. Therefore, we recommend
development of an intensive birth outcomes surveillance system in Garrett and Allegany
counties.
R47. Initiate a longitudinal epidemiologic study of dermal, mucosal, and respiratory
irritation
Skin rashes and eye, nose, throat, and airway irritation symptoms have been associated with
UNGDP (see section 10.3.1.4). Skin, mucosal, and respiratory symptoms such as these can be
early indications of exposure and adverse health effects and can occur relatively soon after the
start of exposure (days to a few months) compared with other effects, especially cancer, that can
have latency periods of years to decades. Therefore, we recommend that the State undertake a
longitudinal epidemiologic study of dermal, mucosal, and respiratory irritation in Garrett and
Allegany counties.
R48. Develop funding mechanism for public health studies
The surveillance and epidemiologic studies will need to be funded. Some funding may be
achieved by collaboration with academic researchers in support of applications for federal
funding. However, the State should also develop its own funding through mechanisms such as
filing fees for the CGDP, well permits, and severance taxes.
12.10 Occupational Health
Based on our evaluations of the limited but emerging studies of UNGD workers exposures to
respirable crystalline silica (frack sand) and what is known from epidemiologic and toxicological
studies of crystalline silica (silicosis, lung cancer), we conclude that there is a High Likelihood
of adverse outcomes among UNGDP workers in Garrett and Allegany Counties. Should
99
Maryland move forward with UNGDP, the following recommendations are made to prevent
most and minimize residual potential negative impacts on occupational health.
R49. Require implementation of NIOSH and OSHA recommended controls for silica
exposure in UNGD operations.
Following the NIOSH study of UNGDP workers, NIOSH and OSHA have provided
extensive recommendation to minimize workers exposure. These recommendations
should be implemented.
R50. Provide MOSH with resources to regularly inspect UNGD workplaces and
monitor worker exposures.
MOSH should provide resources to implement the NIOSH and OSHA recommended
controls for silica exposure as well as workers exposure to other hazards including
noise, VOCs, and PAHs.
R51. Establish community outreach programs to help transient workers feel more
welcome in the community as a means of reducing rates of depression, suicide,
and drug use.
Transient workers suffer from depression, suicide, and drug use to cope with social
isolation. We recommend initiating, to the extent possible, outreach programs
designed to help workers adapt to their new community environment.
R52. Require employers to provide employee assistance programs including
counseling and substance abuse treatment.
In addition to the community, employers should also provide assistance to the
employee to cope with the new community environment. It should also include
counseling services to deal with depression, suicide, and drug use.

100
13 LIMITATIONS
As stated in the MOU the project is designed to provide a baseline assessment of current
regional population health, an assessment of potential public health impacts, and possible
adaptive and public health mitigation strategies in the event that natural gas extraction takes
place within Marylands Marcellus Shale resource. In particular, the project is not designed to
make recommendation about whether or when to allow unconventional natural gas development
and production (UNGDP) in Maryland. Rather this study is designed to inform decisions by
clearly describing the potential public health impacts and make recommendations for minimizing
them, should the decision makers move ahead with UNGDP in Maryland.
Given the short timeframe, the study team made extensive effort to stay within the scope of tasks
identified within the MOU. This section identifies limitations of this report, some of which are
related to the process itself, while others are related to lack of available data.
HIA is a relatively new practice. As such, one major limitation is the lack of consistent
methods that are universally accepted. As such, comparison between HIA, including
ranking of hazards is not straightforward.
There is a lack of monitoring data available in the literature that has evaluated the impact
of UNGDP on air and water quality, based on the measurements taken before UNGDP
related activities as well as during the development and production phase. Limited data
available to date have focused on spatial contrast (i.e., UNGDP impacted areas vs control
sites), as opposed to temporal contrast (data from same site looking at before, during and
after UNGDP related activities).
With the exception of crystalline silica exposure among workers, very little data is
available on individual level exposure to both physical and chemical hazards associated
with UNGDP related activities.
The NIOSH study documenting overwhelming level of occupational exposure to
respirable crystalline silica draws attention to the potential exposure that may be taking
place among nearby residents. Respirable fraction of crystalline silica particles are small
enough to travel to nearby communities, where they may disproportionately impact
vulnerable populations. So far, the scientific literature has overlooked this potential
exposure, and as such we could not evaluate this issue.
Epidemiological investigations of health outcomes related to UNGDP related
activities/hazards is extremely limited, with noted exception of adverse birth outcomes.
Baseline health assessment did not include health survey for population of concern.
We conducted noise monitoring in the UNGDP impacted community in WV. This was
the only primary data we collected. All other evaluations are based on existing data
available through literature review.
Quantitative health risk assessment and cumulative risk assessment were beyond the data
and time resources available to us.

101
14 REFERENCES
1. McKenzie LM, Witter RZ, Newman LS, Adgate JL (2012) Human health risk assessment of
air emissions from development of unconventional natural gas resources. Sci Total Environ
424:7987. doi: 10.1016/j.scitotenv.2012.02.018
2. Mckenzie LM, Guo R, Witter RZ, et al. (2014) Birth Outcomes and Maternal Residential
Proximity to Natural Gas Development in Rural Colorado. Environ Health Perspect. doi:
10.1289/ehp.1306722
3. Steinzor N, Subra W, Sumi L (2013) Investigating Links Between Shale Gas Development
and Health Impacts Through a Community Survey Project in Pennsylvania. New Solut
23:5583. doi: 10.2190/NS.23.1.e
4. Witter RZ, McKenzie L, Stinson KE, et al. (2013) The use of health impact assessment for a
community undergoing natural gas development. Am J Public Heal 103:10021010. doi:
10.2105/AJPH.2012.301017
5. Haggerty J, Gude PH, Delorey M, Rasker R (2011) Oil and Gas Extraction as an Economic
Development Strategy in the American West%: A Longitudinal Performance Analysis ,
1980-2011. 118.
6. Coussens C, Martinez RM (2014) Health Impact Assessment of Shale Gas Extraction:
Workshop Summary. Natl Acad Press 156.
7. Witter R, McKenzie L, Towle M, et al. (2010) Health Impact Assessment for Battlement
Mesa, Garfield County Colorado. 1157.
8. Morhaim DK, Hubbard JW, Bobo E, et al. (2014) House Bill 1030. House of Delegates,
Annapolis, MD
9. National Research Council (2011) Improving Health in the United States: The Role of Health
Impact Assessment. National Academies Press, Washington, D C
10. Regional Economic Studies Institute (2014) Impact Analysis of the Marcellus Shale Safe
Drilling.
http://www.mde.state.md.us/programs/Land/mining/marcellus/Documents/RESI_Marcellus
_Shale_Report_Final.pdf. Accessed 10 Jul 2014
11. Paleontological Research Institution (2012) Understanding Drilling Technology. Sci Beneath
Surf 9.
12. EnergyFromShale.org (2013) Energy from Shale: Marcellus Shale.
http://www.energyfromshale.org/hydraulic-fracturing/marcellus-shale-
gas?gclid=CjkKEQjwh7ucBRD9yY_fyZe398gBEiQAAoy4JHXb24tKVN6hBcjMHqJTW4
kbhV5qZaDB3n7uTIaf_g_w_wcB. Accessed 7 Jul 2014
102
13. Sheir R, Wilson J (2014) Cove Point Serves as Flashpoint for Maryland Debate Over
Fracking. WAMU 88.5
14. Shonkoff SB, Hays J, Finkel ML (2014) Environmental Public Health Dimensions of Shale
and Tight Gas Development. Environ Health Perspect. doi: DOI:10.1289/ehp.1307866
15. Litovitz A, Curtright A, Abramzon S, et al. (2013) Estimation of regional air-quality
damages from Marcellus Shale natural gas extraction in Pennsylvania. Environ Res Lett
8:014017. doi: 10.1088/1748-9326/8/1/014017
16. Maryland Department of the Environment, Maryland Department of Natural Resources
(2013) Marcellus Shale Safe Drilling Initiative Study: Part II Best Practices. 93.
17. Encana Corporation (2014) Wellbore Construction.
http://www.encana.com/sustainability/environment/water/protection/construction.html.
Accessed 27 Jun 2014
18. American Gas Association How Does the Natural Gas Delivery System Work? - Natural
Gas.
http://www.aga.org/KC/ABOUTNATURALGAS/CONSUMERINFO/Pages/NGDeliverySy
stem.aspx. Accessed 9 Feb 2014
19. Kasperson RE, Kasperson JX Climate Change , Vulnerability and Social Justice.
20. DeFur PL, Evans GW, Cohen Hubal E a, et al. (2007) Vulnerability as a Function of
Individual and Group Resources in Cumulative Risk Assessment. Environ Health Perspect
115:817824. doi: 10.1289/ehp.9332
21. Water Systems Council (2012) Wellcare Maryland Fact Sheet. 2.
22. Lichtenberg E, Shapiro LK (1997) Agriculture and Nitrate Concentrations in Maryland
Community Water System Wells. J Environ Qual 26:145.
23. Bhlke JK, Denver JM (1995) Combined Use of Groundwater Dating, Chemical, and
Isotopic Analyses to Resolve the History and Fate of Nitrate Contamination in Two
Agricultural Watersheds, Atlantic Coastal Plain, Maryland. Water Resour Res 31:2319
2339.
24. Kahn SE, Hull RL, Utzschneider KM (2006) Mechanisms linking obesity to insulin
resistance and type 2 diabetes. Nature 444:8406. doi: 10.1038/nature05482
25. FracFocus Chemical Disclosure Registry Chemical Use In Hydraulic Fracturing.
http://fracfocus.org/water-protection/drilling-usage. Accessed 9 Jul 2014
103
26. FracFocus Chemical Disclosure Registry (2013) Why Chemicals Are Used | FracFocus
Chemical Disclosure Registry. http://fracfocus.org/chemical-use/why-chemicals-are-used.
Accessed 30 May 2014
27. U.S. House of Representatives, U.S. House of Representatives Committee on Energy and
Commerce Minority Staff (2011) Chemicals used in hydraulic fracturing. US House
Represent Comm Energy Commer 32.
28. Leidos Incorporated (2014) Marcellus Shale Natural Gas Drilling and Production: Ambient
Air Monitoring. 89.
29. Colborn T, Kwiatkowski C, Schultz K, et al. (2011) Natural Gas Operations from a Public
Health Perspective. Hum Ecol Risk Assess An Int J 17:10391056. doi:
10.1080/10807039.2011.605662
30. Esswein EJ, Breitenstein M, Snawder J, et al. (2013) Occupational exposures to respirable
crystalline silica during hydraulic fracturing. J Occup Environ Hyg 10:34756. doi:
10.1080/15459624.2013.788352
31. Colborn T, Schultz K, Herrick L, Kwiatkowski C (2014) An Exploratory Study of Air
Quality Near Natural Gas Operations. Hum Ecol Risk Assess An Int J 20:86105. doi:
10.1080/10807039.2012.749447
32. Pennsylvania_Department_of_Environmental_Protection (2011) Northeastern Pennsylvania
Marcellus Shale Short-Term Ambient Air Sampling Report. Pennsylvania
33. Pennsylvania_Department_of_Environmental_Protection (2010) Southwestern Pennsylvania
Marcellus Shale Ambient Air Sampling Report. Pennsylvania
34. Mccawley M (2013) Air, Noise, and Light Monitoring Results For Assessing Environmental
Impacts of Horizontal Gas Well Drilling Operations ( ETD 10 Project ) Prepared for%:
West Virginia Department of Environmental Protection Division of Air Quality Charleston ,
WV 2530. 206.
35. Eastern Research Group, Sage Environmental (2011) City of Fort Worth Natural Gas Air
Quality Study: Final Report. Texas
36. Roy AA, Adams PJ, Robinson AL (2014) Air pollutant emissions from the development,
production, and processing of Marcellus Shale natural gas. J Air Waste Manage Assoc
64:1937. doi: 10.1080/10962247.2013.826151
37. Hill E (2013) Hydraulic Fracturing and Infant Health: Evidence From Pennsylvania and
Colorado. Environ. Heal.
104
38. Fryzek J, Pastula S, Jiang X, Garabrant DH (2013) Childhood cancer incidence in
pennsylvania counties in relation to living in counties with hydraulic fracturing sites. J
Occup Environ Med 55:796801. doi: 10.1097/JOM.0b013e318289ee02
39. Morris RD, Naumova EN, Munasinghe RL (1995) Ambient air pollution and hospitalization
for congestive heart failure among elderly people in seven large US cities. Am J Public
Health 85:13611365. doi: 10.2105/AJPH.85.10.1361
40. Venners SA, Wang B, Xu Z, et al. (2003) Particulate matter, sulfur dioxide, and daily
mortality in Chongqing, China. Environ Health Perspect 111:5627.
41. Brook RD, Rajagopalan S, Pope CA, et al. (2010) Particulate matter air pollution and
cardiovascular disease: An update to the scientific statement from the American Heart
Association. Circulation 121:233178. doi: 10.1161/CIR.0b013e3181dbece1
42. Adar SD, Kaufman JD (2008) Cardiovascular disease and air pollutants: Evaluating and
improving epidemiological data implicating traffic exposure.
43. Hoffmann B, Moebus S, Stang A, et al. (2006) Residence Close to High Traffic and
Prevalence of Coronary Heart Disease. Eur Heart J 27:2696702.
44. Gan WQ, Tamburic L, Davies HW., et al. (2010) Changes in Residential Proximity to Road
Traffic and the Risk of Death From Coronary Heart Disease. Epidemiology 21:642649.
45. Jerrett M, Burnett RT, Renjun Ma CAPI, et al. (2005) Spatial Analysis of Air Pollution and
Mortality in Los Angeles. Epidemiology 16:72736. doi:
10.1097/01.ede.0000181630.15826.7d
46. Wellenius GA, Schwartz J, Mittleman MA (2005) Air Pollution and Hospital Admissions for
Ischemic and Hemorrhagic Stroke Among Medicare Beneficiaries. Stroke 36:25492553.
doi: 10.1161/01.STR.0000189687.78760.47
47. Hong Y-C, Lee J-T, Kim H, Kwon H-J (2002) Air pollution: a new risk factor in ischemic
stroke mortality. Stroke 33:21659.
48. Kettunen J, Lanki T, Tiittanen P, et al. (2007) Associations of Fine and Ultrafine Particulate
Air Pollution With Stroke Mortality in an. Air Pollut Stroke Mortal. doi: DOI:
10.1161/01.STR.0000257999.49706.3b
49. Franklin M, Zeka A, Schwartz J (2007) Association between PM2.5 and all-cause and
specific-cause mortality in 27 US communities. J Expo Sci Environ Epidemiol 17:27987.
50. Sinclair AH, Tolsma D (2004) Air Pollution and acute cardio-respiratory visits in an
ambulatory care setting: Two year and preliminary four year results. Epidemiology 15:
105
51. Ostro B, Broadwin R, Green S, et al. (2006) Fine particulate air pollution and mortality in
nine California counties: results from CALFINE. Environ Health Perspect 114:2933.
52. Pereira G, Belanger K, Ebisu K, Bell ML (2014) Fine particulate matter and risk of preterm
birth in Connecticut in 2000-2006: a longitudinal study. Am J Epidemiol 179:6774. doi:
10.1093/aje/kwt216
53. Brauer M, Lencar C, Tamburic L, et al. (2008) A cohort study of traffic-related air pollution
impacts on birth outcomes. Environ Health Perspect 116:6806.
54. Ritz B, Yu F (1999) The effect of ambient carbon monoxide on low birth weight among
children born in southern California between 1989 and 1993. Environ Health Perspect
107:1725.
55. Wilhelm M, Ritz B (2005) Local Variations in CO and Particulate Air Pollution and Adverse
Birth Outcomes in Los Angeles County, California, USA. Environ Heal Perspect 113:1212
1221.
56. Slama R, Morgenstern V, Cyrys J, et al. (2007) Traffic-Related Atmospheric Pollutants
Levels during Pregnancy and Offsprings Term Birth Weight: A Study Relying on a Land-
Use Regression Exposure Model. Environ Health Perspect 115:12831292.
57. Woodruff TJ, Parker JD, Schoendorf KC (2006) Fine particulate matter (PM2.5) air pollution
and selected causes of postneonatal infant mortality in California. Environ Health Perspect
114:78690.
58. Woodruff TJ, Darrow LA, Parker JD (2008) Air pollution and postneonatal infant mortality
in the United States, 1999-2002. Environ Health Perspect 116:1105.
59. U.S. Environmental Protection Agency (1974) Safe Drinking Water Act. US Congress
60. Vengosh A, Jackson RB, Warner N, et al. (2014) A critical review of the risks to water
resources from unconventional shale gas development and hydraulic fracturing in the
United States. Environ Sci Technol. doi: 10.1021/es405118y
61. Schmidt CW (2013) Estimating wastewater impacts from fracking. Environ Health Perspect
121:A117. doi: 10.1289/ehp.121-a117
62. Soeder BDJ, Kappel WM (2009) Water Resources and Natural Gas Production from the
Marcellus Shale. 16.
63. Lutz BD, Lewis AN, Doyle MW (2013) Generation, transport, and disposal of wastewater
associated with Marcellus Shale gas development. Water Resour Res 49:647656. doi:
10.1002/wrcr.20096
106
64. Adgate JL, Goldstein BD, McKenzie LM (2014) Potential public health hazards, exposures
and health effects from unconventional natural gas development. Environ Sci Technol. doi:
10.1021/es404621d
65. Vidic RD, Brantley SL, Vandenbossche JM, et al. (2013) Impact of shale gas development
on regional water quality. Science 340:1235009. doi: 10.1126/science.1235009
66. Warner NR, Christie C a, Jackson RB, Vengosh A (2013) Impacts of shale gas wastewater
disposal on water quality in western Pennsylvania. Environ Sci Technol 47:1184957. doi:
10.1021/es402165b
67. Brown D (2014) Radionuclides in Fracking Wastewater: Managing a Toxic Blend. Environ
Health Perspect 122:5055.
68. Rozell DJ, Reaven SJ (2012) Water pollution risk associated with natural gas extraction from
the Marcellus Shale. Risk Anal 32:138293. doi: 10.1111/j.1539-6924.2011.01757.x
69. Farag AM, Harper DD (2014) A review of environmental impacts of salts from produced
waters on aquatic resources. Int J Coal Geol 126:157161. doi: 10.1016/j.coal.2013.12.006
70. Mitka M. (2012) Rigorous evidence slim for determining health risks from natural gas
fracking. J Am Med Assoc 307:21352136. doi: 10.1001/jama.2012.3726
71. Schmidt CW (2011) Blind rush? Shale gas boom proceeds amid human health questions.
Environ Health Perspect 119:A34853. doi: 10.1289/ehp.119-a348
72. FracFocus Chemical Disclosure Registry (2013) What Chemicals Are Used.
http://fracfocus.org/chemical-use/what-chemicals-are-used. Accessed 30 May 2014
73. Aminto A, Olson MS (2012) Four-compartment partition model of hazardous components in
hydraulic fracturing fluid additives. J Nat Gas Sci Eng 7:1621. doi:
10.1016/j.jngse.2012.03.006
74. Myers T, Saiers JE, Barth E (2012) Potential Contaminant Pathways from Hydraulically
Fractured Shale Aquifers. Ground Water 50:87282. doi: 10.1111/j.1745-
6584.2012.00933.x
75. Rvsz KM, Breen KJ, Baldassare AJ, Burruss RC (2010) Carbon and hydrogen isotopic
evidence for the origin of combustible gases in water-supply wells in north-central
Pennsylvania. Appl Geochemistry 25:18451859. doi: 10.1016/j.apgeochem.2010.09.011
76. Flewelling SA, Sharma M (2014) Constraints on upward migration of hydraulic fracturing
fluid and brine. Ground Water 52:919. doi: 10.1111/gwat.12095
107
77. Gradient Corporation (2009) Evaluation of Potential Impacts of Hydraulic Fracturing
Flowback Fluid Additives on Microbial Processes in Publicly-Owned Treatment Works
(POTWs). Cambridge, MA
78. Hladik ML, Focazio MJ, Engle M (2014) Discharges of produced waters from oil and gas
extraction via wastewater treatment plants are sources of disinfection by-products to
receiving streams. Sci Total Environ 466-467:108593. doi:
10.1016/j.scitotenv.2013.08.008
79. Haluszczak LO, Rose AW, Kump LR (2013) Geochemical evaluation of flowback brine
from Marcellus gas wells in Pennsylvania, USA. Appl Geochemistry 28:5561. doi:
10.1016/j.apgeochem.2012.10.002
80. Skalak KJ, Engle MA, Rowan EL, et al. (2014) Surface disposal of produced waters in
western and southwestern Pennsylvania: Potential for accumulation of alkali-earth elements
in sediments. Int J Coal Geol 126:162170. doi: 10.1016/j.coal.2013.12.001
81. Rich AL, Crosby EC (2013) Analysis of reserve pit sludge from unconventional natural gas
hydraulic fracturing and drilling operations for the presence of technologically enhanced
naturally occurring (TENORM). J Environ Occup Heal Policy 23:117135. doi:
10.2190/NS.23.1.h
82. Osborn SG, Vengosh A, Warner NR, Jackson RB (2011) Methane contamination of drinking
water accompanying gas-well drilling and hydraulic fracturing. Proc Natl Acad Sci U S A
108:81726. doi: 10.1073/pnas.1100682108
83. Jackson RB, Vengosh A, Darrah TH, et al. (2013) Increased stray gas abundance in a subset
of drinking water wells near Marcellus shale gas extraction. Proc Natl Acad Sci U S A
110:112505. doi: 10.1073/pnas.1221635110
84. Fontenot BE, Hunt LR, Hildenbrand ZL, et al. (2013) An evaluation of water quality in
private drinking water wells near natural gas extraction sites in the Barnett Shale formation.
Environ Sci Technol 47:1003240.
85. Molofsky LJ, Connor JA, Wylie AS, et al. (2013) Evaluation of methane sources in
groundwater in northeastern Pennsylvania. Ground Water 51:33349. doi:
10.1111/gwat.12056
86. Li H, Carlson KH (2014) Distribution and origin of groundwater methane in the Wattenberg
oil and gas field of northern Colorado. Environ Sci Technol 48:148491. doi:
10.1021/es404668b
87. Sang W, Stoof CR, Zhang W, et al. (2014) Effect of Hydrofracking Fluid on Colloid
Transport in the Unsaturated Zone. Environ Sci Technol. doi: 10.1021/es501441e
108
88. Adams MB (2011) Land application of hydrofracturing fluids damages a deciduous forest
stand in West Virginia. J Environ Qual 40:13401344. doi: 10.2134/jeq2010.0504
89. Adams MB, Edwards PJ, Ford WM, et al. (2011) Effects of development of a natural gas
well and associated pipeline on the natural and scientific resources of the Fernow
Experimental Forest.
90. Kassotis CD, Tillitt DE, Wade Davis J, et al. (2013) Estrogen and Androgen Receptor
Activities of Hydraulic Fracturing Chemicals and Surface and Ground Water in a Drilling-
Dense Region. Endocrinology 155:en20131697.
91. Bamberger M, Oswald RE (2012) Impacts of Gas Drilling on Human and Animal Health.
New Solut 22:5177.
92. Passchier-Vermeer W, Passchier WF (2000) Noise Exposure and Public Health. Environ
Health Perspect 108 Suppl:12331.
93. Van Kamp I, Davies H (2013) Noise and health in vulnerable groups: a review. Noise Health
15:1539. doi: 10.4103/1463-1741.112361
94. New York State Department of Environmental Conservation (NYSDEC) (2011) Revised
Draft Supplemental Generic Environmental Impact Statement on the Oil, Gas, and Solution
Mining Regulatory Program: Well Permit Issuance for Horizontal Drilling and High-
Volume Hydraulic Fracturing to Develop the Marcellus Shale and Other Low-Perme. 1
1537.
95. U.S. Environmental Protection Agency O of NA and C Information on Levels of
Environmental Noise Requisite to Protect Public Health and Welfare with an Adequate
Margin of Safety. http://www.nonoise.org/library/levels74/levels74.htm. Accessed 9 Feb
2014
96. Swiss Noise Database, Environment FO of the (2009) Noise Pollution in Switzerland:
Results of the SonBase National Noise Monitoring Programme.
97. Murphy E, King EA (2014) An assessment of residential exposure to environmental noise at
a shipping port. Environ Int 63:20715. doi: 10.1016/j.envint.2013.11.001
98. Babisch W, Pershagen G, Selander J, et al. (2013) Noise annoyance--a modifier of the
association between noise level and cardiovascular health? Sci Total Environ 452-453:507.
doi: 10.1016/j.scitotenv.2013.02.034
99. Haralabidis AS, Dimakopoulou K, Vigna-Taglianti F, et al. (2008) Acute effects of night-
time noise exposure on blood pressure in populations living near airports. Eur Heart J
29:65864. doi: 10.1093/eurheartj/ehn013
109
100. Dratva J, Phuleria HC, Foraster M, et al. (2012) Transportation noise and blood pressure in
a population-based sample of adults. Environ Health Perspect 120:505. doi:
10.1289/ehp.1103448
101. Kheirbek I, Ito K, Neitzel R, et al. (2014) Spatial Variation in Environmental Noise and Air
Pollution in New York City. J Urban Heal Bull New York Acad Med 117. doi:
10.1007/s11524-013-9857-0
102. Floud S, Blangiardo M, Clark C, et al. (2013) Exposure to aircraft and road traffic noise and
associations with heart disease and stroke in six European countries: a cross-sectional study.
Environ Health 12:89. doi: 10.1186/1476-069X-12-89
103. Huang J, Deng F, Wu S, et al. (2013) The impacts of short-term exposure to noise and
traffic-related air pollution on heart rate variability in young healthy adults. J Expo Sci
Environ Epidemiol 23:55964. doi: 10.1038/jes.2013.21
104. Allen RW, Davies H, Cohen MA, et al. (2009) The spatial relationship between traffic-
generated air pollution and noise in 2 US cities. Environ Res 109:33442. doi:
10.1016/j.envres.2008.12.006
105. Ellsworth W, Robertson J, Hook C (2014) Man-Made Earthquakes Update. In: USGS.
http://www.usgs.gov/blogs/features/usgs_top_story/man-made-earthquakes/. Accessed 19
May 2014
106. Ellsworth WL (2013) Injection-Induced Earthquakes. Science 341:1225942. doi:
10.1126/science.1225942
107. Keranen KM, Savage HM, Abers GA, Cochran ES (2013) Potentially induced earthquakes
in Oklahoma, USA: Links between wastewater injection and the 2011 Mw 5.7 earthquake
sequence. Geology 41:699702.
108. Sumy DF, Cochran ES, Keranen KM, et al. (2014) Observations of static Coulomb stress
triggering of the November 2011 M 5.7 Oklahoma earthquake sequence. J Geophys Res
Solid Earth 119:19041923. doi: 10.1002/2013JB010612.Received
109. Rutqvist J, Rinaldi AP, Cappa F, Moridis GJ (2013) Modeling of fault reactivation and
induced seismicity during hydraulic fracturing of shale-gas reservoirs. J Pet Sci Eng
107:3144. doi: 10.1016/j.petrol.2013.04.023
110. Kim W-Y (2013) Induced seismicity associated with fluid injection into a deep well in
Youngstown, Ohio. J Geophys Res Solid Earth 118:35063518. doi: 10.1002/jgrb.50247
111. Balcerak E (2013) Linking earthquakes and hydraulic fracturing operations. Trans Am
Geophys Union 94:3.
110
112. U.S. Environmental Protection Agency Classes of Wells.
http://water.epa.gov/type/groundwater/uic/wells.cfm. Accessed 15 Jul 2014
113. Food and Water Watch (2013) The Social Costs of Fracking: A Pennsylvania Case Study.
16.
114. Covey S (2011) Local Experiences Related to the Marcellus Shale Industry.
115. Levy M (2013) Fracking towns see crime, carousing surge amid gas boom. Assoc. Press
116. Jacquet JB (2012) Landowner attitudes toward natural gas and wind farm development in
northern Pennsylvania. Energy Policy 50:677688. doi: 10.1016/j.enpol.2012.08.011
117. Goldenberg S, Shoveller J, Koehoorn M, Ostry A (2008) Barriers to STI testing among
youth in a Canadian oil and gas community. Heal Place 14:718729. doi:
10.1016/j.healthplace.2007.11.005
118. Forsyth C, Luthra A, W.B. B (2007) Framing perceptions of oil development and social
disruption. Soc Sci J 44:287299. doi: 10.1016/j.soscij.2007.03.015
119. Wernham A (2007) Inupiat Health and Proposed Alaskan Oil Development: Results of the
First Integrated Health Impact Assessment/Environmental Impact Statement for Proposed
Oil Development on Alaskas North Slope. Ecohealth 4:500513. doi: 10.1007/s10393-007-
0132-2
120. Ortega F, Lee D, Sui X (2010) Psychological well- being, cardiorespiratory fitness, and
long-term survival. Am J Prev Med 39:440448. doi: 10.1016/j.amepre.2010.07.015.
121. Kubzansky L, Mendes W, Appleton A, Adler G (2009) Protocol for an experimental
investigation of the roles of oxytocin and social support in neuroendocrine, cardiovascular,
and subjective responses to stress across age and gender. BMC Public Health 9:481498.
doi: 10.1186/1471-2458-9-481
122. Kubzansky L, Adler G (2010) Aldosterone: a forgotten mediator of the relationship between
psychological stress and heart disease. Neurosci Biobehav Rev 34:8086. doi:
10.1016/j.neubiorev.2009.07.005
123. Kettl P (1998) Alaska native suicide: lessons for elder suicide. Int Psychogeriatr 10:205
211. doi: 10.1017/S1041610298005316
124. Goldenberg SM, Shoveller JA, Ostry AC, Koehoorn M (2008) Sexually Transmitted
Infection (STI) Testing Among Young Oil and Gas Workers: The Need for Innovation,
Place-Based Approaches to STI Control. Can J Public Heal 99:350350.
125. Ferrar KJ, Kriesky J, Christen CL, et al. (2013) Assessment and longitudinal analysis of
health impacts and stressors perceived to result from unconventional shale gas development
111
in the Marcellus Shale region. Int J Occup Environ Health 19:104112. doi:
10.1179/2049396713Y.0000000024
126. Korfmacher KS, Jones WA, Malone SL, Vinci LF (2013) Public Health and High Volume
Hydraulic Fracturing. New Solut 23:1331.
127. (2010) Pennsylvania Department of Transportation and Pennsylvania State Police.
128. Birley M, Abrahams D, Pennington A, et al. (2008) Prospective Rapid Health Impact
Assessment of the Energy from Waste Facility in the States of Jersey: Stage 2. 110.
129. Gramm L, Hutchison L, Dabney B, Dorsey A (2010) Rural Healthy People: A Companion
Document to Healthy People 2010 Volume 1. I:1275.
130. Wise H, Shtylla S (2007) The Role of the Extractive Sector in Expanding Economic
Opportunity. Corp Soc Responsib Initiat 52.
131. Erny-Albrecht K, Brown L, Raven M, Bywood P (2014) Fly-In Fly-Out/Drive-In Drive-Out
Practices and Health Service Delivery in Rural Areas of Australia. Prim Healthc Res Inf
Serv 79.
132. Shandro JA, Veiga MM, Shoveller J, et al. (2011) Perspectives on Community Health
Issues and the Mining Boom-Bust Cycle. Resour Policy 36:178186. doi:
10.1016/j.resourpol.2011.01.004
133. Greaser K (2012) Community Health Needs Assessment. 21.
134. Christopherson S, Rightor N (2011) How Shale Gas Extraction Affects Drilling Localities:
Lessions for Regionals and City Policy Makers. J T City Manag 2:20.
135. Policy Matters Ohio Fracking in Carroll County, Ohio: An impact assessment | Policy
Matters Ohio. http://www.policymattersohio.org/fracking-apr2014. Accessed 30 Jun 2014
136. Occupational Safety and Health Administration (OSHA) (2012) Safety and Health Topics |
Diesel Exhaust. https://www.osha.gov/SLTC/dieselexhaust/. Accessed 29 May 2014
137. U.S. Department of Labor (2010) Oil and Gas Industry Fatal and Nonfatal Occupational
Injuries Fact Sheet. In: Bur. Labor Stat. http://www.bls.gov/iif/oshwc/osh/os/osar0013.htm.
Accessed 30 Jun 2014
138. The Henry J. Kaiser Family Foundation (2013) Key Facts about the Uninsured Population.
http://kff.org/uninsured/fact-sheet/key-facts-about-the-uninsured-population/. Accessed 11
Jul 2014
139. Constantine S, Battye K Mining towns-does the boom mean bust for health services? Natl
Rural Heal Conf 8.
112
140. Malamud GW (1984) Boomtown communities, Environmen. 255.
141. McKell CM, Browne DG, Freudenburg WR, Perrine RL (1984) Paradoxes of Western
Energy Development: How Can We Maintain the Land and the People If We Develop? 340.
142. Gilmore JS (1976) Boom towns may hinder energy resource development. Science
191:53540. doi: 10.1126/science.191.4227.535
143. Callahan MA, Sexton K (2007) If Cumulative Risk Assessment Is the Answer, What Is the
Question? Environ Health Perspect 115:799806. doi: 10.1289/ehp.9330
144. Sexton K (2012) Cumulative risk assessment: An overview of methodological approaches
for evaluating combined health effects from exposure to multiple environmental stressors.
Int J Environ Res Public Health 9:37090. doi: 10.3390/ijerph9020370
145. U.S. Environmental Protection Agency (2003) Framework for Cumulative Risk
Assessment. 129.
146. Zartarian VG, Schultz BD (2010) The EPAs human exposure research program for
assessing cumulative risk in communities. J Expo Sci Environ Epidemiol 20:35158. doi:
10.1038/jes.2009.20
147. Menzie CA, MacDonnell MM, Mumtaz M (2007) A phased approach for assessing
combined effects from multiple stressors. Environ Health Perspect 115:80716. doi:
10.1289/ehp.9331
148. U.S. Environmental Protection Agency (2007) Concepts, Methods and Data Sources for
Cumulative Health Risk Assessment of Multiple Chemicals, Exposures and Effects: A
Resource Document. 412.
149. Clougherty JE, Levy JI, Kubzansky LD, et al. (2007) Synergistic effects of traffic-related
air pollution and exposure to violence on urban asthma etiology. Environ Health Perspect
115:114046. doi: 10.1289/ehp.9863
150. Sexton K, Hattis D (2007) Assessing cumulative health risks from exposure to
environmental mixturesthree fundamental questions. Environ Heal Perspect 115:82532.
doi: 10.1289/ehp.9333
151. Barth JM (2013) The economic impact of shale gas development on state and local
economies: benefits, costs, and uncertainties. New Solut A J Environ Occup Heal Policy
23:85101. doi: 10.2190/NS.23.1.f
152. Black D, McKinnish T, Sanders S (2005) The Economic Impact Of The Coal Boom And
Bust*. Econ J 115:449476.
113
153. Kaufmann RK, Kauppi H, Mann ML, Stock JH (2011) Reconciling anthropogenic climate
change with observed temperature 1998 2008. Proc Natl Acad Sci 108:1179011793. doi:
10.1073/pnas.1102467108/-
/DCSupplemental.www.pnas.org/cgi/doi/10.1073/pnas.1102467108
154. Khn M, Mnch U (2013) CLEAN. Springer Berlin Heidelberg, Berlin, Heidelberg
155. National Research Council (2009) Science and Decisions: Advancing Risk Assessment.
442.
156. Committee on Human and Environmental Exposure Science in the 21st Century, Board on
Environmental Studies and Toxicology, Division on Earth and Life Studies NRC (2012)
Exposure Science in the 21st Century: A Vision and a Strategy.
157. Pataki G (2011) Fracking can bring good jobs to New York without harming the
environment, says former Gov. George Pataki - NY Daily News. In: New York Dly. News.
http://www.nydailynews.com/opinion/fracking-bring-good-jobs-new-york-harming-
environment-gov-george-pataki-article-1.984773. Accessed 29 May 2014
158. Bureau of Labor Statistics USD of L (2011) Nonfatal Occupational Injuries and Illnesses
Requiring Days Away from Work.
http://www.bls.gov/news.release/archives/osh2_11082012.pdf. Accessed 29 May 2014
159. King P (2013) Worker Safety: Oil and gas deaths reached record high in 2012. In: Environ.
Energy Publ. http://www.eenews.net/stories/1059986375. Accessed 30 May 2014
160. U.S. Small Business Administration O of A (2007) Employer firms & employment size of
firm by NAICS codes, 2007. NAICS 27.
161. Urbina I (2012) For Oil Workers, Deadliest Danger Is Driving. In: New York Times.
http://www.nytimes.com/2012/05/15/us/for-oil-workers-deadliest-danger-is-
driving.html?pagewanted=all. Accessed 30 May 2014
162. Occupational Safety and Health Administration (OSHA) Safety and Health Topics | Oil and
Gas Extraction - Health Hazards.
https://www.osha.gov/SLTC/oilgaswelldrilling/healthhazards.html. Accessed 29 May 2014
163. Centers for Disease Control and Prevention (2013) Noise and Hearing Loss Prevention -
NIOSH Workplace Safety and Health Topic.
164. Occupational Safety and Health Administration (OSHA) Safety and Health Topics |
Occupational Noise Exposure. https://www.osha.gov/SLTC/noisehearingconservation/.
165. McFeeley M (2012) State Hydraulic Fracturing Disclosure Rules and Enforcement: A
Comparison. Nat Resour Def Counc Issue Br 116.
114
166. American Public Health Association (APHA). (2012) The Environmental and Occupational
Health Impacts of High-Volume Hydraulic Fracturing of Unconventional Gas Reserves. In:
Am. Public Heal. Assoc.
http://www.apha.org/advocacy/policy/policysearch/default.htm?id=1439. Accessed 29 May
2014
167. International Agency for Research on Cancer (2012) IARC: Diesel engine exhaust
carcinogenic. Int Agency Res Cancer 4.
168. Esswein, EJ, Breitenstein M SJ NIOSH field effort to assess chemical exposures in oil and
gas workers: Health exposures in oil and gas workers: Health hazards in hydraulic
fracturing. http://www.iom.edu/~/media/Files/Activity
Files/Environment/EnvironmentalHealthRT/2012-04-30/Esswein.pdf. Accessed 29 May
2014
169. Occupational Safety and Health Administration (OSHA) Oil and Gas Well Drilling and
Servicing eTool: General Safety and Health - Hydrogen Sulfide Gas.
https://www.osha.gov/SLTC/etools/oilandgas/general_safety/h2s_monitoring.html.
170. Occupational Safety and Health Administration (OSHA) (2005) Hydrogen Sulfide - OSHA
Fact Sheet.
https://www.osha.gov/OshDoc/data_Hurricane_Facts/hydrogen_sulfide_fact.pdf. Accessed
30 May 2014
171. Occupational Safety and Health Administration (OSHA) (2010) US Labor Departments
OSHA cites Chipco LLC in Zanesville, Ohio, following fatality at natural gas well site in
Londonderry, Ohio.
https://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=NEWS_RELEASES
&p_id=17232. Accessed 30 May 2014
172. Occupational Safety and Health Administration (OSHA) (2011) US Department of Labors
OSHA cites 5 companies for exposing workers to hydrogen sulfide at Eustace, Texas, work
site.
https://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=NEWS_RELEASES
&p_id=19575. Accessed 30 May 2014
173. Aguilar-Gaxiola S, Debb-Sossa N, Elliott K, et al. (2009) Building partnerships:
Conversations with Latina/o migrant workers about mental health needs and community
strengths.
174. Lydersen K (2013) Boom in North Dakota Weighs Heavily on Health Care -
NYTimes.com. In: New York Times. http://www.nytimes.com/2013/01/28/us/boom-in-
north-dakota-weighs-heavily-on-health-care.html?pagewanted=2&_r=0. Accessed 29 May
2014
115
175. Lydersen K (2013) U.S. Chambers fracking job boom: Behind the numbers | Midwest
Energy News. In: Midwest Emergy News.
http://www.midwestenergynews.com/2013/01/10/u-s-chambers-fracking-job-boom-behind-
the-numbers/. Accessed 30 May 2014
176. The New York Times (2009) Editorial: The Halliburton Loophole. New York Times A28.
177. Brady W (2012) Hydraulic fracturing regulation in the United States: The laissez-faire
approach of the Federal Government and varying State Regulations. 19.
178. U.S. Environmental Protection Agency (2014) Private Drinking Water Wells. In: Environ.
Prot. Agency. http://water.epa.gov/drink/info/well/. Accessed 30 May 2014
179. Energy Policy Act of 2005. US Congress
180. Clean Water Act of 1972. U.S. Congress
181. U.S. Environmental Protection Agency (2009) Regulation of Oil and Gas Construction
Activities. http://cfpub.epa.gov/npdes/stormwater/oilgas.cfm. Accessed 30 May 2014
182. World Health Organization (2005) Ecosystems and Human Well-Being. World Heal Organ
64.
183. Clean Air Act of 1970. U.S. Congress
184. Climate Change Division USEPA (2014) Overview of Greenhouse Gases.
http://epa.gov/climatechange/ghgemissions/gases/ch4.html. Accessed 4 Apr 2014
185. Climate Change Division USEPA (2014) National Greenhouse Gas Emissions Data.
http://www.epa.gov/climatechange/ghgemissions/usinventoryreport.html. Accessed 30 May
2014
186. Resource Conservation and Recovery Act of 1976. U.S. Congress
187. Solid Waste Disposal Act Amendments of 1980. U.S. Congress
188. Comprehensive Environmental Response, Compensation, and Liability Act of 1980. U.S.
Congress
189. Environmental Defense Center (2011) Fracking Federal law: loopholes & exemptions.
Environ Def Cent 1.
190. (1984) Emergency Planning and Community Right to Know Act. U.S. Congress
191. Soraghan M (2010) In Fracking Debate, Disclosure Is in the Eye of the Beholder. New
York Times
116
192. Konschnik KE, Boling MK (2014) Shale Gas Development: A Smart Regulation
Framework. Environ Sci Technol. doi: 10.1021/es405377u
193. Richardson N, Gottlieb M, Krupnick A, Wiseman H (2013) The State of State Shale Gas
Regulation. 103.
194. Pagano S (2013) Dallas approves gas drilling ordinance, offering protections for homes,
schools. Bloom. BNA Environ. Report.
195. Dallas City Council (2013) DCA 123-003 Amend Gas Drilling and Production Regulations.
10471140.
196. Cusick M (2013) Pennsylvania Supreme Court strikes down controversial portions of Act
13. In: StateImpact Pennsylvania. http://stateimpact.npr.org/pennsylvania/2014/02/21/pa-
supreme-court-will-not-reconsider-act-13-decision/. Accessed 9 Jul 2014
197. Cusick M (2014) Pa. Supreme Court will not reconsider Act 13 decision. In: StateImpact
Pennsylvania. http://stateimpact.npr.org/pennsylvania/2014/02/21/pa-supreme-court-will-
not-reconsider-act-13-decision/. Accessed 9 Jul 2014
198. Taylor K, Kaplan T (2014) New York Towns Can Prohibit Fracking, States Top Court
Rules. New York Times A16.
199. Campbell J (2014) N.Y. towns can ban fracking, state court rules. USA Today
200. Plagakis S (2014) Wyoming Supreme Court Advances Disclosure of Fracking Chemicals |
Center for Effective Government. In: Cent. Eff. Gov.
http://www.foreffectivegov.org/wyoming-supreme-court-advances-disclosure-fracking-
chemicals. Accessed 7 Jul 2014
201. General Assembly of North Carolina (2014) Energy Modernization Act. 6:126.
202. California Department of Conservation (2014) SB 4 Well Stimulation Treatment
Regulations First Revised Text of Proposed Regulations. 29.
203. McFeeley M (2014) Falling through the cracks: Public informaiton and the patchwork of
hydraulic fracturing disclosure laws. Vt Law Rev 38:849901.
204. Konschnik K, Holden WM, Shasteen A (2013) Legal Fractures in Chemical Disclosure
Laws. 16.
205. Beinecke F, Bras RL, Deutch J, et al. (2014) Secretary of Energy Advisory Board Task
Force Report on FracFocus 2.0. 24.
206. Government of California (2014) SB 4 Well Stimulation. California
117
207. Baker Hughes (2014) Hydraulic Fracturing Chemical Disclosure Policy.
http://public.bakerhughes.com/shalegas/disclosure.html. Accessed 7 Jul 2014
208. Field T, Fink E, Ghosh S, et al. (2013) Letter to Commissioner Cathy P. Foerster of the
Alaska Oil & Gas Conservation Commission. 19.
209. Eshleman KN, Elmore A (2013) Recommended Best Management Practices for Marcellus
Shale Gas Development in Maryland. 172.
210. U.S. Environmental Protection Agency (2012) Oil and Natural Gas Sector: New Source
Performance Standards and National Emission Standards for Hazardous Air Pollutants
Reviews. Fed Regist 77:49489 49600.
211. U.S. Environmental Protection Agency Oil and Natural Gas Air Pollution Standards.
http://www.epa.gov/airquality/oilandgas/. Accessed 13 Jul 2014
212. 2014 Maryland Report Card. http://www.mdreportcard.org/index.aspx?K=110512.
Accessed 16 Jul 2014
213. Gee GC, Payne-Sturges DC (2004) Environmental health disparities: a framework
integrating psychological and environmental concepts. Environ Health Perspect 112:1645
1653. doi: 10.1289/ehp.7074
214. Malats N, Camus-Radon A-M, Nyberg F, et al. (2000) Lung Cancer Risk in Nonsmokers
and GSTM1 and GSTT1 Genetic Polymorphism. Cancer Epidemiol Biomarkers Prev
9:827833.
215. Saintot M, Malaveille C, Hautefeuille A, Gerber M (2004) Interaction between genetic
polymorphism of cytochrome P450-1B1 and environmental pollutants in breast cancer risk.
Eur J cancer Prev 13:8386.
216. Hollis S, Lecky F, Yates DW, Woodford M (2005) The Effect of Pre-Existing Medical
Conditions and Age on Mortality After Injury. J Expo Anal Environ Epidemiol 61:1255
1260.
217. Luthar SS, Masten A, Powell J (2003) Resilience and Vulnerability: Adaptation in the
Context of Childhood Adversities. 574.
218. Brown DL (2008) African American Resiliency: Examining Racial Socialization and Social
Support as Protective Factors. J Black Psychol 34:3248. doi: 10.1177/0095798407310538
219. Ross CE, Wu C (2014) The Links Between Education and Health THE LINKS BETWEEN
EDUCATION AND HEALTH * Chia-ling Wu. 60:719745.
220. Bearer CF (1995) How are children different from adults? Environ Health Perspect 103
Suppl :712.
118
221. Bellinger DC (2008) Neurological and behavioral consequences of childhood lead
exposure. PLoS Med 5:e115. doi: 10.1371/journal.pmed.0050115
222. Black K, Shalat SL, Freeman NCG, et al. (2005) Childrens mouthing and food-handling
behavior in an agricultural community on the US/Mexico border. J Expo Anal Environ
Epidemiol 15:24451.
223. U.S. Environmental Protection Agency (2013) Americas Children and the Environment,
3rd Edition.
224. Tulve NS, Suggs JC, McCurdy T, et al. (2002) Frequency of mouthing behavior in young
children. J Expo Anal Environ Epidemiol 12:25964.
225. Reigart JR, Roberts JR (2001) Pesticides in children. Pediatr Clin North Am 48:118598,
ix.
226. Zahm SH, Ward MH (1998) Pesticides and childhood cancer. Environ Health Perspect 106
Suppl :893908.
227. Clean Air Assistance Program, Environmental Science and Services Division MD of EQ
(2004) Calculating Air Emissions For The Michigan Air Emissions Reporting System
(Maers).
228. Wilson SM, Fraser-Rahim H, Williams E, et al. (2012) Assessment of the distribution of
toxic release inventory facilities in metropolitan Charleston: an environmental justice case
study. Am J Public Health 102:19741980. doi: 10.2105/AJPH.2012.300700
229. Wilson S, Zhang H, Jiang C, et al. (2014) Being overburdened and medically underserved:
assessment of this double disparity for populations in the state of Maryland. Environ Heal
13:26.
230. U.S. Environmental Protection Agency NPDES Home.
http://water.epa.gov/polwaste/npdes/. Accessed 16 Jul 2014
231. Revitalization UEOO of B and L Brownfields Definition.
232. Greenberg M, Lowrie K, Solitare L, Duncan L (2000) Brownfields, Toads, and the Struggle
for Neighborhood Redevelopment: A Case Study of the State of New Jersey. Urban Aff
Rev 35:717733.
233. Freeland WTD (2004) Environmental Justice and the Brownfields Revitalization Act of
2001: Brownfields of Dreams or a Nightmare in the Making. J. Gender, Race Justice 8:
234. Wilson S, Jiang C, Burwell K, et al. (2013) Assessment of spatial disparities in the burden
of underground storage tanks in Maryland (20012011). Environ Justice 6:219225. doi:
10.1089/env.2013.0029
119
235. (1980) Basic Data Report No. 11, Garrett County Gas Well records.
236. Wilson SM (2009) An Ecologic Framework to Study and Address Environmental Justice
and Community Health Issues. Environ Justice 2:1524.
237. Payne-Sturges D, Gee GC (2006) National environmental health measures for minority and
low-income populations: tracking social disparities in environmental health. Environ Res
102:15471.
238. Justice. I of M (US) C on E (1999) Toward Environmental Justice. National Academies
Press (US), Washington (DC)
239. Brulle RJ, Pellow DN (2007) ENVIRONMENTAL JUSTICE: Human Health and
Environmental Inequalities. MIT Press. doi: 10.1146/annurev.publhealth.27.021405.102124
240. Mohai P, Saha R (2006) Reassessing Racial and Socioeconomic Disparities in
Environmental Justice Research. Demography 43:383399.
241. Evans GW, Kantrowitz E (2002) Socioeconomic status and health: the potential role of
environmental risk exposure. Annu Rev Public Health 23:30331.
242. Houston D, Wu J, Ong P, Winer A (2004) Structural Disparities of Urban Traffic in
Southern California: Implications for Vehicle-Related Air Pollution Exposure in Minority
and High-Poverty Neighborhoods. J Urban Aff 26:565592.
243. Sapkota AR, Curriero FC, Gibson KE, Schwab KJ (2007) Antibiotic-resistant enterococci
and fecal indicators in surface water and groundwater impacted by a concentrated Swine
feeding operation. Environ Health Perspect 115:10405.
244. Mccall C (2012) Marcellus Shale Gas Development in Maryland%: A Natural Resource
Analysis. 32.
245. Samet JM, Zeger SL, Dominici F, et al. (2000) The National Morbidity, Mortality, and Air
Pollution Study Part II: Morbidity and Mortality from Air Pollution in the United States.
Heal Eff Inst 570.
246. Pope CA, Dockery DW (2006) Health Effects of Fine Particulate Air Pollution: Lines that
Connect. J Air Waste Manag Assoc 70942.
247. Laden F, Neas LM, Dockery DW, Schwartz J (2000) Association of Fine Particulate Matter
from Different Sources with Daily Mortality in Six U.S. Cities. Environ Health Perspect
108:9417.
248. Braga ALF, Zanobetti A, Schwartz J (2001) The Lag Structure Between Particulate Air
Pollution and Respiratory and Cardiovascular Deaths in 10 US Cities. J Occup Environ Med
927933.
120
249. Dockery DW, Pope CA, Xu X, et al. (1993) An Association between Air Pollution and
Mortality in Six U.S. Cities. N Engl J Med 17539. doi: 10.1056/NEJM199312093292401
250. Samet JM, Dominici F, Curriero FC, et al. (2000) Fine Particulate Air Pollution and
Mortality in 20 U.S. Cities, 1987-1994. N Engl J Med 174249. doi:
10.1056/NEJM200012143432401
251. Dominici F, Peng RD, Bell ML, et al. (2006) Fine Particulate Air Pollution and Hospital
Admission for Cardiovascular and Respiratory Diseases. J Am Med Assoc 295:11271134.
doi: 10.1001/jama.295.10.1127
252. Wellenius GA, Bateson TF, Mittleman MA, Schwartz J (2005) Particulate Air Pollution and
the Rate of Hospitalization for Congestive Heart Failure among Medicare Bene&ciaries in
Pittsburgh, Pennsylvania. Am J Epidemiol 10301036. doi: 10.1093/aje/kwi135
253. Zanobetti A, Schwartz J, Samoli E, et al. (2003) The temporal pattern of respiratory and
heart disease mortality in response to air pollution. Environ Health Perspect 111:1188
1193.
254. Schwartz J, Laden F, Zanobetti A (2002) The Concentration-Response Relation Between
PM(2.5) and Daily Deaths. Environ Health Perspect 110:10259.
255. Villeneuve PJ, Goldberg MS, Krewski D, et al. (2002) Fine Particulate Air Pollution and
All-Cause Mortality within the Harvard Six-Cities Study: Variations in Risk by Period of
Exposure. Ann Epidemiol 12:56876.
256. Gauderman WJ, McConnell R, Gilliland F, et al. (2000) Association Between Air Pollution
and Lung Function Growth in Southern California Children. Am J Respir Crit Care Med
162:138390. doi: 10.1164/ajrccm.162.4.9909096
257. Peters A, Dockery DW, Heinrich J, Wichmann HE (1997) Short-term effects of particulate
air pollution on respiratory morbidity in asthmatic children. Eur Respir J 10:8729.
258. Ostro B, Lipsett M, Mann J, et al. (2001) Air pollution and exacerbation of asthma in
African-American children in Los Angeles. Epidemiology 12:2008.
259. Norris G, YoungPong SN, Koenig JQ, et al. (1999) An association between fine particles
and asthma emergency department visits for children in Seattle. Environ Health Perspect
107:48993.
260. Rabinovitch N, Strand M, Gelfand EW (2006) Particulate levels are associated with early
asthma worsening in children with persistent disease. Am J Respir Crit Care Med
173:1098105.
261. Schildcrout JS, Sheppard L, Lumley T, et al. (2006) Ambient air pollution and asthma
exacerbations in children: an eight-city analysis. Am J Epidemiol 164:50517.
121
262. Tolbert PE, Mulholland JA, MacIntosh DL, et al. (2000) Air Quality and Pediatric
Emergency Room Visits for Asthma in Atlanta, Georgia, USA. Am J Epidemiol 151:798
810.
263. Lin M, Chen Y, Burnett RT, et al. (2002) The influence of ambient coarse particulate matter
on asthma hospitalization in children: case-crossover and time-series analyses. Environ
264. Delfino RJ, Gong H, Linn WS, et al. (2003) Asthma symptoms in Hispanic children and
daily ambient exposures to toxic and criteria air pollutants. Environ Health Perspect
111:64756.
265. Delfino RJ, Quintana PJE, Floro J, et al. (2004) Association of FEV1 in asthmatic children
with personal and microenvironmental exposure to airborne particulate matter. Environ
266. Timonen KL, Pekkanen J (1997) Air pollution and respiratory health among children with
asthmatic or cough symptoms. Am J Respir Crit Care Med 156:54652. doi:
10.1164/ajrccm.156.2.9608044
267. Mar TF, Koenig JQ, Primomo J (2010) Associations between asthma emergency visits and
particulate matter sources, including diesel emissions from stationary generators in Tacoma,
Washington. Inhal Toxicol 22:4458. doi: 10.3109/08958370903575774
268. Akinbami LJ, Moorman JE, Bailey C, et al. (2012) Trends in asthma prevalence, health care
use, and mortality in the United States, 2001-2010. In: NCHS Data Brief.
file:///Users/carlybrody/Downloads/cdc_12331_DS1.pdf. Accessed 1 Jul 2014
269. Oraka E, King ME, Callahan DB (2010) Asthma and serious psychological distress:
prevalence and risk factors among US adults, 2001-2007. Chest 137:60916.
270. Gilbert A, Chakraborty J (2011) Using geographically weighted regression for
environmental justice analysis: Cumulative cancer risks from air toxics in Florida. Soc Sci
Res 40:273286.
271. U.S. Environmental Protection Agency Taking Toxics out of the Air.
272. Technology Transfer Network, Air Toxics Website USEPA National Air Toxics
Assessments. http://www.epa.gov/nata/. Accessed 13 Jul 2014
273. U.S. Department of Health and Human Services Community Health Status Indicators
(CHSI). http://wwwn.cdc.gov/CommunityHealth/homepage.aspx?j=1. Accessed 11 Jul
2014
274. Centers for Disease Control and Prevention CDC - BRFSS Annual Survey Data.
http://www.cdc.gov/brfss/annual_data/annual_data.htm. Accessed 12 Jul 2014
122
275. Robert Wood Johnson Foundation, Institute U of WPH County Health Ranking &
Roadmap. http://www.countyhealthrankings.org/.
276. National Cancer Institute State Cancer Profiles Home Page.
http://statecancerprofiles.cancer.gov/index.html. Accessed 12 Jul 2014
277. Centers for Disease Control and Prevention (2013) Cancer - United States Cancer Statistics
(USCS) Data - 2010 Top Ten Cancers. http://apps.nccd.cdc.gov/uscs/toptencancers.aspx.
Accessed 2 Jul 2014
278. Centers for Disease Control and Prevention Compressed Mortality File on CDC WONDER.
In: Mortal. 1999 - 2010 with ICD 10 codes. http://wonder.cdc.gov/mortSQL.html. Accessed
11 Jul 2014
279. Ulmer C, Bruno M, Burke S (2010) Committee on Future Directions for the National
Healthcare Quality and Disparities Reports Board on Health Care Services. 247.
280. Bodenheimer T, Pham HH (2010) Primary care: current problems and proposed solutions.
Health Aff (Millwood) 29:799805.
281. Liu JJ (2007) Health professional shortage and health status and health care access. J Health
Care Poor Underserved 18:5908.
282. Chin MH, Walters AE, Cook SC, Huang ES (2007) Interventions to reduce racial and ethnic
disparities in health care. Med Care Res Rev 64:7S28S.
283. Zuvekas SH, Taliaferro GS (2003) Pathways To Access: Health Insurance, The Health Care
Delivery System, And Racial/Ethnic Disparities, 1996-1999. Health Aff 22:139153.
284. U.S. Department of Health and Human Service (2014) HRSA Data Warehouse.
http://hpsafind.hrsa.gov/HPSASearch.aspx.
285. Allegany County Health Department (2013) Allegany County community health needs
assessment. 24.
286. Centers for Disease Control and Prevention Behavioral Risk Factor Surveillance System
Survey Questionnaire.

Page 123
15 APPENDIX 1: Baseline Health Assessment
15.1 Overview of Allegany and Garrett Counties
15.1.1 Geography
15.1.1.1 Allegany County
Allegany County with a population of 75,087 individuals is located in the northwestern part of
Maryland and is 424.16 square miles. Positioned in the Ridge-and-Valley Country of
the Appalachian Mountains, it is bordered to the north by the Mason-Dixon Line along
with Pennsylvania. To the south, it is surrounded by the Potomac River and West Virginia. To
the west is the Allegheny Front, and to the east is Frostburg, MD. The cities, towns & census
designated places incorporated municipalities the makeup Allegany County and this include:
Barrelville
Barton*
Bel Air
Bier
Bowling Green
Bowmans Addition
Carlos
Clarysville
Corriganville
Cresaptown
Cumberland*
Danville
Detmold
Eckhart Mines
Ellerslie
Flintstone
Franklin
Frostburg*
Gilmore
Grahamtown
Klondike
La Vale
Little Orleans
Lonaconing*
Luke*
McCoole
Midland*
Midlothian
Moscow
Mount Savage
Nikep
Ocean
Oldtown
Pleasant Grove
Potomac Park
Rawlings
South Cumberland
Spring Gap
Vale Summit
Westernport*
Woodland
Zihlman
Incorporated Places have an asterisk (*)
15.1.1.2 Garrett County
Garrett County with a population of 30,097 individuals is the western-most county in Maryland,
and its bordered to the north by the Mason-Dixon Line with Pennsylvania, to the south by
the Potomac River and West Virginia. Garrett County is 647.10 square miles of incorporated and
unincorporated jurisdiction divided into several neighborhoods, the names of which are:
Accident*
Bloomington
Crellin
Deer Park*
Finzel
Friendsville*
Gorman
Grantsville*
Hutton
Jennings
Kitzmiller*
Loch Lynn Heights*
Mountain Lake Park*
Oakland
Swanton
Incorporated Places have an asterisk (*)
DRAFT Final Report
Page 124

Figure 15-1: Major Cities and Towns in Allegany and Garrett Counties
Figure 15-1 displays the locations of major cities and towns in Allegany and Garrett Counties in
relation to population density. We observe that there are almost 15 major cities and towns in
Allegany Counties with several having a population density greater 525 people/km
2
. The two
largest cities (Frostburg and Cumberland) that have high population densities are located in the
central and northwestern parts of the county. In contrast, there are only a few major towns in
Garrett County with most located in low population density census tracts (< 26.7 persons/km
2
).
15.1.2 Schools
Allegany County has fourteen elementary schools, four middle schools, three high schools, one
technical education school, and one alternative program in the county [212]. They include:
Elementary School

Beall Elementary School (451)
Bel Air Elementary School (215)
Cash Valley Elementary School (320)
Cresaptown Elementary School (362)
Flintstone Elementary School (227)
Frost Elementary School (233)
George's Creek Elementary School (316)
John Humbird Elementary School (291)
Northeast Elementary School (315)
Parkside Elementary School (256)
South Penn Elementary School (497)
West Side Elementary School (381)
DRAFT Final Report
Page 125
Westernport Elementary School (277)

Middle School

Braddock Middle School (573)
Washington Middle School (695)
Westmar Middle School (282)
Mount Savage School (K-8) (401)

High School

Allegany High School (672)
Fort Hill High School (754)
Mountain Ridge High School (840)
Center for Career & Technical Education (304)
Eckhart Alternative Program (67)

Garrett County has eight elementary, two middle, and two high schools. They currently include:

Elementary
Accident Elementary School (235)
Broad Ford Elementary School (631)
Crellin Elementary School (109)
Friendsville Elementary School (132)
Grantsville Elementary School (227)
Swan Meadow Elementary School (36)
Yough Glades Elementary School (329)

Middle
Northern Middle School (323)
Southern Middle School (538)

High
Northern High School (495)
Southern Garrett High School (695)
15.1.3 Hospitals
Currently in Allegany County, the Western Maryland Health System (WMHS) offers a
continuum of care ranging from primary care to nursing home services. Services include acute
and chronic care, community health and wellness, clinical prevention, care coordination, home
care, community health workers, and provider recruitment. In addition, WMHS is the only
licensed hospice care facility in Allegany County and operates a regional medical center
consisting of a 275-bed hospital in Cumberland, along with two diagnostic centers, a nursing and
rehabilitation center in Frostburg, a community health and wellness center, two urgent care
DRAFT Final Report
Page 126
centers, and three primary care centers. WMHS is also a Level III trauma center, the only trauma
center in Western Maryland.
In Allegany County, the State of Maryland also owns and operates the Thomas B. Finan Center,
an inpatient psychiatric facility with 80 beds, in Cumberland. It provides services to those 18
years of age and older and includes inmates with criminal histories, non-criminals who have
been involuntarily committed, and voluntary patients.
In Garrett County, Garrett County Memorial Hospital (GCMH) runs a 55-bed, not-for-profit,
acute care hospital facility, including a 10-bed sub-acute rehabilitation unit. GCMH is the only
hospital in the region, serving a population of 31,000, including residents of Garrett County and
communities in the surrounding West Virginia counties. Services at the Hospital include a 24-
hour emergency department; inpatient care; observations services; obstetrics; pediatrics;
medical/surgical intensive care unit; operating room: radiology; lab; cardiopulmonary services;
as well as community and worksite wellness; safe sitter; and CPR programs and other ancillaries.
15.1.4 Important Landmarks
Garrett County has over 76,000 acres of parks, lakes, and publicly accessible forestland.
Nicknamed Marylands Mountaintop Playground," the county has the states highest elevation
at 3,360 feet, as well as its largest inland body of water (Deep Creek Lake). Garrett County is
home to the state's only sub-arctic wetlands and is the only county in the state to produce natural
gas.
According to the National Register of Historic Places listings, Garrett County
has 20 historic landmarks. These include:
The Anderson Chapel
Baltimore and Ohio Railroad Station, Oakland
Bloomington Viaduct
Borderside
Casselmans Bridge, National Road
Creedmore
James Drane House
Fuller-Baker Log House
Garrett County Courthouse
Glamorgan
Hoye Site
Inns on the National Road
Kaese Mill
Mercy Chapel at Mill Run
Meyer Site
Mountain Lake Park Historic District
Oakland Historic District
Pennington Cottage
Stantons Mill
Tomlinson Inn and the Little Meadows

According to the National Register of Historic Places there are 44 historic landmarks in Allegany
County. They include:
16 Altamont Terrace
200-208 Decatur Street
African Methodist Episcopal Church
Ber Chavim Temple
Barton Village Site
Bell Tower Building
Big Bottom Farm
Borden Mines Superintendents House
Breakneck Road Historic District
Wright Butler House
Canada Hose Company Building
Chapel Hill Historic District
Chesapeake and Ohio Canal National Historical
Park
City Hall
Michael Cresap House
DRAFT Final Report
Page 127
Cumberland YMCA
Decatur Heights Historic District
Downtown Cumberland Historic District
First Baptist Church
Folcks Mill
Footers Dye Works
Frostburg Historic District
Greene Street Historic District
Francis Haley House
Hocking House
Inns on the National Road
Klots Throwing Company Mill
Thomas Koon House
La Vale Tollgate House
Lonaconing Furnace
Lonaconing Historic District
Mount Savage Historic District
Old National Pike Milestones
Phoenix Mill Farm
Public Safety Building
Rolling Mill Historic District
Shaw Mansion
Shawnee Old Fields Village Site
Town Clock Church
George Truog House
Union Grove Schoolhouse
Washington Street Historic District
Waverly Street Bridge
Western Maryland Railroad Right-of-Way,
Milepost 126-Milepost 160
Western
Maryland Railway Station
DRAFT Final Report
Page 128

Figure 15-2: Location of Community Assets and Sensitive Human Receptors
Figure 15-2 shows the distribution of community assets and sensitive human receptors including
hospitals, schools, churches, and parks in Allegany and Garrett. There are approximately 153
churches and 87 schools in both counties. In Allegany county, the majority of the schools and
churches are located in central and western part of the county where there is a higher population
density compared to the eastern part of the county. For Garrett County, schools and churches are
dispersed somewhat evenly throughout the county, which has very few people as illustrated by
the low population density on the map. On the map, we also observe that there are four hospitals
in the western part of Allegany County in or near census tracts with high population density.
There is only one hospital in Garrett County located in the southwestern part of the county.
Figure 15-2 is important because it illustrates the spatial distribution of sensitive non-residential
land uses in the two counties. These land uses are important features of community ecosystems
in both counties and act as health promoting elements of the local infrastructure. It is important
to note that at these sensitive human receptor locations, there will be vulnerable populations
including children, elderly, and individuals with underlying disease who could be at risk from
UNGDP activities particularly air pollution near well pads and pollution emitted from diesel
truck traffic. In addition, there are small and large parks in both counties that act as ecologic
amenities for local populations. These parks act as recreational resources, contribute to local
aesthetics, and contribute to health particularly mental health and quality of life for residents.
UNGDP activities have the potential to reduce air quality near the parks, have negative ecologic
impacts, and reduce the use of the parks for recreational use.
DRAFT Final Report
Page 129
15.2 Demographics
The 2012 U.S. Census was used to obtain the most accurate population counts as well as
information on age, gender, and racial composition of Garrett County. The population of Garrett
County was defined from the population living in the following zip codes: 21520, 21521, 21522,
21523, 21531, 21532, 21536, 21538, 21539, 21541, 21550, 21561, and 21562.

Figure 15-3: Map of Zip Codes in Allegany and Garrett Counties
According to the 2012 U.S. Census, 50.4% of the populations were female and 49.6% were
male. 27.1% of the population were under the age of 18, while 17.7% of adults were 65 years
and older. Of those reporting race in Garrett County, 97.2% identified themselves White, 1% as
African-American, 0.8% as Hispanic and 1% as other. 3.7 percent of the population was
unemployed and 13% of the residents were living in poverty. The median income for resident is
approximately $45,354, which is higher than the regional average of $39,026.
In Allegany County, the population was defined using the following zip codes: 21502, 21521,
21530, 21532, 21539, 21540, 21545, 21545, 21555, 21557, 21562, 21766, 21501, 21503, 21504,
21505, 21524, 21528, 21529, 21542, 21543, 21556, and 21560. The 2012 U.S. Census was also
used to obtain the most accurate population counts as well as information on age, gender, and
racial composition of Allegany County. In 2012, the US Census estimated that 48% of county
residents were female, and 52% were male. In addition, 18% of the population were under the
age of 18, while 18.1% were 65 years and older. For those who reported their race, 88.3%
identified themselves as white, 7.6% as African-American, 1.5% as Hispanic and 2.6% as other.
In comparison to Garrett County 13%, 16.1% of residents live at or below poverty. The median
income in Allegany County is $39,087, compared to the Maryland state average of $68,559.
Table 15-1: Demographics, US Census 2012
!"#$%&'()*+ ,--"%'./ 0'&&"11 2'&/-'.3 4"%*$.
5'1"%$&/ % Total % Total % Total % Total
DRAFT Final Report
Page 130
,%" 7 8 4.6 S,S99 S.1 1,SS7 6.S S6S,2S8 4.9 2S,S29
,%" 7 9: 18 1S,4u4 21.7 6,SSu 2S.S 1,SSu,7uS 2u.1 94,998
,%" ; <8 18.1 1S,SS7 17.7 S,S2S 12.4 71S,726 18.1 8S,S1S
=$.>?*@('.*+
A)*1"
88.S 6S,921 97.2 29,24S S4.7 S,16S,29S 9S.9 444,S92
=$.>?*@('.*+
B-'+C
7.6 S,6S6 1 288 29 1,67S,SS2 S.4 16,u21
?*@('.*+ 1.S 1,u89 u.8 227 8.2 472,28S 1 4,7u7
D1)"& 2.6 1,999 1 S19 8.2 474,S84 1.7 7,98S
E"@@ 1)'. ?F 1S.1 6,772 1S.1 S,17S 11.S 44S,826 1S.S S2,S72
G."#(-$/#"
.1
4.9 S,u84 S.7 897 S.4 246,72u 4.7 18,47S
H$I"&1/ 16.1 1u,74u 1S S,797 9.4 SS2,116 1S.6 7u,719
G.*1@ BJ*-1
(&">9K8L
S7.9 19,27u 27.1 S,1u7 Su.8 7S1,SSS 44.7 1uS,168
D++J(*"3 B/
DM."&
8S.7 28,SS7 6S.S 12,SS4 89.9 2,1S8,8u6 81.S 188,1uu
2"3*'. ??
N.+$#"
$S9,u87 $4S,SS4 $68,99S $S9,u26
H"& 5'(*1'
N.+$#"
$21,677 $24,9u4 $S2,S2u $2u,9S6

When comparing Allegany and Garrett Counties, Garrett County had the highest number of
residents with less than a high school education (15%). In addition 57.9% of Allegany County
residents resided in units built before 1950, which is higher than Garrett County (27.1%), the
state (30.8%) and the region (44.7%). 85.7 percent of homes were occupied by the owner in
Allegany County compared to the region whose average was slightly lower at 81.5%. Garrett had
the lowest number of percentage of owners at 65.5%.
15.3 Vulnerable Populations
Garrett and Allegany counties. Vulnerability has been defined as how individuals or groups of
individuals or organisms respond to and recover from stressors inadequately or not as well as the
average [19, 20]. Vulnerability factors include characteristics, individual level and/or community
level that moderate the effect of environmental hazards on community health and well-being.
Individual level vulnerability factors influence the individuals response to stressors.
Demographic factors of interest when assessing vulnerability include race, ethnicity, age (e.g.,
children, elderly), and sex [7]. Some biologic factors include genetic make-up and pre-existing
medical conditions [213]. Genetic polymorphisms have been implicated in the etiology of
carcinogenesis when exposed to toxic pollutants [214, 215]. Pre-existing conditions and age have
also been associated with reduced response to stressors [216].Non-biologic factors such as
resilience, a pattern of positive adaptation in the context of significant risk or adversity [217]

has
DRAFT Final Report
Page 131
been identified as protective against stress [218]. Other individual level vulnerability factors,
low socioeconomic status, low educational attainment [219], and psychosocial stress [213] have
also been associated with negative health outcomes. Psychosocial stressors act synergistically to
raise levels of stress, increase vulnerability, and limit capacity of burdened populations to
overcome disease and improve health status [213]. Health behaviors also play a role in
increasing or decreasing an individuals vulnerability. For example, individuals who smoke, use
alcohol, consume unhealthy foods, or lead sedentary lifestyles have a higher risk of cancer,
diabetes, obesity, and lower life expectancy compared to other groups. In this HIA, we are
limited to assessing vulnerability using sociodemographic data and some county level health
data. We will not have access to individual health data including family history of disease for
populations in both counties.

Figure 15-4: Age Distribution for Allegany and Garrett Counties, Maryland, the Region, and the
U.S., Source: U.S. Census 2012
15.3.1 Age
In Figure 15-4, we observe a small percentage of children less than age 5 in Garrett and Allegany
counties compared to Maryland and the United States. We also observe a `high percentage of
children less than age 18 in Garrett County compared to the region and Allegany County. These
percentages were somewhat lower than the percentages for Maryland and the United States. For
both Allegany and Garrett counties, there is a high percentage of the population (approximately
18%) over the age of 65. Elderly residents may be more vulnerable to exposure to chemicals in
air and water due to compromised immune systems and comorbidities. Compared to adults,
young children are more susceptible to the potential effects of environmental contaminants
because of their higher consumption, metabolic and ventilation rates relative to their body mass
[220223]. Chemical exposures in early childhood could pose long-term health consequences
DRAFT Final Report
Page 132
because young children are in a state of rapid growth [220223]. Exposures in these sensitive
developmental windows could lead to poor health outcomes during their non-adult years and
chronic health outcomes over their life course. Additionally, young children have unique
behavioral and activity patterns that may predispose them to higher exposures. Other factors that
make them susceptible to the potential effects of environmental exposures include the fact that:
they may explore their world by mouthing objects [222, 223]; they are closer to the ground
where they may come into contact with contaminated surfaces; they eat, drink, and breathe more
per unit body weight compared to adults; their body systems including nervous, immune, and
reproductive systems are still developing; and their detoxification mechanisms may not be fully
developed [220, 223225]. Children can be exposed to a wide range of potentially toxic
compounds including metals, PAHs, PCBs, PBDEs, pesticides among other chemicals at their
homes, gardens/yards, and school/child cares and these exposures have a longer time to manifest
into adverse outcomes [226]. Both elderly populations and children less than age 18 should be
viewed as sensitive human receptors in Western Maryland.

Figure 15-5: Children Less than Age 5 and Adults Greater than 65 in Allegany and Garrett
Counties, Source: U.S. Census 2012
In Figure 15-5, we observe that there is a cluster of individuals over the age of 65 who live in
central, northwestern, and southwestern parts of Allegany County. We also observe a high
percentage of individuals over the age of 65 in the northwestern, central, and southern parts of
Garrett County. This population may be more vulnerable and have higher health risks due to
their health status, weakened immune systems, and co-morbidities. Additional steps must be
taken to ensure that this population has access to appropriate medical care and other resources
needed to improve health and quality of life.

DRAFT Final Report
Page 133

Figure 15-6: Comparison of Percent Poverty and Percent Less than High School Education for
Allegany and Garrett Counties, Source: U.S. Census 2012
15.3.2 Socioeconomic Status

Figure 15-7: Percent Poverty and Unemployment for Allegany and Garrett Counties, Maryland,
the Region, and the U.S., Source: U.S. Census 2012
DRAFT Final Report
Page 134
Figure 15-7 shows the spatial relationship between percent poverty and percent low educational
attainment (e.g., less than high school education). We observe clusters of high percentages of
persons in poverty and with less than HS education in Central and Western Allegany County. We
also observe high percentage of individuals without a high school diploma in the north central
and south central portions of Garrett County with a large percentage of persons in poverty in the
northwestern part and south central portion of the county. Overall, there are a large number of
individuals with less than a HS education. The map illustrates the area has an underserved
population that lacks economic opportunities. Although both Garrett and Allegany counties had
lower unemployment rates compared to the region, the state of Maryland, and the United States,
limited educational attainment indicates that the population may only have access to low wage
jobs and the population has limited economic mobility since the poverty rates for these counties
are high. Over 15% of the population in Allegany County is below the federal poverty level
which was higher than the poverty rate for Garrett, the region, Maryland, and the US.
Individuals living below the federal poverty line may have access to fewer resources such as
insurance and health care, higher exposure to social stressors, and may not have opportunity to
move away from industrial pollution sources. Additional steps must be taken to ensure that this
population has access to appropriate medical care and other resources needed to improve health
and quality of life [213].
15.4 Environmental Health
The U.S. Environmental Protection Agency (EPA), under the authority of the Emergency
Planning and Community Right-to-Know Act (42 USC 11004-11049 [1986]) established the
toxic release inventory (TRI) database through section 313 [227229]. It requires that major
industrial facilities that use more than 10,000 pounds or process more than 25,000 pounds of any
of the 650 TRI chemicals report their releases and waste management strategies [227]. We
extracted 2013 TRI data from an EPA database by using the EPAs TRI Explorer. Superfund
data was obtained from the USEPAs Comprehensive Environmental Response, Compensation,
and Liability Information System (CERCLIS) public access database, which contains non-
enforcement confidential information on hazardous waste sites, potentially hazardous waste
sites, and remedial activities as well as those noted on the National Priority List. As authorized
by the Clean Water Act, the National Pollutant Discharge Elimination System (NPDES) permit
program controls water pollution by regulating point sources that discharge pollutants into waters
of the United States. Industrial, municipal, and other facilities must obtain permits if their
discharges go directly to surface waters [230]. The EPA defines a brownfield as a property, the
expansion, redevelopment, or reuse of which may be complicated by the presence or potential
presence of a hazardous substance, pollutant, or contaminant [231]. The EPA estimates that
more than 450,000 brownfields exist in communities across the US with many in economically
depressed rural and urban neighborhoods [231233]. The EPA defines an Underground Storage
Tank (UST) as any underground piping connected to a chemical storage tank with at least 10%
of its combined volume underground [234]. Although there are many types of USTs classifiable
by their contents, only sites containing hazardous substances are regulated by both the EPA and
state agencies such as MDE [234]. Despite the EPAs efforts to manage USTs, 95% of all
regulated USTs contain petroleum derivatives. When an UST leaks, it is then known as a
Leaking Underground Storage Tank (LUST) [234]. In the event of a leak, air, groundwater, and
DRAFT Final Report
Page 135
soil contamination may become potential hazards for residents who live near these sites [234].
We obtained 2013 point location data for TRI facilities, Superfund sites, brownfields, LUSTs,
and NPDES permitted facilities from the EPA. We also obtained information on the point
location of conventional wells in Garrett County from a 1980 report [235].
We mapped all of the point locations for these facilities and land uses using ArcGIS. In addition,
we used ArcGIS to construct overlays of the facility and land use data in relation to population
density (persons/km
2
) at the census tract level using 2012 American Community Survey five-
year estimates (US Census).
Figure 15-8 shows the spatial distribution of conventional gas wells, Superfund sites,
brownfields, LUSTs and TRI facilities in Allegany and Garrett counties. The illustration shows
the overall burden of the facilities and land uses in both counties. Approximately 210
conventional gas wells were mapped in Garrett County. The wells are unevenly distributed
throughout the county. Two large clusters are located in the northwestern quadrant of the county
near Friendsville and Accident (see Figure 15-8) in an area with the lowest population density
(less than 26.7 persons/km
2
). While a heavily concentration of wells is clustered around the south
central region of the county near Oakland, Mountain Lake Park, and Loch Lynn Heights. This
area is more populated (26.8-65.7 persons/km
2
) in comparison to the Accident/Friendship region.

Figure 15-8: Spatial Distribution of Conventional Gas Wells, NPDES-Permitted Facilities,
Superfund Sites, Brownfields, LUSTs, and TRI Facilities in Allegany and Garrett Counties
A total of 101 facilities with NPDES permits are located in Allegany County, while 74 are
located in Garrett County. In Allegany, NPDES permitted facilities are primarily located in the
central, western, and southwestern parts of the county. Major clusters of the NPDES permitted
facilities were in Frostburg and Cumberland. Thirteen TRI facilities are located in Allegany
County, while three are located in Garrett. Similar to the spatial distribution of the NPDES
DRAFT Final Report
Page 136
permitted facilities, the majority of the TRI facilities are dispersed in the central, western, and
southwestern portions of Allegany County in or near local cities and towns such as Frostburg,
Cumberland, Midland, and Barton. Three Superfund sites are located in Garrett County with two
in Oakland in the southwestern part of the county and one in Grantsville in the northern part of
the county. Two Superfund sites were found in Allegany with one located in Frostburg and one
located in the Cumberland area. Three LUSTs were located in Allegany County, while two
LUSTs were located in Garrett County. A total of six brownfields were found in Allegany
County with zero located in Garrett County. The majority of the brownfields were located in a
high population density census tract in the Cumberland area (central Allegany).
Figure 15-8 illustrates the cumulative burden of various facilities that emit toxic compounds to
the air, water, or soil or if there is a breach or leak could have a negative impact on the
environment and the health of the public. Previous research has shown that low-income
populations, marginalized, and underserved groups such as some of the populations in Allegany
and Garrett counties, live in communities that experience a disproportionate risk from the burden
of and exposure to environmental hazards including noxious land uses such as landfills,
incinerators, brownfields, Publicly Owned Treatment Works (POTWs) (e.g., sewer and water
treatment plants), Superfund sites, TRI facilities, energy production facilities, chemical plants,
heavily trafficked roadways, LUSTs, and other locally unwanted land uses (LULUs) [213, 236
242]. This disproportionate burden and proximity to one or more pollution sources may lead to
an increase in exposure to adverse environmental conditions and contaminants for impacted
populations and communities.
15.4.1 Drinking Water
A large proportion of Marylanders currently rely on unregulated private wells as sources of
drinking water. An estimated 1.1 million Maryland residents draw drinking water from private
wells [21]. Since private wells that serve less than 25 people are not regulated by the federal Safe
Drinking Water Act, residents who rely on a private well system for their home drinking water
supply have the responsibility of managing the quality of their private well to ensure that it meets
drinking water safety standards [59]. As a result of improper well maintenance and testing, a
significant proportion of Maryland well owners could be exposed to elevated concentrations of
microbiological, chemical and or heavy metal contaminants in their drinking water. Moreover,
previous research conducted in our group has provided evidence that groundwater drawn from
Maryland aquifers in the Monocacy River basin in Western Maryland are impacted by elevated
levels of fecal indicator bacteria, including antibiotic-resistant Enterococcus spp [243]. Elevated
levels of nitrates and other chemicals have also been noted in Maryland groundwater; however,
as mentioned above, comprehensive data regarding the quality of groundwater consumed by
private well owners (over 1 million Marylanders) does not exist [22, 23]. Figure 15-9 illustrates
that private wells are concentrated most heavily around McHenry, Grantsville and Oakland. Over
14,200 will location records are currently available for Garrett County [244]. Approximately,
8,250 or 58% of well records occur in grid cells that contain Marcellus shale gas leases [244].
Previous studies indicate that private and public wells in close proximity to active gas wells may
pose a risk to the health of residents who rely on wells as their primary drinking water source
[83, 84].
DRAFT Final Report
Page 137

Figure 15-9: Location of Private Wells in Garrett County

Figure 15-10: Average Daily PM2.5 Concentrations, 2011
DRAFT Final Report
Page 138
15.4.2 Air
A wealth of scientific literature has shown relationships between PM exposure (e.g., coarse or
fine particles, acute or chronic) and increased respiratory and cardiovascular health end points
including increased mortality, hospital admissions, and emergency department visits [45, 46,
245255]. Scientific literature suggests that exposure to PM may be associated with decreased
lung function and increases in respiratory disease and symptoms such as asthma in children and
children with asthma may have the greatest risk to PM
2.5
[256266]. Other studies have shown
that particulate matter contributes to higher cardiovascular mortality risks in elderly patients and
sensitive populations particularly those with co-morbidities.
While no direct asthma data was collected, the average daily PM
2.5
concentrations were gathered
(Figure 15-10). Studies have shown that PM
2.5
levels are associated with asthma development
and increased asthma admissions to hospital emergency departments [267], so PM
2.5
concentrations may be an important issue for populations with persistent asthma. Across the
groups, PM
2.5
concentrations were very high, with Allegany and Garrett counties almost equal to
each other and the region PM
2.5
concentrations, all of which are higher than the PM
2.5

concentrations across Maryland. This is in line with the national trend that indicates asthma
incidence nationally is on the rise [268, 269].

Figure 15-11: Total TRI Releases for 2000, 2005, and 2010
TRI facilities must report releases to air, water, and land, as well as the quantities of chemicals
they recycle, treat, incinerate, or dispose of these chemicals on-site and off-site. In Figure 15-11
we see that total releases of chemicals reported to TRI have decreased from 2000 to 2010 for
Allegany, Garrett, the region, and the state of Maryland. In 2000, total TRI releases in Allegany
DRAFT Final Report
Page 139
County (4,585,316 pounds) were 40 times higher than the total TRI releases in Garrett County
(112,785). In 2005, total TRI releases in Allegany County (2,879,309 pounds) were more than 10
times higher than the total TRI releases in Garrett County (224,787). In 2010, total TRI releases
in Allegany County (2,312,628 pounds) were more than 10 times higher than the total TRI
releases in Garrett County (269,727). TRI releases in Allegany were significantly higher than
TRI releases in Garrett County over the ten-year period. Many of the chemicals reported by
industry are known carcinogens, genotoxins, developmental toxins, reproductive toxins,
mutagens, and can have chronic and acute health effects for exposed populations. Thus, there
may presently be a high potential for adverse health outcomes for cancer due to environmental
exposures and other endpoints in Allegany County compared to Garrett County.
15.4.3 National Scale Air Toxics Assessment (NATA)
The US Environmental Protection Agency (USEPA) National-Scale Air Toxics Assessment
(NATA) estimates the cancer and respiratory risk of hazardous air pollutants (HAPs). HAPs are
known to cause or are suspected of causing cancer or other serious health problems such as
damage to the immune system, and neurological, reproductive, developmental, and respiratory
problems [270272] All 187 HAPs are addressed within NATA dataset, most of which are
defined in the Clean Air Act.
In addition, for the air toxics for which information on chronic risks exists, the exposure
concentration estimates are used to quantify potential health effects (cancer and non-cancer)
from inhalation of air toxics using EPA's risk assessment and characterization framework [270].
Non-cancer risks are categorized as either respiratory or neurological hazards [270]. The census
tract is the smallest analytical unit for which exposure and health risk estimates are provided in
the NATA dataset. The cancer risk estimation is calculated from personal exposure. The relation
of likelihood of contracting cancer and the exposure level is quantified by an USEPA developed
cancer dose-response curve [270]. The NATA estimates cancer risk on the basis of the inhalation
unit risk (IUR) factor, a measure of the cancer-causing potential of each air toxic [270]. The
concentration of each pollutant in a given census tract is multiplied by its IUR in order to
estimate individual lifetime cancer risk. Cancer risks are assumed to be additive and lifetime
cancer risk from all air toxics present in a tract are summed to obtain the total estimated lifetime
cancer risk for the tract. Estimated lifetime cancer risks are expressed by number of people per
million, where "N" is the likelihood of contracting cancer out of one million people exposed to a
specific concentration of an air pollutant continuously (24 hours/day) over a lifetime (defined as
seventy years) [270, 271].
Respiratory risks are estimated using the concentration of the pollutant in the air believed to have
no adverse effect on the lungs and air passages with constant exposure, referred to as the
inhalation reference concentration [270]. To estimate respiratory risk for each census tract, a
hazard quotient is calculated by dividing the ambient concentration of each pollutant in each tract
by its inhalation reference concentration. A composite respiratory hazard index is then obtained
by summing the hazard quotients of all air toxics present within that particular census tract.
Index above one indicates the potential for respiratory problems over a lifetime of exposure
while an index below one means a lifetime of exposure is not expected to cause adverse effects
to the lungs and air passages [270].
DRAFT Final Report
Page 140

Figure 15-12: NATA Cancer Risk, 2002 and 2005
Figure 15-12 displays the lifetime cancer risk from air toxics. The estimated cancer risk for
Allegany County was higher than then lifetime cancer risk for Garrett County and state of
Maryland in 2002 and 2005. The estimated cancer risk changed significantly from 2002 to 2005
which could be due to a decrease in HAP levels or changes in how the estimated lifetime cancer
risk from air toxics was calculated. Figure 15-13 displays respiratory risk for Garrett and
Allegany, the region, and the state of Maryland. We observe that in 2002, the respiratory hazard
score was above 1 for Allegany, Garrett, and the state of Maryland. This means that populations
were at risk of respiratory problems including asthma, COPD, bronchitis, and other issues. We
see that in 2005, the respiratory score decreased across the board with only Allegany and the
state of Maryland receiving a score above 1. This decrease could be due to changes in the
calculation or improvements in air quality. However, in 2005, citizens in Allegany were still at
risk for negative respiratory health outcomes including vulnerable groups who have pre-existing
conditions such as asthma who would be at a higher risk of hospitalization and emergency
department visits. Additionally, respiratory health risks due to proximity to one or more pollution
sources may be of concern to populations in the two counties currently burdened by air pollution
from the oil and gas industry.
DRAFT Final Report
Page 141

Figure 15-13: Respiratory Hazard Index, 2002 and 2005
15.5 Physical Health Indicators
The health profile of the residents of this region was compiled by using data collected on overall
life expectancy, poor physical days, preventable hospital stays, chronic disease, major causes of
morbidity and mortality, and birth outcomes. The collected data for Allegany and Garrett
Counties (where UNGDP activities may take place) was compared to the health data of the
region (Allegany and Garrett Counties in Maryland, Bedford, Fayette, and Somerset Counties in
Pennsylvania, and Grant, Hampshire, Mineral, Preston, and Tucker Counties in West Virginia),
and the State of Maryland for an overall health profile. We defined the region as Garrett and
Allegany and other counties in neighboring states of West Virginia and Pennsylvania because the
team thought the counties in the neighboring states had more in common with Western Maryland
(culturally, sociodemographically including racial composition, occupational opportunities,
geology, hydrology, topography, economy, history of oil and gas industry including conventional
wells and UNGDP) than Allegany and Garrett have with other counties in Maryland.
15.5.1 Life Expectancy
Data on life expectancy was obtained from the CDCs Community Health Status Indicators
website [273]. As displayed in Table 15-2, Garrett County has the highest average life
expectancy at 78.2, compared to Allegany County (77.4) and the state of Maryland (67.8). The
DRAFT Final Report
Page 142
regional average life expectancy of Maryland, West Virginia, and Pennsylvania was lower than
both Allegany County and Garrett County at 76.7.
Table 15-2: Life Expectancy, 2009
,--"%'./ 5$O 0'&&"11 5$O 2'&/-'.3 4"%*$.
Life Expectancy 77.4 78.2 67.8 76.7
15.5.2 Poor Physical Health Days
Data on the number of poor physical health days in the past 30 days was obtained from the
Behavioral Risk Factor Surveillance System (BRFSS) for 2006-2012 [274] As displayed in
Table 15-3, Allegany County had the highest number of poor physical health days at 4.8, while
Maryland had the lowest number (3.1). Allegany County had more poor physical health days
than both Garrett County and Maryland.
Table 15-3: Poor Physical Health Days, 2006-2012
,--"%'./ 5$O 0'&&"11 5$O 2'&/-'.3 4"%*$.
Pooi Physical
Bealth Bays
4.8 (4.2-S.S) S.7 (S.1-4.2) S.1 (S.u-S.2) 4.S (S.6-S.S)

Figure 15-14: Preventable Hospital Stays, 2011
We obtained 2011 data from the University of Wisconsin County Health Indicators Project
[275]. Figure 15-14 displays the Ambulatory Care Sensitive Conditions (ACSC) rate for
DRAFT Final Report
Page 143
preventable hospital stays in Allegany (88.0) and Garrett (67.6) counties was higher than the
overall state rate (60.2). The ACSC rate for Allegany was higher than the rate for the both the
region (85.6) and Garrett County.
15.5.3 Chronic Diseases
Chronic diseases examined in this study include adult hypertension, adult obesity, diabetes, and
adult smoking.
15.5.3.1 Adult Hypertension
We obtained data on adults with high blood pressure for 2006-2012 from the Behavioral Risk
Factor Surveillance System (BRFSS) [274]. Figure 15-15 illustrates that Both Allegany and
Garrett counties had higher percentages of adults with high blood pressure (37% and 31%,
respectively) in comparison to the State of Maryland (30%). In comparison to the region
(34.3%), Allegany County has a higher percentage of adults with high blood pressure while
Garrett Countys percentage was lower.

Figure 15-15: Percent of Adults with High Blood Pressure, 2006-2012
15.5.3.2 Adult Obesity and Diabetes
Other serious issues facing this area of Maryland are adult obesity and diabetes. Figure 15-16
shows the percent of obese adults and adults with diabetes in Allegany and Garrett counties in
Maryland compared to the region and Maryland. This data was obtained for years 2006-2012
from BRFSS [274]. The percent of obese adults in Allegany and Garrett was 21% and 30%,
respectively. While, percent with diabetes in Allegany and Garrett was 12% and 11%,
respectively. The trends are seen mirroring each other, not surprisingly since obesity has been
linked to the development of Type 2 Diabetes [24]. While both counties have lower percentages
DRAFT Final Report
Page 144
of obese adults and either equal or lower percentages of diabetic adults compared to the region
(12%), they are both higher compared to the state of the Maryland (9.7%).

Figure 15-16: Percent of Obese Adults and Percent of Adults with Diabetes, 2006-2012
15.5.3.3 Adult Smoking
15.5.3.3.1 Adult Smoking
We obtained adult smoking data for the years 2006-2012 from BRFSS [274]. In both Allegany
(23%) and Garrett (19.5%) counties, the percent of adults who smoke was much higher than the
percent of adults who do the same across the state (15.4%). However, only the smoking rate for
Allegany was higher than the smoking rate for the region (22.7%) (Figure 15-17).

Figure 15-17: Percent of Adult Smokers, 2006-2012
DRAFT Final Report
Page 145
15.5.4 Major Causes of Morbidity and Mortality
The morbidity and mortality of the residents of this region was compiled by using data collected
on cancer, mortality, and birth rates. The collected data for Allegheny and Garrett Counties was
compared to the health data of the State of Maryland for an overall health profile.
15.5.4.1 Cancer
We obtained cancer incidence data from the National Cancer Institutes (NCI) State Cancer
Profile site [276]. As displayed in Table 15-4, prostate cancer, breast cancer, and colorectal
cancer were the cancers with the highest incidence rates. Allegany County had the highest
incidence rate for Non-Hodgkins Lymphoma, while the Maryland, West Virginia, and
Pennsylvania region had the lowest incidence rate. Allegany County also had the highest
incidence rate for leukemia (16.2) and Garrett County had the lowest (9.1). For melanoma,
Maryland had the highest incidence rate (21.2) and Garrett County had the lowest incidence rate
(16.3). For breast cancer, Maryland had the highest incidence rate (128.0), while the Maryland,
West Virginia, and Pennsylvania region had the lowest incidence rate (111.8). For prostate
cancer, Maryland had the highest incidence rate (157.2) while Garrett County had the lowest
(113.3). The Maryland, West Virginia, and Pennsylvania region had the highest incidence rate,
24.7, for bladder cancer, while Maryland had the lowest incidence rate, 19.2. For colorectal
cancer, Allegany County had the highest incidence rate (52.1), and Maryland had the lowest
incidence rate (41.5).

Table 15-4: Cancer Incidence Rates, 2006-2010
5'.+"&
P/("
,--"%'./ 5$O 0'&&"11 5$O 2'&/-'.3 4"%*$.
Non-
Bougkin's
Lymphoma
2S.6 (19.4- 28.6) 2u.S (14.7-28.1) 17.8 (17.S-18.S) 16.2 (12.6-2u.S)
Leukemia 16.2 (12.6-2u.S) 9.1 (S.2-14.9) 11.2 (1u.8-11.6) 1S.S (9.4-18.S)
Nelanoma 17.1 (1S.4-21.6) 16.S (1u.7-2S.8) 21.2 (2u.6-21.7) 17.1 (12.S-2S.u)
Bieast
Cancei
114.u (1uu.7-128.8) 118.9 (98.u-
14S.S)
128.u (126.2-
129.7)
111.8 (92.u-
1S6.2)
Piostate
Cancei
146.6 (1S1.S-16S.4) 11S.S (9S.1-
1S7.u)
1S7.2 (1SS.u-
1S9.S)
1S7.8 (11S.S-
164.7)
Blauuei
Cancei
2u.1 (16.4- 24.4) 21.6 (1S.S-29.6) 19.2 (18.7-
19.7)
24.7 (19.4-S1.4)
Coloiectal
Cancei
S2.1 (4S.9- S9.u) 4S.1 (S4.2-SS.7) 41.S (4u.7-42.2) Su.S (4u.6-62.S)

Cancer Mortality
Cancer data chosen for analysis was based on a known relationship between a particular cancer
and an exposure of concern that occurs during the UNGDP process and community concerns.
Cancer mortality data was chosen from the following:
DRAFT Final Report
Page 146
Colorectal cancer
Leukemia
Melanoma
Multiple myeloma
Non-Hodgkins lymphoma
Prostate cancer
Bladder cancer
Breast cancer
Cancer deaths per 100,000
For this part of the baseline health assessment, we obtained data on mortality for various cancers
using the following ICD-10 codes: Non-Hodgkin's lymphoma (C82-C85); Multiple myeloma
and immunoproliferative neoplasms (C88,C90); Leukemia (C91-C95); Malignant melanoma of
skin (C43); Malignant neoplasm of breast (C50); Malignant neoplasm of prostate (C61);
Malignant neoplasm of bladder (C67); and Malignant neoplasms of colon, rectum and anus
(C18-C21).

a. Colorectal Cancer b. Leukemia

c. Melanoma d. Multiple Myeloma

DRAFT Final Report
Page 147

e. Non-Hodgkins Lymphoma f. Prostate Cancer

g. Bladder Cancer h. Breast Cancer
Figure 15-18: Number of Deaths from Various Cancers per 100,000 (Age-Adjusted) in Allegany
and Garrett Counties Compared to Maryland and the Region (2000-2010), Source: National
Cancer Institute
Figure 15-18 shows the number of deaths per 100,000 from the cancers of interest previously
mentioned in Allegany and Garrett counties compared to the region and all of Maryland. Overall,
the top three cancers in Allegany and Garrett counties combined that result in the highest
numbers of deaths were colorectal, breast, and prostate cancers. Deaths from these cancers were
higher in these counties compared to the region and State of Maryland overall. Furthermore,
compared to the leading causes of death from cancer nationwide, these counties rates of
colorectal cancer deaths were higher [277].
We obtained data on various cancers from CDC Wonder on cancer mortality for Allegany,
Garrett, the region, and the state of Maryland [278]. The ICD-10 codes for the cancers included:
Malignant neoplasms of lip, oral cavity and pharynx (C00-C14), Malignant neoplasm of
esophagus (C15), Malignant neoplasm of stomach (C16), Malignant neoplasms of colon, rectum
and anus (C18-C21), Malignant neoplasms of liver and intrahepatic bile ducts (C22), Malignant
DRAFT Final Report
Page 148
neoplasm of pancreas (C25), Malignant neoplasm of larynx (C32), Malignant neoplasms of
trachea, bronchus and lung (C33-C34), Malignant melanoma of skin (C43), Malignant neoplasm
of breast (C50), Malignant neoplasm of cervix uteri (C53), Malignant neoplasms of corpus uteri
and uterus, part unspecified (C54-C55), Malignant neoplasm of ovary (C56), Malignant
neoplasm of prostate (C61), Malignant neoplasms of kidney and renal pelvis (C64-C65),
Malignant neoplasm of bladder (C67), Malignant neoplasms of meninges, brain and other parts
of central nervous system (C70-C72), Hodgkin's disease (C81), Non-Hodgkin's lymphoma (C82-
C85), Leukemia (C91-C95), Multiple myeloma and immunoproliferative neoplasms (C88,C90),
Other and unspecified malignant neoplasms of lymphoid, hematopoietic and related tissue (C96),
All other and unspecified malignant neoplasms (C17,C23-C24,C26-C31,C37-C41,C44-
C49,C51-C52,C57-C60,C62-C63,C66,C68-C69,C73-C80,C97), In situ neoplasms, benign
neoplasms and neoplasms of uncertain or unknown behavior (D00-D48).
Overall, the combined numbers of deaths from cancer in Allegany and Garrett counties are
slightly higher than the total cancer deaths in the region and Maryland, with Allegany County
having a higher number of deaths compared to Garrett.

Figure 15-19: Total Cancer Deaths per 100,000, 2000-2010
15.5.4.2 Other Mortality Data
Mortality data was analyzed by examining chronic respiratory disease deaths, flu deaths,
cardiovascular disease deaths, cerebrovascular disease deaths, septicemia deaths, and all- cause
mortality.
15.5.4.2.1 Chronic respiratory disease deaths
We obtained data on chronic respiratory deaths from CDC Wonder using the following ICD-10
codes (Bronchitis, chronic and unspecified (J40-J42), Emphysema (J43), Asthma (J45-J46), and
other chronic lower respiratory diseases (J44, J47) [278]. The number of deaths in Allegany
(54.5/100,000) and Garrett counties (51.4/100,000) due to chronic respiratory disease were
DRAFT Final Report
Page 149
higher than both the number of deaths per 100,000 in the region and state for the same disease
(Figure 15-20).

Figure 15-20: Total Chronic Respiratory Deaths per 100,000, 2000-2010
15.5.4.2.2 Flu deaths
We obtained data on influenza and pneumonia mortality from CDC Wonder using the following
ICD-10 codes (Influenza (J09-J11), Pneumonia (J12-J18) [278]. The number of deaths
contributed to flu in Allegany County is higher than the number of flu deaths in Garrett County,
yet both are lower than the number of deaths from flu in the state (Figure 15-21).
DRAFT Final Report
Page 150

Figure 15-21: Total Flu and Pneumonia Deaths per 100,000, 2000-2010
15.5.4.2.3 Cardiovascular disease deaths
We obtained heart disease mortality data using CDC Wonder [278] and the following ICD-10
codes: Hypertensive heart disease (I11); Hypertensive heart and renal disease (I13); Acute
myocardial infarction (I21-I22); Other acute ischemic heart diseases (I24); Atherosclerotic
cardiovascular disease, so described (I25.0); All other forms of chronic ischemic heart disease
(I20,I25.1-I25.9); Acute and sub-acute endocarditis (I33); Diseases of pericardium and acute
myocarditis (I30-I31,I40); Heart failure (I50); and All other forms of heart disease (I26-I28,I34-
I38,I42-I49,I51). We found that the cardiovascular disease mortality rate for Allegany County
(275.6) was significantly higher than the rate for Garrett, the region, and the state of Maryland.
This disparity in cardiovascular disease mortality could be due to a number of factors including
exposure to air pollution, health behaviors, lifestyle, or genetic factors [45].
DRAFT Final Report
Page 151

Figure 15-22: Cardiovascular Disease Deaths per 100,000, 2000-2010
15.5.4.2.4 Cerebrovascular disease deaths
We obtained data on cerebrovascular disease mortality from CDC Wonder using the following
ICD-10 codes (I60-I69) [278]. The rate of stroke-related mortality for Allegany County
(59/100,000) was higher than the mortality rates for Garrett, the region, and the state of
Maryland.

Figure 15-23: Cerebrovascular Disease Deaths per 100,000, 2000-2010
DRAFT Final Report
Page 152
15.5.4.2.5 Septicemia deaths
Data on Sepsis (septicemia) mortality was obtained through CDC Wonder using the following
ICD-10 codes (A40-A41) [278]. Septicemia is an illness that affects all parts of the body that can
happen in response to an infection and can quickly become life-threatening. In severe cases of
sepsis, one or more organs fail. During the worst case scenario, sepsis causes a decrease in blood
pressure, the heart to weaken, and septic shock which can lead to organ failure and death.
Patients who develop sepsis have an increased risk of complications and death and face higher
healthcare costs and longer treatment. Sepsis is a response to an infection. When there is an
infection, the immune system releases chemicals to fight the infection. The chemicals sometimes
cause body-wide inflammation, which can lead to blood clots and leaky blood vessels. This
impairs blood flow, which damages the bodys organs by depriving them of nutrients and
oxygen. People with weakened immune systems, infants and children, elderly citizens, and
people with chronic diseases are at risk from this condition. We found that the septicemia
mortality rate for Allegany County was 21/100,000. This rate is twice as high as the rate of
Garrett County and also higher than the rates of the region and the state of Maryland.

Figure 15-24: Septicemia Deaths per 100,000, 2000-2010
15.5.4.2.6 All-Cause mortality
All-cause mortality rates for Allegany (853) and Garrett (808) were higher than the rate for
Maryland (768).
DRAFT Final Report
Page 153

Figure 15-25: All-Cause Mortality, 2000-2010
15.5.4.3 Birth Outcomes
Birth outcomes were analyzed by examining low birth weight, premature births, and infant
mortality.
15.5.4.3.1 Low birth weight and premature births
We obtained data on percent low birth weight (< 2800 grams) and premature births for Allegany,
Garrett, the region, the state of Maryland, and the U.S. from the Health Indicators Warehouse
and National Vital Statistics System for 2006-2012. The percentage of premature births in
Allegany (13%) was higher than the percentages for Garrett (12%), MD (12.9%), region
(11.6%), and the United States (12.2%). Percentage of babies born with low birth weight (LBW)
in Allegany (9.1%) was higher than % low birth weight for Garrett (7.5%), MD (9%), region
(8%), and the United States (8.2%).
15.5.4.1 Infant mortality
We obtained data on infant mortality for Allegany, Garrett, the region, the state of Maryland, and
the US from the Health Indicators Warehouse. Infant mortality rates of 8.4 deaths/1000 births
(Allegany) and 10.8 deaths/1000 births (Garrett) were higher than the rates for the MD (7.2
deaths/1000 births), and US (6.9 births/1000 deaths).

DRAFT Final Report
Page 154

Figure 15-26: Percent Low Birth Weight, 2006-2012

Figure 15-27: Percent Premature Births and Low Birth Weight, 2006-2012
DRAFT Final Report
Page 155

Figure 15-28: Infant Mortality, 2006-2010
Many factors work together to affect the health of individuals and populations. Factors such as
genes and biology, individual behavior, physical and social environments, and health care
services have considerable impact on health. However, there is growing evidence that the genetic
composition of individuals and populations and the lifestyle choices, such as diet and exercise
that comprise individual behavioral determinants have considerably less impact on health than
factors related to social circumstances, physical environments, and access and quality of health
care services (WHO, 2008). Health services, the social environment, and the physical
environment are significant drivers of population health outcomes and constitute what we call
the social determinants of health (SDH).
SDH are the complex, integrated, and overlapping social structures and economic systems that
include the social environment, physical environment, and health services. In order to determine
the baseline health of citizens of Allegany and Garrett counties, we obtained available
information regarding rates of sexually transmitted infections (STIs), violent and nonviolent
crime, injuries, substance abuse, mental health, and suicide.
15.6.1 Sexually Transmitted Infections (STIs)
Information regarding STIs for 2011 was obtained from County Health Rankings (Chlamydia)
and the Health Indicators Warehouse (Gonorrhea). In Allegany County, the incidence of
chlamydia was 236 per 100,000 population and 190 per 100,000 for gonorrhea; in Garrett
DRAFT Final Report
Page 156
County, the incidence of chlamydia was 143 per 100,000 and 29.9 per 100,000 for gonorrhea.
These rates are low when compared to the State of Maryland. The prevalence of HIV was 179.6
per 100,000 in Allegany County and 23.2 per 100,000 in Garrett County. HIV rates in both
counties are well below the 2011 Maryland state average of 632.9 per 100,000.
Table 15-5. Sexually Transmitted Infections (STIs), 2011
Area Chlamydia Rate
(per 100,000)
Gonorrhea Rate
(per 100,000)
HIV Rate
(per 100,000)
Allegany 235.6 41.5 179.6
Garrett 143.1 29.9 23.2
Maryland 466.9 110.8 632.9
Region 137.9 214.4 66.6

Figure 15-29: Chlamydia Rate, 2011 and HIV Rate, 2010 per 100,000
15.6.2 Crime
Information regarding violent and property crime was obtained from the Maryland Governors
Office on Crime Control and Prevention Crime Statistics Report for 2000, 2005, and 2010. Data
regarding homicides was obtained from County Health Rankings and the National Center for
Health Statistics for 2010. In Maryland, violent crime included murder, rape, robbery,
aggravated assault and property crime included offenses such as breaking and entering, larceny
theft, and motor vehicle theft. The State average for 2010 was approximately 3,549.2 reported
incidents per 100,000 population, and marked the lowest ever reported crime rates for
Maryland. For the years obtained, total arrests for violent and property crimes peaked in 2010 for
Allegany County, with a total of 2,878 incidents reported, contributing to an overall crime rate of
3,957.6 reported incidents per 100,000, slightly higher than the state average. Crime rates are a
DRAFT Final Report
Page 157
lower in Garrett County, fluctuating between 530 and 550 incidents reported per year with a
crime rate between 1,742.8 per 100,000 and 1,856.2 per 100,000.
In Garrett County, crime rates across all categories remain steady and lower than the Maryland
State averages, fluctuating slightly over the 10-year period between 2000 and 2010. In Allegany
County, similar to regional data, there is a slow but steady increase in most crime categories in
this same period. This increase runs counter to statewide trends, which demonstrate major
decreases in crime rates across all categories in the last decade. In Allegany County, the rate of
reported property crimes increases from 2107 in 2000, to 3528.6 in 2010, an increase from 2,812
per 100,000 to 3,528.6 per 100,000. The Maryland State average during this same time period
witnessed a decrease from 4,048.6 incidents per 100,000 in 2000 to 3,001.8 incidents per
100,000 in 2010. Allegany County also witnessed similar increases in violent crime rates, in
2000 there were 271 incidents report at a rate of 361.7 per 100,000, and in 2010 this number had
increased to 312 incidents and 429 per 100,000. These rates are still lower than the Maryland
State average of 547.4 incidents per 100,000 but are steadily increasing while the statewide
numbers are decreasing. Allegany County violent crime rates run parallel to regional rates: in the
Western Maryland, West Virginia, and Pennsylvania region, the violent crime rate per 100,000
was highest in 2010, at 621.7, and lowest in 2000, at 492.4. Homicide rates, as reported in the
County Rankings Data shows that rates in both counties are quite low, much lower than the
Maryland State average of 9.3 homicides per 100,000. In Garrett County, the rates were so low
that the data was reported as unreliable. Maryland Crime Statistics estimates the murder rate to
be approximately 2 per 100,000 for 2011. In Allegany County the homicide rate for 2011 was
2.3 per 100,000 (1.2-4.0 95 CI).
Table 15-6. Total Crime, 2010
Area Total Crime Incidents (#) Total Crime Rate (per
100,000)
Allegany 2,878 3957.6
Garrett 532 1791.5
Maryland 204,916 3549.2
Region 1154.9 2139.6

Table 15-7. Violent and Property Crime, 2010
Violent Crime Property Crime
Area Incidents (#) Rate
(per 100,000)
Incidents (#) Rate
(per 100,000)
Allegany 312 429.0 2566 3528.6
Garrett 49 165.0 483 1626.5
Maryland 31,604 547.4 173,312 3001.8
Region 197 621.7 957.8 1531.8
DRAFT Final Report
Page 158

15.6.3 Injuries
Data for deaths resulting from unintentional injuries were obtained from Health Indicators
Warehouse, National Vital Statistics System for the years 2006-2010. Injuries include
unintentional injury mortality, accidental poisonings, alcohol-impaired driving deaths, motor
vehicle traffic deaths, fall deaths, and drownings all per 100,000. The total mortality rate from
unintentional injury was 41 per 100,000 for Allegany County and 40.5 per 100,000 for Garrett
County, both counties have much higher rates than the Maryland State average of 25.5 per
100,000, yet lower than the overall region. Both are slightly higher than the national average of
39.9 per 100,000. Information collected on accidental poisonings, drownings, and fall deaths
revealed very low mortality rates that were too small to be reliably reported. Mortality from
motor vehicle traffic deaths were also measured, with a mortality rate of 12.1 per 100,000 for
Allegany County and a rate of 21.6 per 100,000 in Garrett County. Maryland averages 10.7
traffic deaths per 100,000.
Table 15-8. Unintentional Injuries, 2006-2010
Area Unintentional Injury Mortality
Rate (per 100,000)
6

Motor Vehicle Death Rate
(per 100,000)
Allegany 41.0 12.1
Garrett 40.5 21.6
Maryland 25.5 10.7
Region 51.2 20.3

6
2000-2010 rate
DRAFT Final Report
Page 159

Figure 15-30: Total Accidental Deaths and Motor Vehicle Deaths per 100,000, 2006-2010
Information on alcohol impaired driving deaths was obtained from the 2014 County Health
Rankings Information and the Fatality Analysis Reporting System. The percentage of driving
deaths that were a result of alcohol impairment has been estimated 29.4 % for Allegany County
(15/51 deaths) and 41% for Garrett County (16/39 deaths). Statewide estimates suggest that 33%
or 867/2626 deaths in Maryland can be attributed to alcohol-impairment while the Western
Maryland, West Virginia, and Pennsylvania region had the largest percentage of alcohol-
impaired drivers, at 42.1% or 196/469 deaths.
Table 15-9: Alcohol-Impaired Driving Deaths, 2008-2012
Area Alcohol-Impaired
Driving Deaths (#)
Driving Deaths (#) % Alcohol-
Impaired Driving
Deaths
Allegany 15 51 29.4
Garrett 16 39 41.0
Maryland 867 2626 33.0
Region 196 469 42.1

,--"%'./ 5$O 0'&&"11 5$O 2'&/-'.3 4"%*$.
Alcohol-Impaiieu
Biiving Beaths (#)
1S 16 867 196
Biiving Beaths (#) S1 S9 2626 469
% Alcohol-
Impaiieu Biiving
Beaths
29.4 41.u SS.u 42.1

DRAFT Final Report
Page 160
Data on suicide including intentional self-harm by discharge of firearms and intentional self-
harm by other and unspecified means and their sequelae were obtained from CDC Wonder
Mortality from 2000-2010. The total mortality rate from intentional self-harm (suicide) for
Allegany County was 12.1 (9.7-14.1 95%CI) per 100,000 and 11.5 (8.2-15.8 95%CI) per
100,000 for Garrett County. These rates are significantly higher than the state average of 8.7
(8.2-15.8 95% CI) per 100,000 and lower than the regional average of 13.9 per 100,000.
Table 15-10. Suicide, 2000-2010
Area Suicide Mortality
(per 100,000)
Allegany 12.1
Garrett 11.5
Maryland 8.7
Region 13.9
15.6.4 Mental Health
Data on mental health specific to residents of Allegany and Garrett counties were obtained
through the County Health Rankings Database and the Health Indicators Warehouse. Mental
health was measured by the County Health Rankings as mentally unhealthy days (or the number
of reported mentally unhealthy days per month among adults over age 18). A related measure
tracks the percentage of adults over 18 who report not having sufficient social-emotional support.
In the period 2006-2012, Allegany reports an average of 3.8 mentally unhealthy days per month,
and Garrett County reports 3.6 mentally unhealthy days per month. Rates for both counties are
higher than the Maryland average of 3.2 mentally unhealthy days per month. Data based on the
Behavioral Risk Factor Surveillance System measured adult responses to the question How
often do you get the social and emotional support you need? In Allegany County 18.7% of
adults felt that they did not receive enough support, and in Garrett County this number
approximately 20.0%, while the Maryland State average is approximately 19.8%.
Table 15-11. Mental Health, 2006-2012
Area Mentally Unhealthy Days
(days/month)
Perceived Social Support
(%)
Allegany 3.8 18.0
Garrett 3.6 20.0
Maryland 3.2 19.8
Region 3.9 19.2
DRAFT Final Report
Page 161
15.6.5 Substance Abuse
Substance abuse data were extracted from the Health Indicators Warehouse, with measures for
adult binge drinking and excessive drinking, collected from the period 2006-2012. The
Behavioral Risk Factor Surveillance System was used for self-reported data on binge drinking
7

and excessive drinking.
8
In Allegany County 16.5% of adults over 18 years of age report binge
drinking and 17.3% report excessive drinking, slightly higher than the state averages. In Garrett
County, these rates were similar, 16.0% adults admit to binge drinking and 18.2% report
excessive drinking. Although both counties report slightly higher rates when compared to
Maryland State averages (14.4% binge drinking and 15.7% for excessive drinking), wide
margins of error could account for these differences. Information on other types of substance
abuse are more difficult to obtain. However measures from the National Community Health
Status Indicators for 2009 includes a count of the number of recent drug users, estimating that
4,597 people in Allegany County and 1,758 people in Garrett County are recent drug users.
Table 15-12. Substance Abuse, 2006-2012
Area Binge Drinking
(%)
Excessive
Drinking (%)
Recent Drug Use
(#)
Allegany 16.5 15.5 4,597
Garrett 16.0 17.3 1,758
Maryland 14.4 15.4 N/A
Region 13.5 13.2 N/A

7
Sample respondents age 18+ who drank 5 or more drinks for men, 4 or more drinks for women, at one or more
occasions in the past 30 days [286].
8
Sample respondents age 18+ who drank more than two drinks per day on average (for men) or more than one drink
per day on average (for women) or who drank 5 or more drinks during a single occasion (for men) or 4 or more
drinks (for women) during a single occasion [286].
DRAFT Final Report
Page 162

Figure 15-31: Percent Adult Excessive Drinking
15.7 Health Care Infrastructure
Health care infrastructure is fundamental to the provision and execution of health services and a
well-coordinated, high-quality infrastructure provides the capacity to prepare for and respond to
both acute (emergency) and chronic (ongoing) issues related to a communitys health. A robust
healthcare infrastructure includes a capable, well-distributed, and culturally competent
workforce; qualified institutional agencies such as private and public medical services, hospitals,
and emergency transport services capable of assessing and responding to public health needs;
and high performance and coordinated informational systems to support quality patient care and
clinical communication [279]. Availability, access, and quality of local clinical and public health
services can be limited in some communities, due to low population density, low rates of insured
patients, and limited public resources.
15.7.1 Providers
To assess the healthcare infrastructure of Allegany and Garrett counties, the team obtained
information regarding rates and ratios of primary care physicians, dentists, and mental health
providers to the population from the 2014 County Health Indicators Project, University of
Wisconsin. Allegany County has 44 primary care providers (at a rate of 58.9 per 100,000 and a
ratio of 70.7 to the population), 50 dentists (rate 61.0 and ratio of 68.3), and 82 mental health
providers (rate of 100.1 and ratio of 41.6). Garrett County has 15 primary care providers (rate of
49.9 and ratio or 83.5), 11 dentists (rate of 36.2, ratio of 115.0), and 12 mental health providers
(rate of 39.5, ratio of 105.5). These rates are, on average, much lower than the statewide
DRAFT Final Report
Page 163
averages, especially for mental health providers (rate 146.7, ratio 28.4), indicating a critical
shortage of providers in both Allegany and Garrett counties.
Table 15-13. Health Care Infrastructure
Area Primary Care
Providers (rate)
Mental Health
Providers (rate)
Dental Health
Providers (rate)
Allegany 58.9 100.1 61.0
Garrett 49.9 39.5 36.2
Maryland 88.2 146.7 67.9

Figure 15-32: Number of Dentists, 2012 and Primary Care Physicians, 2011 per 100,000
DRAFT Final Report
Page 164
15.7.1.1 Health Professional Shortage Area/Medically Underserved Area
According to HRSA, Health Professional Shortage Areas (HPSAs) are federal designations for
shortages of primary medical care, dental or mental health providers [280282]. These
designations may be geographic (a county or service area), demographic (low income
population) or institutional (comprehensive health center, federally qualified health center or
other public facility) [283]. Allegany County is a designated HPSA for primary care for low-
income populations, mental health care for Medical Assistance populations, and dental care for

Figure 15-33: HPSA Designations in Allegany and Garrett Counties, 2013
Medical Assistance populations [284]. Allegany County has a critical need for specialty
providers including vascular surgery, urology, as well as dentists willing to provide care for
adults with no insurance or Medical Assistance [285]. Garrett County is a designated HPSA for
primary and mental health care, and dental care for Medical Assistance populations.
DRAFT Final Report
Page 165

Figure 15-34: Uninsured Populations, 2011
Medically Underserved Areas (MUA) are locations designated by HRSA in which residents have
a shortage of personal health services. All of Garrett County is considered an MUA, while
substantial portions of Allegany County (Orleans, Lonaconing, Oldtown, and Cumberland) also
qualify as MUA [284].
15.7.2 Insurance Status
The team also obtained information on insurance status of individuals living in Garrett and
Allegany counties from the County Health Rankings Database. In 2011, there were an estimated
6,532 uninsured individuals living in Allegany County, approximately 11.9% of total population,
including 4% of children. In Garrett County, an estimated 3,473 individuals were uninsured,
approximately 14% of the total population. In the State of Maryland, an average of 12% of the
total population is uninsured, with most counties having between 8-16% of the total population
uninsured. Compared to the region (16.9%), the percent uninsured in Maryland and in the two
counties was lower; this could be due to the states health care exchange program.
DRAFT Final Report
Page 166
16 APPENDIX 2
Table 16-1: Health Effects Associated with Chemicals Used During UNGDP
Chemical Name CAS Number IARC EPA IRIS CAL EPA ATSDR Cancer
Effects
ATSDR Target Organs
1-methoxy-2-propanol 107-98-2 Alimentary
Acetaldehyde 75-07-0 2B B2 Respiratory,
Eyes

Acetone 67-64-1

Hematological,
Neurological
Acetophenone 98-86-2 D
Acrylamide 79-06-1 2A Likely Reasonably
anticipated to be
a human
carcinogen
Neurological,
Reproductive
Aluminum 7429-90-5 Neurotoxicity,
Immunotoxicity
None Musculoskeletal,
Neurological,
Respiratory
Aluminum chloride 1327-41-9 Musculoskeletal,
Neurological,
Respiratory
Aluminum oxide (alpha-
Alumina)
1344-28-1 Musculoskeletal,
Neurological,
Respiratory
Aluminum sulfate
hydrate
10043-01-3 Musculoskeletal,
Neurological,
Respiratory
DRAFT Final Report
Page 167
Effects
ATSDR Target Organs
Ammonia 7664-41-7 None Dermal, Ocular,
Respiratory
Ammonium acetate 631-61-8 D
Antimony pentoxide 1314-60-9 None Cardiovascular,
Respiratory
Antimony trichloride 10025-91-9 None Cardiovascular,
Respiratory
Barium sulfate 7727-43-7 D None None Cardiovascular,
Gastrointestinal,
Reproductive
Benzene 71-43-2 1 A Carcinogenicity
(leukemia)
Known to be a
human
carcinogen
Hematological,
Immunological,
Neurological
Benzene, C10-16, alkyl
derivatives
68648-87-3
Benzenemethanaminium 3844-45-9 3
Benzoic acid 65-85-0 D
Benzyl chloride 100-44-7 2A B2
Boric acid 10043-35-3 None None Cardiovascular,
Development
Boric oxide 1303-86-2 None Cardiovascular,
Development
Butanol 71-36-3 D
Coconut fatty acid 68603-42-9 2B
DRAFT Final Report
Page 168
Effects
ATSDR Target Organs
diethanolamide
Copper 7440-50-8 D Digestive None Gastrointestinal,
Hematological, Hepatic
Copper iodide 7681-65-4 None Endocrine
Copper sulfate 7758-98-7 None Gastrointestinal,
Hematological, Hepatic
Crotonaldehyde 123-73-9 C
Crystalline silica - quartz
(SiO2)
14808-60-7 1
Cumene 98-82-8 2B D
Cyclohexane 110-82-7
Cyclohexanone 108-94-1 3
d-Limonene 5989-27-5 3
Di (2-ethylhexyl)
phthalate
117-81-7 2B B2 Carcinogenicity Reasonably
anticipated to be
a human
carcinogen
Reproductive
Dibromoacetonitrile 3252-43-5 2B
Diesel 68334-30-5 None Dermal, Hepatic,
Neurological, Ocular,
Respiratory
Respiratory
DRAFT Final Report
Page 169
Effects
ATSDR Target Organs
Respiratory
Diethanolamine (2,2-
iminodiethanol)
111-42-2 2B
Dimethyl formamide 68-12-2 3
Ethanol (Ethyl alcohol) 64-17-5 #
Ethylbenzene 100-41-4 2B D Hepatoxicity None Developmental,
Neurological
Ethylene glycol (1,2-
ethanediol)
107-21-1 None Developmental, Renal
Ethylene glycol
monobutyl ether (2-
butoxyethanol)
111-76-2 3 Not likely
to be
carcinogeni
c to
humans
None Hematological, Hepatic
Ethylene oxide 75-21-8 1 Known to be a
human
carcinogen
Dermal, Developmental,
Renal
Formaldehyde 50-00-0 1 B1 Known to be a
human
carcinogen
Dermal, Gastrointestinal,
Immunological,
Respiratory
Furfural 98-01-1 3
Furfuryl alcohol 98-00-0
Hydrocarbon mixtures 8002-05-9 None Dermal, Hematological,
DRAFT Final Report
Page 170
Effects
ATSDR Target Organs
Neurological
Hydrogen chloride
(Hydrochloric acid)
7647-01-0 3 None
Hydrogen fluoride
(Hydrofluoric acid)
7664-39-3 None None
Hydrogen peroxide 7722-84-1 3
Hydrogen sulfide 7783-06-4 Neurological, Ocular,
Respiratory
Inorganic salt 7446-70-0 Musculoskeletal,
Neurological,
Respiratory
Iron oxide (Ferric oxide) 1309-37-1 3
Isopropanol (Isopropyl
alcohol, Propan-2-ol)
67-63-0 3
Kerosene 8008-20-6 Dermal, Hepatic,
Respiratory
Lead 7439-92-1 2B B2 Carcinogenicity
Developmental
neurotoxicity,
Cardiovascular
Reasonably
Anticipated to be
a Human
Carcinogen
Cardiovascular,
Developmental,
Gastrointestinal,
Hematological,
Musculoskeletal,
Renal, Reproductive
Magnesium silicate
hydrate (talc containing
14807-96-6 1
DRAFT Final Report
Page 171
Effects
ATSDR Target Organs
asbestiform fibers)
Medium aliphatic solvent
petroleum naphtha
64742-88-7
Morpholine 110-91-8 3
Mullite 1302-93-8
N-heptane 142-82-5 D
Naphthalene 91-20-3 2B C Reasonably
anticipated to by
a human
carcinogen
Hematological, Hepatic,
Respiratory
Nitrilotriacetic acid 139-13-9 2B
Nylon fibers 25038-54-4 3
Octyltrimethylammoniu
m bromide
57-09-0 Carcinogenicity
p-Xylene 106-42-3 Neurotoxicity None Developmental, Hepatic,
Neurological, Renal
Phenol 108-95-2 3 D Neurotoxicity None Dermal, Hematological
Phosphoric acid 7664-38-2 Respiratory
Potassium carbonate 584-08-7 Respiratory
Potassium iodide 7681-11-0 None Endocrine
Propylene oxide 75-56-9 2B B2 Reproductive/
Development,
Respiratory,
Eye irritation

DRAFT Final Report
Page 172
Effects
ATSDR Target Organs
Silica (crytalline) 7631-86-9 1 Respiratory
Sodium bromate 7789-38-0 B2
Sodium chlorite 7758-19-2 3 D Hematotoxicity,
Neurotoxicity
None Ocular, Respiratory
Sodium hydroxide
(Caustic soda)
1310-73-2 Eyes, Skin,
Respiratory
None None
Sodium tetraborate
decahydrate
1303-96-4 None Cardiovascular,
Developmental
Stabilized aqueous
chlorine dioxide
10049-04-4 D Hematotoxicity,
Neurotoxicity
None Dermal, Neurological,
Ocular, Respiratory
Stannous chloride
dihydrate
10025-69-1
Straight run middle
petroleum distillates
64741-44-2 None Dermal, Gastrointestinal,
Neurological,
Respiratory
Sulfuric acid 7664-93-9 1 Respiratory Known to be a
human
Carcinogen
Dermal, Respiratory
Thiourea 62-56-6 3
Toluene 108-88-3 3 Reproductive/
Development,
Headache,
Dizziness,
Sensory
irritation
None Cardiovascular,
Neurological
DRAFT Final Report
Page 173
Effects
ATSDR Target Organs
Xylene 1330-20-7 3 Central nervous
system
impairment,
Respiratory,
Eye irritation
None Developmental, Hepatic,
Neurological, Renal
Zinc chloride 7646-85-7 D None Gastrointestinal,
Hematological,
Respiratory
Zinc oxide 1314-13-2 D None Gastrointestinal,
Hematological,
Respiratory
#
Ethanol is Group 1 carcinogen through ingestion route.

Final Report 08 15 2014

Uploaded by

Copyright:

Available Formats

Final Report 08 15 2014

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Final Report 08 15 2014

Uploaded by

Copyright:

Available Formats

Potential Public Health Impacts of Natural

Gas Development and Production in the

You might also like