Cancer Incidence

Cancer Incidence

Cancer Incidence

Fryzek et al. (2013) conducted a retrospective assessment of the potential for an
association between childhood cancer incidence and HVHF in Pennsylvania, and
reported no increase in childhood cancers after HVHF commenced. Study limitations
included the insensitivity of the methods employed, the rarity of childhood cancers, and
the absence of adequate lag time between most HVHF activities and most of the study’s
childhood cancer diagnoses. These raise some uncertainty about the strength of the
study conclusions.

Non-peer-reviewed Information

In addition to investigating information in the peer-reviewed scientific literature, DOH
has maintained an ongoing effort to follow news reports and other non-peer-reviewed
sources for emerging information related to HVHF and potential public health impacts.10
Many findings reported through such non-peer-reviewed sources are from informal or
anecdotal health evaluations that have significant limitations such as self-selected
symptoms reports, non-specific symptoms, lack of exposure data, lack of baseline
health information, lack of unexposed comparison groups, and lack of controls for bias
and confounding. Reports of this sort cannot be used to draw conclusions about
associations between reported health symptoms or complaints and any specific
potential environmental exposure source such as HVHF shale-gas development.
However, these types of reports suggest hypotheses for associations between health
outcomes and shale-gas activities that could be tested with proper environmental
epidemiology methods.

HVHF Environmental Studies

Studies investigating HVHF impacts on environmental media such as air or water were
included in the review if they provided information about the potential for human
exposures from HVHF activity.
Air Quality Impacts
Maintaining good air quality is obviously vital for promoting public health; poor air quality
can affect large populations of people, and therefore can contribute to significant
morbidity and mortality. DOH programs promote clean outdoor air quality by developing
health comparison values for use by DEC and by investigating and helping to correct
conditions that contribute to poor indoor air quality. NYS was the first state in the

country to establish indoor smoking prohibitions in public spaces under the NYS Clean
Indoor Air Act.
The National Institute for Occupational Safety and Health (NIOSH) has assessed
potential risks to workers associated with chemical exposure at natural gas drilling sites
(NIOSH, 2012). In field studies conducted at 11 sites, respirable crystalline silica and
diesel particulates were measured at levels with the potential to pose health hazards.
NIOSH has proposed several controls and recommended proper use of personal
protective equipment to minimize exposures. NIOSH has also reported that the
occupational fatality rate among oil and gas industry workers is seven times higher than
the average rate for all US industries (Retzer, 2011). On August 23, 2013, the federal
Occupational Safety and Health Administration (OSHA) announced that it intended to
propose a revised standard (called a permissible exposure limit) to protect workers from
exposure to respirable crystalline silica.11 OSHA's Notice of Proposed Rulemaking for
Occupational Exposure to Respirable Crystalline Silica was published in the Federal
Register on September 12, 2013.12 If enacted, the new regulation would reduce the
permissible exposure limit for crystalline silica and would establish certain other
requirements related to measuring levels of silica in workplace air, controlling dust,
providing respiratory protection, training of workers, and offering medical exams. While
the NIOSH assessment focused on worksite air quality, this report is suggestive that
uncontrolled silica emissions could affect the air quality of residences or businesses
near well pads.
In 2010, the Texas Department of State Health Services collected blood and urine
samples from 28 people, living in and near the town of Dish, to determine whether
people there had higher levels of volatile organic compounds (VOCs) in their blood than
95% of the general United States (U.S.) population. Community residents had raised

concerns that they were experiencing exposure to air contaminants from nearby gas
wells and compressor stations. Measuring the presence of chemicals in biological fluids
(i.e., biomonitoring) is a technique that can demonstrate that exposure occurred to
those chemicals, but does not necessarily identify the source of the exposure, or when
exposure occurred. Based on the pattern of VOC values found in the samples, the
information obtained from this investigation did not provide evidence that communitywide
exposures from gas wells or compressor stations were occurring in the sample
population. Other sources of exposure such as cigarette smoking, disinfectant
byproducts in drinking water and consumer or occupational/hobby related products
could explain many of the findings.
In 2010, the Colorado Department of Public Health and Environment released a public
health consultation evaluating the potential public health hazards of ambient air pollution
in areas of Garfield County in close proximity to oil and natural gas development
activities. This report summarized results from enhanced air quality monitoring
implemented following a 2008 public health consultation13 which found air
concentrations near the upper end of EPA’s acceptable range for benzene-associated
cancer risk at one monitoring site. In this study, air monitoring was used to measure
concentrations of chemical contaminants in the air near HVHF activities, and then those
measured levels were compared to health-based comparison values for the chemicals.
Health comparison values are a risk-assessment tool and are set at levels to be
protective of public health. If comparison values are exceeded, it does not imply that
adverse health impacts will occur, but it indicates that further investigation of potential
exposures is warranted.
In the 2010 report, the investigators concluded that it could not be determined if
breathing ambient air in those areas of Garfield County that were monitored could harm

people’s health. This conclusion was reached because the cancer risks and noncancer
hazards for 65 out of 86 contaminants could not be quantitatively estimated due to the
unavailability of chronic inhalation toxicity values. Although the evaluation suggests that
exposures are not likely to result in significant cancer and noncancer effects (the levels
measured are much lower than those known to cause health effects), cumulative health
effects from synergistic interactions are unknown. Where quantitative evaluations were
possible, increased risks of cancer, long-term (chronic) noncancer hazards and shortterm
(acute) noncancer hazards (where data were available) were low, although for the
latter there is uncertainty because insufficient data are available to evaluate intermittent
short-term peak exposures.
A similar risk-assessment study of air-quality monitoring in the Barnett Shale region of
Texas was published in 2014 by Bunch et al. (2014). The study summarized airmonitoring
data for volatile organic chemicals collected at six fixed monitoring locations
in Wise, Denton and Tarrant counties in north-central Texas including areas in and
around the city of Fort Worth. The monitoring network is operated by the Texas
Commission on Environmental Quality (CEQ) and is described in the report as the most
extensive air monitoring network in place in any U.S. shale play. The network includes
both real-time monitors and 24-hour average samples analyzed in the laboratory,
covers regions of the Barnett shale producing both dry and wet gas, and spans areas of
urban and suburban development where the potential for community exposure to any
shale-gas air emissions could be significant. The analysis of these data included
assessing potential health risks of short-term and long-term exposure to all chemicals
measured by the monitoring network using existing health comparison values (for
example, Texas CEQ air monitoring comparison values or US EPA reference
concentrations). Many of the chemicals measured by the existing network are unrelated
to shale-gas development. Therefore, the authors also conducted more refined

quantitative risk assessments for a subset of volatile organic chemicals thought to be
most likely to be associated with shale gas production.
The Bunch et al. study summarized the results of over 4.6 million data points collected
over more than 10 years for up to 105 different volatile organic chemicals per monitor.
Only one observed short-term value exceeded an applicable odor-based comparison
value.14 None of the measured short-term (one hour or 24-hour average) air levels for
the entire panel of chemicals exceeded an applicable short-term health-based
comparison value. Only one chemical (1,2-dibromoethane) had any annual average
concentrations that exceeded its applicable long-term health comparison value.15 The
authors noted that the analytical detection limit for 1,2-dibromoethane is substantially
higher than its chronic comparison value and about 90% of the 1,2-dibromoethane
results that contributed to the exceedances were non-detects. This suggests the true
annual average concentrations could have been substantially lower than the reported
estimates. The authors also did not consider 1,2-dibromoethane to be a chemical
reasonably expected to be associated with shale-gas production. According to the
authors, it is used as a lead-scavenger in aviation fuel. The two monitoring locations
where the 1,2-dibromoethane 2011 annual averages exceeded applicable comparison
values are located near airports. More refined deterministic and probabilistic quantitative
risk assessments for annual average concentrations found that estimates of cumulative
noncancer and cancer health risks were below levels of concern at all monitoring
locations. The authors concluded that their analysis demonstrated that shale gas
operations in the monitored region of the Barnett play have not resulted in communitywide
exposures to the measured volatile organic chemicals at levels that would pose a
health concern.

Macey et al. (2014) analyzed data from grab and passive air samples that were
collected in Arkansas, Colorado, Ohio, Pennsylvania and Wyoming by trained
volunteers at locations identified through systematic observation of industrial operations
and air impacts over the course of residents’ daily routines. The investigators reported
that concentrations of eight volatile chemicals exceeded risk-based comparison values
under several operational circumstances. Benzene, formaldehyde, and hydrogen sulfide
were the most common compounds to exceed acute and other risk-based values.
However, it was not always clear that the authors employed appropriate risk-based
comparison values given the nature of the samples that were collected. For example,
the use of comparison values based on lifetime (long-term) cancer risk levels may have
substantially overstated cancer risks associated with exposures to short-term levels of
air pollutants that were measured. Moreover, retrospective source apportionment efforts
are not possible based on study data because the investigators did not collect the
necessary control samples, such as upwind air samples, or wind direction data. This
complicates evaluation of the study data because, at least in some urban and industrial
settings, it is not unusual for atmospheric concentrations of benzene and formaldehyde
to exceed some of the comparison values that were employed by the authors (Weisel,
2010).
The Pennsylvania Department of Environmental Protection (PA DEP) conducted shortterm,
screening-level air quality sampling initiatives in various parts of the
Commonwealth where a majority of the Marcellus Shale operations have been
undertaken.16 Sampling windows often captured pollutant concentrations during the
early morning hours and late evening hours, to reflect the predominate times when
complaints related to Marcellus gas exploration activities are received by the DEP.
Following the completion of a comparative analysis, which will consider data from
separate surveys conducted in four Pennsylvania regions, the DEP will determine
whether additional, longer-term sampling is warranted.