Purpose of Review
Epidemiologic studies have observed elevated health risks in populations living near unconventional oil and gas development (UOGD). In this narrative review, we discuss strengths and limitations of UOG exposure assessment approaches used in or available for epidemiologic studies, emphasizing studies of children’s health outcomes.
Exposure assessment challenges include (1) numerous potential stressors with distinct spatiotemporal patterns, (2) critical exposure windows that cover long periods and occur in the past, and (3) limited existing monitoring data coupled with the resource-intensiveness of collecting new exposure measurements to capture spatiotemporal variation. All epidemiologic studies used proximity-based models for exposure assessment as opposed to surveys, biomonitoring, or environmental measurements. Nearly all studies used aggregate (rather than pathway-specific) models, which are useful surrogates for the complex mix of potential hazards.
Simple and less-specific exposure assessment approaches have benefits in terms of scalability, interpretability, and relevance to specific policy initiatives such as set-back distances. More detailed and specific models and metrics, including dispersion methods and stressor-specific models, could reduce exposure misclassification, illuminate underlying exposure pathways, and inform emission control and exposure mitigation strategies. While less practical in a large population, collection of multi-media environmental and biological exposure measurements would be feasible in cohort subsets. Such assessments are well-suited to provide insights into the presence and magnitude of exposures to UOG-related stressors in relation to spatial surrogates and to better elucidate the plausibility of observed effects in both children and adults.
Unconventional oil and gas development (UOGD) involves the production of oil and natural gas from deep geologic formations such as shale using horizontal drilling and high-volume hydraulic fracturing. Shale gas production in the USA increased more than eightfold over the past decade, from 3,110 to 25,226 billion cubic feet from 2009 to 2019 . In 2019, there were approximately 150,000 active UOG wells in the USA , with an estimated 1.5 to 4 million people living within 1.6 km of a UOG well . Considerable community concerns and public debate have accompanied the rapid growth of the UOG industry regarding the multitude of potential human health hazards, including air pollution, noise, odors, water contamination, radioactive releases, seismic activity, traffic, and psychosocial stress [4, 5]. Potential benefits of UOGD include increased employment opportunities and wages, increased income for landowners who lease their land, lower energy prices, and domestic energy independence . However, benefits are not distributed equally, may be short-term, and should be considered in the context of local and global environmental health impacts from fossil fuel extraction .
Epidemiologic studies have examined the human health impacts of UOGD, with the majority focused on children’s health . While the bulk of these studies provide evidence of increased risk of adverse health outcomes in children (e.g., low birth weight, preterm birth, congenital anomalies, childhood asthma hospitalizations, and childhood cancer), studies have not been entirely consistent across endpoints. Limitations in exposure assessment methods can contribute to uncertainty about associations, and heterogeneity in methods can lead to different conclusions. In this narrative review, we (i) discuss the process and hazards of UOGD, (ii) describe the strengths, limitations, and challenges of environmental exposure assessment approaches relevant for UOGD health studies, and (iii) review the application of such methods in the epidemiologic literature, with a focus on studies of pediatric populations.
Process of Unconventional Oil and Gas Development (UOGD)
UOGD, considered upstream oil and gas activity (versus downstream transport and refinement processes), is a complex process with multiple phases that vary over time and location . The process starts with application and approval of the necessary permits. Next, the UOG well pad is prepared, which involves building access roads and clearing land where infrastructure for one or more UOG wells is stationed. Next is the drilling phase in which a borehole is drilled vertically approximately 1.6 to 3.2 km (1–2 mi) below ground, before the trajectory of the drill bit is turned to bore horizontally through the hydrocarbon-bearing rock layers. Steel pipes (casing) and cement are installed to maintain the integrity of the borehole and to isolate the upper portions of the borehole from the surrounding aquifer. Hydraulic fracturing then occurs, during which millions of gallons of water, chemical additives, and sand are injected into the well under high pressure and expelled through perforations in the casing to create fractures in the rock. When the pressure is released, the oil and/or gas, together with a portion of the injected fracturing fluids (flowback), move from the newly created fractures into the well and up to the surface for collection . The production phase begins as oil and gas that flows up the well is separated at the surface from formation fluids (produced water) and transmitted through gathering pipelines from the wellhead to a storage facility or a processing plant. The approximately 7 to 14 million gallons of wastewater (flowback and produced water) generated during the operating lifetime of an UOG well  may be discharged to surface water following treatment, injected in class II injection wells, reused, or disposed of in unlined percolation pits, practices that vary by state . Solid waste (e.g., drill cuttings) is commonly disposed of in municipal landfills . At the end its production life, the well should be permanently sealed, although some wells are kept idle or abandoned and become increasingly likely to lose integrity as they age .
Chemical Contaminants in Water
With more than 9 million people in the USA relying on drinking-water sources located within 1.6 km (1 mile) from a UOG well , water contamination from UOGD has been a major community concern [15, 16]. Hundreds of chemicals have been reportedly used in injection water or detected in wastewater, including known and suspected endocrine disruptors and carcinogens, such as metals, volatile organic compounds, polycyclic aromatic hydrocarbons, phthalates, and per- and poly-fluoroalkyl substances (PFAS) [9, 17,18,19,20,21,22,23]. Water contamination may occur due to surface spills of fracturing or wastewater fluids, release of improperly treated wastewater, structural failures, and well leaks [9, 12, 24,25,26,27,28,29]. Groundwater monitoring studies conducted thus far do not support widespread contamination [30,31,32,33]. However, groundwater and surface water impairments, spills, and violations have been documented across multiple states [15, 24, 26, 34,35,36,37,38].
UOGD generates air pollutants from various sources, including well and road construction; use of diesel-powered construction, drilling, and transportation equipment and vehicles; dust generation during drilling; intentional flaring of natural gas; and volatilization of wastewater constituents [39,40,41]. Emission sources have different temporal profiles and may be continuous over varying time frames (e.g., diesel equipment, leaks) or be intermittent (e.g., flaring, venting). They may also occur at the well pad or off-site from transportation or associated UOG infrastructure . Studies of airborne emissions or concentrations have identified several pollutants associated with UOGD activities, including carcinogens (e.g., diesel particulate matter, polycyclic aromatic hydrocarbons), and reproductive and/or developmental toxicants (e.g., ethylbenzene, toluene) [18, 39, 43,44,45,46,47].
Radiation, Radioactivity, and Radon
Technologically enhanced naturally occurring radioactive compounds (TENORMS) have been detected in a variety of wastes generated by UOGD, including produced water [48,49,50], lateral drill cuttings , and impoundment sediments , and co-occur with sludges and mineral scales that accumulate inside UOGD equipment . TENORMS have also been found in streambed sediments next to facilities managing UOGD wastewaters . Additionally, higher indoor radon levels were observed in homes with more or closer UOG wells compared to those with fewer or farther UOG wells [53, 54]. Health effects of radiation exposure include adult and childhood cancers , impaired lung function , increased blood pressure , and oxidative stress .
Sensory stressors include noise, artificial light at night, and odors [59,60,61,62]. Noise pollution can activate the sympathetic nervous system; lead to annoyance, stress, and sleep disturbance; and potentially contribute to cardiovascular disease and adverse birth outcomes [62, 63]. Odorous compounds can reflect mixtures of volatile organic or sulfuric compounds, which have neurotoxic and respiratory effects, and can also trigger annoyance, concern, and anxiety at levels below established toxicity thresholds [64, 65].
Social and environmental disturbances include, but are not limited to, deterioration of roadway infrastructure , increased traffic accidents , visible changes to the ecological landscape , and earthquakes . Social impacts such as disruption of community cohesion , increased crime [71, 72], and decreased property values [73, 74] also threaten communities near UOGD.
Methane, a potent greenhouse gas, is emitted to the air throughout UOGD from leaks and intentional releases (e.g., venting) . Although methane is not directly toxic to nearby communities, except at very high levels in tap water, at which it poses a flammability danger, its indirect effects on future health risks through its role in climate change is an important threat to children’s health. This aspect is not generally considered in the current UOGD-related epidemiologic literature, which tends to be retrospective.
Exposure Assessment Methods and Review of Epidemiologic Studies
We review the features, strengths, and limitations of five categories of methods for assessing exposure to UOGD-related stressors : surveys, environmental measurements, aggregate proximity-based models, pathway-specific models, and biological monitoring (Table 1, Fig. 1). We identified published epidemiologic studies designed to assess an exposure–response relationship between UOGD (or both conventional and UOGD) and human health conducted in the USA and Canada through searches in PubMed and Google Scholar using keywords related to oil and gas and health outcomes and reference lists of identified studies, with the last search conducted on January 31, 2022. While of public health value, we excluded small community surveys lacking clear comparison groups and studies of physical outcomes such as injuries or traffic accidents. Our search yielded 42 epidemiologic studies, of which 29 studies included a pediatric population or results related to children’s health (Table 2). We describe the exposure assessment methods used in each publication (Table 2).
Surveys are used to gather self-reported, individual-level information on a wide variety of exposure-related factors, such as participant behavior, activities, observations, and perception of UOGD. None of the epidemiologic studies identified in this review utilized this method (Table 2).
Surveys are comparatively inexpensive tools that can be administered to many individuals across a vast geographic scale and allow for acquisition of current and historical information. Surveys have been useful for assessing perceived exposures such as the felt experience of earthquakes, which may be as or more relevant for health outcomes as compared to objective measures of seismicity. For example, the US Geological Survey (USGS) Did You Feel It (DYFI) is a crowd-sourced database where individuals submit an online report to the USGS about the time, location, and felt impact of an earthquake . USGS uses these reports to estimate a community decimal intensity for each seismic event at the ZIP code level. In addition, participant self-reported housing damage has shown strong correlation with cumulative ground motion assessed from shakemaps (i.e., visual displays of ground shaking occurring at different locations immediately after an earthquake). In terms of UOGD, residents have also reported on observable changes to their air or water, such as odors, change in the color, appearance, turbidity, or foaming of their drinking water [16, 30, 78].
Limitations and Challenges
The utility of exposure assessment surveys is limited to hazards that are directly knowable and observable by residents and therefore are not suitable for most UOGD-related chemicals and radioactive materials, which are odorless, colorless, and not detectable by human senses. Moreover, even observable changes (e.g., color change, foaming, turbidity, odor) cannot readily be translated into quantitative exposure estimates for specific agents.
Models and Metrics
Human exposure models are mathematical expressions of attributes of the environment, community, or individual that are expected to influence exposure. These can take several forms, such as geostatistical models that map proximity to a contaminant source or physically based models that draw upon established principles governing how contaminants move through various media (e.g., air, water, soil) to predict the transport of a stressor from source to receptor. Here, we group models into aggregate proximity-based models, which treat UOGD as a collective entity, and pathway-specific models which focus on a particular source or exposure pathway related to UOGD. All epidemiologic studies published to date have used models to assess exposure to UOGD, with the majority using aggregate approaches (Table 2).
Aggregate Proximity-Based Models and Metrics
Counts or Geographic Aggregation
These metrics consider presence or number of UOGD facilities per ZIP code, county, or circular buffer around a residence, used in 21 epidemiologic studies to date (Table 2). This metric assumes that residents of geographic units with higher numbers of UOGD facilities experience higher levels of exposure. Because environmental exposures do not respect administrative boundaries, use of these types of spatial units is subject to edge effects (i.e., wells located just across a Census tract border). These metrics have recently been extended to estimate counts of UOG wells around other receptors, such as drinking water sources .
The mapped Euclidean distance to the UOG well nearest to a home or other receptor is another simple metric that assumes that homes closer to UOG wells are more likely to be exposed or face higher levels of exposure. This calculation is based upon a single well—the closest well—and does not incorporate the density of wells in the area surrounding a residence.
Inverse-distance or inverse distance-squared weighted UOG well counts improve upon simple distance measures by accounting for the number of UOG wells within a buffer zone (generally 1, 2, 5, or 10 km) while weighting the closer wells more heavily than the distant wells by assuming a decline in exposure potential as an inverse (Eq. 1) or inverse-squared (Eq. 2) function of distance.
where d is the distance between the ith UOG well and a residence, and n the number of UOG wells.
These models assume all wells have the same maximum exposure potential regardless of production volume, wastewater produced, well phase, topography, or meteorology. Distance-weighted metrics have been commonly used in epidemiologic analyses (n = 9 studies, Table 2).
Activity-based models extend distance-based metric by also including attributes of UOG wells expected to influence exposure potential, such as vertical well depth or production volume [41, 80, 81•]. Nine health studies to date have used some form of an activity-based metric (Table 2).
Aggregate proximity-based models offer the advantage of being scalable (i.e., feasible to apply to large geographic areas or populations) and thus practicable for assignment of exposure to thousands of wells across the thousands of subjects sometimes included in health studies. Proximity-based spatial surrogates offer the opportunity to assess exposures retrospectively, which is important if the etiologically relevant time windows occur in the past. They can also be calculated to align windows of health vulnerability (e.g., trimesters) with the phases of UOGD to create phase-specific metrics (e.g., UOG exposure during production). These temporal considerations are of particular importance for adverse birth outcomes and childhood cancer, because effect estimates can vary based on the timing of exposure [82, 83]. Conversely, other studies of adverse birth outcomes have observed strong correlations in exposure assessments across trimesters, making it difficult to examine potential differences [84•, 85].
These approaches also serve as an aggregate measure of the myriad of physical, chemical, and social stressors potentially associated with UOGD. This aggregation is a useful feature for this complex industrial process, where exposures to multiple hazards are likely and the dominant stressor is not known and may differ from well to well. Additionally, the linkage of health data or risks with relatively simple metrics like distance between home and UOG well is directly relevant and actionable for policy makers and stakeholders; for example, such data can be used to inform the establishment of setback distances, the minimum allowable distance between an oil and gas well and sensitive human receptors (e.g., residences, schools, or hospitals).
Limitations and Challenges
Proximity-based spatial surrogates do not allow for identification of etiologic agents underlying observed associations . Additionally, the effects of processes affecting chemical fate and transport, such as advection, dispersion, and transformation reactions that alter contaminant toxicity, are unlikely to be approximated by simple spatial metrics. Moreover, the accuracy of the underlying assumption that UOG-related exposures diminish predictably with distance likely varies by exposure type. For example, the impacts of increased traffic, ecological disturbance, or formation of secondary air pollutants may not be appropriately captured by a model that presumes a linear attenuation of exposure intensity with increasing distance. Approaches that are based on buffers (e.g., 1 km radius around a residence) are limited by the fact that the actual exposures do not follow perfect circular patterns. Because these metrics have only been compared to environmental measurements in a small number of studies (e.g., 30, 41, 94), our understanding of their ability to capture specific etiologic agents is still unfolding.
The epidemiologic studies conducted to date have predominantly relied on secondary data sources, such as birth, hospitalization, or electronic medical records, without participant contact. This has necessitated the use of single address (e.g., maternal residence at birth) to assign exposures, which could introduce exposure misclassification due to unaccounted residential mobility [87, 88]. While residential mobility is an important consideration in studies of children’s health, risk estimates have generally been robust to the impact of residential mobility on exposure assessments [89, 90]. The potential to introduce error is compounded if evaluating diseases with longer exposure windows or latency periods (e.g., adult cancer).
Pathway-Specific Proximity-Based Models and Metrics
In contrast to aggregate models, approaches have been developed to capture specific stressors such as air emissions , flaring , drinking water vulnerability , radioactivity [93•], earthquakes [94, 95], and ancillary UOG infrastructure . The strengths and limitations for these approaches are discussed below. To date, five epidemiologic studies have applied a pathway-specific exposure model (Table 2).
Air Pathway Models
An activity model included weighting factors based on air emissions data that accounts for changes in emissions between and within phases of well development was constructed for Colorado  and subsequently applied in an epidemiologic study of congenital malformations and adult cardiovascular endpoints [96•, 97]. Limitations include the requirement of specific UOG data (e.g., production volumes and number of tanks per well pad) that may not be readily available in all states. Application to other regions would require assumptions that the emissions are correlated with the measured factors and that these are similar across producers and formations. In addition, while targeting the air pathway, it does not provide information on specific pollutants of concern.
Li et al. (2020) constructed a downwind proximity-based model and compared it to radioactivity measurements available from a national network of environmental radiation monitors [93•]. They created circular sectional buffer around each residence. To isolate emissions from upwind industrial activities, they sliced a 90° circular sector centered on the daily upwind direction for each circular buffer and summed the number of UOGD wells in each wedge-shaped buffer based on daily wind fields. They found an association between UOGD wells and downwind radiation measurements. This downwind model was applied in one health study of all-cause mortality among Medicare beneficiaries, which observed significant higher mortality risk associated with living downwind of UOGD wells .
A flaring-specific model was developed for Texas using data from the Visible Infrared Imaging Spectroradiometer (VIIRS) Nightfire satellite instrument (which detects thermal sources) alongside reported estimates of vented and flared gas from the Railroad Commission of Texas . The model was used to identify flaring events and to estimate the gas volume of each flare in a study of birth outcomes in Texas, which found that exposure to a higher number of flaring events was associated with increased risk of preterm birth [84•]. This model provides high resolution estimates of flared locations and volumes using publicly available VIIRS data, which covers the entire planet (albeit asynchronously). The model may overestimate the number of flares in a region due to artifacts of VIIRS . Willis et al. (2020) used flaring volumes by zip code data to classify exposures (among other metrics) in a study of pediatric asthma hospitalizations in Texas . While there was an association between oil and gas well density and increased hospitalizations, the results specific to flaring volumes were inconclusive.
Metrics for Injection-Induced Earthquakes
Exposure to earthquakes can be based on location of the epicenter, timing of the earthquake, and magnitude of the seismic event. Generally, earthquakes greater than (or equal to) magnitude 4 have been considered relevant for human exposure and health. These have been used to assess short-term exposure windows for acute health outcomes, such as 1 month prior to health events , and longer-term windows such as summing cumulative peak ground acceleration from shakemap data over a 6-year period . Prior work has connected earthquakes to adverse perinatal and child health [100,101,102,103], but no studies have evaluated the role of UOGD-induced earthquakes and adverse birth outcomes.
Ancillary UOGD Infrastructure
Few studies have considered UOGD infrastructure beyond the well pad. Koehler et al. (2018) incorporated compressor station locations into an inverse-distance squared exposure metric that also included pad development, drilling, hydraulic fracturing, and gas production . This metric performed similarly to a previously used inverse distance squared metric that did not incorporate compressor stations in evaluations of respiratory outcomes . The authors hypothesized this occurred because compressor stations were co-located in space and time with the other phases of UOGD activity measured.
Hydrological Models and Metrics
Physics-based estimates of water-well vulnerability (i.e., the likelihood of contamination) that draw on hydrological principles to simulate groundwater flow rates and patterns can improve exposure assignments in human-health studies focusing on the drinking-water pathway. Soriano et al. (2020) used hydrologic models to simulate the capture zones of residential drinking-water wells and estimated vulnerability based on the probability that a capture zone encompassed UOG infrastructure (i.e., well pads) . Physics-based vulnerability estimates can also predict the likelihood of future exposures, thereby enabling households of greatest risk to be targeted for prospective monitoring and preventative actions to minimize exposure to UOG contaminants. Physics-based models can be computationally intensive and are challenging to scale to large geographic areas required of many population-based epidemiologic studies. To circumvent this challenge, Soriano et al. (2021) used vulnerability estimates of a physics-based model of groundwater flow and chemical transport to train a machine-learning model that classified the vulnerability of household wells on the basis of predictors readily computable from a geographic information system . A predictor combining information on topography, hydrology, and proximity to contaminant sources (inverse distance to nearest upgradient UOG source) was found to be highly important for accurate machine-learning model predictions of vulnerability. This new metric was recently applied as an exposure indicator in a groundwater monitoring study by Clark et al. (2022) [106•] and has not yet been applied in any published epidemiologic studies.
Pathway-specific models may offer improved accuracy over simple proximity models due to their incorporation of chemical transport, source characteristics, land use, meteorology, and other factors. This can reduce exposure misclassification in epidemiologic studies and help disentangle the effects of different exposure routes or hazards, providing important information for additional monitoring studies or exposure mitigation strategies.
Limitations and Challenges
Pathway-specific models may be computationally intensive and require substantial information for parameterization that may be difficult to acquire. The increased complexity can make the models less readily scalable to large geographical areas, although recent advances in computational and statistical approaches are helping to circumvent this issue. In addition, more complex models may be less readily interpretable to the public and policy makers. If the models predict exposures at a single residential address, they are prone to the same error as the simple proximity models in terms of residential mobility.
Environmental measurements are quantitative measurements of samples collected from various media (e.g., air, water, soil, and dust) that can be used to calculate exposure. Environmental measurements can be collected directly at the point of contact (e.g., a personal air monitor worn by an individual) or from area or ambient settings (e.g., a stationary air monitor in a home). No large-scale epidemiologic analyses have used environmental measurements to assess UOG exposure to date (Table 2).
Measurement of targeted analytes often has standard methods and can be robustly quantitative. Repeated measurements could help quantify duration and temporal variations in exposures. Measurements can be incorporated into models as weighting factors or can be used to evaluate or calibrate models. Measurements can be used to establish exposure pathways or assess the dominant pathways, demonstrating whether agents reportedly used in UOG development are reaching human receptors and by which mechanisms. Environmental measurements can be highly relevant to individual exposure and can be reported to stakeholders and residents. These can also be compared to health-based standards and guidelines where available. Untargeted analyses can reveal the presence of unexpected contaminants.
Limitations and Challenges
Measurements (particularly those with sufficient spatiotemporal resolution to be informative at an individual level) may not be feasible or may be prohibitively costly in a large-scale epidemiologic study. Furthermore, monitoring networks do not exist for many relevant environmental stressors. Given the lack of timely and publicly available reporting of UOG-related incidents (e.g., spills, leaks) and the resource-intensive nature of environmental sampling, it is difficult to coordinate sampling efforts with environmental releases. As a result, if monitoring is conducted at only a single time point, it is unlikely to represent the full range of past exposures or peak exposure at that location. This issue is particularly salient for groundwater measurements given the lack of monitoring and regulatory oversight of domestic groundwater wells [104, 107] which may be shallow and made even more vulnerable to contamination as their casing and cement seals deteriorate with age. Monitoring is further complicated by the challenging and non-random access to such groundwater (i.e., permission must be granted by a property owner) [105, 108]. For both air and water, it is impossible to measure everything of interest because there are numerous potential pollutants of public health concern [17, 18]. Environmental sampling also requires carefully designed protocols, participant cooperation and consent, quality control procedures, and appropriate analytical capabilities and materials to properly collect and quantify target analytes.
Identification of target analytes for measurement is challenging because constituents of fracturing fluids are reported in an inconsistent and voluntary manner [23, 109, 110]. Naturally occurring compounds mobilized by the UOG process and transformation products that are not used deliberately may evade targeting for analysis [111, 112]. There are also limited data for comparison, although there are efforts to build publicly available datasets of detections to complement disclosure databases . In terms of water measurements, publicly available pre-drill data are sparse [33, 114], and many contaminants of emerging concern do not have health standards or guidelines. As such, it is difficult to identify exposures that are “elevated” or present a definitive health risk. For air pollution, there are fewer federal monitoring stations in rural areas that tend to host UOGD compared to heavily populated metropolitan areas, and a small number of atmospheric pollutants are systematically measured.
Finally, because true exposure is a combination of environmental concentrations and human behaviors and activities, collected measurements may not be representative of actual human exposure or reflect heterogeneity in exposures across persons. Individuals are rarely stationary, and their exposure profile will be highly influenced by the different microenvironments where they spend time as well as their behavior and activities in those microenvironments. Direct personal air sampling devices can overcome this issue, or surveys can complement measurements with individual-level information about behaviors and activities. For example, with respect to water-based exposures, people may consume water or other beverages from multiple locations or may have treatment systems that alter the water chemistry. A study of 255 homes in Pennsylvania and Ohio located in areas with UOGD found that more than 20% of their participants reported drinking primarily bottled water, which could come from many sources, instead of water from their private well [106•].
Biological Monitoring of Exposure
Biological monitoring is a direct measure of chemical exposure that integrates all potential routes of exposure into one measurement of internal dose. No large-scale epidemiologic studies have used biological monitoring to assess UOG exposure (Table 2), and to our knowledge, only two pilot studies have measured exposure biomarkers among individuals living in a community with UOG activity [115, 116]. These studies observed elevated levels of benzene metabolites and manganese in urine and manganese, barium, aluminum, and strontium in the hair of pregnant women living atop the Montney Formation in Northeastern British Columbia, Canada, as compared to reference populations in both Canada and France. Individual measurements were not compared to proximity to or density of UOG activity, meaning it is possible the population was exposed to other sources of VOCs and trace metals.
Biological measurements provide an integrated and direct measure of total body burden that reflect multiple exposure pathways and routes and the different activities and microenvironments of individuals. Biological monitoring may also be useful in identifying etiologic agents associated with health outcomes.
Limitations and Challenges
The broad range of compounds associated with UOGD complicates targeted biomonitoring. Because UOG chemicals typically have numerous indoor and outdoor sources, biological monitoring cannot provide insights into the dominant source or route of exposure. Many biomarkers have short half-lives and therefore new measurements may not reflect prior exposures during etiologically relevant time windows. Finally, to collect biosamples, researchers must obtain participant consent, adhere to ethical guidelines and protocols, and train sample collectors in the necessary methodology (e.g., phlebotomy), all of which raise cost and prevent barriers to broad, systematic study. Although children often have a higher body burden of environmental chemicals compared to adults, collection of biospecimens may be particularly challenging or not possible among pediatric populations, for reasons of practicality (e.g., compliance, ability to achieve sufficient sample volume) and ethics (e.g., the invasive nature, obtaining informed assent) [117, 118].
In this review, we describe the features of UOG exposure assessment methods, which are summarized in Fig. 1. We also enumerate the many challenges facing UOGD exposure assessment, including limited information on the identity and intensity of chemical and non-chemical stressors, sparse monitoring data in areas with active UOGD, the episodic or infrequent release of chemicals or other hazards, and the complexity of the different stressors with regard to temporal, spatial, release, and dispersion patterns. Despite these issues, published epidemiologic studies generally relying on relatively simple, aggregate proximity-based models, have consistently identified increased risk of health problems, particularly in children. Improved methods could potentially reduce uncertainty and provide evidence for specific exposure pathways. While the substantial analytic capabilities required for large-scale environmental measurements may make them impractical for use in a traditional epidemiologic study, such studies are highly valuable and could shed light on pathways of exposure, particularly in a cohort subset. Not all stressors lend themselves to the approaches discussed in this review, such as estimating the health impacts of greenhouse gases (e.g., methane), which may not immediately or directly affect health but will hold global consequences for future generations. As epidemiologic investigation of UOGD is still a relatively nascent research area, UOGD exposure assessment methods have already progressed substantially and will continue to be refined as new information comes to light and researchers tackle these challenges.
Papers of particular interest, published recently, have been highlighted as: • Of importance
US Energy Information Administration. U.S. annual estimated shale gas production. Available: https://www.eia.gov/dnav/ng/NG_ENR_SHALEGAS_A_EPG0_R5302_BCF_A.htm. Accessed August 2021.
US Energy Information Administration. U.S. oil and natural gas wells by production rate. 2020. Viewable at: https://www.eia.gov/petroleum/wells/. Accessed August 2021.
Czolowski ED, Santoro RL, Srebotnjak T, et al. Toward consistent methodology to quantify populations in proximity to oil and gas development: a national spatial analysis and review. Environ Health Perspect. 2017;125(8):086004.
Hays J, Shonkoff SB. Toward an understanding of the environmental and public health impacts of unconventional natural gas development: a categorical assessment of the peer-reviewed scientific literature, 2009–2015. PLoS One. 2016;11(4):e0154164.
Adgate JL, Goldstein BD, McKenzie LM. Potential public health hazards, exposures and health effects from unconventional natural gas development. Environ Sci Technol. 2014;48(15):8307–20.
Sovacool BK. Cornucopia or curse? Reviewing the costs and benefits of shale gas hydraulic fracturing (fracking). Renew Sustain Energy Rev. 2014;37:249–64.
McKenzie LM, Allshouse WB, Burke T, et al. Population size, growth, and environmental justice near oil and gas wells in Colorado. Environ Sci Technol. 2016;50(21):11471–80.
Deziel NC, Brokovich E, Grotto I, et al. Unconventional oil and gas development and health outcomes: a scoping review of the epidemiological research. Environ Res. 2020;182:109124.
US Environmental Protection Agency. Hydraulic fracturing for oil and gas: impacts from the hydraulic fracturing water cycle on drinking water resources in the United States (final report). Washington, DC, EPA/600/R-16/236F, 2016.
Kondash AJ, Albright E, Vengosh A. Quantity of flowback and produced waters from unconventional oil and gas exploration. Sci Total Environ. 2016;574:314–21.
Getzinger GJ, O’Connor MP, Hoelzer K, et al. Natural gas residual fluids: sources, endpoints, and organic chemical composition after centralized waste treatment in Pennsylvania. Environ Sci Technol. 2015;49(14):8347–55.
Burgos WD, Castillo-Meza L, Tasker TL, et al. Watershed-scale impacts from surface water disposal of oil and gas wastewater in western Pennsylvania. Environ Sci Technol. 2017;51(15):8851–60.
Maloney KO, Yoxtheimer DA. Production and disposal of waste materials from gas and oil extraction from the Marcellus Shale play in Pennsylvania. Environ Pract. 2012;14(04):278–87.
Davies RJ, Almond S, Ward RS, et al. Oil and gas wells and their integrity: implications for shale and unconventional resource exploitation. Mar Pet Geol. 2014;56:239–54.
Clark CJ, Warren JL, Kadan-Lottick N, et al. Community concern and government response: Identifying socio-economic and demographic predictors of oil and gas complaints and drinking water impairments in Pennsylvania. Energy Res Soc Sci. 2021;76:102070.
Turley B, Caretta MA. Household water security: an analysis of water affect in the context of hydraulic fracturing in West Virginia, Appalachia. Water. 2020;12(1):147.
Elliott EG, Ettinger AS, Leaderer BP, et al. A systematic evaluation of chemicals in hydraulic-fracturing fluids and wastewater for reproductive and developmental toxicity. J Exp Sci Environ Epidemiol. 2016;27(1):90–9.
Elliott EG, Trinh P, Ma X, et al. Unconventional oil and gas development and risk of childhood leukemia: Assessing the evidence. Sci Total Environ. 2017;576:138–47.
Kassotis CD, Tillitt DE, Lin CH, et al. Endocrine-disrupting chemicals and oil and natural gas operations: potential environmental contamination and recommendations to assess complex environmental mixtures. Environ Health Perspect. 2016;124(3):256–64.
Physicians for Social Responsibility. Fracking with “Forever Chemicals”. 2021. Accessed from: https://www.psr.org/wp-content/uploads/2021/07/fracking-with-forever-chemicals.pdf. Accessed 20 Jan 2022.
Hoelzer K, Sumner AJ, Karatum O, et al. Indications of transformation products from hydraulic fracturing additives in shale-gas wastewater. Environ Sci Technol. 2016;50(15):8036–48.
Sumner AJ, Plata DL. Exploring the hydraulic fracturing parameter space: a novel high-pressure, high-throughput reactor system for investigating subsurface chemical transformations. Environ Sci Process Impacts. 2018;20(2):318–31.
Stringfellow WT, Domen JK, Camarillo MK, et al. Physical, chemical, and biological characteristics of compounds used in hydraulic fracturing. J Hazard Mater. 2014;275:37–54.
Brantley SL, Yoxtheimer D, Arjmand S, et al. Water resource impacts during unconventional shale gas development: the Pennsylvania Experience. Int J Coal Geol. 2014;126:140–56.
Cozzarelli IM, Kent DB, Briggs M, et al. Geochemical and geophysical indicators of oil and gas wastewater can trace potential exposure pathways following releases to surface waters. Sci Total Environ. 2021;755:142909.
DiGiulio DC, Jackson RB. Impact to underground sources of drinking water and domestic wells from production well stimulation and completion practices in the Pavillion, Wyoming, field. Environ Sci Technol. 2016;50(8):4524–36.
Ferrar KJ, Michanowicz DR, Christen CL, et al. Assessment of effluent contaminants from three facilities discharging Marcellus Shale wastewater to surface waters in Pennsylvania. Environ Sci Technol. 2013;47(7):3472–81.
Drollette BD, Hoelzer K, Warner NR, et al. Elevated levels of diesel range organic compounds in groundwater near Marcellus gas operations are derived from surface activities. Proc Natl Acad Sci. 2015;112(43):13184–9.
Harkness JS, Dwyer GS, Warner NR, et al. Iodide, bromide, and ammonium in hydraulic fracturing and oil and gas wastewaters: environmental implications. Environ Sci Technol. 2015;49(3):1955–63.
Elliott EG, Ma X, Leaderer BP, et al. A community-based evaluation of proximity to unconventional oil and gas wells, drinking water contaminants, and health symptoms in Ohio. Environ Res. 2018;167:550–7.
Fontenot BE, Hunt LR, Hildenbrand ZL, et al. An evaluation of water quality in private drinking water wells near natural gas extraction sites in the Barnett Shale formation. Environ Sci Technol. 2013;47(17):10032–40.
McMahon PB, Lindsey BD, Conlon MD, et al. Hydrocarbons in upland groundwater, Marcellus shale region, northeastern Pennsylvania and southern New York, U.S.A. Environ Sci Technol. 2019;53(14):8027–35.
Wen T, Niu X, Gonzales M, et al. Big groundwater data sets reveal possible rare contamination amid otherwise improved water quality for some analytes in a region of Marcellus shale development. Environ Sci Technol. 2018;52(12):7149–59.
Clancy SA, Worrall F, Davies RJ, et al. The potential for spills and leaks of contaminated liquids from shale gas developments. Sci Total Environ. 2018;626:1463–73.
Maloney KO, Baruch-Mordo S, Patterson LA, et al. Unconventional oil and gas spills: Materials, volumes, and risks to surface waters in four states of the U.S. Sci Total Environ. 2017;581–582:369–77.
Akob DM, Mumford AC, Orem W, et al. Wastewater disposal from unconventional oil and gas development degrades stream quality at a West Virginia injection facility. Environ Sci Technol. 2016;50(11):5517–25.
Lauer NE, Harkness JS, Vengosh A. Brine spills associated with unconventional oil development in North Dakota. Environ Sci Technol. 2016;50(10):5389–97.
Warner NR, Jackson RB, Darrah TH, et al. Geochemical evidence for possible natural migration of Marcellus Formation brine to shallow aquifers in Pennsylvania. Proc Natl Acad Sci. 2012;109(30):11961–6.
Allshouse WB, McKenzie LM, Barton K, et al. Community noise and air pollution exposure during the development of a multi-well oil and gas pad. Environ Sci Technol. 2019;53(12):7126–35.
Garcia-Gonzales DA, Shonkoff SBC, Hays J, et al. Hazardous air pollutants associated with upstream oil and natural gas development: a critical synthesis of current peer-reviewed literature. Annu Rev Public Heatlh. 2019;40(1):283–304.
Allshouse WB, Adgate JL, Blair BD, et al. Spatiotemporal industrial activity model for estimating the intensity of oil and gas operations in Colorado. Environ Sci Technol. 2017;51(17):10243–50.
Koehler K, Ellis JH, Casey JA, et al. Exposure assessment using secondary data sources in unconventional natural gas development and health studies. Environ Sci Technol. 2018;52(10):6061–9.
McKenzie LM, Blair B, Hughes J, et al. Ambient nonmethane hydrocarbon levels along Colorado’s Northern Front Range: acute and chronic health risks. Environ Sci Technol. 2018;52(8):4514–25.
Goetz JD, Floerchinger C, Fortner EC, et al. Atmospheric emission characterization of Marcellus Shale natural gas development sites. Environ Sci Technol. 2015;49(11):7012–20.
Macey GP, Breech R, Chernaik M, et al. Air concentrations of volatile compounds near oil and gas production: a community-based exploratory study. Environ Health. 2014;13(1):1–18.
Colborn T, Schultz K, Herrick L, et al. An exploratory study of air quality near natural gas operations. Hum Ecol Risk Assess. 2014;20(1):86–105.
Paulik LB, Hobbie KA, Rohlman D, et al. Environmental and individual PAH exposures near rural natural gas extraction. Environ Pollut. 2018;241:397–405.
Torres L, Yadav OP, Khan E. Risk assessment of human exposure to Ra-226 in oil produced water from the Bakken Shale. Sci Total Environ. 2018;626:867–74.
Brown VJ. Radionuclides in fracking wastewater: managing a toxic blend. Environ Health Perspect. 2014;122(2):A50–5.
Eitrheim ES, May D, Forbes TZ, et al. Disequilibrium of naturally occurring radioactive materials (NORM) in drill cuttings from a horizontal drilling operation. Environ Sci Technol Lett. 2016;3(12):425–9.
Rich AL, Crosby EC. Analysis of reserve pit sludge from unconventional natural gas hydraulic fracturing and drilling operations for the presence of technologically enhanced naturally occurring radioactive material (TENORM). New Solut. 2013;23(1):117–35.
Lauer NE, Warner NR, Vengosh A. Sources of radium accumulation in stream sediments near disposal sites in Pennsylvania: Implications for Disposal of Conventional Oil and Gas Wastewater. Environ Sci Technol. 2018;52(3):955–62.
Casey JA, Ogburn EL, Rasmussen SG, et al. Predictors of indoor radon concentrations in Pennsylvania, 1989–2013. Environ Health Perspect. 2015;123(11):1130–7.
Xu Y, Sajja M, Kumar A. Impact of the hydraulic fracturing on indoor radon concentrations in Ohio: a multilevel modeling approach. Front Public Health. 2019;7:76.
Gilbert ES. Ionising radiation and cancer risks: what have we learned from epidemiology? Int J Radiat Biol. 2009;85(6):467–82.
Nyhan MM, Rice M, Blomberg A, et al. Associations between ambient particle radioactivity and lung function. Environ Int. 2019;130:104795.
Nyhan MM, Coull BA, Blomberg AJ, et al. Associations between ambient particle radioactivity and blood pressure: the NAS (Normative Aging Study). J Am Heart Assoc 2018;7(6):e008245.
Li W, Nyhan MM, Wilker EH, et al. Recent exposure to particle radioactivity and biomarkers of oxidative stress and inflammation: the Framingham Heart Study. Environ Int. 2018;121(Pt 2):1210–6.
Fisher MP, Mayer A, Vollet K, et al. Psychosocial implications of unconventional natural gas development: quality of life in Ohio’s Guernsey and Noble Counties. J Environ Psychol. 2018;55:90–8.
Ferrar KJ, Kriesky J, Christen CL, et al. Assessment and longitudinal analysis of health impacts and stressors perceived to result from unconventional shale gas development in the Marcellus Shale region. Int J Occup Environ Health. 2013;19(2):104–12.
Boslett A, Hill E, Ma L, et al. Rural light pollution from shale gas development and associated sleep and subjective well-being. Resour Energy Econ. 2021;64:101220.
Hays J, McCawley M, Shonkoff SBC. Public health implications of environmental noise associated with unconventional oil and gas development. Sci Total Environ. 2017;580:448–56.
Basner M, Babisch W, Davis A, et al. Auditory and non-auditory effects of noise on health. Lancet. 2014;383(9925):1325–32.
Axelsson G, Stockfelt L, Andersson E, et al. Annoyance and worry in a petrochemical industrial area—prevalence, time trends and risk indicators. Int J Environ Res Public Health. 2013;10(4):1418–38.
Shusterman D. Critical review: the health significance of environmental odor pollution. Arch Environ Health. 1992;47(1):76–87.
Abramzon S, Samaras C, Curtright A, et al. Estimating the consumptive use costs of shale natural gas extraction on Pennsylvania roadways. J Infrastruct Syst. 2014;20(3):06014001.
Graham J, Irving J, Tang X, et al. Increased traffic accident rates associated with shale gas drilling in Pennsylvania. Accid Anal Prev. 2015;74:203–9.
Elser H, Goldman-Mellor S, Morello-Frosch R, et al. Petro-riskscapes and environmental distress in West Texas: community perceptions of environmental degradation, threats, and loss. Energy Res Soci Sci. 2020;70:101798.
Ellsworth WL. Injection-induced earthquakes. Science. 2013;341(6142):1225942.
Morrone M, Chadwick AE, Kruse N. A community divided: hydraulic fracturing in rural Appalachia. J Appalach Stud Assoc. 2015;21(2):207–28.
James A, Smith B. There will be blood: Crime rates in shale-rich U.S. counties. J Environ Econ Manag. 2017;84:125–52.
Komarek TM. Crime and natural resource booms: evidence from unconventional natural gas production. Ann Reg Sci. 2018;61(1):113–37.
Muehlenbachs L, Spiller E, Timmins C. The housing market impacts of shale gas development. Am Econ Rev. 2015;105(12):3633–59.
Burnett JW, Mothorpe C. Human-induced earthquakes, risk salience, and housing values. Resour Energy Econ. 2021;63:101212.
Caulton DR, Shepson PB, Santoro RL, et al. Toward a better understanding and quantification of methane emissions from shale gas development. Proc Natl Acad Sci U S A. 2014;111(17):6237–42.
US Environmental Protection Agency. Guidelines for human exposure assessment. (EPA/100/B-19/001). Washington, D.C.: Risk Assessment Forum. 2019. Accessible athttps://www.epa.gov/risk/guidelines-human-exposureassessment. Accessed 20 Jan 2022.
Quitoriano V, Wald DJ. USGS “Did You Feel It?”—Science and lessons from 20 years of citizen science-based macroseismology. Front Earth Sci. 2020;8(120).
Willow AJ. Troubling water: shale energy and waterscape transformation in a North American extraction zone. Anthropologica. 2016;58(2):166–78.
Hill EL, Ma L. Drinking water, fracking, and infant health. J Health Econ. 2022;82:102595.
Casey JA, Savitz DA, Rasmussen SG, et al. Unconventional natural gas development and birth outcomes in Pennsylvania, USA. Epidemiology. 2016;27(2):163–72.
• Tran KV, Casey JA, Cushing LJ, et al. Residential proximity to oil and gas development and birth outcomes in California: a retrospective cohort study of 2006–2015 births. Environ Health Perspect. 2020;128(6):067001. This study of nearly 3 million births in California used an activity metric of combined production volume from oil and gas and observed associations with adverse perinatal outcomes.
Walker Whitworth K, Kaye Marshall A, Symanski E. Drilling and production activity related to unconventional gas development and severity of preterm birth. Environ Health Perspect. 2018;126(3):037006.
Gonzalez DJX, Sherris AR, Yang W, et al. Oil and gas production and spontaneous preterm birth in the San Joaquin Valley, CA: a case–control study. Environ Epidemiol. 2020;4(4):e099.
• Cushing LJ, Vavra-Musser K, Chau K, et al. Flaring from unconventional oil and gas development and birth outcomes in the Eagle Ford Shale in South Texas. Environ Health Perspect. 2020;128(7):077003. This study in Texas applied an innovative metric to assess exposure to flaring from oil and gas and identified an increased risk of preterm birth.
Casey JA, Savitz DA, Rasmussen SG, et al. Unconventional natural gas development and birth outcomes in Pennsylvania, USA. Epidemiology. 2016;27(2):163–72.
Deziel NC. Invited perspective: oil and gas development and adverse birth outcomes: what more do we need to know? Environ Health Perspect. 2021;129(7):071301.
Lupo PJ, Symanski E, Chan W, et al. Differences in exposure assignment between conception and delivery: the impact of maternal mobility. Paediatr Perinat Epidemiol. 2010;24(2):200–8.
Warren JL, Son J-Y, Pereira G, et al. Investigating the impact of maternal residential mobility on identifying critical windows of susceptibility to ambient air pollution during pregnancy. Am J Epidemiol. 2018;187(5):992–1000.
Bell ML, Belanger K. Review of research on residential mobility during pregnancy: consequences for assessment of prenatal environmental exposures. J Expo Sci Environ Epidemiol. 2012;22(5):429–38.
Bell ML, Banerjee G, Pereira G. Residential mobility of pregnant women and implications for assessment of spatially-varying environmental exposures. J Expo Sci Environ Epidemiol. 2018;28(5):470–80.
Franklin M, Chau K, Cushing LJ, et al. Characterizing flaring from unconventional oil and gas operations in south Texas using satellite observations. Environ Sci Technol. 2019;53(4):2220–8.
Soriano MA, Siegel HG, Johnson NP, et al. Assessment of groundwater well vulnerability to contamination through physics-informed machine learning. Environ Res Lett. 2021;16(8):084013.
• Li L, Blomberg AJ, Spengler JD, et al. Unconventional oil and gas development and ambient particle radioactivity. Nat Commun. 2020;11(1):1–8. This study introduced a new wind-weighted model of unconventional oil and gas wells.
Casey JA, Goldman-Mellor S, Catalano R. Association between Oklahoma earthquakes and anxiety-related Google search episodes. Environ Epidemiol 2018;2(2):e016.
Casey JA, Elser H, Goldman-Mellor S, et al. Increased motor vehicle crashes following induced earthquakes in Oklahoma, USA. Sci Total Environ. 2019;650:2974–9.
• McKenzie LM, Allshouse W, Daniels S. Congenital heart defects and intensity of oil and gas well site activities in early pregnancy. Environ Int. 2019;132:104949. This study applied an intensity-adjusted inverse distance weighted oil and gas well model and observed associations with several types of coronary heart defects.
McKenzie LM, Crooks J, Peel JL, et al. Relationships between indicators of cardiovascular disease and intensity of oil and natural gas activity in Northeastern Colorado. Environ Res. 2019;170:56–64.
Li L, Dominici F, Blomberg AJ, et al. Exposure to unconventional oil and gas development and all-cause mortality in Medicare beneficiaries. Nat Energy. 2022;7(2):177–85.
Willis M, Hystad P, Denham A, et al. Natural gas development, flaring practices and paediatric asthma hospitalizations in Texas. Int J Epidemiol. 2020;49(6):1883–96.
Lian Q, Ni J, Zhang J, et al. Maternal exposure to Wenchuan earthquake and prolonged risk of offspring birth outcomes: a natural experiment study. BMC Pregnancy Childbirth. 2020;20(1):1–9.
Menclova AK, Stillman S. Maternal stress and birth outcomes: evidence from an unexpected earthquake swarm. Health Econ. 2020;29(12):1705–20.
Kim B, Carruthers CK, Harris MC. Maternal stress and birth outcomes: evidence from the 1994 Northridge earthquake. J Econ Behav Organ. 2017;140:354–73.
Palmeiro-Silva YK, Orellana P, Venegas P, et al. Effects of earthquake on perinatal outcomes: A Chilean register-based study. PLoS One. 2018;13(2):e0191340.
Rasmussen SG, Ogburn EL, McCormack M, et al. Association between unconventional natural gas development in the Marcellus Shale and asthma exacerbations. JAMA Intern Med. 2016;176(9):1334–43.
Soriano MA Jr, Siegel HG, Gutchess KM, et al. Evaluating domestic well vulnerability to contamination from unconventional oil and gas development sites. Water Resour. 2020;56(10):e2020WR028005.
• Clark CJ, Xiong B, Soriano MA, et al. Assessing unconventional oil and gas exposure in the Appalachian basin: comparison of exposure surrogates and residential drinking water measurements. Environ Sci Technol. 2022. This study used a water-specific metric accounting for groundwater flow paths as an exposure surrogate in comparison to drinking water measurements.
Roy AA, Adams PJ, Robinson AL. Air pollutant emissions from the development, production, and processing of Marcellus Shale natural gas. J Air Waste Manag Assoc. 2014;64(1):19–37.
Lee D, Murphy HM. Private wells and rural health: groundwater contaminants of emerging concern. Curr Environ Health Rep. 2020;7(2):129–39.
Sumner AJ, Plata DL. A geospatially resolved database of hydraulic fracturing wells for chemical transformation assessment. Environ Sci Process Impacts. 2020;22(4):945–55.
Elsner M, Hoelzer K. Quantitative survey and structural classification of hydraulic fracturing chemicals reported in unconventional gas production. Environ Sci Technol. 2016;50(7):3290–314.
Sumner AJ, Plata DL. Halogenation chemistry of hydraulic fracturing additives under highly saline simulated subsurface conditions. Environ Sci Technol. 2018;52(16):9097–107.
Sumner AJ, Plata DL. Oxidative breakers can stimulate halogenation and competitive oxidation in guar-gelled hydraulic fracturing fluids. Environ Sci Technol. 2019;53(14):8216–26.
Shi W, Plata DL. Environmental Chemistry Oil & Gas Application. 2020. https://www.ecogap.mit.edu. Accessed 26 Apr 2022.
Down A, Schreglmann K, Plata DL, et al. Pre-drilling background groundwater quality in the Deep River Triassic Basin of central North Carolina, USA. Appl Geochem. 2015;60:3–13.
Caron-Beaudoin É, Bouchard M, Wendling G, et al. Urinary and hair concentrations of trace metals in pregnant women from Northeastern British Columbia, Canada: a pilot study. J Expo Sci Environ Epidemiol. 2019;29(5):613–23.
Caron-Beaudoin É, Valter N, Chevrier J, et al. Gestational exposure to volatile organic compounds (VOCs) in Northeastern British Columbia, Canada: a pilot study. Environ Int. 2018;110:131–8.
Pedersen M, Merlo DF, Knudsen LE. Ethical issues related to biomonitoring studies on children. Int J Hyg Environ Health. 2007;210(3–4):479–82.
National Research Council. Human biomonitoring for environmental chemicals. Washington, DC: The National Academies Press; 2006.
Ma Z, Sneeringer KC, Liu L, et al. Time series evaluation of birth defects in areas with and without unconventional natural gas development. J Epidemiol Public Health Rev. 2016;1(4):10.16966.
Tran KV, Casey JA, Cushing LJ, et al. Residential proximity to hydraulically fractured oil and gas wells and adverse birth outcomes in urban and rural communities in California (2006–2015). Environ Epidemiol. 2021;5(6):e172.
Busby C, Mangano JJ. There’sa world going on underground—infant mortality and fracking in Pennsylvania. J Environ Prot. 2017;8(04):381.
Tang IW, Langlois PH, Vieira VM. Birth defects and unconventional natural gas developments in Texas, 1999–2011. Environ Res. 2021;194:110511.
Apergis N, Hayat T, Saeed T. Fracking and infant mortality: fresh evidence from Oklahoma. Environ Sci Pollut Res. 2019;26(31):32360–7.
Hill EL. Shale gas development and infant health: evidence from Pennsylvania. J Health Econ. 2018;61:134–50.
Jemielita T, Gerton GL, Neidell M, et al. unconventional gas and oil drilling is associated with increased hospital utilization rates. PLoS One. 2015;10(7):e0131093.
Willis MD, Jusko TA, Halterman JS, et al. Unconventional natural gas development and pediatric asthma hospitalizations in Pennsylvania. Environ Res. 2018;166:402–8.
Denham A, Willis M, Zavez A, et al. Unconventional natural gas development and hospitalizations: evidence from Pennsylvania, United States, 2003–2014. Public Health. 2019;168:17–25.
Denham A, Willis MD, Croft DP, et al. Acute myocardial infarction associated with unconventional natural gas development: a natural experiment. Environ Res. 2021;195:110872.
Peng L, Meyerhoefer C, Chou S-Y. The health implications of unconventional natural gas development in Pennsylvania. Health Econ. 2018;27(6):956–83.
Johnston JE, Enebish T, Eckel SP, et al. Respiratory health, pulmonary function and local engagement in urban communities near oil development. Environ Res. 2021;197:111088.
Fryzek J, Pastula S, Jiang X, et al. Childhood cancer incidence in Pennsylvania counties in relation to living in counties with hydraulic fracturing sites. J Occup Environ Med. 2013;55(7):796–801.
Finkel ML. Shale gas development and cancer incidence in southwest Pennsylvania. Public Health. 2016;141:198–206.
Deziel NC, Humeau Z, Elliott EG, et al. Shale gas activity and increased rates of sexually transmitted infections in Ohio, 2000–2016. PLoS One. 2018;13(3):e0194203.
Beleche T, Cintina I. Fracking and risky behaviors: evidence from Pennsylvania. Econ Hum Biol. 2018;31:69–82.
Johnson NP, Warren JL, Elliott EG, et al. A multiregion analysis of shale drilling activity and rates of sexually transmitted infections in the United States. Sex Transm Dis. 2020;47(4):254–60.
Currie J, Greenstone M, Meckel K. Hydraulic fracturing and infant health: new evidence from Pennsylvania. Sci Adv. 2017;3(12):e1603021.
• Willis MD, Hill EL, Boslett A, et al. Associations between residential proximity to oil and gas drilling and term birth weight and small-for-gestational-age infants in Texas: a difference-in-differences analysis. Environ Health Perspect. 2021;129(7):077002. This paper used a difference-in-different study design, for which the reported associations adverse birth outcomes were based on both the difference during versus prior to drilling activity and living nearer versus further from an oil or gas well site.
Stacy SL, Brink LL, Larkin JC, et al. Perinatal outcomes and unconventional natural gas operations in Southwest Pennsylvania. PLoS One. 2015;10(6):e0126425.
McKenzie LM, Guo R, Witter RZ, et al. Birth outcomes and maternal residential proximity to natural gas development in rural Colorado. Environ Health Perspect. 2014;122:412–7.
Whitworth KW, Marshall AM, Symanski E. Maternal residential proximity to unconventional gas development and perinatal outcomes among a diverse urban population in Texas. PLoS One. 2017;12(7):e0180966.
Janitz AE, Dao HD, Campbell JE, et al. The association between natural gas well activity and specific congenital anomalies in Oklahoma, 1997–2009. Environ Int. 2019;122:381–8.
Caron-Beaudoin É, Whitworth KW, Bosson-Rieutort D, et al. Density and proximity to hydraulic fracturing wells and birth outcomes in Northeastern British Columbia, Canada. J Expo Sci Environ Epidemiol. 2021;31(1):53–61.
McKenzie LM, Allshouse WB, Byers TE, et al. Childhood hematologic cancer and residential proximity to oil and gas development. PLoS One. 2017;12(2):e0170423.
Casey JA, Goin DE, Rudolph KE, et al. Unconventional natural gas development and adverse birth outcomes in Pennsylvania: the potential mediating role of antenatal anxiety and depression. Environ Res. 2019;177:108598.
McAlexander TP, Bandeen-Roche K, Buckley JP, et al. Unconventional natural gas development and hospitalization for heart failure in Pennsylvania. J Am Coll Cardiol. 2020;76(24):2862–74.
Tustin AW, Hirsch AG, Rasmussen SG, et al. Associations between unconventional natural gas development and nasal and sinus, migraine headache, and fatigue symptoms in Pennsylvania. Environ Health Perspect. 2016.
Casey JA, Wilcox HC, Hirsch AG, et al. Associations of unconventional natural gas development with depression symptoms and disordered sleep in Pennsylvania. Sci Rep. 2018;8(1):11375.
This publication was developed under Assistance Agreement No. CR839249 awarded by the U.S. Environmental Protection Agency to Yale University. It has not been formally reviewed by EPA. The views expressed in this document are solely those of the authors and do not necessarily reflect those of the Agency. EPA does not endorse any products or commercial services mentioned in this publication. JAC was funded by the National Institutes of Health R00 ES027023. CJC was also supported by the National Institute of Environmental Health Sciences under the National Institutes of Health [F31ES031441], the Yale Cancer Center [T32CA250803], and the Yale Institute for Biospheric Studies.
Conflict of Interest
The authors declare no competing interests.
Human and Animal Rights and Informed Consent
This review includes descriptions of previously published studies conducted by the authors that involved human subjects and for which IRB approval was obtained. This article does not contain any studies with animal subjects performed by any of the authors.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This article is part of the Topical Collection on Early Life Environmental Health
About this article
Cite this article
Deziel, N.C., Clark, C.J., Casey, J.A. et al. Assessing Exposure to Unconventional Oil and Gas Development: Strengths, Challenges, and Implications for Epidemiologic Research. Curr Envir Health Rpt 9, 436–450 (2022). https://doi.org/10.1007/s40572-022-00358-4
- Unconventional oil and gas
- Exposure assessment
- Children’s health
- Epidemiologic studies