Health Impact Assessment of Shale Gas Extraction: Workshop Summary (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Occupational Health and Community Impacts This chapter provides a summary of presentations focused on the communities impacted by shale gas extraction, concerning the occupational opportunities and hazards for populations and the everyday consequences of living near a hydraulic fracturing site. The presentations describe separate formal assessments conducted at or around shale gas production sites in Colorado and other states, while the final presentation focuses on an overview of the economic impacts in Pennsylvania. The presentations are followed by a summary of the discussion that took place. NIOSH FIELD EFFORT TO ASSESS CHEMICAL EXPOSURES IN OIL AND GAS WORKERS: HEALTH HAZARDS IN HYDRAULIC FRACTURING Eric J. Esswein, M.S.P.H. Senior Industrial Hygienist, National Institute for Occupational Safety and Health Centers for Disease Control and Prevention Eric J. Esswein began his presentation by noting that occupational safety hazards in oil and gas extraction (including hydraulic fracturing) are known but the same cannot be said for occupational health hazards, especially risks for chemical exposures. Attention has typically focused more on safety—work practices, policy and procedures to prevent acute injury—than on health (understanding risks for chemical exposures and prevention of chronic disease). The National Institute for Occupational Safety and Health (NIOSH) implemented the NIOSH Field Effort to Assess Risks for Chemical Exposures in Oil and Gas Workers1 to better understand risks for chemical exposures in workers involved in oil and gas extraction activities. 1 See http://www.cdc.gov/niosh/docs/2010-130 (accessed May 30, 2013). 27 

28 HEALTH IMPACT ASSESSMENT OF SHALE GAS EXTRACTION Operators within the oil and gas industry have been voluntarily participating with NIOSH since 2008 and since 2010 in the NIOSH Field Effort. Participation occurs through memoranda of understanding between NIOSH and various companies. NIOSH researchers’ review work practices and chemical products used on site, and conduct industrial hygiene exposure assessments to determine if occupational health risks might be present. Depending on the outcome of the field research studies NIOSH provides recommendations to control exposures if risks are determined. Mr. Esswein provided a list of some chemicals that may pose exposure risks during oil and gas extraction including respirable crystalline silica, diesel particulate, volatile organic compounds (VOCs), hydrogen sulfide, acid gases, aldehydes, and elements (i.e., lead) from lead-based pipe thread grease. Although the NIOSH field work in 2010 and 2011 excluded physical agents (i.e., noise, radiation, etc.), the research team looked at chemical exposures across a wide range of oil and gas extraction basins in the United States. NIOSH visited 11 hydraulic fracturing sites in five states, during different seasons and at varying elevations. Respirable crystalline silica (from quartz sand) was identified as a potential occupational exposure hazard at hydraulic fracturing sites. Sand is used as a proppant to hold open fissures and cracks in the hydrocarbon-bearing formation that are created by hydraulic fracturing. NIOSH determined that respirable crystalline silica presented an occupational exposure hazard likely greater than exposures to chemicals used during hydraulic fracturing. Crystalline silica is the known causative agent for the lung disease silicosis and is associated with lung cancer. Silicosis is preventable by eliminating exposure, but discontinuing exposure does not stop progression of the disease, which is incurable, irreversible, and progressive. Industry wide, in the United States, exposures to crystalline silica are associated with 160–200 worker deaths annually (Rosenman et al., 2003); consequently occupational exposures are regulated by the Occupational Safety and Health Administration (OSHA). Millions of pounds of quartz-containing sand are used during hydraulic fracturing. Sand-moving transport operations result in respirable crystalline becoming airborne during handling and pneumatic transport by a variety of machines used during the hydraulic fracturing process (e.g., sand movers, sand transport belts, and sand and water blender trucks). Windy conditions on- site can also contribute to dust generation and exposures to unprotected workers. During 2010–2011 the NIOSH Field Effort research team collected 111 full-shift personal breathing zone samples for respirable crystalline silica during active hydraulic fracturing operations. NIOSH determined that more than 50 percent of the samples exceeded a calculated OSHA permissible exposure limit, 68 percent were greater than the NIOSH recommended exposure limit; in some cases by factors of 10 or more, 

OCCUPATIONAL HEALTH AND COMMUNITY IMPACTS 29 which exceed the assigned protection factor for the typical type of respirator (half-face elastomeric or filtering face-piece) used by the workers at the sites visited. As part of the study, NIOSH identified seven primary points of dust generation and developed control recommend- ations that included both passive and active controls. The NIOSH team also described nine possible interventions to control dust, including  consideration of Prevention through Design2 on future versions of sand moving equipment that has built in dust control;  use of remote operations to keep workers out of areas of high dust concentrations;  substitution of sand with ceramic or other proppant materials;  installation of active controls such as the NIOSH-designed mini- baghouse retrofit assembly (technology to reduce amount of dust released), over the thief hatches (access ports over sand movers), and consideration of enclosed sand transport mechanism such as a screw augur retrofit assembly to replace belt transport of sand;  use of enclosures around bottoms of sand movers (such as stilling or staging curtains) and the dragon tail (end of the sand belt) on sand movers;  minimizing the distance that sand falls from the dragon tail;  use of end caps on fill nozzles on sand movers;  use of amended water for dust control on site; and  implementing an effective respiratory protection program. Mr. Esswein reiterated that NIOSH determined that respirable crystalline silica is a significant occupational health hazard associated with hydraulic fracturing; diesel particulate is also a likely occupational health hazard. Discussion Following the presentation from Mr. Esswein, Roundtable members started the discussion by asking about silica exposures at mines during the loading and unloading of trucks. Mr. Esswein noted that NIOSH’s assessment was limited to evaluating exposures only at oil and gas sites, not any further upstream. Nsedu Witherspoon inquired about exposures to pregnant workers and Mr. Esswein replied that he only saw two female oil and gas workers, and the NIOSH team did not evaluate their exposures. Bernard Goldstein asked about a 2008 article showing an increase in injuries per shale gas well and asked about the status of injuries. Dr. Goldstein went on to comment on integrating workers’ health and 2 Prevention through Design is a NIOSH-led effort to address occupational safety and health needs in the design process to prevent or minimize the work- related hazards and risks (http://www.cdc.gov/niosh/topics/ptd [accessed May 30, 2013]). 

30 HEALTH IMPACT ASSESSMENT OF SHALE GAS EXTRACTION environmental health, and asked whether there is any indication that companies that are less careful with the environment are also less careful about worker health, that is, determining whether worker health violations track with the same companies that are fined for environmental risks. Mr. Esswein replied that oil and gas extraction and production have about seven times the fatality rate of general industry (27–28 per 100,000 compared with 4 per 100,000), and that most injuries are attributable to motor vehicle accidents, although there are additional deaths from causes such as being struck by equipment, being caught between parts of equipment, and falling from heights. Mr. Esswein also acknowledged the challenges represented by a worker population that is highly transient, which makes surveillance and follow-up efforts difficult. Mr. Esswein responded to a question about dissemination of findings by noting that NIOSH participates in the oil and gas workgroup of the American Industrial Hygiene Association, and that the NIOSH team will participate in the OSHA Safety Conference, which will focus on oil and gas. They have submitted an article for the Journal of Occupational and Environmental Health (Esswein et al., 2013). An audience member inquired whether NIOSH researchers monitored the exposures of female security officers stationed outside hydraulic fracturing project sites. Mr. Esswein stated that the security staff he has seen are typically stationed on the lease road some distance away from the wellhead, and NIOSH had not researched these exposures. Another person asked whether NIOSH’s visits are announced; Mr. Esswein replied that NIOSH is a research agency, not a regulatory agency, and site visits are scheduled, not unannounced. Moreover, it seems implausible that operators at a hydraulic fracturing site could clean up before a NIOSH visit because they have a performance contract to fulfill, and they need to pressure up and pump, making it difficult if not impossible to somehow stop and spruce things up. One audience member inquired whether the NIOSH researchers have reviewed data to show whether hydraulic fracturing site workers are experiencing silica-related diseases, and secondly, whether they have looked at other types of silica-related illnesses, including autoimmune, chronic respiratory, or kidney diseases. Mr. Esswein stated that NIOSH has not done any biological monitoring of workers, and reiterated that the work was an exposure assessment study, not a health study, and the study was still progressing. There is no systematic surveillance for oil and gas extraction workers and no known sentinel cases of accelerated or acute silicosis in the workers. Companies have informed NIOSH that they are expanding preemployment physicals to assess baseline pulmonary function and improving respiratory protection programs. Some comp- anies have begun considering or even implementing some of the NIOSH recommended controls to limit exposures to crystalline silica dust. Another audience member asked if silica dust is going offsite or if this was an issue, and if NIOSH had collected air samples offsite; Mr. Esswein responded that sampling was limited to workers in the immed- 

OCCUPATIONAL HEALTH AND COMMUNITY IMPACTS 31 iate area around the well pad. Mr. Esswein noted that he considered the sampling conducted to be fairly representative of the workforce at hydraulic fracturing sites (from the “company man” to the variety of contractors on site), because most workers were willing to be sampled. Another audience member asked whether any workers were represented by unions, and Mr. Esswein answered that they were not. COMMUNITY IMPACTS OF NATURAL GAS DEVELOPMENT AND HUMAN HEALTH Roxana Witter, M.D., M.S.P.H., M.S. Assistant Research Professor, Environmental and Occupational Health Colorado School of Public Health Roxana Witter noted that she led a team from the Colorado School of Public Health that conducted a health impact assessment (HIA) of a proposed natural gas project in a small community in western Colorado. The research team identified three possible types of exposures: chemical exposures, nonchemical exposures such as industrial activities (e.g., noise, traffic), and community changes. The team made more than 70 recommendations to reduce the risk of effects, including noise, traffic, and community changes (Witter et al., 2011). Dr. Witter explained that some of the data reviewed in the HIA were provided by operators who collected noise data, such as the noise emitted by diesel generators with and without noise blankets. Although it is unclear what the cumulative noise of 20 wells being drilled might be, and the effects of sustained noise over a period of time, noise levels can disturb sleep, affecting cognition, mood, and school performance. In thinking about the potential effects of truck traffic, the researchers recognized that it meant not only traffic around a well pad and its neighboring homes, but also a broader range of effects dispersed around the community, including exposure to exhaust, vibration, and dust, and safety risks, especially along haul routes that might be traveled by children. In assessing the community changes, the research team examined the social disruption, with the potential to affect quality of life, elicit stress and anxiety, and pose other risks to physical and mental health, as well as safety. The data reviewed in the HIA were from nearly a decade of natural gas projects in nearby areas of western Colorado. Between 2005 and 2008, well drilling was scaled up rapidly, but a steep decline took place in 2009, perhaps due to the drop in natural gas prices. The researchers reviewed rates of police arrests and sexually transmitted diseases, finding patterns that coincided with the increase in the introduction, expansion, and then decrease in drilling activity. 

32 HEALTH IMPACT ASSESSMENT OF SHALE GAS EXTRACTION Dr. Witter noted that although there are some gaps in knowledge, and making the link between natural gas drilling and social effects is not entirely straightforward, there is no doubt that there are some lessons in the available evidence, and that the social sciences and social epidemiology are useful in examining these relationships. For example, the sociology literature on boomtowns in the 1970s’ energy development is instructive. In 2009, sociologist Jeffrey Jacquet reviewed this literature and documented a similar “boomtown model” in Wyoming. People living in natural gas areas of Colorado, North Dakota, Pennsylvania, and Wyoming will likely find aspects of the model familiar (Jacquet, 2009). As the energy industry rapidly expanded, such communities saw rapid population influx, unprepared local governments, burgeoning resentment between old and new residents, a need for tax increases to support new and greater infrastructure, boom and bust cycles, inflation, mixed economic effects, and industry monopolization of information, leading to distrust of industry and a sense that there was a power grab away from the community. Other sociology and social epidemiology literature beginning in the 1920s and 1930s document the effects on health of the social circumstances. For example, highly mobile, socially isolated groups (e.g., the families that follow industrial workers to new sites of work) experience high rates of disease associated with social factors such as lack of social cohesion, social capital, trust, and shared values. Dr. Witter stated that evidence indicates that the ties that facilitate collective action confer benefits on health. Multilevel statistical analyses can be conducted to measure the impact of environment on health, and to show the effects of psychosocial stress on the body and community vulnerability contributions to health problems. The research team explored several potential solutions, for example, determining what boomtown characteristics may be present around natural gas projects, and what characteristics could be amenable to interventions. Are there ways to shape population influx, perhaps through controlled or slower development, involving local governments, facilitating community engagement, and providing jurisdictional control? The Colorado School of Public Health research team recognized that the environment around natural gas drilling sites is changing and difficult, and that it is hard for HIA to assess all impacts up front, in advance of actual implementation. Establishing monitoring mechanisms and putting in place adaptive management plans can be incorporated into the recommendations of HIA and are important for addressing un- expected and ongoing impacts. 

OCCUPATIONAL HEALTH AND COMMUNITY IMPACTS 33 ECONOMIC AND COMMUNITY IMPACTS OF GAS SHALE IN PENNSYLVANIA Timothy Kelsey, Ph.D. Professor of Agricultural Economics, State Program Leader Economic & Community Development The Pennsylvania State University Timothy Kelsey noted that the Marcellus Shale is the second largest natural gas field in the world according the U.S. Geological Survey, covering hundreds of thousands of acres. It has considerable economic effects, including local economic benefits, which have led to a great level of interest. Natural gas is a nonrenewable natural resource, and any discussion about jobs and economic development related to natural gas needs to acknowledge that when the gas is gone, the basis for economic activity is gone as well. There is much uncertainty about health effects and other aspects of unconventional gas development, stated Dr. Kelsey. In a context of great public interest, both positive and negative, there is a quest for scientific answers, and when they are unavailable, people are forced to rely on anecdotal information. This is a difficult basis on which to hold cogent policy discussions. Dr. Kelsey suggested that gas activity should be thought about holistically, across all three phases: development, production, and reclamation. The development phase is short-lived and labor-intensive, the production phase is long-lived with small and steady labor force requirements, and in the reclamation phase, the employment needs decline. There are two estimates of the duration of drilling in Pennsylvania, either 30 years or 50 years of active drilling, but the number of workers will depend on the number of drilling rigs and number of wells being dug. Different jobs are needed at each phase, stated Dr. Kelsey, and when the development phase begins, large numbers of jobs will be created, but when the drilling ends, many of the jobs will go away. Gas development is more than just wells; it involves multiple well pads, supporting many locations, multiple specialized companies and workers, changing transportation patterns, and a major need for coordination and communication. In addition to “short-run” facilities, specifically well pads, there is a need for “long-run” facilities. The former include staging areas; worker housing; office areas; storage, maintenance, and compressor stations; and water withdrawal facilities and treatment areas. Pennsylvania’s economic experience through March 2012 includes hiring, lower unemployment, and higher average wages in Marcellus counties. The state government has seen some increased revenue (taxes), but there are no estimates yet of the total costs to them. 

34 HEALTH IMPACT ASSESSMENT OF SHALE GAS EXTRACTION Dr. Kelsey described in some detail the business activity surrounding shale gas extraction in Pennsylvania. Drilling each well requires about 420 individuals across 150 different occupations (Kelsey et al., 2011). For the first well on a pad, there are 13.1–13.3 full-time direct jobs, and 9.65–9.85 full-time direct jobs for each subsequent well (MSETC, 2011). Once the wells are drilled, every 100 dry gas wells generate 19 long-term full-time jobs, and every 100 high-BTU (“wet”) gas wells generate 30 long-term full-time jobs (MSETC, 2011). The natural gas workforce is occupationally diverse, ranging from general labor to drivers and from engineers to geologists. About 75 percent of the jobs do not require a college education, although many require some type of certification (MSETC, 2011). Dr. Kelsey stated that in the decade since drilling began in Pennsylvania, many questions about community and family effects have arisen, but there are many questions yet to be asked, and many answers yet to be found. For example, there is a great deal of discussion about infrastructure needs and how these should be met and about the effects of the shale gas boom on housing affordability and on youth migration. Rents have tripled and even quadrupled in some of the affected counties. This is creating great burdens on vulnerable populations. Schools are reporting a decrease in the number of subsidized lunches, not due to increased income levels but due to families moving out of an area they can no longer afford. Local infrastructure issues include damage to roads, highway safety, housing costs and availability, sewerage and water requirements, and need for police, emergency services, and schools. On a positive note, stated Dr. Kelsey, development (such as hotels) has also been noted in the Marcellus counties. The shale gas drilling has dramatically increased traffic, with studies indicating that in 2010, some drilling locations averaged 3,100 to 3,900 tractor trailers at each location, a 10-fold increase over several years of drilling. Tractor trailer traffic can have considerable effects on local communities, including implications for road safety. From a sociologic perspective, researchers are observing more conflict in communities, including polarization, social service demands, early leasers versus late leasers, and newcomers versus oldtimers. From an economic perspective, there are effects on taxes and politics. Dr. Kelsey emphasized the importance of local engagement, and stated that there is little opportunity for local governments to weigh in on what is happening in their communities. The state recently enacted a law to address this. In his concluding remarks, Dr. Kelsey emphasized the importance of looking beyond the boom and examining what will be needed after the boom. 

OCCUPATIONAL HEALTH AND COMMUNITY IMPACTS 35 DISCUSSION Following the presentations by Dr. Witter and Dr. Kelsey, Roundtable and audience members were invited to ask questions and engage in discussion. Richard Fenske began by asking Dr. Kelsey why remediation was not on the job chart and how this is handled in the permitting process in Pennsylvania. Dr. Kelsey replied that they are not at that stage yet. Patricia Verduin asked whether anything can be learned from the boom–bust experiences of other industries. Dr. Witter said that potential lessons could be gleaned from the experience of the energy industry in the West in the 1970s and 1980s. She acknowledged, however, that there are geographic and other differences that should be noted. For example, areas that are less populous have had to bring in a labor force, and in areas with limited infrastructure, workers have had to live next to the well pads. Also, when production sites are dispersed, as in mining, there are lessons that could be learned about what has and has not worked and how to avoid mistakes that were made. Lynn Goldman restated the assertion made by several speakers that there are gaps in research on the health and social effects of shale gas extraction, and asked about the level of funding available to support research in the field. Dr. Kelsey stated that he has not lacked funding and that much of his work has been supported by the Pennsylvania state government. There are issues related to actual or perceived conflict of interest if industry or an advocacy organization provides the support, and that can taint how the research is received no matter how well conducted or objective. Dr. Kelsey stated his belief that there is not a stable source of funding for long-term monitoring, and the importance of this type of research to better understanding the effects of shale gas extraction. Dr. Witter agreed, and stated that the Colorado HIA relied on available data sources because that was the best option for the funding available. The field is so new that there is no track record that could serve as the basis for funding searches. Al McGartland remarked that the comment about school turnover made him wonder why companies appeared not to have hired locally and thus helped to maintain community stability. He asked whether they are attracting outsiders because they are paying more and because the skills are not available locally. Dr. Witter stated that some skills are necessary for the work, and she is aware that some companies have been unable to find the skills locally, but she is also aware that in some places in Colorado and Wyoming some companies offered training to local people. Dr. Kelsey found similar company training in Pennsylvania as part of an effort to hire locally, but he also learned that some local businesses were hurt by having employees hired away by the better-paying gas companies. To a follow-up question about why hotels needed to be built, Dr. Kelsey responded that many of the Marcellus counties have old 

36 HEALTH IMPACT ASSESSMENT OF SHALE GAS EXTRACTION housing stock, and not a lot of new housing, leading even some companies to build hotels to meet housing needs. To a question about environmental justice (i.e., for low socio- economic status or minority population) and considerations pertaining to the placement of the wells, Dr. Kelsey replied that much of the drilling is occurring on state forest or game land, or on farms. Many of these regions do not have significant minority populations, but he has not seen a study that examines these issues. An audience member suggested that noise levels of 60–70 decibels 1,000 feet away are very high and would likely cause elevated cortisol levels and blood pressure—some objective findings that could be identified by research in this area. The person added that during a 30- year work boom it seems clear how one would go about minimizing some of the negatives such as traffic safety and road damage, but the more important question seems to be how to maximize the benefits that arise from the economic boom? Dr. Kelsey replied that this is the central concern of many Pennsylvania communities and landowners who would like to ensure that the dollars that are coming in can be invested for long- term benefit of the community. Dr. Kelsey and colleagues urge local government leaders to ensure that their infrastructure investments have a dual purpose in mind. An audience member asked about the hurdles to preparing HIA for shale gas extraction, and asked Dr. Witter why her team’s HIA, the only one on shale gas extraction to date, was pulled at the last moment. Dr. Witter responded that the process is very political, and that the Colorado School of Public Health researchers were hampered by not having good access to the decision makers, which affected the clarity of their commun- ication. When commissioners were concerned or wanted answers, it was difficult for the HIA team to communicate with them. It was also never made clear to the team what decisions their recommendations would come to bear on, and moreover, information was offered to them in a piecemeal fashion over a long period of time, making it difficult for them to evaluate and reflect on its implications. A final audience question to Dr. Witter inquired whether she planned to follow up with and study the medical records of people she found had respiratory exposures related to shale gas drilling. The audience member also asked whether the nondisclosure agreements some people sign are proving to be a barrier to obtaining health data. Dr. Witter responded that the questioner’s work in Ithaca is farther ahead in looking at individual health effects. The Colorado research team looked at population-level effects and does not have specific plans to move in the direction the questioner indicated. 

OCCUPATIONAL HEALTH AND COMMUNITY IMPACTS 37 REFERENCES Jacquet, J. 2009. Energy boomtowns & natural gas: Implications for Marcellus Shale local governments & rural communities. http://energy.wilkes.edu/ PDFFiles/Issues/Energy%20Boomtowns%20and%20Natural%20Gas.pdf (accessed May 30, 2013). Kelsey, T. W., M. Shields, J. R. Ladlee, and M. Ward. 2011. Economic impacts of Marcellus Shale in Pennsylvania: Employment and income in 2009. Williamsport, PA: Pennsylvania College of Technology and Penn State Extension. MSETC (Marcellus Shale Education & Training Center). 2011. Pennsylvania statewide Marcellus Shale workforce needs assessment. Williamsport, PA: Marcellus Shale Education & Training Center. Rosenman, K. D., M. J. Reilly, and P. K. Henneberger. 2003. Estimating the total number of newly recognized silicosis cases in the United States. American Journal of Industrial Medicine 44:141–147. Witter, R., L. McKenzie, M. Towle, K. Stinson, K. Scott, L. Newman, and J. Adgate. 2011. Health Impact Assessment for Battlement Mesa, Garfield County Colorado. February 2011 Draft. Denver, CO: Colorado School of Public Health, University of Colorado Denver.