During the workshop, a panel of scientists was convened to discuss the research opportunities and needs for understanding health impacts of hydraulic fracturing. The presenters were asked to reflect on the evidence presented the previous day and identify the next steps to minimize health effects particularly as technology evolves. Further, the panel was asked to identify in their opinions where there is uncertainty in the available evidence. The purpose of the panel was to elicit a range of viewpoints, but not to reach a consensus. The following section is a summary of the initial presentations and the discussion among the panelists.
David Carey, Ph.D.
Director, Weis Center for Research
Geisinger Health Center
David Carey provided background on the Geisinger Health System, located in Danville, Pennsylvania—part of an area affected by hydraulic fracturing. He described that leadership and staff have a vested interest in the health outcomes occurring regionally related to the use of hydraulic fracturing. In response to increased activity, the health system assembled a coalition of stakeholders and experts from academia, other health care systems, government, and industry to conduct a coordinated, multidisciplinary research project. Carey stated that the focus of the project is to collect data to inform a scientifically rigorous assessment of regional health impacts. At one level Geisinger is creating a database of health information on patients in the region that could be used for broad surveillance and analysis to identify areas of concern and “hotspots” in terms of adverse health outcomes. These data could then be used as the basis for performing more focused hypothesis-driven studies. Combining
this research with geographic information systems mapping, patient and clinical data can be located in both space and time.
Geisinger has a number of key system attributes that position it to be ideal to play a central role in this effort, including being a highly integrated health care system and providing patient care to a large number of patients covering 31 counties, including areas where intense drilling activity is occurring. The integrated system comprises 4 hospitals and 40 community-based clinics that provide both primary care and specialty care, and Geisinger also operates as a health care insurance company. This degree of integration allows for cross-disciplinary research along the health care front and to better serve the health needs of the patients in the region. Further, Geisinger was one of the first adopters of electronic health records for both inpatient and outpatient data in the country. The system has comprehensive data on hundreds of thousands of patients beginning in 2004. The data are stored in a clinical data warehouse, a more easily searchable and mineable platform.
Additionally, the Geisinger Health System has become a leader in the Keystone Health Information Exchange and the Beacon Community Project which electronically links health care providers in the region. Dr. Carey said that Geisinger is initiating an effort to expand this collaboration further to engage other health care systems and other providers in the region to participate through this health information exchange network. With the ability to collect larger sets of data, standardize the data, and blend them into the searchable database, longitudinal data mining could be possible. Understanding the health impacts is a complex problem. Dr. Carey described that although the focus of the presentation at this workshop is on health impacts, he believes that there is also a need to incorporate information on environmental assessment research.
Rob Donnelly, M.B.Ch.B., MFOM
Vice President of Health
Royal Dutch Shell
Royal Dutch Shell is engaged in discussions such as this workshop, according to Rob Donnelly, in order to be at the table and participate in learning of potential health impacts related to its work. Dr. Donnelly provided an overview of the approaches Shell is using to engage in discussions with Shell communities, who they recognize have real concerns.
Dr. Donnelly noted that as stated earlier, natural gas is a part of the energy mix for the future. Royal Dutch Shell’s focus, therefore, must be on conducting its work right. The company has issued a set of operating principles for tight gas production where hydraulic fracturing is used. The operating principles were shared publicly in 2011 and continue to be adopted across Shell’s global onshore operations. The principle focus is
on safety and well integrity, water, air, footprint, and community. Dr. Donnelly stated that although there is a need for additional research, these focus areas highlight what can be done currently to protect human health based on available information while working to ensure growing energy needs.
Incidents of the last 2 years have highlighted the primacy of well integrity for any hydrocarbon operation according to Dr. Donnelly. Each of Royal Dutch Shell’s wells is individually designed and pressure tested before being put into production. He went on further to state that the company focuses on water recycling and protection. Dr. Donnelly emphasized that Royal Dutch Shell works with local communities, to understand water supply issues in the area and to ensure that corporate decisions take into account the entire water resource in an area. For air quality and reducing air emissions, he described that the company’s effort to add fitted catalytic technology to diesel generators in some areas. According to Dr. Donnelly, the company establishes an interactive dialogue early on with the community to discuss local concerns, for example, rerouting delivery trucks to limit traffic around schools or through the middle of town. The company’s community principle utilizes health impact assessment to identify socioeconomic impacts. Incorporating these simple things in addition to research studies is a key approach to obtaining gas safely, according to Dr. Donnelly.
Bernard D. Goldstein, M.D.
Professor Emeritus, Department of Environmental and Occupational Health
Graduate School of Public Health, University of Pittsburgh
Bernard Goldstein began by describing, in his opinion, the unfortunately typical progression of environmental and occupational issues related to human health. A new technology is not adequately evaluated for potential adverse health consequences. Public concern follows that results in calls for investigation of a potential causal relationship between the new technology and adverse health consequences, which is then usually hampered by inadequate exposure and toxicity information to perform a retrospective analysis. Dr. Goldstein identified barriers: many changes occur rapidly over time, disease clusters occur whether causal or not, litigation occurs, and message control can be harmful in terms of getting research efforts started. The end result is usually that solutions are delayed. Over time, industry will find technologies to reduce pollutants, which is ultimately in their best interest.
Specific to hydraulic fracturing, Dr. Goldstein agrees with earlier statements that the public is confused and concerned. For instance, is hydraulic fracturing old or new technology and does it cause groundwater
contamination? Some statements from the government and industry might suggest that it is a new technology that now allows extraction of natural gas. However, in response to questions about health and safety issues, the message from the same sources is not to worry because the technology has been around for more than 40 years. It cannot be both, Dr. Goldstein remarked. Second, the public receives conflicting information about whether hydraulic fracturing causes groundwater contamination. The focus too often is on whether the successful release of hydraulic fracturing agents 5,000 feet underground will cause groundwater contamination, which is not responsive to the public’s question of whether water pollution will occur over the entire process from the development phase through 30 years from now when the well is no longer active. Dr. Goldstein suggested that the process likely will not affect surface water if chemicals are released 5,000 feet underground, but there is a possibility of chemicals seeping into groundwater when casings blow, drums leak, and trucks spill. Dr. Goldstein noted that three advisory committees, established in 2011 by President Obama and the governors of Maryland and Pennsylvania to examine the issue of unconventional shale gas drilling, have in their executive orders requests for advice on the protection of public health. But, of the 52 members appointed to the three commissions, not one has a health background. Not surprisingly, very little has come out of their advice regarding health research. For example, in Pennsylvania, a bill passed after advice from this commission includes 17 different state agencies that will receive funding allocations from the Pennsylvania impact fee, but not one penny went to the Pennsylvania Department of Health to develop standard public health surveillance or to fund research. Similarly, in President Obama’s Advisory Commission, the lead is given to the Department of Energy with input from the U.S. Environmental Protection Agency and from the Department of the Interior. The Department of Health and Human Services is not included (Goldstein et al., 2012).
Finally, Dr. Goldstein expressed that the disclosure of the chemicals used in hydraulic fracturing has not been transparent. There has been a significant increase in information released about the hydraulic fracturing chemicals with the adoption of the recent laws. However, there are still inappropriate exceptions for confidential business information. Of greater toxicological concern than the hydraulic fracturing chemicals are the chemicals brought up from underground that need to be disposed. These include naturally occurring brine constituents, radionuclides, and other potentially toxic agents. In Pennsylvania, the exceptions to releasing chemical information are stated as follows:
“Notwithstanding any other provision of this chapter, a vendor, service provider, or operator shall not be required to do any of the following:
- Disclose chemicals that are not disclosed to it by the manufacturer, vendor, or service provider.
- Disclose chemicals that were not intentionally added to the stimulation fluid.
- Disclose chemicals that occur incidentally or are otherwise unintentionally present in trace amounts, may be the incidental result of a chemical reaction or chemical process or may be constituents of naturally occurring materials that become part of a stimulation fluid.”1
These exceptions result in health scientists not having access to necessary information. In summary, the issues that Dr. Goldstein described as being of most concern to toxicologists are the agents used in hydraulic fracturing; natural gas hydrocarbons, the naturally present agents brought to the surface in flowback water (e.g., arsenic, brine components, radionuclides), and the effects of mixtures of these agents and reactants.
Roxana Witter, M.D., M.S.P.H.
Assistant Research Professor,
Environmental and Occupational Health
Colorado School of Public Health
Roxana Witter described a number of areas where community impact research could be done. There are preliminary reports that link community impacts to natural gas developments, but the effects on public health are uncertain. Sociological and socioepidemiological literature demonstrates that social environment can affect health and this literature may suggest appropriate methodologies for future research. She added that there is some initial evidence for social impacts as a result of population influx, including traffic, noise, rise in sexually transmitted infections, changes in economic conditions, and quality-of-life effects that will likely affect individual and community health. Dr. Witter suggested that looking into the longitudinal impacts pre- and post-extraction by looking at similar metrics, such as sexually transmitted infections, crime, substance abuse, and so forth, and looking at impacts from population influx would be revealing. Further, surveys could be used to understand social cohesion and social capital impacts at a community level and to direct community interventions. Another method she suggested could utilize the stress end points and identify potential markers, such as heart rate, cortisol, and Creactive protein. The purpose of these surveys could be to assess perceived stress (measuring acute stress) and affect (measuring chronic stress). Further, surveys could be used to look at decreased exercise, increased substance abuse, and so forth to understand the exposures that cause stress. Finally, Dr. Witter suggested that a variety of individual
1 Pennsylvania Consolidated Statutes, Title 58, Oil and Gas, § 3222.1(c), Disclosures not required.
health end points, such as cardiovascular and pulmonary, cancer, mental health, and integrating stress metrics and disrupted community metrics, could be incorporated into these studies to understand the link between environment, stress, and health.
After hearing the panel discussion, the individuals were asked to comment on suggested next steps. Dr. Goldstein noted that environmental health scientists need to be more involved in policy making and that they must be more emphatic about the importance of human health. This includes gathering data and documenting the fact that the public has a concern about their health and beginning a discussion with all stakeholders. Dr. Carey emphasized that there is a need to harness clinical data and combine this information with location. This approach could identify the hotspots for further investigations. He put forth that there is a need to move beyond anecdotal case reports to collecting systematic data that can be analyzed rigorously. Dr. Goldstein added the critical need for a National Institutes of Health study section to fund this type of research as opposed to enforcing new laws that would not inform researchers about the toxicological effects. Dr. Jackson added that funding needs to be front loaded so that when the inevitable clusters are identified (such as a cancer cluster, birth defect cluster, or neurological cluster) public health researchers will have initial data to launch investigations. He expressed concern that waiting until the clusters happen and trying to do a retrospective study is far too difficult and often inconclusive.
The discussion turned to the value of electronic health records and the problem that occupational history is often not recorded by health care professionals. Dr. Carey noted that even in the Geisinger Health System, employment data are collected, but not systematically, and therefore, there are gaps. He suggested that one way to fill the gaps is to use supplementary data collection modes. One participant noted that another weakness of the health records is that they are not the best source to assess the linkages between stress and health outcomes because some of these items are preclinical.
Dr. Donnelly discussed the role of industry and the need to create partnerships between vested parties. He sees a need to have a continued dialogue with the community and other stakeholders to define the research questions. He noted the need to have credible third parties to conduct research. Dr. Jackson and others noted that the Health Effects Institute and the Public Health Institute are good models that should be explored to begin research in this area. Dr. Donnelly added that Royal Dutch Shell is committed to safe operating practices, and they are also open to being informed by the research and dialogue in order to make changes. He indicated that there is also a need to continue to identify
alternative chemicals that may offer better solutions and safer community-based solutions. Dr. Donnelly concluded that Royal Dutch Shell has called for transparency of the chemical composition of hydraulic fracturing fluid. The company participates in the fracfocus.org website where individuals can review the chemicals that are being used in their wells.
Goldstein, B. D., J. Kriesky, and B. Pavliakova. 2012. Missing from the table: Role of the environmental public health community in governmental advisory commissions related to Marcellus Shale drilling. Environmental Health Perspectives 120(4):483–486.