Enabling Rapid and Sustainable Public Health Research During Disasters: Summary of a Joint Workshop by the Institute of Medicine and the U.S. Department of Health and Human Services (2015)

8

Improving the Role of Extramural Research Networks

David Abramson of Columbia University’s National Center for Disaster Preparedness at the Earth Institute facilitated a discussion on models of extramural research networks that could be used to advance disaster science. Participants discussed the essential role for the academic and clinical research community and other partners in collecting data, data sharing, communications, and other priorities to enable timely research; multiple institutions working together as one entity; and the characteristics of an ongoing, sustained research network.

CONVERGENCE OF MULTIPLE TYPES OF DISASTER RESEARCH

Lori Peek, codirector of the Center for Disaster and Risk Analysis at Colorado State University, stated that social science disaster research likely emerged in the 1920s. The first known empirical study was a doctoral dissertation by Columbia University student Samuel Henry Prince, on the 1917 explosion of a ship in Halifax, Nova Scotia, Canada. His dissertation evaluated the behavioral response to the disaster that claimed many lives and resulted in great environmental destruction. The field was then dormant for about two decades, until World War II and the start of the Cold War sparked an increased interest in human behavior in the face of disaster. From 1949 to 1954, university-based field research teams, led primarily by sociologists, were funded by the military to study a variety of behavioral questions such as: Would people panic in the face of disaster?; Would citizens become so demoralized that they would become incapacitated and unable or unwilling to act?; Would there be civil unrest?; Would there be a

need for increased social control? The field teams observed that the answers to all of these questions were generally no, and they wrote extensively about how communities came together and how they shared information to build on preexisting community networks. These early field teams then moved on to study acute onset disasters. The military lost interest in funding these academic research centers, Peek said, but the National Academy of Sciences (NAS) recognized the importance of the work and supported these field research teams through the work of the NAS Committee on Disaster Studies from 1951 through 1962. In 1963, sociologists and disaster researchers, E. L. Quarantelli and Russell Dynes founded the Disaster Research Center, which, Peek noted, recently celebrated its 50th anniversary with a workshop on the state of disaster research and challenges for the future.¹

Development of Disaster Research Fields

Peek reported on some of the substantive consequences of the development of disaster research that were discussed at the May 2014 Disaster Research Center workshop. Overall, the field has been heavily focused on rapid onset disasters (and correspondingly there has been less focus on slower onset, more chronic types of disasters that also affect many people). The research has been predominantly U.S. focused (although the past decade has shown wider scope). In general, disaster research has used classical social science research methods to understand collective responses to disaster. The social science disaster research community has been focused on applied concerns and policy outcomes, Peek said. The theoretical base that is available is predominately grounded in sociology, with a focus on collective behavior and organizational response, and on demographic disparities and social vulnerability.

Peek noted that there has been tremendous growth in the field of social science disaster research over the past five or six decades. The field now incorporates natural hazards research; engineering, atmospheric science, computer science, and other technical fields; public health research; and science, technology, and society research. Having attended several recent meetings on disaster research, Peek observed that the same conversations and the same calls for action are occurring in each of these domains, in particular, the need for websites to compile information datasets, ready-made research protocols, and lists of experts (rosters) so that teams can be more rapidly assembled after a disaster (similar needs experts were asking at this current workshop). The question is how all of these domains can

_______________

¹See http://www.udel.edu/DRC (accessed December 18, 2014).

mobilize together in such a way as to influence national, state, and local policy to reduce disaster losses.

As an example of mobilizing diverse partners, Peek described the Social Science Research Council Task Force on Hurricane Katrina and Rebuilding the Gulf Coast. The thinking of the task force from the beginning was that Hurricane Katrina was too big of an event to conduct the usual one-off case studies and short-term studies that provide only a snapshot of a particular element of a tremendous event, she said. Hurricane Katrina was such a multidimensional event that there was no one person and no one discipline alone that could truly understand it. Similarly, disaster research could benefit from multiple disciplines working together, combining both clinical and social sciences.

Following a series of meetings along the Gulf Coast with researchers, workers on the ground, and community members, the task force developed a program of 10 distinct studies spanning an array of topics from risk communication, to environmental impacts, to displaced populations. Peek explained that the studies were independent but done in conversation with one another. The Social Science Research Council provided support for a website where researchers posted profiles of their work so others could find who was doing what research on Hurricane Katrina by a keyword search. The council also provided briefings or short bulletins to inform the practice communities about the research findings in a timely manner. The long-term vision is to produce the Katrina Bookshelf, including books on each of the independently funded projects together in an edited collection. Peek noted that the first book on population displacement after Hurricane Katrina has been released, three more are in press (children in Katrina, cultural trauma, environmental and community impacts), and the remaining books are in development.²

_______________

²See http://utpress.utexas.edu/index.php/books/series/series/The-Katrina-Bookshelf (accessed December 18, 2014).

ADVANCING SCIENCE DURING CRISIS

Gary Machlis, coleader of the Strategic Sciences Group (SSG) at the U.S. Department of the Interior (DOI), discussed characteristics of, and recommendations for, advancing science during crisis (i.e., while it is happening, not in preparation for or in response to). As background, he explained that the SSG was formed by secretarial order following an experimental deployment during the Deepwater Horizon disaster. The mission of the SSG is limited to conducting interdisciplinary, science-based assessments and scenario building during a major environmental crisis and delivering the results and potential interventions to decision makers. The SSG uses interdisciplinary teams of both federal and nonfederal personnel. Machlis explained that the SSG drew some of its organizational principles from the research and development division of the World War II intelligence agency, the Office of Strategic Services. Those principles include focus on the mission, not the process; build operational teams based on expertise, not representation, and on skill, not rank; and have direct access to leaders and decision makers. Machlis also noted the value of having personnel with both physical and mental strength when working in disaster research.

Characteristics in Crisis to Include

Based on the work of the SSG, Machlis outlined six distinctive characteristics of science during crisis that could be included in a research framework focused on health.

• Coupled human and natural systems. Science in crisis is to inform response, and responses with significant consequences require coupled human and natural systems approaches. A purely biophysical or purely sociocultural approach is ineffective. The ability to deliver actionable recommendations is limited by science in silos because real-world decisions transcend human and natural systems.
• Collaboration and interdisciplinary teams. A coupled human and natural systems approach relies on interdisciplinary teams. However, a disaster often brings together people who do not know each other or who do not have a long history of working together. The critical challenge is to create, within hours, a working interdisciplinary team of the best expertise available. Team

• members must work using the same vocabulary; suspend their own territorial claims, disciplinary turf, and paradigms of thought; and work for the common good.

• Uncertainties and limitations. It is not sufficient to deliver findings to decision makers unless they are accompanied by a clear presentation of the uncertainties and limitations of the findings, Machlis said. The SSG uses a scientific scale of uncertainty based on that of the Intergovernmental Panel on Climate Change.
• Cascading consequences and assessing impacts. Crises often have an immediate need for tactical science (e.g., how to cap a leaking oil well), but it is important to remember that decisions regarding the emergency response will influence long-term restoration as well. There is a need to understand the cascade of consequences of each decision, each of which has its own uncertainty.
• Sense of place. While crises have commonalities, every crisis is distinct from the next, and all disasters are local. Having a sense of place is vital to success, Machlis stressed. It is not enough, for example, to have demographic data on the populations or neighborhoods that might be affected. What is required is an understanding of cultural history and a visceral sense of place.
• Communicating science during crisis. Communication during crisis requires extraordinary clarity and concise explanation of the findings, uncertainties, and implications. This is much more important than the literature review, background, or methods. Compelling visualization is also essential to convey the message (e.g., maps, graphs, figures, charts). In addition, when communicating with leadership, researchers must speak the truth and be transparent, without ambushing them through public attention seeking.

Machlis concluded by offering four recommendations for advancing science in crisis:

1. Identify best practices of science during crisis.
2. Advance systematic rostering by learning from others who use it (e.g., the U.S. Forest Service for fighting wildfires).
3. Seek legislative relief from the Federal Advisory Committee Act (FACA) provisions that prohibit or impede federal/non-federal scientific collaboration in a disaster. A presidential disaster

4. declaration should trigger a FACA exemption for certain forms of scientific work.

5. Prepare leadership before a crisis to be ready to integrate science into decision making during a crisis.

PHARMACEUTICAL RESEARCH DURING A DISASTER

Paul Seligman, executive director for Global Regulatory Policy at Amgen, offered a pharmaceutical company perspective on disaster research. A pharmaceutical company that sponsors a product for U.S. regulatory approval first conducts or sponsors clinical studies of the product and then uses that data (and possibly data from others) to assemble a dossier and submit an application for marketing approval to FDA. Sponsors develop the product labeling and conduct additional postapproval studies. With regard to disaster preparedness, sponsors provide and maintain products for the Strategic National Stockpile (SNS), both prespecified quantities purchased by the SNS as well as product for surge capacity and stock rotation to maintain maximum shelf life. He reminded participants that under the Animal Rule, FDA can approve a drug or biologic product based on substantial evidence of efficacy from studies in animals when efficacy studies in humans are not ethical or feasible.³

Seligman noted that companies have a lot of information about their products that may not be in the public domain, for example, analytics particular to measuring levels of the product; additional data on product parameters (e.g., pharmacokinetics, pharmacodynamics, genetic testing/susceptibility, biomarkers); or data from clinical trials regarding other potential indications, specific treated populations, or other comparators, including historical controls. This means, Seligman suggested that pharmaceutical sponsors may have baseline information of interest to disaster researchers, albeit in the context of the development of their product for an indication that is used in medicine generally.

The immediate response to any event will be managed by public health, the medical field, law enforcement, and the disaster and response infrastructure, Seligman concluded. While the sponsor’s role is generally to ensure supply of medicine, there are a variety of opportunities for collaboration with pharmaceutical product sponsors before, during, and after a disaster. A sponsor’s role extends, for example, to the development of the postevent study protocols, contribution of unique analytic

_______________

³21 CFR 314.600 (drugs) and 21 CFR 601.90 (biologics).

capabilities, involvement in the conduct and analysis of studies, and patient communications, but they may not realize this until engaged and asked for assistance.

RESPONSE TO ENVIRONMENTAL HEALTH IMPACTS OF DISASTERS

Sharon Croisant, director of the Community Outreach and Education Core at the University of Texas Medical Branch (UTMB) Center in Environmental Toxicology, shared several examples of UTMB’s evolving involvement in research efforts in the field in direct response to disasters. The research efforts of the Community Outreach and Education Core after Hurricanes Isaac, Katrina, and Rita started simply, she said, by contacting Red Cross and community shelters to compile lists of supplies that people needed (e.g., water, mosquito repellant, adhesive bandages, antibiotics) and then collecting and delivering those supplies. This grew into informal needs assessments in the communities, which turned into a year-long project to assess community resiliency and preparedness. Similar needs assessments and relief and response efforts were done in Galveston after Hurricane Ike. In addition, there was a community-based participatory research project focused on assessing toxins in the posthurricane sediment sludge. Croisant noted that when the waters receded, sludge was left behind in three-quarters of the residential and commercial buildings on Galveston Island. Much of the floodwaters came from the bay, which she said is contaminated with pollutants from Superfund sites, an old creosote plant, a sewage plant, a sulfur repository, and the port. The findings (toxins including arsenic, cadmium, chromium, and others) were important not only for the residents, but also for those involved in the cleanup (tens of thousands of people on the island mucking out homes and doing repairs, including 5,000 college students who came to Galveston over a period of 2 years to help clean up). Safety training sessions were held for the volunteers, and an educational series on mold and lead paint was also developed. Croisant added that materials were translated into Spanish, as many of the workers and families did not speak English.

Creating a Consortium

As a result of its work in the aftermath of the hurricanes, UTMB had established relationships with the Gulf Coast communities of Texas.

Croisant said that after the Deepwater Horizon explosion and oil spill, those communities began calling and expressing concern about the conflicting information they were hearing from the White House, the media, the National Oceanic Atmospheric Administration, and others. UTMB conducted scoping visits to better understand their fears. Efforts were made to find and report back with answers, and an NIEHS U19 consortium project was established to study the health risks related to the spill. The consortium included academic institutions as well as community groups from the impacted areas (e.g., Vietnamese Fisherfolk, United Houma Nation, Alabama Fisheries Cooperative, and Bayou Interfaith Shared Community Organizing). The mission of the consortium was to explore the health impacts and the community resiliency related to the Deepwater Horizon disaster by fostering collaborative interactions among multidisciplinary, multi-institutional, basic, and clinical investigators, supported by active involvement of community partners. Because the project involved seafood sampling, those partners included, for example, shrimpers, commercial fishermen, and people running the seafood processing market. Croisant noted that around 350 fishermen were trained to collect samples. A total of 24 community meetings revealed that people were most concerned about the safety of the seafood. They needed to know definitively if it was contaminated and could not be sold or eaten, or if it was safe and how to get that message out to everyone. The consortium is studying the presence and toxicity of petrogenic polycyclic aromatic hydrocarbons, evaluating exposures and longitudinal outcomes, and disseminating findings to the community stakeholders.

Barriers Encountered

Croisant highlighted several challenges to getting out into the field in a timely manner. Funding must be repurposed or solicited and may involve multiple industry and agency stakeholders. Bureaucracy is an impediment, she said; for example, getting a new vendor approved when a suitable vendor is not already in the system takes too much time. The lack of a communications infrastructure means that information is fragmented and intermittent. Coordination among local, state, and federal agencies sometimes lacks infrastructure and thus impedes communication and integration of efforts. Communities are rarely prepared for disasters, and poor communities are the least resilient, Croisant added.

Another concern is that emergency responders generally lack training on possible environmental exposures (e.g., toxins in the sediment sludge

after Hurricane Ike). In addition, emissions or spills resulting from flooding or accidents are frequently not identified until much too late to protect public health. Croisant also noted that there can be tension between and among community members and groups, industry, science, and government that can intensify in emergency situations. Technology can be a barrier, particularly when it is not available as a result of the disaster. For example, cell phone service is often interrupted, and fewer and fewer people have landlines. Internet service can also be disrupted, limiting access to emergency information generally found online (e.g., Material Safety Data Sheets) or limiting the ability of people to sign up for services and assistance online.

Moving forward, UTMB has been working on developing a partnership with the Galveston County network of regional Emergency Management representatives, leveraging emergency management training and integrating UTMB Center in Environmental Toxicology resources.

CHALLENGES AND OPPORTUNITIES

In summarizing the panel discussions, Abramson said that a framework for an extramural research community collaborative could include multidisciplinary strategic science teams, consortia, suppliers, and intra- and interdisciplinary research networks (see Box 8-1). It is important to recognize that many networks already exist, Croisant said, and she recommended working to integrate those. There is also already a cadre of scientific experts at NIEHS core centers, Clinical and Translational Science Award institutes, and other centers. The question is how to identify those that would be appropriate and willing to participate in a disaster response research network. As part of networking for preparedness, Machlis said, researchers need to learn about the scales larger and smaller than the one at which they work. For example, an ecologist might work with a watershed area, or an ecosystem, or a region, or large landscape, or a sociologist might work at the level of a family, or neighborhood, or larger community.

When a crisis happens, it is valuable to know how we are connected to those working at the scale above and below. A key challenge is identifying who can bring all of these entities together and coordinate efforts. It was discussed that coordinating bodies could be the mission agencies (e.g., CDC or DOI), funders (e.g., NIH, NSF, foundations), or regulators (e.g., FDA).

Seligman raised a concern that it is not clear what agency or group a pharmaceutical company could engage to ensure that adequate information on benefit and safety is going to be collected when the company’s product is used in a disaster response. Industry is developing medical countermeasures with an FDA requirement to have protocols in place so that when those products are used, data are captured regarding effectiveness and safety of the product. Where in the public health community or in the academic community would a pharmaceutical company turn to ensure that capturing that kind of information is included in protocols and instruments that are

being developed? It was suggested by some that there be a more focused discussion, perhaps a follow-up workshop, on the issue of collecting postmarket product data in a crisis response.

Irwin Redlener, director of the National Center for Disaster Preparedness at Columbia University, said there is no public health infrastructure in the United States that has an authoritative overview of what the country’s priorities are relative to nearly every aspect of preparedness, including how to ensure the collection of data on the use of pharmaceutical products in a crisis. There is no infrastructure that allows the government to answer a broad, important public health question simply, clearly, definitively, and quickly. He suggested that one reason we keep repeating the same mistakes and not learning lessons is a lack of a fundamental structural organization for dealing with disasters.

Moving forward, a participant called for regulatory relief from FACA and other provisions that might impede scientific collaboration in a disaster. Abramson also called for pilot-testing the system before the disaster occurs, setting up small collaboratives at multiple levels and working on fictional scenarios and directed tabletop exercises to become well practiced at rapidly bringing partners together across disciplinary spans. It was also suggested that it might be possible to crowdsource solutions, engaging not only the science community but the greater public.

This page intentionally left blank.