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)

5

Partnering with the Community to Enable Research

As mentioned previously in Chapter 3, engaging the community to become involved in disaster research is a goal of many researchers, yet successful access still has many challenges. Often these are due to outside research teams unfamiliar with the communities entering after a disaster with their own agendas in mind and a lack of communication with and commitment to the people they are studying. Jack Herrmann, senior advisor and chief of public health programs at the National Association of County and City Health Officials (NACCHO), and Joseph “Chip” Hughes, program director for the Worker Education and Training Branch at NIEHS, facilitated discussion on engaging community and citizen scientists in disaster research. Participants considered strategies to strengthen the interface and collaborations with first responders and emergency management, health departments, workers, and others to promote successful disaster research.

PARTNERING FOR EPIDEMIOLOGY RESEARCH

Texas leads the nation in the number of presidentially declared disasters, said David Lakey, commissioner of the Texas Department of State Health Services (DSHS). From 2008 through 2013, response events in Texas have included, for example, Hurricanes Alex, Dolly, Eduard Gustav, and Ike; the Yearning for Zion Ranch evacuation of children from polygamist families because of abuse allegations; the West fertilizer plant explosion; wildfires; the H1N1 pandemic; West Nile virus; food-borne outbreaks; epidemiological investigations (e.g., tuberculosis,

mercury poisoning from cosmetics); investigation of compounding pharmacies; and unaccompanied minors crossing the Texas–Mexico border. The Texas DSHS is responsible for the co-ordination of health and medical response.

Timely information after a disaster is essential, and Texas DSHS uses the Community Assessment for Public Health Emergency Response (CASPER) tool, an epidemiologic technique to rapidly provide low-cost, household-based information about community needs. Following a disaster, teams conduct door-to-door household surveys about public health needs and emergency management issues such as health status, access to utilities, access to health care, mental health issues, evacuation behaviors, messaging, and recovery needs. Lakey noted that the CASPER system has been critical for improvement of disaster planning and response in Texas. In response to a question about the IRB review, Lakey noted that CASPER is a tool being used in realtime to get information for disaster response improvement. This is an example of “Disaster Epidemiology,” which can be done more quickly and span topic areas more widely than traditional research often does. For other questions that are longer term, full IRB approval is received beforehand. A participant added that there is a process before the survey that explains who is doing the survey, why, what they are doing, and what will be done with the results; people can then choose whether to participate.

State Health Department Research

With the goal of continuous quality improvement in mind, the Texas DSHS has published nearly 50 peer-reviewed articles on public health preparedness over the past decade, Lakey said.¹ These are written by DSHS regional and central office staff from a variety of programs, often with academic partners as coauthors. Lakey stressed the value of partnering with academia in the community, noting that Texas DSHS has collaborated with Texas A&M, University of Texas, and University of North Texas. Academic partners can help to provide baseline data, conduct studies, facilitate stakeholder meetings, provide surge capacity (e.g., staff, space for shelters), assess training needs, provide technical assistance, and collaborate on publications. Texas DSHS has also developed formal internship opportunities for students, and a residency program so that preventive medicine residents can gain experience in the

_______________

¹See http://www.dshs.state.tx.us/commprep/publications.aspx (accessed December 12, 2014).

field of public health during a disaster. Lakey added that after-action reports are also an essential component of continuous quality improvement, but they are not always accessible to a broader audience. He shared an example of what he described as a negative research-related experience. In 2008 during Hurricane Ike, he received phone calls from people he did not know who were seeking information for articles they wanted to publish in the media. It was clear to Lakey that these requesters had obvious biases for how they planned to present their story.

Disaster Epidemiology

Lakey also shared several examples of how disaster epidemiology research has been effective and helpful in response (also discussed further in Chapter 6). A surveillance study tracking injury deaths related to Hurricane Ike was intended to identify strategies to prevent or reduce hurricane-related mortality in the future. Working with partners, data were collected from a variety of sources, including medical examiners, justices of the peace, coroners, forensic centers, hospitals, and regional epidemiologists. The majority of the 74 deaths reported were indirectly related to the hurricane, suggesting that the evacuation of people out of the disaster zone worked well. However, many succumbed to carbon monoxide poisoning in their temporary quarters (from inappropriate use of stoves and heaters) or were hit by falling trees as they tried to clear their land. This is important information for how we approach disasters, Lakey said, because it demonstrates the need to not only evacuate people, but to give them information on how they can remain healthy in the aftermath.

The epidemiological investigation of the West fertilizer plant explosion in 2013 is an example of collaborations among local, regional, state, and federal agencies. The local public health department was the lead agency, which helped to ensure that the community was involved in the research being done in their community, Lakey explained. Texas DSHS provided the technical expertise, and additional resources and best practices were shared by CDC and other agencies. Using data from the medical examiner, death certificates, hospitals, urgent care clinics, and survivor interviews, Lakey and colleagues characterized the injuries, resource needs and distribution, and communication priorities for affected individuals to help inform practice in future disasters.

Success in these cases and others is dependent on relationships. As many responders know, to be most effective, partners need to get to

know each other before an event, Lakey said. This applies to disaster research as well, and he noted that, based on experience, he is very cautious about engaging with a previously unknown outside entity that appears during disaster response wanting to do research. He offered several relationship-building lessons:

• Partner early with stakeholders in the investigation-forming process to solicit expectations.
• Combine efforts to make an investigation more useful to stakeholders and maximize resources.
• Continue and expand pre-event partnerships to aid in collaboration during and after an incident.
• Develop nontraditional partnerships for public health (e.g., with the Texas State Fire Marshal).

In closing, Lakey said that in conducting research during disasters, it is important to partner with trusted researchers and to have clear, agreed-upon objectives, developed with input from the appropriate stakeholders (public, private, local, state, and federal). Research should be coordinated with the ICS, which can help avoid duplication of effort. There are lessons to be learned from community-based participatory research, including understanding what is important to the community and, as Abramson previously noted, the need to disseminate results back to the community. Researchers also need to be aware of the politics that may play into the overall response. Overall, be sensitive and respectful to those we are trying to serve, he concluded.

ESTABLISHING TRUST IN THE COMMUNITY

Stephen Bradberry, executive director of the Alliance Institute, expanded on the concept of maintaining an ongoing connection with the community, previously emphasized by Goldfrank during his remarks about hospital populations during Hurricane Sandy. When a disaster happens and researchers need the help of the community, Bradberry also reiterated that it is important to have trust established with the community. He pointed out that distrust comes from years of people coming into the community to collect data and then leaving, never to be heard from again. Bradberry noted that even well-intentioned researchers often focus on high-quality medical services and community-centered

health homes while neglecting the community environment. The community environment has many resources, including the residents, patients, and networks, and their experience with, for example, advocacy, community meetings, and mobilizations. The community has a wealth of knowledge regarding how they, as a community, can best be engaged, and what serves their needs in their environment. The people living in a particular area are often very knowledgeable about the dangers they face by living in that region (e.g., oil leaking from wells), and about the best ways to deal with a situation, but they are rarely asked. Instead, they are given recommendations and told how things will be done, which can lead to resistance and resentment.

As an example, Bradberry recalled the comment by Birnbaum that the NIEHS GuLF Study (on the health impacts of the Deepwater Horizon oil spill) has had challenges recruiting participants and is about 20,000 participants short of its goal. Bradberry suggested this is due to “an extreme lack of trust.” He said that during the community meetings when the study was outlined, the community suggested that their immediate need was for health centers because people were sick and needed care. However, that was not the path that was chosen, he said. In addition, people were aware that BP Global had full control of the media and of the recovery. They were being asked to trust NIEHS, which is part of the same government that had put the party responsible for the accident (i.e., BP) in charge of the recovery.

Regarding presentation of information to the community, Bradberry said people in the community are most interested in learning what actions they need to take. The suggested actions should be up front, he said, with the data following after for those who may be interested. This is opposite of how scientists usually present information (with data first and recommendations at the end). Including the community in the work and sharing the resulting information can alleviate many of the challenges associated with data collection and obtaining baseline information in past disasters.

PARTNERING WITH UNIONS, WORKERS’ ORGANIZATIONS, AND WORKERS

Depending on the disaster setting, multiple sets of worker populations are involved, said Craig Slatin, principal investigator and director of The New England Consortium, based in the College of Health

Sciences at the University of Massachusetts Lowell. If the site is a plant (e.g., the fertilizer plant in West, Texas), facility workers may be on-site. There will also be a broad range of emergency responders; emergency management, public health, medical, and social workers; skilled support workers; workers involved in the cleanup and in construction labor; and others.

Although much of the focus in a community impacted by a disaster is on the safety and health of the residents and businesses, those involved in cleanup should be monitored as well. Recovery and remediation work is often done by low-wage workers. For example, following Hurricane Katrina, much of the work was done by immigrant day laborers. Slatin described a study conducted jointly by the University of California, Los Angeles, Labor and Occupational Safety Health program, and the National Day Laborer Organizing Network (NDLON) of health and safety issues for the Latino migrant laborers in the region. The study found that immigrant workers were gutting buildings, cleaning up debris, and tearing out moldy sheetrock from flooded houses, mostly without any protective gear. They were then going “home” after work (often an abandoned car or a shelter with nowhere to wash) to sleep in their work clothes, or paying to sleep in soaking-wet tents pitched in a muddy field at City Park, where they had to pay an extra fee to use the shower. Without this study, it is likely much of this would have gone on unnoticed. While they may not have been affected by the immediate disaster, recovery workers present an important demographic that could also benefit from environmental health monitoring to ensure they are able to safely perform their job duties.

Understanding Worker Needs and Leveraging Local Knowledge

When performing research in a community, it is important to have workers who understand the needs and nuances within that community in order to access important information, as Lakey alluded to previously, especially following a disaster. However, recruiting these types of local workers presents difficulties, so it is valuable to understand the needs workers may have throughout the process. In conducting research to evaluate the health and safety hazards affecting immigrant workers in cleanup, Tomas Aguilar from NDLON faced a variety of challenges. Those challenges included being unfamiliar with the area and dealing with initial distrust and wariness from the workers (who were facing

harassment from the police, immigration agents, and other workers). Slatin relayed that Aguilar tried to supplement worker interview data by observing worksites, applying for jobs, attempting to obtain protective equipment, and generally putting himself in similar situations as the workers. He discovered that the Red Cross/FEMA site was only for local residents, so he could not get assistance. Regardless, they did not have any of the protective equipment he was seeking. A nearby relief organization also had no equipment to spare. Contractors were generally hostile and suspicious and would not talk with him, and he faced obstructions when trying to observe the worksites.

Worker Safety and Sensitivities

As another example, Slatin referred to the DR2 tabletop exercise introduced by Birnbaum in the plenary session. The activity, held at the port of Los Angeles in April 2014, was designed to develop a concept of operations for NIEHS in preparation for deployment of a disaster research team. During discussions on transition (i.e., how to sustain research efforts or undertake longer-term research), those at the exercise emphasized the need to maintain the collaboration among trainers, researchers, local and state agencies, workers, and communities. Workers and their representatives participating in the DR2 exercise cautioned that workers might be partners and help to gather information for disaster research, but they also may be at risk of retaliation for disclosing information that employers do not want disclosed. Workers are also sensitive to how the research might impact their jobs. They live in these impacted communities and want their families and community to be safe, but they also want their jobs to be secure.

During the exercise in Los Angeles, participants identified several worker/community research priorities relative to the exercise scenario (earthquake-induced tsunami leading to refinery fires): have clear knowledge of what chemicals are released and an active monitoring system to protect the health of workers and community members; ensure that workers, first responders, and community members have the appropriate protective equipment; know the health and safety issues for workers and how to appropriately train them; understand the physical and psychological impacts that the event may have on communities, response workers, and refinery workers; and create a registry of those exposed.

Slatin noted that the NIEHS Worker Education and Training Program has a network of worker trainers. These worker trainers and organized labor can assist with connecting to the community and collecting data and can help build capacity and confidence in communities.

In closing, Slatin concurred with others that a key challenge for disaster research is establishing relationships before a disaster. We do not know where disasters are going to occur, and there is not funding for the development of partnerships and relationships, Slatin added, which can be fundamental for conducting a collaborative, successful study. Based on his personal experience, it is difficult to get the support of one’s institution to spend the time and resources to go into the community and build these relationships in the case that an incident occurs to precipitate the need for research.

UNDERSTANDING THE NEEDS OF THE COMMUNITY: THE DISASTER PSYCHOSOCIAL ASSESSMENT AND SURVEILLANCE TOOLKIT (DISASTER-PAST)

Anthony Speier, associate professor at the Louisiana State University Health Sciences Center, described the Disaster Psychosocial Assessment and Surveillance Toolkit (Disaster-PAST), developed to better understand the “who, what, where, when, and how” of recovering communities through surveillance of community mental health and psychosocial functioning following disasters.² Tracking basic demographic information, such as where people are living before a disaster, can help in planning for where services will be needed when people start returning after being displaced by the disaster. He concurred with Bradberry that disasters are highly politicized, and it helps to have a more quantitative and objective method of assessing where the needs are, and what services should be added to address those needs.

Data-informed knowledge helps determine what level of service is needed and allows funding sources to direct an appropriate level of services to those needs, Speier said. Disaster-PAST can also help to identify which populations are most in need of mental health services after a disaster and can identify risk factors for developing certain types

_______________

²The toolkit was designed for use by any agency or entity and is free and publicly available. See http://www.medschool.lsuhsc.edu/psychiatry/disasterpast_contents.aspx (accessed November 10, 2014).

of mental illness following a disaster. The toolkit also provides recommendations of when to conduct the assessment and surveillance following a disaster, including guidance for ongoing evaluation and long-term surveillance of mental health needs over time. The main purpose of the toolkit is to provide guidance on how to conduct psychosocial surveillance after a disaster (e.g., screening tools, sampling) and how to use that information to inform provision of services to help the community.

Speier concluded by highlighting the key elements of preparing a community for participation in disaster impact research, such as Disaster-PAST, that overlapped with key points from previous speakers. He recommended the following:

• A prearranged community research advisory board strategy (e.g., knowing who the stakeholders are and how those people can be encouraged to become involved in the research design and data gathering);
• Observable, tangible actions to foster community trust (e.g., following through on promises made to the community);
• An easily understood script regarding direct benefits of the research for the community; and
• Clinical support for participants embedded within the research design and data gathering.

CHALLENGES AND OPPORTUNITIES

A recurring theme throughout the community discussion was the need to reach out and partner with other groups and establish trusting relationships and grassroots connections among public health, academia, and all appropriate community stakeholders before the disaster, which Herrmann summarized in his report of the panel session (see Box 5-1). A challenge is identifying the “right” partners and finding the resources and mechanisms to develop and maintain these relationships, as academic institutions, are not funded for relationship building.

A suggestion was made that local and state health departments especially need to be engaged as partners before a disaster, as they are often overwhelmed during the response and bombarded with requests. With regard to funding predisaster research and relationship building, it

was noted that local health departments often have internships for students in public health programs. How can the resources of universities be leveraged to help the health department gather data prior to an event?

It was also noted that many local health departments are working toward accreditation by the Public Health Accreditation Board, and among the measures are building partnerships, working with students, and being involved in research. Participants also noted the need to identify the unofficial “mayors,” the people who are recognized as leaders within the community, Herrmann summarized. Additionally, there is a lack of clarity around the benefit of conducting research in the community, Herrmann explained. A participant noted the value of risk communication in helping people who are emotionally stressed to understand how the research affects them and the health of their community without inundating them with data, as Bradberry mentioned previously.

A participant said there is an opportunity to work with first responder organizations, community leaders, and others to find the commonalities across the country and start thinking about “plug and play” protocols. Participants discussed balancing respect for the diversity and distinct cult-ures of communities with the development of standardized, ready-made research protocols. Additionally, further concerns were raised about researchers who impose their own ideas on a community, rather than take the time to understand what the community needs, and the impact of the onslaught of researchers into a community after a disaster. A participant said that people descend into the Gulf region when disasters happen, but nobody comes when the region is talking about enhancing the infrastructure and the capacity beforehand. The region is the poorest and has the most health needs, but the only time researchers come is when there is a disaster, they said.

This page intentionally left blank.