Health Monitoring, Assessment, and Response
RAPID ASSESSMENT FOR IDENTIFICATION, MANAGEMENT, AND PREVENTION OF ENVIRONMENTALLY RELATED DISEASE
Coordinated effort and prioritization of health risks is critical in a disaster aftermath. It can become a challenge when a relatively rapid response is needed and there is virtually no time to prepare for it, noted Kellogg Schwab
A straightforward, relevant, ongoing health evaluation conducted by health professionals is needed during disasters.
of the Johns Hopkins University’s Bloomberg School of Public Health. This was true for Hurricane Katrina, which had a disaster area of 90,000 square miles, creating community-wide and regional issues. Communication is very important for a successful rapid public health assessment. Communication in the Gulf Coast region during the Hurricane Katrina aftermath was inadequate, creating one of the biggest challenges, observed Schwab.
Prioritization of Health Risks
A straightforward, relevant, ongoing health evaluation conducted by health professionals is necessary during disasters, said Schwab. It provides information on the prevalence and incidence of potential diseases and targets limited resources to evaluate acute health issues such as infectious diseases, chemical exposure, heat exhaustion and heat stroke, lack of medications, and mental illness. Evaluating these immediate versus long-term risks through rapid assessment during a disaster is challenging, noted Schwab. There is a real need for precise assessment tools and technology to address chemical and biological exposures and acute
versus chronic concerns. These health risks need to be determined in multiple media, including the air, water, and food supplies.
Biological health risk assessment in contaminated areas needs to include viruses and protozoa in addition to bacteria. Because many viruses and protozoa can persist in the environment for much longer periods of time than bacteria, they can contribute to morbidity and mortality in humans, noted Schwab. He noted that routine biological monitoring does not usually include viruses and protozoa and that this would need to be addressed in the future.
For chemical exposures, such as volatile and semivolatile organic compounds, total metals, pesticides, herbicides, total petroleum hydrocarbons, and polychlorinated biphenyls, researchers need to include other specific exposure routes, such as dermal and airborne exposure routes, in addition to ingestion via drinking water.
Challenges of Rapid Response Assessment
Multiple pitfalls prevented assessment success after Hurricane Katrina, said Schwab. One of the major issues was damage to the telecommunication infrastructure, which limited the ability to communicate within the region. Land lines and the cell phone towers were damaged, and satellite phones were not working reliably. In addition, roads and bridges were impassable in some areas, and the shortage of gas supplies limited people’s ability to travel in order to perform rapid assessments (Figure 5-1). In addition to infrastructure damage limiting access, there were governance and training barriers. Multiple jurisdictions may have meant that assessors could not enter all places of interest (e.g., shelters), which slowed down health assessments. The lack of trained personnel who could rapidly adapt hindered the successful response as well, said Schwab.
Targeted health surveys applicable to the situation, using field-tested methods versus laboratory prototypes for agent identification during the assessments, were challenging in the very rapid response mode. For example, delay in sample analysis and ineffective dissemination of findings were problematic for rapid assessments. During the unfolding disaster, multiple contaminants—both microbes and chemicals—originating from multiple sources—municipal, industrial, and small businesses—and multiple media, including air, water, and sediment, were constantly changing, creating overlaying scenarios that had to be addressed during the assessments, noted Schwab. For the field of environmental health, this area suggests the need for further discussion.
Despite the many challenges to conducting health risk assessment, there were some positive outcomes, said Schwab. For example, Harvard University and Johns Hopkins University, teaming up with the Centers for Disease Control and Prevention, set up a toll-free hotline in Mississippi to provide the public with direct access to public health professionals who could provide information on
the Mississippi Public Health Department’s response. Callers could also leave voicemail for nonemergency issues.
Environmental Monitoring Detection Strategies
One of the issues for environmental monitoring is determining what detection strategies will be used and what will be designated as the gold standard. For example, the Environmental Protection Agency (EPA) collected air samples from multiple locations across the New Orleans metropolitan area on September 11 and 13, 2005. These data were collected with portable, battery-powered monitors that are often used in an emergency response because they give immediate readings; however, the data obtained from these monitors could not easily be compared with the EPA standards. EPA does not use data from these types of monitors either for compliance purposes or for generating routine air quality advisories, noted Schwab. Even so, to provide the public with a point of reference, EPA compared the results with its air quality index for inhalable coarse particles, also known as PM 10.
Schwab noted that assessments need to be broad in focus for infectious diseases such as dysentery, cholera, and gastroenteritis. He noted that one of the
largest outbreaks in a shelter was a norovirus in the Houston sports complex that housed 24,000 evacuees. Approximately 1,000 of the evacuees were infected with the norovirus, which is a resistant microorganism that easily transmits from person to person. Health officials need to understand that even though water, sediment, and food are negative for bacteria, they can still contain other pathogens of potential health concern for both morbidity and mortality.
Microbial analysis in the field is technically challenging, and the detection assays must be sensitive, specific, and capable of detecting low concentrations of target agents without interference from background materials. Sample inhibition is very challenging, said Schwab, because of false negatives due to the inability of the assay to work effectively using that detection technique. This problem needs to be addressed with appropriate quality controls and quality assurance during the sampling. In addition, current sample matrixes are
Providing information to frontline healthcare providers, including shelter managers and local and regional coordinators, during telecommunication gridlock was challenging.
very complex. Water samples are usually concentrated from large volume to small volume, which concentrates the inhibitors. Large volumes of air samples are concentrated either on a filter or into a liquid medium for subsequent analysis. These concentration steps are not 100 percent efficient; thus, assessors need to take into account the ability to know what the numbers actually mean with respect to the exposure levels during the sampling. More importantly, the nucleic acid–based or antibody-based molecular detection techniques usually do not determine the infectious nature of the microorganism.
Dissemination of Accurate Information
As events were unfolding, health officials struggled with actual versus perceived risk, and providing accurate information to frontline healthcare providers, including shelter managers and local and regional coordinators, was hampered by the telecommunication gridlock, said Schwab. Disseminating accurate and verified information to agencies and the news media, in addition to the frontline staff, is vital. According to Schwab, the absence of authority for implementing public health measures can limit effectiveness, thus causing disease surveillance and preventive measures to fall through the cracks.
Response personnel faced a dangerous environment with multiple exposure hazards, and their own mental health was very important to monitor as well. Agencies and groups that send their personnel to an area need to consider and implement effective strategies for pre-deployment and post-deployment debriefings, such as pre-deployment blood draws and basic infectious disease characterization.
Rapid Assessment: Themes for Future Discussion
Rapid health response is a critical component of any disaster response. On the basis of his experience in the region, Schwab highlighted some areas for further discussion in order to prepare for future disasters:
Enhancing communication to assist in rapid health assessment,
Involving the public health community in articulating health issues,
Preparing assessors prior to an event and assisting them in adapting to changing situations,
Developing simple and meaningful target goals,
Developing effective strategies to provide targeted and timely results, and
Providing concise and accurate public health information and advice.
Schwab noted that additional work is needed to ensure effective communication strategies and prepare responders for health assessment. Although the next large disaster may be different from Hurricane Katrina, the same concepts of public health, infrastructure, and basic needs will still be present.
POST-KATRINA MEDICAL SURVEILLANCE
Before deciding what medical surveillance projects need to be established for tracking the health impacts of Hurricane Katrina, scientists and policy makers need to answer some questions:
What are the questions that need to be answered?
What resources are available to answer these questions?
What approaches can be used?
What barriers can be foreseen, and which can we attempt to overcome?
Answering these questions will help to minimize the impact on the victims and serve to inform future efforts under other disaster conditions. Most of these questions involve several components, such as the population of interest, the exposure of interest, and the outcomes of interest, said David Goff of Wake Forest University. Once scientists know the components of the question, it will be easier to determine which research approach to undertake.
How to Choose the Right Question
There may be some particular exposures or outcomes that can be examined on the basis of data from other similar types of events. In the case of Hurricane Katrina, there are many unique aspects of this natural calamity and the exposed population. Questions could thus be guided by some consideration of those
unique aspects, noted Goff. Ideally, the focus would be on the public health burden of the outcome by looking at issues that have significant impacts on population health through morbidity, mortality, or cost, rather than on rare aspects. The preventability of the outcome is also worth considering. It is far more important to track and study issues that we think we have knowledge about for risk mitigation if a similar type of event happens in the future than to get distracted by aspects that are less known, said Goff.
Populations of Interest
Researchers need to answer the question of population of interest first, because it has implications regarding how a monitoring system or a research project would be set up. Determining what defines the population of interest is complex, said Goff. It could be either all the people exposed to Hurricane Katrina, evacuees only, or first and subsequent responders. This is not a straightforward question and quickly becomes complicated. Considering individual exposure to Hurricane Katrina, more questions arise about who was exposed and what was the exposure. The population of interest could be defined as either the individuals who evacuated or those who remained in the New Orleans area. In addition, some evacuees relocated to other cities and did not return to the Gulf Coast, which complicates trying to determine if they are part of the group of interest.
Exposures of Interest
The exposures of interest may be mold, air, water, soil, changes to the built environment, housing quality, community characteristics and resources, and other exposures that will have some impact on human health. While researching exposures of interest, it is important to take into consideration the time perspective for the exposure. If a researcher is interested in acute exposures, the data collection window is going to be narrower than for monitoring long-term exposures, said Goff. This is not to say that acute exposures do not have potential for long-term effects.
Outcomes of Interest
Medical conditions have an impact on the type of surveillance system that will need to be set up, said Goff. Infectious disease in the area is one of the outcomes of interest, considering the potential contamination of the water supply and the crowded conditions under which people have been living and may continue to live. It is not immediately clear from the initial surveillance which chronic diseases should be researched as outcomes of Hurricane Katrina. Categorizing these will be important. Any medical surveillance program following a disaster needs to include mental health conditions, occupational disorders, and traumas.
It is not immediately clear which chronic diseases should be researched as outcomes of Katrina. Any medical surveillance program needs to include mental health conditions, occupational disorders, and traumas.
The program needs to be holistic in its understanding of the complexity of the exposures. For example, an event such as Hurricane Katrina can lead to substantial morbidity caused by strokes and cardiovascular conditions, which have been related to stressful life events.
The choice of indicators of health has an impact on the type of surveillance program that will be established. If the researchers’ interest is hospitalizations, that implies a set of projects in which they could monitor hospitalizations, other health care resource utilization, emergency department visits, and outpatient visits. The tracking of ambulatory care visits as well as hospital visits in the days immediately after Hurricane Katrina has already been done, said Goff.
The time horizon is also important for outcomes. The current data available from New Orleans provide a very-short-term snapshot of what might have happened immediately after Hurricane Katrina—for example, whether there were any immediate outbreaks of infectious diseases, insect and snake bites, and so forth. Interest in medium- and long-term outcomes should not be undervalued, noted Goff.
Patient-centered outcomes, such as the functional status of the people who have been exposed to and hurt by Hurricane Katrina, are an important long-term issue, noted Goff. To identify hospitalizations and outpatient visits, researchers will need to be able to interact with the exposed people, ask them about their functional status, and perhaps perform some hands-on examinations. The same is true for health-related quality of life, a particularly important patient-centered outcome.
Potential Medical Surveillance Approaches
Many different approaches can be used in research and in public health practice. One approach is to use an existing model and perform cross-sectional population surveys. Another approach is to establish a cohort that is assessed at baseline and followed over time. These approaches should be designed to be representative of the population, so that appropriate inferences can be made about the health condition of the population. The approaches should also enable direct examination of people and include the collection of questionnaire data.
Several surveillance methods are currently used in the United States. The use of a method in its current or modified form may be useful during disasters such as Hurricane Katrina, said Goff. For example, previously used national surveillance
methods—such as the National Health and Nutrition Examination Survey, mobile examination and survey units, the Nationwide Inpatient Sample, the National Hospital Discharge Survey, surveillance of hospitalizations, and the National Ambulatory Medical Care Survey surveillance of outpatient encounters—could be used. There are certain strengths to this approach, noted Goff. The existing surveillance systems could be expanded in the Gulf Coast area for oversampling—because the expertise, methods, and infrastructure are already available. Because these surveys are national, multiple conditions are tracked and comparison data are available. Thus, it is possible to compare the disaster experience with experiences in other parts of the country. The drawbacks of using these national surveys following a disaster are that (1) there are no direct examination data, (2) there is no follow-up in any of these surveys, and (3) there is limited ability to control data collection, noted Goff.
Goff observed that medical surveillance for multiple conditions is possible in assessing the health impacts of the victims of Hurricane Katrina. There are multiple surveillance models readily available—for example, the Reasons for Geographic and Racial Differences in Stroke (REGARDS) cohort study being conducted by the University of Alabama, Birmingham. This 30,000-person cohort study has representatives from 48 states, a unique feature that could be useful to track the health impacts of Hurricane Katrina because evacuees were scattered around the country.
In the cohort study, direct examinations are done in the subjects’ homes through a contract with Examination Management Services Inc. (EMSI), a company that performs insurance physicals. EMSI trained its staff in research methodology and human subject protection in order to participate in this study. The use of this surveillance system has a number of strengths. It enables the researchers to perform direct examination and follow-up, researchers have substantive control over the data elements collected, multiple conditions and exposures can be assessed, internal comparison data can be generated, and people who do not move back to the Gulf Coast area can be included in this type of design because the home visits can occur anywhere in the country. One drawback of this study design is that, although this model has been previously used, some additional infrastructure would have to be developed, said Goff.
Building on existing national surveillance programs is also appealing, observed Goff. Scientists should define the questions of interest as soon as possible so that the right approach to pursue those questions is chosen. Although there are multiple barriers, they are foreseeable and most of them are manageable.
Challenges to Collecting Medical Information About Individuals from the Affected Region
Many clinics and hospitals in the area were condemned, severely damaged, or destroyed by the storm surge. The status of their clinical records is unknown
but presumably destroyed in many cases. Records in clinics that had paper records stored in file cabinets no longer exist to be able to determine for historical purposes what the health condition of the population was prior to the event. Even hospitals with electronic health records that were not backed up off-site lost their records because their hard drives were not functional after being under water for some time.
A valuable lesson that the medical community has learned from
A valuable lesson that the medical community has learned from Katrina is that electronic health records should be backed up off-site on web servers.
Katrina is that electronic health records should be backed up off-site on web servers, said Goff.
ETHICAL HARMS IN COMMUNITY HEALTH RESEARCH
The protection of human subjects is the focus of a substantial number of articles and publications on medical ethics for individual patients, but not all of these principles are applicable to community-based research. Researchers need to look beyond the Belmont principles* to more community-centered ethical frameworks, such as virtue and communitarian ethics and the ethics of care, as well as postmodern ethics, which deals with power issues, otherness, and cultural diversity, noted Dianne Quigley of Syracuse University. In addition to these, non-Western models, such as Native American ethical philosophies, with their emphasis on protecting communal values in knowledge production, need to be considered, said Quigley.
The Collaborative Initiative for Research Ethics and Environmental Health project, funded by the National Institutes of Health, includes an interdisciplinary project team of public health, social science, biomedical, behavioral, and humanities researchers from Syracuse University and four other collaborating universities. The project focuses on ethical issues surrounding community-based research collaborations between researchers and communities in the fields of environmental and community health research. It represents a unique experience in dealing with research ethics concerns for Native American, African American, Hispanic, and Southeast Asian populations in environmental and community health research. The project is trying to move research ethics from focusing on ethical harms to individual human subjects to the whole community as a subject. According to Quigley, the ethics field lags behind in looking at the community as a subject of research. The project has developed courses at universities exploring community-based and multicultural ethical dimensions of the community as a research subject, working with the multiple voices and problems of community
members and the research conditions of multiple community contexts and what they mean for research ethics.
Quigley singled out six common ethical problems in community research that can harm communities and cause distrust in the relationships between scientific researchers and community members.
1. Irrelevance to Community Needs
Irrelevance to community needs can occur when research approaches are academically controlled, research teams are inexperienced with the community’s needs and values, and there are limited provisions for community participation. Although research designs and methods are scientifically interesting to academics, they are irrelevant and sometimes damaging to community needs. If researchers bring too many research efforts into an area that has been affected by contamination, they run the risk of the research being irrelevant to the community, said Quigley. Examples of “parachute research” demonstrate how it can end up being damaging to community needs when the community is not engaged from the beginning.
2. Exploitation of Community Members
Exploitation of community members may create serious inequities in the research process, whereby community members are burdened with research activities without compensation or funding for community expenses, leading to exploitation of community members and resources. Many communities complain about researchers using their time or about helping with recruitment of subjects or performing actual research activities, for which they do not get compensated. This is a problem, particularly in underserved communities that are already burdened by a number of other needs. Researchers have to be aware of exploitation of community resources and ensure that they provide funding for any research activities they want to perform in the community.
3. Community Stigmatization
Community stigmatization is often caused by a lack of attention to or development of group or community needs and values in scientific research practices. This produces ethical inadequacies in the research obligations of community consent, involvement, comprehension, and risks or benefits from a research effort. Without obtaining community consent or approval for research efforts, researchers can put communities at risk for community stigmatization in publications, said Quigley. In addition, researchers should report results back to communities before publishing them, so that the community is allowed to provide rebuttals or alert researchers to the harm that might come to them from publication of
the research findings. This issue illustrates the need for developing community approval and consent and community research protections, noted Quigley.
4. Lack of Comprehension by the Community
Without full comprehension of a research intervention and discussion of the risks and benefits of research designs, communities can suffer from these more specific ethical harms:
Research findings that bring no public health benefit to the community and may be used as justification for no further follow-up of research activities in a community (i.e., studies that often yield statistically insignificant findings in small populations),
Unintended social or cultural harms (treatment of tissue samples, violations of cultural practices, overriding communal norms), and
Researchers’ indifference that may intimidate or demoralize community members.
Communities can be educated on a number of health risk methodologies, and they should be there with the researchers to understand what the investigation entails, whether the community wants it or not, and whether it is going to be beneficial to them or not, said Quigley. Researchers need to help the community understand the methods of their research. Although comprehension of high-level technical methods can be very hard for disadvantaged communities, they can understand the information if researchers give them enough support and infrastructure to evaluate these health risk methodologies, said Quigley.
5. Exclusion of Community Contextual Knowledge
Exclusion of community contextual knowledge occurs when research designs exclude the observations, local knowledge, and experiences reported by community members. This can lead to inadequate information about diet, lifestyle, and other relevant exposure information. It can lead to inadequate recruitment and participation of research subjects. If the community was not involved in collecting the data about diet, lifestyle, subsistence, or other relevant aspects of the community’s experience with the research question, researchers may not get actual exposure data from the community’s embedded conditions. Researchers run the risk of overlooking important data sources when they make assumptions about lifestyle scenarios without being in the field or working with communities to get actual contextual information, asserted Quigley.
6. Exploitation of Community Data
Community approval and consent procedures are not well developed for research dissemination, publication, or uses of community tissue samples, archives of local knowledge, or other community data. This may lead to the exploitation of community data. Quigley asserted that protocols could be developed with approvals and consent from the community. Furthermore, these protocols need input from the community to avoid situations in which researchers take community data, use them, and transfer them to other institutions or use them inappropriately. This may lead to further stigmatization of the community, noted Quigley.
BEST PRACTICES IN COMMUNITY HEALTH RESEARCH
There are many examples of best practices in engaging the community as a partner in research in areas that may relate to the post-Katrina situation, such as studies of air monitoring, indoor air pollution, and fish contamination and pesticides. These case studies, many of which were conducted with federal support from the National Institute of Environmental Health Sciences, EPA, or federal health agency commitments to community-based participatory research, illustrate innovative methods for how researchers can engage communities and create many positive outcomes for both the researchers and the community.
Some of the examples of best practices include developing community advisory committees or stakeholder steering committees and allowing for a process of continuing involvement of affected groups. Such measures as funding, paying for community involvement, participation stipends, transportation, day care, and the training of community research staff help researchers to share research and ethical decisions that they are facing with the community research investigation and at the same time help to build trust with the community advisory committees. The committees are very helpful in terms of setting research designs and ensuring the community’s partnership or ownership of the project. At the same time, these community advisory committees facilitate building community research experience and decision-making capacities, because the communities are going to face the environmental problems longer than researchers will be there, noted Quigley. Communities should be given an opportunity to be trained and to build capacities with research administrative issues. Community members can be recruited and trained as lay health advisers or community researchers, and they can help design and implement questionnaires and identify participants for research projects.
In places with no infrastructure or strong community leadership for environmental health, the advisory committees can be replaced by community health organization representatives, environmental groups, church groups, existing public health or medical organizations, physicians, or various networks, such as environmental justice or community health networks, noted Quigley. For example, in a study of Southeast Asians and fish contamination in Massachusetts, local researchers learned about culturally appropriate research methods from a national
refugee organization in San Francisco that had a great deal of culturally sensitive research experience with this population group. The organization’s expertise contributed to the building of a community-based research infrastructure and effective outreach interventions.
Culturally appropriate research and outreach strategies, such as educational methods that focus on the community’s languages, graphics, and teaching methods that incorporate ethnic values and traditions in the research activities, are all very important, noted Quigley. When community members are involved in working with researchers side by side, they develop a commitment to dealing with the community harms that might be found from the research investigation. The community will then take it to the social action level or the policy action level, which is an important feature of community-based research, said Quigley. If the community feels that it can own some of the management of the research problem and it is given funding and training, it will be there to work on the problem in the long term.
According to Quigley, communities can develop multidimensional types of outcomes and benefits from a research effort. They may help not only with identifying ways to reduce exposures, but also with other diet, lifestyle, and recreational areas of community life that can improve health conditions. Community members build the contextual and local community knowledge for determining and assessing exposure pathways; they should therefore have a strong role in interpreting results and designing and implementing interventions, asserted Quigley.
Another important outcome is that involving the community in research can actually improve the scientific research analysis in terms of recruitment and interviewing community members and involving workers, migrants, and transitory groups that scientists cannot reach on their own. Community involvement can improve questionnaires by ensuring cultural and regional relevancy. Community involvement facilitates interview processes, providing culturally appropriate listening skills and engagement with people who are being interviewed (RTI International–University of North Carolina, 2004).
Cultural competence can often be an overlooked aspect of training in the environmental health field, said Quigley. Scientists cannot really move that far ahead with monitoring and technical research without knowing the context of the community in question. Developing bicultural models for research, which take into consideration traditions and values of people involved in the research process, and taking cultural sensitivity courses before researchers even start would be valuable, noted Quigley. At the same time, researchers should be more conscious of their own perceptions and experience and how these may collide with the traditions and values of cultural groups.
Quigley concluded by saying that researchers need to improve their cultural competence and learn more about exchanges of cultural knowledge and values in the research process. Diverse cultural views and community-based knowledge are key understandings that researchers should have.