Hurricane Maria hit Puerto Rico on September 20, 2017, and the island sustained significant damage and loss of property and life. To date, the public health response has not been adequate given the level of severity of the damages sustained. More recently, there have been additional extreme weather events in the Bahamas after Hurricane Dorian, along with wildfires in Russia and California and the tropical typhoon that hit Japan in early October 2019. Upon reviewing APHA policies, the authors identified that there is a policy gap regarding the issue of the long-term public health impacts of natural disasters. This policy statement responds to the public health challenges of high-impact natural disasters. APHA calls for a medium- and long-term public health response and research to assess the long-term effects of high-impact disasters, including effects on environmental health and well-being in vulnerable populations.
Relationship to Existing APHA Policy Statements
This policy statement builds upon APHA Policy Statement 20157 (Public Health Opportunities to Address the Health Effects of Climate Change). It is also consistent with the following APHA policies that reference disasters and public health responses:
- APHA Policy Statement 6211(PP): The Role of State and Local Health Departments in Planning for Community Health Emergencies
- APHA Policy Statement 9116: Health Professionals and Disaster Preparedness
- APHA Policy Statement 20069: Response to Disasters: Protection of Rescue and Recovery Workers, Volunteers, and Residents Responding to Disasters
- APHA Policy Statement 20126: Anticipating and Addressing Sources of Pollution to Preserve Coastal Watersheds, Coastal Waters, and Human Health
- APHA Policy Statement 20078: Addressing the Urgent Threat of Global Climate Change to Public Health and the Environment
- APHA Policy Statement 9510: Global Climate Change
- APHA Policy Statement 201713: Establishing Environmental Public Health Systems for Children at Risk or with Environmental Exposures in Schools
- APHA Policy Statement 201712: Advancing a ‘One Health’ Approach to Promote Health at the Human-Animal-Environment Interface
As the 10th most intense Atlantic hurricane on record and the most intense tropical cyclone worldwide in 2017, Hurricane Maria is regarded as one of the worst natural disasters to occur in the United States. It was the deadliest storm of the active 2017 Atlantic hurricane season. An estimated 2,000 to 5,000 human lives were lost to post-Maria conditions.[2,3] (According to the official government estimate, 2,975 Puerto Ricans were killed by the hurricane or its aftermath. This figure was raised from 1,427 fatalities only a few weeks earlier.) Indeed, there has been ongoing controversy as to the final mortality figures. In May 2018, Harvard University released a study claiming that 4,645 Puerto Ricans had perished from reasons related to Maria. It seems prudent to encourage comprehensive, long-term epidemiological surveillance in postdisaster settings involving high levels of mortality.
Localities often suffer catastrophic damage from storms, including destruction of electrical grids. For weeks, there may be sustained flooding and lack of resources, often compounded by a relatively slow relief process. Flooding and wind can cause significant damage to electrical grids, houses, and roads, all of which necessitate costly repairs.
Health inequities may worsen after hurricanes and magnify the impact of social determinants of health, including housing, health care services, and access to clear water and sanitation.[4,5] Moreover, with many localities facing austerity measures, the availability of local resources may not adequately address the public health situation, with limited or temporarily off-line health care resources.[5,6]
According to APHA Policy Statement 6211, “In the event of a disaster, state and local public health officials have primary responsibility for responding to the health needs of the victims. Moreover, it is these public agencies’ appropriate role to work with other responding agencies by actively coordinating the various response efforts that impact physical and psychological health.” However, it is not clear which governmental agencies have responsibility for long-term health surveillance and research. Moreover, often there is confusion related to multiple agencies having relevant responsibilities in supporting postdisaster communities. With ambiguous governmental responsibilities, the question of “Who is in charge?” or “Who is coordinating relief?” also needs to be addressed. As a result of the various agencies involved, they also have different funding streams and competing interests. For example, in terms of surveillance, there should be better coordination of existing systems, including systems that could be used to support surveillance after a disaster (e.g., Behavioral Risk Factor Surveillance System, Youth Risk Behavior Survey, National Violent Death Reporting System, National Intimate Partner and Sexual Violence Survey, Pregnancy Risk Assessment Monitoring System).
Here we describe a proposed long-term public health approach to these types of disasters in the future. Although we focus primarily on the effects of tropical storms, this statement could well serve other high-impact, postdisaster settings (e.g., flash floods and, in a few cases, extreme forest fires).
The adverse health outcomes of high-impact disasters will necessitate not only reestablishment of more sustainable infrastructure but also more robust, long-term responses to psychological and physical health issues, along with management of policy factors. This statement proposes using a life course approach to disaster response, including surveillance and longitudinal research. The life course approach is based on Bronfenbrenner’s socioecological theory of childhood development. According to this approach, individuals grow up and live within several contexts, including neighborhood, familial, cultural, and social contexts. The life course approach relies on three phases for determining health outcomes: the critical period, the latency period, and postdisaster resilience.
Natural disaster as a critical period: The critical period represents a time in which an exposure, such as a hurricane and the subsequent flooding, has a stronger effect on human development and, hence, disease risk than it would at other times. Exposures, in the broadest sense of the term, could be compounded as a result of the stressful effects of disasters, especially in the face of climate change. Exposures of these types involve different risk and protective factors as part of the One Health Framework. Indeed, the One Health Framework proposes that human-animal-environment connections are important when considering the mid- to long-term impacts of these exposures (e.g., when considering rodent-borne infections such as leptospirosis). With increasing effects of climate change, these types of zoonoses (diseases that can be transmitted from animals to humans) will only worsen. Similarly, exposures are of many varieties: chemical, spatial, economic, social, and racial, among others. One specific biological exposure from raw sewage recorded in Puerto Rico was Escherichia coli bacteria in drinking water, which can lead to diarrheal infections. Another chemical exposure that continues to pose a problem in post-Maria Puerto Rico is heavy-metal-laden coal ash contamination. In any disaster setting, these types of chemical exposures will lead to short- and long-term adverse health outcomes such as developmental delays and, possibly over time, cancer. Specifically, in terms of children’s postdisaster health, there are no mandated federal, state, or local inspections of damaged educational facilities for safe reoccupancy by children. It has been common practice that health agencies do not log or track children’s exposures in these postdisaster settings.
The latency period after natural disasters: The latency model of the life course approach theorizes that discrete events have a strong independent effect later in life. The latency period is the time between exposure and the emergence of the first symptoms of illness. In the Puerto Rican example and other postdisaster settings, there may be an increase in certain adverse health impacts, such as posttraumatic stress disorder (PTSD), birth defects, and/or cancers, subsequent to a 5- to 15-year latency period.
The latency period also reflects the stress-vulnerability theory. For example, in looking at the World Trade Center disaster, Adams et al. found that the affected communities in New York City recorded higher levels of alcohol abuse, domestic violence, and other related social tensions, and these levels were particularly high among communities of color. Also, a recent study by Ratnapradipa et al. highlights the unique health risks that vulnerable populations faced in Houston during Hurricane Harvey. Vulnerable populations, including people of color, emergency workers, pregnant women, undocumented immigrants, people living in lower-income housing, children, individuals living with functional disabilities or chronic medical conditions, and LGBT (lesbian/gay/bisexual/transsexual) individuals, are most at risk of experiencing stress during the latency period. Specifically, emergency and recovery workers should be included in any postdisaster surveillance program. Pregnant women, in particular, may be vulnerable due to poor prenatal and perinatal care in the postdisaster period.
The extreme emotional stress that individuals may experience during high-impact disasters can speed up the usual latency period and may be part of a synergistic effect that leads to negative health outcomes over a much shorter period of time in postdisaster settings. These secondary stressors should not be overlooked in the postdisaster period, as studies indicate that long-term mental health is affected during this period. We can help disaster survivors by not only offering mental health first aid but also by proposing their involvement in designing postdisaster surveillance or research studies.
Postdisaster resilience: After disaster hits, some level of distress among inhabitants is nearly a universal reaction. Such reactions can be grouped into twin maladaptive and adaptive reactions to the disaster’s stress that evolve as a function of vulnerability and resilience factors. Researchers have shown that the single most important factor encouraging resilience is social support.
In postdisaster communities, information, infrastructure, and communication resources are often poor, unreliable, or untrustworthy. This perceived lack of adequate communication may be detrimental to helping a community heal its psychological wounds in the postdisaster period.
For these reasons, we need to better understand the effects of vulnerability and resilience factors on the advent of posttraumatic stress in postdisaster settings. Some postdisaster projects focusing on traumatic events have developed conceptual models around adaptive and maladaptive factors such as disaster-zone stressors, predisaster risk factors, stressful life events, social support, and alienation.
Evidence-Based Interventions to Address the Problem
This section is divided into three subsections: long-term postdisaster research, long-term postdisaster surveillance, and extending current short-term postdisaster responses.
Long-term postdisaster research: In terms of future public health interventions, we identified only a couple of federal plans for studying the long-term effects of Hurricane Maria in Puerto Rico. (It is unknown whether studies are planned to identify the long-term impacts of Hurricane Dorian, which occurred in 2019.) Such studies, however, are not stand-alone studies but rather are “tag-ons” to existing cohort investigations. An analogous example, Hurricane Katrina in 2005, has been widely studied by public health researchers at Louisiana State University, Tulane University, and other universities, including with substantial National Institutes of Health (NIH) funding. Specifically, the NIH Disaster Research Response Program, the national framework for public health research on disasters and public health emergencies, has offered funding for short- to medium-term postdisaster health research. However, it is not clear whether an extant longitudinal cohort is being followed to determine the existence of unique outcomes among Katrina survivors. From this Katrina-focused research, the most unusual finding has been “Katrina brain,” defined as a condition wherein patients struggle with short-term memory loss and cognitive impairment. In the Puerto Rico context, researchers could consider investigating the possibility of short-term memory loss and cognitive impairment among survivors of the hurricane.
Another retrospective cohort study of the effects of Hurricane Katrina on the incidence of acute myocardial infarction (AMI) in New Orleans showed that the mean age of onset of AMI decreased from 62 years before Katrina to 59 years after Katrina. According to additional post-Katrina studies, there have been many other negative health outcomes including asthma, allergies, birth defects, learning delays, PTSD, and cancers.[17,18]
Long-term postdisaster surveillance: Another example in support of more long-term surveillance involved Typhoon Haiyan (2013), which struck the Philippines. Haiyan’s death toll was estimated at 6,300 individuals. Although immediate medical needs were not necessarily injury related, lack of prevention of infectious diseases and worsening of noncommunicable diseases due to a lack of access to food, water, housing, and medicine suggest that death toll figures could still be rising.
A final example of postdisaster surveillance is the tsunami that struck the eastern Japanese coast in March 2011. In this example, the Fukushima nuclear plant suffered a massive meltdown and caused radioactive releases. Beyond an increase in allergies, affected individuals reported suffering from increases in vertigo and dizziness during the first few years after the disaster. According to the World Health Organization (WHO), there was a sharp increase in mortality among elderly people who were placed in temporary housing. In addition, there was an increased risk of noncommunicable diseases such as diabetes and mental health problems. A lack of adequate health care access further contributed to these health concerns.
Extending current short-term postdisaster responses: Postdisaster research could be supported through specific funding mechanisms for short-term responses, such as existing evaluation funding from the Centers for Disease Control and Prevention (CDC). The Bipartisan Budget Act of 2018 includes $200 million in one-time funding appropriated to the CDC for response, recovery, preparation, mitigation, and other expenses directly related to the consequences of Hurricanes Harvey, Irma, and Maria.
The CDC has already drafted and implemented a disaster health response cooperative agreement. Although this agreement provides support for biosurveillance, time frames are not specified. The agreement does not provide support for postdisaster research activities. It is hoped that additional postdisaster cooperative agreements will emerge from future federal funding appropriations.
Given the urgency associated with postdisaster settings, preexisting cooperative agreements between federal agencies and local research institutions (e.g., universities) often involve cost sharing and/or task sharing to ensure evaluation. In the future, this evaluation component could include research and/or surveillance, which may already be included in short- to mid-term response projects.
Prohibitive cost: The main opposing argument would be that long-term public health surveillance and research will be prohibitively costly. In a recent NIH request for application for a cancer epidemiological cohort study, the budget was $1,250,000 per annum, or $12.5 million for a 10-year cohort study. With uncertain sources of funding for long-term, federally funded health research, projects almost exclusively focus on surveillance and monitoring of negative health outcomes.
New technologies to clean water or provide rapid testing for various diseases could be a way to reduce costs. In terms of ensuring a well-trained workforce, health care, emergency, or information technology workers may already possess the requisite skills and could be retrained for surveillance or research. Also, additional educational opportunities offered for research and data analysis could reduce the costs of long-term research and surveillance. Such opportunities could include cost-benefit analyses that would help quantify why these postdisaster approaches are both important and necessary. A final counterargument is that federally funded surveillance could help track postdisaster health effects, lead to earlier diagnosis and treatment, and therefore save money.
Difficult-to-locate displaced populations: Another opposing argument would be that, after high-impact disasters, populations of interest are often displaced. For example, many Puerto Ricans (an estimated 130,000 individuals), especially the most vulnerable, have already left the island and migrated to the mainland United States. A similar exodus of hurricane victims occurred after Hurricane Katrina. Therefore, a cohort study may not accurately capture the long-term effects on the Puerto Rican population as a whole.
Future postdisaster studies should better track displaced people affected by the disaster under investigation. These studies may need to involve multisite designs in order to enroll displaced individuals. In addition, these individuals should be linked up with the appropriate social services.
Protecting the rights of vulnerable populations: Another concern is the recruitment of postdisaster vulnerable populations and how they might be targeted by a postdisaster cohort study. Due to historical bioethical violations (e.g., abuses against African American men during the Tuskegee Syphilis Study), these populations harbor some suspicion of these types of biomedical studies and therefore may be hesitant about participating.
To encourage these populations to participate, it would be helpful to explain the long-term benefits of such research and to highlight that there is a gap in the scientific literature that urgently needs to be addressed. Furthermore, special and deliberate efforts should be made to protect these populations by ensuring that the study burden is minimized. Another counterargument to recruiting participants would be that national security may be threatened by natural disasters. Notwithstanding other massive, man-made disasters such as the accidents in Chernobyl (1986) and Bhopal (1984), study teams could emphasize to cautious participants that additional long-term, disaster-focused cohort studies are needed to prevent future disasters and/or minimize the impacts of future disasters on their health and their own communities.
Competing resources: We do not believe that new funds for a more coherent, long-term public health response will replace or compete for resources devoted to rebuilding infrastructure and humanitarian aid.
The following evidence indicates that funds for long-term responses are much less than would be expected:
The Overseas Development Institute think-tank report on worldwide spending on disaster expenditures showed that very little funding goes toward disaster prevention and preparedness that can build resilience to cope with emergencies.
According to the Inter-Agency Standing Committee of the World Bank, in general government expenditures are lower in terms of prevention than in terms of relief spending, which rises after a disaster and remains high for several subsequent years.
In the United States, because the federal government and states do not know how much they spend on mitigation in total, we lack the information to accurately compare proactive investments with postdisaster response and recovery expenditures. Although outside the scope of this policy, all levels of government need a more comprehensive understanding of federal and state investments in order to better target funding to help manage the growing costs of catastrophic events.
This policy statement represents APHA’s strong commitment to supporting surveillance and research in high-impact, postdisaster settings. As a means of reaching this goal, APHA:
- Urges that the Environmental Protection Agency (EPA) and the CDC, in partnership with research universities and faculty, engage in medium- to long-term health surveillance of disaster-affected populations. This action should include creating a health risk registry as well as implementing a tracking system for residents’ exposures to chemicals and zoonoses via the One Health Framework. Such a registry could be created, updated, and maintained on a regular basis in coordination with international agencies (e.g., WHO, United Nations Office for the Coordination of Humanitarian Affairs, United Nations Development Programme) and federal agencies including the EPA, the CDC, the U.S. Department of Health and Human Services (DHHS), the Agency for Toxic Substances and Disease Registry, and the Substance Abuse and Mental Health Services Administration (as well as the Council of State and Territorial Epidemiologists). One federal agency, for example the CDC, could take the lead in establishing and maintaining a surveillance system for tracking children’s exposures to environmental hazards in the 12-to 48-month postdisaster period. The system would include the complaints received by state and local agencies as well as information from local/regional hospital emergency departments, emergency responders, Medicaid claims, and other sources that might become available through electronic medical records. In order to facilitate retention, e-health technologies, such as smartphone apps, could be used. Any resulting reports or studies would be widely disseminated to affected communities for their express benefit.
- Urges that the long-term health impacts from high-impact disasters warrant large-scale, longitudinal public health cohort studies. The relevant research-mandated agencies (e.g., WHO, CDC, DHHS/NIH) invest long-term resources in monitoring, preventing, and treating postdisaster-related illnesses, especially using a trauma-informed (e.g., carrying out interventions with knowledge of how trauma manifests and avoiding retraumatization of affected individuals) and life course approach. As seen in previous emergency situations, this includes linking exposures to poor sanitation and lack of stable housing as well as psychological consequences of these disasters. These types of postdisaster health studies will help quantify human health sequelae, particularly among vulnerable, underrepresented, and underserved populations. These populations are disproportionately affected by disasters. These studies should also include health workers (e.g., first responders) and those involved in disaster relief and clean-up efforts. Due to previous work indicating long-term negative health outcomes in postdisaster settings, a cohort can help assess interventions wherein community resilience was paramount, uncover gaps when it was not, and define long-term strategies for preparedness in the wake of more intense and more frequent climate change–related weather events. Any resulting reports or studies would be widely disseminated to affected communities for their express benefit.
- Urges future or existing short-term postdisaster response projects (e.g., responses from the CDC or the Federal Emergency Management Agency), including local, state, and/or federal disaster-specific cooperative agreements, to take advantage of existing health care infrastructure. Indeed, if this infrastructure is not damaged during the disaster itself, agencies can invest some of the extant funding (i.e., beyond appropriations for conventional relief work) into mid-term evaluations. This evaluative component would be used to provide the groundwork and initial support for future medium- or long-term research and/or surveillance activities, as outlined in the first two action steps.
1. Watts N, Amann M, Ayeb-Carlsson S, et al. The Lancet countdown on health and climate change: from 25 years of inaction to a global transformation for public health. Lancet. 2018;391:P581–P630.
2. Newkirk VR. The true scope of the disaster in Puerto Rico. Available at: https://www.theatlantic.com. Accessed December 22, 2019.
3. Santos-Burgoa C, Sandberg J, Suarez E, et al. Differential and persistent risk of excess mortality from Hurricane Maria in Puerto Rico: a time-series analysis. Available at: https://www.thelancet.com. Accessed December 22, 2019.
4. Kishore N, Marques D, Mahmud A, et al. Mortality in Puerto Rico after Hurricane Maria. N Engl J Med. 2018;379:162–170.
5. American Public Health Association. Policy statement on the role of state and local health departments in planning for community health emergencies. Available at: https://www.apha.org. Accessed December 22, 2019.
6. Kuh D, Shlomo YB, eds. A Life Course Approach to Chronic Disease Epidemiology. Oxford, England: Oxford University Press; 2004.
7. Panditharatne M. New data: 2 million Puerto Ricans risk water contamination. Available at: https://www.nrdc.org. Accessed December 22, 2019.
8. Brown P, Vega CMV, Murphy C, et al. Hurricanes and the environmental justice island: Irma and Maria in Puerto Rico. Environ Justice. 2018;11:148–153.
9. American Public Health Association. Establishing environmental public health systems for children at risk or with environmental exposures in schools. Available at: https://www.apha.org. Accessed December 22, 2019.
10. Adams RE, Boscarino JA. Differences in mental health outcomes among Whites, African Americans, and Hispanics following a community disaster. Psychiatry. 2005;68:250–265.
11. Ratnapradipa D, Cardinal C, Ratnapradipa K, Scarbrough A, Xie Y. Implications of Hurricane Harvey on environmental public health in Harris County, Texas. Environ Health. 2018;81:24–32.
12. Powell T, Thompson SJ. Enhancing coping and supporting protective factors after a disaster: findings from a quasi-experimental study. Res Soc Work Pract. 2016;26:539–549.
13. Chi PC, Urdal H, Umeora O, Sundby J, Spiegel P, Devane D. Improving maternal, newborn and women’s reproductive health in crisis settings. Cochrane Database Syst Rev. 2018;8:CD011829.
14. Lock S, Rubin J, Murray V, et al. Secondary stressors and extreme events and disasters: a systematic review of primary research from 2010–2011. Available at: http://currents.plos.org. Accessed December 22, 2019.
15. Bonanno GA. Loss, trauma, and human resilience: have we underestimated the human capacity to thrive after extremely aversive events? Am Psychol. 2004;59:20.
16. Somerfield MR, McCrae RR. Stress and coping research: methodological challenges, theoretical advances, and clinical applications. Am Psychol. 2000;55:620.
17. Tulane University. Demographic and health disparities in recovery from Hurricane Katrina. Available at: https://sph.tulane.edu. Accessed December 22, 2019.
18. Vasterling JJ. The aftermath of Hurricane Katrina: a trauma researcher’s perspective. Traumatology. 2008;14:21–26.
19. Jiao Z, Kakoulides S, Moscona J, et al. Effect of Hurricane Katrina on incidence of acute myocardial infarction in New Orleans three years after the storm. Am J Cardiol. 2012;109:502–505.
20. Lagmay AMF, Racoma BA, Aracan K, Alconis-Ayco J, Saddi IL. Disseminating near-real-time hazards information and flood maps in the Philippines through Web-GIS. J Environ Sci. 2017;59:13–23.
21. Egawa S. International Research Institute of Disaster Science: fact finding mission to the Philippines. Available at: http://irides.tohoku.ac.jp. Accessed December 22, 2019.
22. Hasegawa J, Hidaka H, Kuriyama S, et al. Change in and long-term investigation of neuro-otologic disorders in disaster-stricken Fukushima prefecture: retrospective cohort study before and after the Great East Japan Earthquake. PLoS One. 2015;10:e0122631.
23. World Health Organization. FAQs: Fukushima five years on. Available at: https://www.who.int. Accessed December 22, 2019.
24. Centers for Disease Control and Prevention. Cooperative agreement for emergency response: public health crisis response—2017 hurricane recovery crisis cooperative agreement interim guidance. Available at: https://www.cdc.gov. Accessed December 22, 2019.
25. National Cancer Institute. Cohort studies in cancer epidemiology. Available at: https://epi.grants.cancer.gov. Accessed December 22, 2019.
26. US Census Bureau. Vintage population estimates. Available at: www.census.gov. Accessed December 22, 2019.
27. Freimuth VS, Quinn SC. African Americans’ views on research and the Tuskegee Syphilis Study. Soc Sci Med. 2001;52:797–808.