2

Value and Use of Mortality and Morbidity Data

The act of quantifying mortality and morbidity following a traumatic event, such as a large-scale disaster, holds deep emotional, societal, financial, and logistical value and serves a multitude of different uses for different stakeholders. Accurately quantifying disaster-related mortality and morbidity is a complex and challenging endeavor. The meaningful use of morbidity and mortality data is often undermined by the fact that no uniform framework or standard vocabulary for conceptualizing disaster-related mortality and morbidity data is in use across all jurisdictions, federal and state, local, tribal, and territorial (SLTT) agencies, and professional domains. Data are captured inconsistently, and data that are collected are not being used to their fullest potential due to the siloing of agencies and systems. The first part of this chapter lays out a potential framework for conceptualizing disaster-related mortality and morbidity and introduces updated case definitions developed by the committee. The second part of the chapter discusses the value and meaningful use of mortality and morbidity data by various stakeholders across the disaster lifecycle and provides examples of how these data are currently used or could be used. Chapters 3 and 4 will focus on the analytical and operational challenges and practices related to the collection, reporting, and recording of individual counts and population estimates of mortality and morbidity.

CONCEPTUALIZING ALL-CAUSE MORTALITY AND MORBIDITY

Significant confusion and disagreement persist across systems and stakeholders regarding what counts as a disaster-related death or morbidity,

which profoundly affects the ability to use mortality and morbidity data in meaningful ways. Resolving this discordance and moving toward consistently applied standards and harmonized practices will require more than merely developing simple case definitions, however. It will require taking a broader understanding of disaster-related mortality and morbidity data that considers the context, timing, and methods by which data are collected and recorded as well as considering the methods used to assess and use the data to protect the health of the public. Although a framework with common definitions can be helpful in catalyzing and supporting the adoption of this more comprehensive type of approach, no uniform framework is widely used in current practice.

A framework to guide the assessment of mortality and morbidity would provide a methodological structure for more accurately and completely categorizing and reporting those outcomes in a consistent manner. Ideally, such a framework would strike a balance between uniformity and flexibility, would be applicable to all disasters (e.g., small- or large-scale, human-induced or naturally occurring), and would include case definitions that are designed to capture all mortality and morbidity related to the event while also excluding cases that are unrelated (Combs et al., 1999). A uniform approach for quantitatively describing a disaster’s health impacts would also enable analyses of the effectiveness of disaster management activities across different disasters. A uniform approach could also support:

• More consistent assessment of the human impact of disasters across all jurisdictions;
• Delivery of adequate resources for recovery;
• Forecasting of needs for similar incidents in the future;
• Identification of behavioral contributors to disaster-related mortality and morbidity to inform interventions to modify future behavior;
• Exploration of population impacts to promote health and offer relevant services and support for prevention and recovery; and
• Identification of vulnerable populations and their specific needs to improve services and reduce additional morbidity, injury, and death.

To address the need for a uniform approach for conceptualizing and assessing mortality and mortality data following large-scale disasters, the committee developed a framework that can be adopted across all systems and jurisdictions (see Table 2-1). This framework incorporates the two primary methodological approaches for estimating disaster-related mortality and morbidity and builds on a body of literature of analytical prospective and retrospective methodologies (Kishore et al., 2018; Santos-Burgoa et al., 2018; Stephens et al., 2007).

TABLE 2-1 Proposed Framework of Approaches for Defining Mortality and Morbidity Following Large-Scale Disasters

This proposed framework builds on and provides standard meaning to the disaster mortality definitions promoted by the Centers for Disease Control and Prevention and works in tandem with designations for natural and unnatural deaths used by medical examiners and coroners. It is important to note that the case definitions proposed by the committee are aligned with—and not intended to replace—the language already being used in the medical examiner and coroner community to categorize different manners and causes of death. Most states offer five options for coding the

manner of death: natural, homicide, suicide, accidental, and undetermined (NAME, 2002). The manner of death is how that injury or illness led to the death, while the cause of death refers to the specific illness or injury that led to death (Washoe County Regional Medical Examiner’s Office, 2020). Natural deaths are those that have internal physiological causes. The term unnatural death is used by medical examiners and coroners to categorize a death that did not occur due to natural causes (IOM, 2003). As discussed below, the committee’s case definition for a direct death will capture only unnatural disaster-related deaths, while the case definitions for indirect deaths and partially attributable deaths can capture both natural and unnatural deaths. Another potential advantage of using the committee’s uniform case definitions and framework is that it offers medical certifiers of individual cases of mortality (e.g., doctors, nurse practitioners) greater autonomy in categorizing a death as being indirectly related or partially attributable to a disaster without triggering a mandatory review by the

medical examiner. Chapter 3 provides more detail about the roles of these stakeholders in mortality data collection and reporting.

Most significantly, this framework shifts the paradigm for defining a disaster’s health impacts from a singular death toll toward a more inclusive understanding of the complex impact of a disaster on human life. In a major disaster, for example, the total mortality estimate can remain dynamic for years, as individuals succumb to injuries or health conditions that occurred as a result of their exposure to the disaster. Even the number of deaths that can be directly attributed to the force of the disaster (i.e., direct deaths) can change over time, because people injured in the event may eventually die of those same injuries or morbidities years later. Insufficient research exists to define a clear minimum timeline by which disaster-related mortality and morbidity should be tracked; however, setting such guidelines for the capture of data is critical (see Recommendations 3-2 and 3-3 on standards for collection of mortality and morbidity data). Box 2-1 provides an overview of some additional stakeholder considerations that provide further rationale

for the use of a uniform approach to assessing disaster-related mortality and morbidity.

Methodological Approaches for Assessing Total Disaster-Related Mortality and Morbidity

Disasters are complex events with such multifactorial health consequences that no single number can sufficiently describe a disaster’s health impact. This precludes the possibility of any universal methodological approach that can be used across all disasters to generate an estimation of mortalities or morbidities related to that disaster. This complexity also gives rise to a persistent challenge in the assessment of mortality and morbidity, which is the widespread conflation of the outputs of two different but complementary methodological approaches for estimating disaster-related mortality and morbidity: (1) individual counts, which are numbers derived from individual administrative case reports, and (2) population estimates, which are based on statistical approaches. The committee’s proposed framework includes both of these essential methodological approaches. Both approaches provide essential information in the face of a disaster, but they differ in their assumptions, data requirements, strengths, weaknesses, and appropriate uses (see Table 2-2). They are similar in that they pertain to a defined point in time and to a geographical area, and both can be refined over time as the situation evolves or as new data become available. When applied appropriately, the two approaches can help answer different questions, elucidate different sets of risk factors, and uncover different potential points of intervention. Therefore, both individual counts and population estimates contribute to a comprehensive picture of a disaster’s health impacts, which can be used to inform response and recovery and to prepare for future events.

Individual counts are estimates that are based on counts of deaths recorded in administrative systems and are valuable for understanding the immediate impact of disasters. However, the accuracy of this approach depends on the completeness with which individual cases are recorded and reported. Depending on the strength and precision of the data collection system to capture information about each bit of data accurately and consistently, individual count methods often fail to capture certain types of disaster-related deaths (e.g., individuals who die of natural causes during a disaster and would not have died but for the disaster). However, individual counting methods, if deployed successfully, can provide an early estimate of the number of reported individual deaths, injuries, and cases of illness that are considered to be directly or indirectly caused by the disaster or partially attributable to it. Operational considerations for the collection, recording, reporting, and use of individual counts can be found in Chapter 3.

TABLE 2-2 Strengths, Weaknesses, and Uses for Individual Counts and Population Estimate Approaches

SOURCE: Adapted from Appendix C.

Unlike individual counts, population estimates of total disaster-related mortality and morbidity are derived by estimating the number of mortalities and morbidities using statistical means, such as representative and complex sampling, survey-based methods or using a variety of excess mortality and morbidity methods (e.g., comparing deaths or illness rates in the disaster-affected population to rates observed in the same population during the previous year or during a relevant time period). Population-based estimation methods are crucial for capturing a full understanding of the impacts of a disaster on health and mortality. These methods are often reported in ways that convey the appearance of less precision (i.e., they provide a point estimate with confidence intervals) compared with reports of individual

counting methods (which provide a single number, but no confidence intervals, thereby implying greater certainty around the estimate) and in some applications (e.g., estimates of excess deaths) they cannot distinguish individuals who would have survived in the absence of the disaster from those who would have died during the period regardless. Chapter 4 will explore the landscape of population estimation methods and identify potential best practices for conducting and using these analyses.

It is critical to recognize that individual counts are not always superior to population estimates based on samples or vice versa. For some audiences, the term “count” might imply greater precision than the term “estimate,” but this assumption is incorrect and both approaches can produce valuable estimates of the true effect of a disaster, which are useful for different purposes. Importantly, both are estimates in the sense that they are ways of gauging total impact that are inherently incomplete and subject to variability over time and according to the specific methods used. Therefore, focusing solely on individual counts limits the scope of an investigation. A more effective strategy is to apply both approaches to assessing mortality and morbidity, which makes the assessment more valuable in terms of understanding the complex nuances of the disaster’s impact and the population’s vulnerabilities.

Multifactorial Problem

Another concept critical to understanding disaster-related mortality and morbidity is that all quantitative assessments developed using either approach represent a description of the disaster’s impact at a specific point in time based on a unique set of conditions and assumptions. These estimates can change over time as more data are gathered, additional mortality and morbidity occur, new assumptions are developed, and updated analyses are performed. Regardless of the methodological approach applied, the assessment of mortality and morbidity is a complex multifactorial problem that is influenced by time, resources, capability, and the health of the affected population. It is impossible to definitively know the true impact of a disaster on human life. Instead, this report attempts to highlight how the administrative, organizational, logistical, and analytical components associated with each of these approaches can be improved to make counts and estimates more accurate and complete reflections of the disaster’s true effect.

As a data recording and reporting system matures and is able to more accurately count individual deaths, it will capture progressively more of the deaths that are included in the population-based estimates. This will consequently lessen the disparity between the population estimates of mortality and the number of individual deaths reported, for example. Still, certain categories of indirect and partially attributable mortality and morbidity

may always be more difficult to detect in practice at the individual level (e.g., heart attack, stroke) and will tend to be better captured through population estimation methods.

Conclusion 2-1: Current terminology and case definitions used to describe disaster-related mortality and morbidity fail to capture the differences in assessment methods used and the totality and temporality of disaster-related deaths and significant morbidity. The lack of a uniform framework for assessing disaster-related health impacts undermines the quality and usability of these data in informing disaster management.

Accuracy of Individual Counts and Degree of Attribution

The precision of estimates of disaster-related mortality and morbidity made using individual counts depends on the accuracy of decisions that are made about the strength of association of an individual outcome to a disaster. In the case of mortality, medical examiners, coroners, or other medical certifiers¹ must consider for each individual death (1) the type of death, (2) the degree to which the death can be attributed to a disaster, and (3) the temporality of the death—“the timescale over which the death is expected and can be attributed to a disaster in the context of different types of disasters” (Green et al., 2019, p. 452). Given the multidimensionality of these judgments, the different parameters that the responsible parties use to assess the degree of attribution drive variation in the types and quantities of outcomes that are collected and recorded. See Chapter 3 for a discussion of the variation in data collection and recording practices throughout the medicolegal system. Multiple terms have been used to denote the presence and degree of a relationship between a death or injury and a disaster. These are often conflated, resulting in misunderstandings about disasters’ impacts and also poor comparability between mortality and morbidity assessments over time and across disasters. Another fundamental disadvantage of the individual counting approach is the risk of failing to count difficult-to-capture cases, such as natural deaths that would not have occurred but for the disaster.

To improve consistency and provide guidance for attributing mortality and morbidity to a disaster using an individual-count approach, the committee developed a set of terms for use throughout the report: namely, an individual reported disaster-related mortality or morbidity can be categorized

___________________

¹ Medical certifiers are medical professionals with the authority to record a cause of death and sign a death registration following a death, typically in a hospital or a health care setting. Medical certifiers are commonly physicians, nurse practitioners, or physician assistants, although this varies by state based on state law (see Chapter 3 for more information on medical certifiers and death registration).

as directly related, indirectly related, or partially attributable to a disaster. These terms are intended to be flexible enough yet also to be precise enough to support accurate and consistent decision making on whether a case is related to the disaster’s impact or other consequences and, if so, to what extent (see Table 2-1). Consistent use of these three terms is also intended to facilitate more effective communication to the media and public about the different ways that an individual reported death or morbidity can be related to a disaster.

Direct Mortality and Morbidity

Direct deaths or morbidities are directly attributable to the disaster itself, such as deaths due to blunt force injury or drowning as a consequence of a disaster’s physical force. In the language of cause of death reporting, a direct death is always an unnatural death (Green et al., 2019). Direct deaths tend to be the primary focus of data collection efforts, particularly in the medicolegal death investigation system, because these outcomes are generally the most straightforward to collect and record. However, exclusively capturing direct deaths and failing to account for direct morbidities or for indirect or partially attributable deaths underestimates the actual short- and long-term impacts of the disaster. Additionally, the attribution of individual direct deaths to a disaster does not have an obvious natural end point. As discussed above, a person who dies as a direct consequence of a disaster should always be considered a direct death or morbidity, even if the death occurs many years later. Box 2-2 provides an overview of data sources for individual counts of disaster-related deaths and morbidity.

Indirect Mortality and Morbidity

Indirect disaster-related mortality and morbidity data capture natural and unnatural deaths and morbidities that are associated with—but not directly caused by—the event. Essentially, the criteria for an indirect death require that the death would not have occurred “but for” the conditions present due to the disaster. Indirect deaths include outcomes due to unsafe or unhealthy conditions around the time of the disasters or during any phase of the disaster lifecycle that contribute to a death (Combs et al., 1999) (see Table 2-1). For example, carbon monoxide poisoning due to unsafe generator use during a power outage resulting from a hurricane, or deaths due to lack of access to essential medications or treatments, such as dialysis, should be recorded and reported as indirect deaths. Indirect mortality and morbidity data can offer a wealth of information about a disaster’s impact and provide actionable evidence to inform disaster response and on how to prevent deaths during future events. However, accurately

capturing and recording indirect deaths can be hampered by subjectivity in determinations by medical examiners, coroners, or medical certifiers and often by incomplete evidence to support attribution at the time the death is certified. Strategies that could help prevent the loss of these valuable data on indirect mortality and morbidity include (1) establishing a common policy and philosophical approach for collecting and recording these data, (2) providing professional training to reduce variation in practice, and (3) developing improved tools and methods for capturing these data. In addition, providing a third category for individuals for whom their mortality or morbidity might be partially attributable to the disaster should help mitigate the risk of losing valuable data on these individuals’ outcomes.

Partially Attributable Mortality and Morbidity

Partially attributable mortality and morbidity is an intentionally nebulous category compared to direct or indirect deaths or morbidities. It encompasses those deaths or morbidities that cannot be tied to the disaster with a high degree of certainty but where the death, injury, or illness was more likely than not to be at least partially related to the disaster. In other words, the death would be unlikely to have occurred “but for” the disaster, but it cannot be tied definitively to it. For instance, this category would include a person who dies because the disaster caused stress or exhaustion that exacerbated a pre-existing chronic condition (Green et al., 2019), such as a person with known heart disease who suffered a heart attack while in a shelter environment. At present there is no term commonly in use to define these types of deaths or morbidities, and because these deaths can be

variably labeled as indirect, the lack of a common terminology generates inconsistency and possible bias across jurisdictions or individual certifiers. The committee aims to address this problem with the category of partially attributable deaths. Importantly, this category is not static, and partially attributable deaths or morbidities can be reclassified as indirect deaths or morbidities as more evidence of causation becomes available.

Population Estimation Methods

Population estimates quantify mortality and morbidity related to a disaster at a population level through statistical analyses and epidemiological approaches to assess the size and characteristics of the population affected. The analytical approaches used to develop population estimates include surveys using representative or complex sampling of affected populations as well as estimates derived by comparing observed deaths or morbidities during the disaster time to what was observed in a prior time frame or to a comparison population (Green et al., 2019; Kishore et al., 2018; Stephens et al., 2007). Population estimates often include a broader range of disaster-associated effects than are captured by individual counts, which frequently undercount the true number of cases, because population estimates inherently include indirect and partially attributable deaths and morbidities. Thus, for example, population estimates of disaster deaths will often be larger than the number of direct or indirect deaths captured through individual count methods, sometimes by very large margins (see Appendix C for examples). Population estimates have been developed by researchers following many previous disasters. For instance, modeling excess mortality is a statistical approach for conducting population estimates. Models of excess mortality developed following Hurricane Maria in 2017 in Puerto Rico demonstrate the value of comparing immediate all-cause mortality reported in a community during a disaster to a baseline number of deaths that would be expected in the same community from a disaster of that magnitude (Santos-Burgoa et al., 2018). Other analytical approaches, such as those used in other social science fields such as demography and anthropology can be applied to estimate the size of an affected population and develop excess mortality and morbidity estimates, particularly in situations where the affected population is hard to count. For example, mortality estimates following the 1985 Mexico City earthquake were made using a network scale-up method to determine the estimated victim count (Bernard et al., 1991). Sampling approaches and analyses of electronic medical records and related data are especially useful for understanding the long-term, non-fatal consequences of disasters, including mental health issues (see Chapter 4).

ATTRIBUTING MORBIDITY TO A DISASTER

While the cause of death may sometimes be difficult to determine, mortality itself is easier to define than morbidity. Morbidities related to disasters represent an exceptionally broad range of health outcomes that span physical injuries, chronic and infectious conditions, and psychological impacts in addition to having long-term impacts on communities, health systems, and economies. The physical and mental health effects of disasters can be immediate or delayed (Adeola and Picou, 2012) and can occur as a result of the direct forces of the disaster, such as an injury, or as a result of the conditions brought forth by the disaster. The latter includes such scenarios as the interruption of mental health services or the decline of health maintenance activities (e.g., blood pressure testing or access to prescription drugs), which exacerbate existing vulnerabilities and pre-existing co-morbidities to produce additional morbidity and mortality (Bourque et al., 2009; Schnall et al., 2011).² For example, research following Hurricane Katrina found that the interruption of health maintenance activities was an indicator for additional future morbidity since up to one-half of evacuees seeking shelter in the Astrodome and Red Cross facilities lacked access to their prescription medications (Brodie et al., 2006; Greenough et al., 2008). Common disaster-related morbidities include infectious diseases (Kouadio et al., 2012), chronic diseases exacerbated by disaster conditions (Miller and Arquilla, 2008; Mokdad et al., 2005), and mental health problems, including self-harm and substance abuse, caused or worsened by exposure to intense stressors (McFarlane and Williams, 2012) (see Table 2-3 for an overview of select research on disaster-related morbidities).

Further complexity is added by the fact that different disasters produce a different landscape of morbidities (Bourque et al., 2009). A major flood is not expected to produce an increase in burns and a wildfire should not generally lead to an increase in near-drownings. In disasters where large populations are temporarily displaced in close-quartered shelters, gastrointestinal and respiratory infections are likely to be prevalent (Schnall et al., 2011), while terrorist attacks are likely to result in physical and psychological trauma. Chronic diseases such as asthma, cardiovascular conditions, and diabetes and their associated co-morbidities represent a larger proportion of morbidities associated with disasters, particularly environmental disasters, in the United States (Schnall et al., 2011).

What should count as a disaster-related morbidity might also be shaped by the intended uses of the data—for example, to address hospital capacity issues, to assess short- or long-term disabilities versus specialty conditions,

___________________

² See Appendix D for an exploration of the causal links across social determinants of health and disaster-related morbidity and mortality through the lens of COVID-19.

TABLE 2-3 Selected Research on Physical and Psychological Morbidities Associated with Disaster Exposure

or to evaluate the impact on the health system more broadly during the response and recovery periods.

Despite the range of information that is possible to collect, the variation in morbidity across disasters, and the ongoing lack of a standard approach for collecting morbidity data, it is likely that a group of key morbidities could be distilled across common disasters (e.g., hurricanes, tornadoes, floods, wildfires, and extreme temperature events). This standard dataset could provide a starting point for the collection of more standardized data points and approaches for data collection and analysis (see Recommendation 3-3 and Conclusion 3-6). Potential morbidities on target for consistent data collection across a range of common disasters should move beyond the definition of “significant morbidity” and be inclusive of morbidities that are known to be associated with those common socioeconomic and environmental conditions prevalent following most disasters. These include morbidities associated with mass displacement, environmental exposure, extreme stress, and lasting infrastructure damage, among others. These data, even if imperfect and incomplete to start, could provide actionable information for disaster management policy and practice.

In particular it is well documented that socially disadvantaged and underserved communities suffer disproportionally from disasters and morbidities. More research is needed on how to best mitigate these disparities before, during, and after a disaster. To achieve this aim, research and decisive action are needed to develop a consensus around which morbidities and other appropriate indicators would be most useful to collect before, during, and after every disaster or specific types of disasters, and what analytical methods and systems should be developed or enhanced to facilitate the collection, analysis, and use of these data (Bourque et al., 2009). More discussion about gaps and potential opportunities for collecting, recording, and using morbidity data for individual counts can be found in Chapter 3. Further discussion of methods for population estimates can be found in Chapter 4.

OVERLOOKED VALUE OF MORBIDITY DATA

Assessing health outcomes is a critical component of improving rapid response and recovery following a disaster through the allocation of resources and targeted public health messaging and enhancing prevention and mitigation activities during the inter-disaster period (Schnall et al., 2011). The collection and use of morbidity data, however, is an often overlooked component of the disaster management enterprise, which tends to focus on mortality as an indicator of disaster-related health impacts. When acted on appropriately, morbidity data can help to reduce mortality (i.e., by preventing morbidities from becoming mortalities) and can be used to help shape

public health messaging and medical preparedness. For end users in the field of disaster management, in particular, estimates of morbidity resulting from a disaster may actually be of more value than mortality data in informing life-saving mitigation and preparedness activities and in enhancing real-time response. Therefore, an exclusive focus on mortality data, the traditional outcome of interest, at the expense of morbidity data is tantamount to focusing only on the worst case and diverts responders’ attention from efforts that could reduce human suffering and save additional lives.

A recent example of the power of morbidity data to prevent additional suffering is the testing and tracking of individual coronavirus disease 2019 (COVID-19) cases throughout the world. In Singapore and Hong Kong, among other places, surveillance data were used successfully in the early months of the pandemic to identify and isolate cases in order to prevent additional mortalities and avoid a drain on medical resources, especially in intensive care units (Purnell and Solomon, 2020). For hospitals, evidence-based models for quantifying expected morbidities following a disaster may be of much greater value than expected mortality models, because injuries and disease tend to consume a great deal of health system resources.

Evaluation of Health System Access, Capacity, and Cost

Morbidity data can be used to assess health system functions, costs, and access to care over time in order to support the shifting needs of patients with chronic conditions such as diabetes (Lee et al., 2016) during and after a disaster. For example, in the 6 years following Hurricane Katrina, Peters et al. (2014) noted a three-fold increase in the percentage of admissions to Tulane University Hospital for acute myocardial infarction (Peters et al., 2014). Another 2009 modeling study found that Hurricane Katrina also (1) had a significant effect on diabetes management, (2) had exacerbated existing racial/ethnic health disparities, and (3) had an estimated lifetime cost of $504 million for the affected adult population (Fonseca et al., 2009). This research indicates that not only do disaster-related morbidities put pressure on local and regional health care systems but that access to care and the costs of obtaining care are significant and represent variables that could be addressed proactively with better and earlier access to descriptive morbidity data combined with data on related sociodemographic factors. Overall, the disaster management enterprise remains underinvested in understanding morbidity and how using morbidity data can contribute to saving lives, protecting health, and improving health equity, as will be discussed in Chapters 3 and 4.

VALUE AND USE OF DATA ACROSS THE DISASTER LIFECYCLE

Morbidity and mortality data add value to all phases of the disaster lifecycle, from the immediate aftermath of an event through the recovery and inter-disaster periods. The value of these data and how they can be used optimally will shift over time. For instance, quantifying the health impacts of a disaster can help to determine the disaster’s scale and inform resource deployment at the early stages. During later phases, the data can be used for predictive planning, risk mitigation, and other efforts to improve preparedness and strengthen public health systems to perform better during future disasters. Figure 1-1 in Chapter 1 provides a more detailed description of the use of mortality and morbidity data across the disaster lifecycle. The integration of applied epidemiology into disaster management can provide a reliable, actionable evidence base for decision makers and other stakeholders (Malilay et al., 2014). However, extracting the value of mortality and morbidity data is dependent on the right data being collected and having the right methods and systems in place to effectively analyze and use the data. This requires determining the types of data with the most value in ensuring people’s well-being at each phase of the disaster management cycle.

In terms of their functional value, mortality and morbidity data enable the Federal Emergency Management Agency (FEMA), the Department of Health and Human Services (HHS), and other federal and SLTT agencies involved in a response to (CDC, 2016):

• quantify disaster health impacts and ongoing hazards;
• detect and track epidemiological trends;
• limit further health impacts;
• ensure a common operating picture;
• inform resource allocation;
• shape public messaging and control rumors;
• provide support to individuals and families;
• target interventions and other public health responses;
• monitor and evaluate the effectiveness of the response and recovery efforts; and
• evaluate the effectiveness of prevention and mitigation activities and inform preparedness planning.

Value and Use of Data During Response and Recovery Period

The response and recovery periods include the disaster impact and its immediate aftermath and includes emergency response efforts and efforts to restore infrastructure and services (DOI, 2020). The timing of optimal data use during the response and recovery periods depends on the type of

disaster, its severity, and the areas and populations affected. However, the primary value of data during these periods is to track the evolution of the disaster in order to save lives and limit further deaths and health consequences. For example, the HHS emPOWER program draws on data from the Centers for Medicare & Medicaid Services (CMS) to create dynamic maps, tools, and resources to identify and provide support during disasters to CMS beneficiaries who live alone and are dependent on electricity for their health care needs (ASPR, 2020). The program has been deployed successfully during Hurricane Matthew in Florida (2016) and during the severe wildfires in Los Angeles, California (2017) (Finne, 2018).

Accurate real-time and historical mortality and morbidity data are critical for ensuring a common operating picture and real-time situational awareness as a disaster unfolds and recovery efforts begin. All stakeholders need reliable, accurate information with which to inform their decisions. Therefore, a lack of shared information and poor access to data among stakeholders can hinder response efforts. Although mortality and morbidity data present a significant opportunity to better target future planning, mitigation, and response efforts—particularly for vulnerable communities—the usability of these data is reliant on their consistent collection, reporting, and interpretation (see Chapter 3).

Targeting Response and Recovery Efforts and Resource Allocation

During the immediate response period, reliable real-time mortality and morbidity data can be used to assess and respond to the evolving current needs. For example, the data can guide the strategic allocation of resources in response to situational needs, which can help limit future morbidity and mortality. For example, the Assistant Secretary for Preparedness and Response’s Emergency Support Function #8 (ESF8)—Public Health and Medical Services—offers federal support to strengthen SLTT-level disaster response capabilities. Under ESF8, the National Disaster Medical System can respond to spikes in mortality and morbidity to provide human resources in settings where health system capacity and infrastructure have been compromised by a disaster. Disaster medical assistance teams (DMATs) can help triage mass casualties and provide acute care, while disaster mortuary operational response team (DMORT) medical examiner and coroner services can provide standalone morgue operations and human remains identification services (ASPR, 2012). Spikes in mortality can provide real-time evidence for the need for support from DMATs and DMORTs. Rapid increases in initial mortality and morbidity counts could also indicate the need for specific types of health care services or resources. These initial raw data captured immediately following a disaster can also help responders predict and prepare for subsequent waves of mortality and morbidity that

often occur during the following days, especially if health delivery systems and transportation infrastructure are incapacitated (Malilay et al., 2014).

In addition to the use of real-time data to address current needs, historical morbidity data can be used in the immediate aftermath of a disaster to model expected trajectories and outcomes and to identify vulnerable populations. These models can provide essential details for informing the planning and execution of the response that cannot be derived rapidly from other sources. For example, prospective models based on historical data could predict whether local-level capacity is capable of continuing to manage the current threat or if state or federal resources may be needed. These approaches would be particularly valuable for determining current or future access to basic resources such as food, clean water, and energy in vulnerable communities as well as for identifying and addressing critical gaps in service delivery (Malilay et al., 2014). For example, following Hurricane Maria in Puerto Rico, individuals requiring dialysis or diabetes medication were often unable to access them due to a lack of power and clean water and debris blocking access roads (Cordero, 2019).

Financial and Emotional Support for Survivors and Families

Mortality and morbidity data can also help alleviate the devastating impacts of disasters on individuals and families by guiding the provision of resources and support. Mortality data in particular have major financial implications for individuals and families seeking funeral assistance from FEMA, because one of the eligibility requirements for this support is documentation that the death occurred either directly or indirectly as a result of the disaster impact (NCHS, 2017).³ If a death was not attributed as such by the medical examiner, coroner, or other medical certifier, then the death may be ruled ineligible for funeral assistance, or assistance may be delayed (Bowden, 2018). Data are also used to determine eligibility for other types of resources and support, to guide social and medical benefits and payouts in a standardized way, to streamline assistance programs, and to reduce public confusion about how to access support. For instance, FEMA can use data following a disaster to provide timely individual and public assistance to supplement SLTT resources, such as the crisis counseling program that

___________________

³ Under the Other Needs Assistance provision of the Individuals and Households Program, FEMA provides financial assistance to help with the cost of uninsured expenses for a death or interment caused directly or indirectly by a presidentially declared emergency or major disaster (see 44 C.F.R. § 206.119(c)(4)). Applicants must submit an official death certificate clearly indicating that the death was attributed to the emergency or disaster, either directly or indirectly, or a signed statement from an SLTT government-licensed medical official (e.g., medical examiner or coroner) directly or indirectly attributing the death to the emergency or disaster (FEMA, 2019).

offers funds for mental health services to communities in disaster-affected areas as defined by the Robert T. Stafford Disaster Relief and Emergency Assistance Act (Stafford Act) (FEMA, 2016; State of Michigan, 2020).

Public Health Emergency Communication and Response

Effective public health messaging in a disaster-affected area is shaped by accurate and timely mortality and morbidity data. Timely and clear communication following a disaster keeps the public informed about the unfolding event, protects them from ongoing hazards, and dissuades individuals from taking risks due to a lack of awareness or haste. For example, historical and real-time mortality and morbidity data can be used to identify and communicate risks to the public such as carbon monoxide poisonings, unsafe drinking water sources, fallen electrical wires, or hazardous chemical and environmental exposures (Malilay et al., 2014). These data can also be used to inform the timing and frequency of communicating the existence of digital tools and resources that are available to individuals in their immediate area or in an unfamiliar area to which they have been evacuated. For example, the communication of accessible tools like RxOpen’s real-time pharmacy information could enhance access to critical medications for these groups (RxOpen, 2020). For disaster survivors with chronic health conditions, limited access or a lack of access to essential medications and medical equipment can directly affect immediate and future health.

Value and Use of Data During the Inter-Disaster Period

The inter-disaster period encompasses mitigation and preparedness efforts that occur in the interim phase between the end of one disaster and the beginning of the next (Malilay et al., 2014). The core value of mortality and morbidity data during this period is to save lives and protect human health during the next disaster (Green et al., 2019). Such data can be used to identify vulnerabilities and exposure-related risk factors, to improve the allocation of resources on a regional level, and to proactively inform the design of interventions to reduce morbidity and mortality from future events (Malilay et al., 2014). In the case of human-caused disasters (such as terrorism, or climate change-related events), better data on health effects and interactions might not only mitigate disaster impacts but even help to prevent future disasters.

Building Health System Capacity

To support health systems in building response capacity, mortality and morbidity data can be used to predict demands on health care delivery

systems, to guide resource and service usage patterns, to strengthen physical infrastructure, and to organize staff and systems to deploy during a future response (Malilay et al., 2014). Gaps in access to care and gaps along the continuum of care can also be identified to improve health systems during and between disasters. Data can be used to assess costs to the medical system at set intervals following a disaster and to perform cost-effectiveness analyses to compare the costs associated with a large-scale disaster response versus the costs of prevention, mitigation, and planning; this can help to prioritize resource allocation and investment in lower-cost planning activities (Malilay et al., 2014). Additionally, mortality data from previous disasters can inform the windows of coverage for FEMA assistance and other benefits for future disasters.

Improving Policy and Practice

Mortality and morbidity data can be used to muster appropriate levels of support for legislation and funding for programs, infrastructure, and resources. More accurate predictions of excess mortality from certain types of disasters could serve as powerful levers for policy change to prevent such disasters or mitigate their impacts. Furthermore, comparing actual morbidity and mortality data following a disaster to historical data for similar prior disasters can feed back into continuous improvement of preparedness policies and activities (Schnall et al., 2017). Mortality and morbidity data also can be used reflexively to evaluate and improve mortality and morbidity data collection practices. These efforts may include assessing the effectiveness of specific case definitions and data sources with respect to different health outcomes (Malilay et al., 2014) and developing institutional best practices for collecting, using, and sharing data (e.g., standardized practices for assessing morbidity and mortality).

Cultivating Community Resilience

Protecting the public by investing in community resilience⁴ is a long-term public health goal, so data should be collected with sufficient granularity to be relevant at the community level. Data can be used to strengthen community resilience to future disasters in myriad ways. In addition to strengthening community-level health systems and emergency management infrastructure, data can be used to understand biopsychosocial and environmental conditions—including the proximal and distal influencers of disparities—that affect a community’s susceptibility to disaster-related mortality

___________________

⁴ Community resilience is multidimensional, spanning six types of community capital: natural/environmental, buildings and infrastructure, financial and economic capital, human and cultural capital, social capital, and political capital (NASEM, 2019).

and morbidity. Mortality and morbidity data are critical for understanding and mitigating the disproportionate impacts of disasters on vulnerable populations. These populations need to be identified during the inter-disaster period so that their access and functional needs can be supported through targeted interventions and messaging (Browning et al., 2006; Feehan and Salganik, 2016; Lowe et al., 2013; Mace et al., 2018; Wolkin et al., 2015).⁵ For example, in discussions with community leaders, the committee learned that the response to the Camp Fire that struck Paradise, California, revealed that the city was reliant on the Butte County Health Department for access to public health data. When the county was unable to provide needed data during the disaster and recovery, Paradise officials were forced to look to the local hospital to provide data that could have been used to enhance preparedness and response capacities at the local and county levels. Furthermore, data on disaster-related mortality and morbidity are important for helping communities access resources and support services as well as for supporting fundraising and advocacy. In Florida, for example, the committee learned that data from the state’s robust women, infants, and children program showed a significant disaster-related reduction in the number of certain program participants after a hurricane, which informed strategies to reengage with specific communities. Demonstrating the value of data collected during disasters can also help to improve transparency and build trust between emergency management and communities.

Mortality and morbidity data hold important value for monitoring a community’s progress over time in the recovery period, which can last for years. Longitudinal morbidity data can be used to track long-term health sequelae of disasters, including mortality risk.⁶ Some of these outcomes are shaped by age, gender, and other sociodemographic factors that have been established by longitudinal studies conducted after Hurricane Katrina (Adams et al., 2011; Adeola and Picou, 2012), Hurricane Maria (Santos-Burgoa et al., 2018), and the 2004 Indian Ocean tsunami (Frankenberg et al., 2014; Ho et al., 2017). Studying relative mortality over time also allows for charting the effects of cumulative stress and chronic disease as the survivors age (Ho et al., 2017). Going forward, standardizing the collection of longitudinal mortality and morbidity data would enable comparisons of

___________________

⁵ Medically vulnerable groups include people who are hospitalized, people who have electricity grid–sensitive conditions (e.g., hypertension and diabetes) and people living with infectious diseases such as tuberculosis and AIDS (Bernard et al., 2010). Grid-sensitive conditions are especially common among people in nursing homes and people receiving home care (Banks, 2013; DeSalvo et al., 2014). Socially vulnerable groups include people in financially precarious situations who become homeless and unemployed, leading to indirect deaths by overdose or suicide, as well as local homeless populations.

⁶ Longer-term physical health consequences are a particular concern when a disaster causes environmental hazards that can persist for years (e.g., oil spills or radiation) (Frankenberg et al., 2014).

specific communities at selected points in time after incidents in order to develop common 6-month, 1-year, 3-year, and 5-year datasets, for example.

Stress, posttraumatic stress disorder, depression, and anxiety are common mental health morbidities and co-morbidities after a disaster (Adams et al., 2014; Adhikari Baral and Bhagawati, 2019; Beaglehole et al., 2018; Buttke et al., 2012; Mulchandani et al., 2019; Norris et al., 2002). Using data to better understand the risk factors for these mental health sequelae could help to inform screening, prevention, and intervention efforts in communities affected by disasters (Adams et al., 2014).

Ongoing Public Health Communication

Another valuable use of mortality and morbidity data is in evaluating how effectively public health messaging is able to drive behavioral change and reduce risk across the disaster management cycle (Malilay et al., 2014). For instance, data on the health consequences of improper generator use could be used to shape messaging and policy in ways to mitigate similar mortality in subsequent disasters (Schnall et al., 2017).

Value of Multidimensional Mortality and Morbidity Data

As described throughout this chapter, mortality and morbidity data represent a wide variety of uses and values. These data, if accurate and complete, can be used to identify at-risk populations, among other uses, and respond with appropriate actions to support recovery, mitigate root vulnerabilities, and prepare to prevent future harm, which represents great value to the field of disaster management. Critically, mortality and morbidity data alone represent just one category of data and further contextualization of these data with other rich data points, such as race and ethnicity, socioeconomic status, among others, provides for a multidimensional understanding of those same mortality and morbidity data. The integration of these data represents real opportunities to identify the underlying causal pathways and sub-population inequities existing at the intersection of the social determinants of health (SDOH), disaster exposure, and disaster-related mortality and morbidity, which could in turn allow for the improved design and targeting of resources and programs to the sub-population in greatest need. The contextualization of morbidity and mortality using SDOH data adds additional value and evidence to foster a stronger and more responsive disaster management enterprise that prioritizes community resilience.

The contributory role of SDOH on vulnerability should not be deemphasized. For example, Thomas-Henkel and Schulman (2017, p. 1) write, “SDOH can account for up to 40% of individual health outcomes, particularly among low-income populations, [and] their providers are increasingly

focused on strategies to address patients’ unmet social needs (e.g. food insecurity, housing, transportation, etc.).” Co-morbidities are one significant consequence of these unmet social needs (Valderas et al., 2009), which add distinct complexity to the assessment and use of disaster-related mortality and morbidity data. Certain socio-environmental factors—population density, exposure to pollution, outdoor manual labor—are known to increase biological susceptibility to disease, and those impacts seep into individual treatment decisions and care of various medical conditions (McKibben, 2020). During and after a disaster, these influences are even more pronounced. Other SDOH, such as race and economic status, are known to be associated with persistent inequities in health and are critical to examine alongside disaster-related mortality and morbidity data to provide a foundation of evidence for promoting community resilience. Excluding an assessment of SDOH in the overall assessment of post-disaster mortality and morbidity significantly limits the opportunity to prevent disaster-related mortality and morbidity; an oversight that would have especially deleterious consequences for regions with large communities of disadvantaged and underserved populations and that are also susceptible to natural disasters, such as Puerto Rico, the U.S. Gulf Coast, and areas experiencing frequent wildfires (e.g., California).

RECOMMENDATIONS

Mortality and morbidity data related to large-scale disasters represent a poorly tapped resource for critical information to improve the nation’s ability to respond to disasters and save lives. Fundamentally, the lack of a consistently used framework for attributing mortality and morbidity results in the inconsistent collection and reporting of data on the scope and causes of mortality and morbidity over time and across disasters. The committee’s framework responds to this critical gap and is unique in that it balances the value of both individual count and population estimation methods for developing quantitative indicators of total mortality or morbidity and provides updated individual count case definitions to characterize the level of attribution for all deaths related to disasters of all types.

REFERENCES

Bourque, L., J. Siegel, M. Kano, and M. Wood. 2009. Morbidity and mortality associated with disasters. In H. Rodríquez, E. L. Quarantelli, and R. R. Dynes (eds.), Handbook of disaster research. New York: Springer. Pp. 97–112.

Bowden, J. 2018. FEMA has approved 3 percent of requests for Puerto Rico funeral aid. The Hill, September 11. https://thehill.com/homenews/administration/406169-fema-has-approved-just-3-percent-of-requests-for-puerto-rico-funeral (accessed June 26, 2020).

Brodie, M., E. Weltzien, D. Altman, R. J. Blendon, J. M. Benson. 2006. Experiences of Hurricane Katrina evacuees in Houston shelters: Implications for future planning. American Journal of Public Health 96(8):1402–1408. doi: 10.2105/AJPH.2005.084475.

Browning, C. R., S. L. Feinberg, D. Wallace, and K. A. Cagney. 2006. Neighborhood social processes, physical conditions, and disaster-related mortality: The case of the 1995 Chicago heat wave. American Sociological Review 71(4):661–678.

Buttke, D., S. Vagi, A. Schnall, T. Bayleyegn, M. Morrison, M. Allen, and A. Wolkin. 2012. Community assessment for public health emergency response (CASPER) one year following the Gulf Coast oil spill: Alabama and Mississippi, 2011. Prehospital and Disaster Medicine 27(6):496–502.

CDC (Centers for Disease Control and Prevention). 2016. A primer for understanding the principles and practices of disaster surveillance in the United States, 1st ed. Atlanta, GA: Centers for Disease Control and Prevention.

Combs, D. L., L. E. Quenemoen, R. G. Parrish, and J. H. Davis. 1999. Assessing disaster-attributed mortality: Development and application of a definition and classification matrix. International Journal of Epidemiology 28(6):1124–1129.

Cordero, J. 2019. Session 1: Hurricane Maria in Puerto Rico. Presentation at the August 29, 2019, public meeting of the National Academies of Sciences, Engineering, and Medicine’s Committee on Best Practices for Assessing Mortality and Significant Morbidity Following Large-Scale Disasters, Washington, DC.

DeSalvo, K., N. Lurie, K. Finne, C. Worrall, A. Bogdanov, A. Dinkler, S. Babcock, and J. Kelman. 2014. Using Medicare data to identify individuals who are electricity dependent to improve disaster preparedness and response. American Journal of Public Health 104(7):1160–1164.

DOI (Department of the Interior). 2020. Natural disaster response and recovery. https://www.doi.gov/recovery (accessed August 1, 2020).

Feehan, D. M., and M. J. Salganik. 2016. Generalizing the network scale-up method: A new estimator for the size of hidden populations. Sociological Methodology 46(1):153–186.

FEMA (Federal Emergency Management Agency). 2016. Crisis counseling assistance and training program guidance. https://www.samhsa.gov/sites/default/files/images/fema-ccp-guidance.pdf (acccessed June 26, 2020).

FEMA. 2019. Fact sheet—funeral assistance. https://www.fema.gov/media-librarydata/1565189678667-0fdafea4dbca363ed6380cf2ae453aa8/FACTSHEET_FuneralAssistanceFINA2019Compliant.pdf (accessed August 2, 2020).

FEMA. 2020. Rumor control. https://www.fema.gov/rumor-control (accessed June 10, 2020).

Finne, K. P. 2018. HHS emPOWER program: Translating research into practice. PowerPoint presentation. sites.nationalacademies.org/cs/groups/dbassesite/documents/webpage/dbasse_187829.pdf (accessed June 26, 2020).

Fonseca, V. A., H. Smith, N. Kuhadiya, S. M. Leger, C. L. Yau, K. Reynolds, L. Shi, R. H. McDuffie, T. Thethi, and J. John-Kalarickal. 2009. Impact of a natural disaster on diabetes: Exacerbation of disparities and long-term consequences. Diabetes Care 32(9):1632–1638.

Frankenberg, E., M. Laurito, and D. Thomas. 2014. The demography of disasters. In International Encyclopedia of the Social and Behavioral Sciences (Area 3), 2nd ed. Amsterdam, The Netherlands: North Holland. Pp. 1–22.

Goldmann, E., and S. Galea. 2014. Mental health consequences of disasters. Annual Review of Public Health 35(1):169–183. https://doi.org/10.1146/annurev-publhealth-032013-182435.

Green, H. K., O. Lysaght, D. D. Saulnier, K. Blanchard, A. Humphrey, B. Fakhruddin, and V. Murray. 2019. Challenges with disaster mortality data and measuring progress towards the implementation of the Sendai Framework. International Journal of Disaster Risk Science 10:449–461.

Greenough, P. G., M. D. Lappi, E. B. Hsu, S. Fink, Y.–H. Hsieh, A. Vu, C. Heaton, and T. D. Kirsch. 2008. Burden of disease and health status among Hurricane Katrina-displaced persons in shelters: A population-based cluster sample. Annals of Emergency Medicine 51(4):426–432.

Ho, J. Y., E. Frankenberg, C. Sumantri, and D. Thomas. 2017. Adult mortality five years after a natural disaster. Population and Development Review 43(3):467–490.

Horney, J. 2017. Disaster epidemiology: Methods and applications. London: Academic Press.

IOM (Institute of Medicine). 2003. Medicolegal death investigation system: Workshop summary. Washington, DC: The National Academies Press. https://doi.org/10.17226/10792.

Kishore, N., D. Marqués, A. Mahmud, M. V. Kiang, I. Rodriguez, A. Fuller, P. Ebner, C. Sorensen, F. Racy, J. Lemery, L. Maas, J. Leaning, R. A. Irizarry, S. Balsari, and C. O. Buckee. 2018. Mortality in Puerto Rico after Hurricane Maria. New England Journal of Medicine 379(2):162–170.

Kouadio, I. K., S. Aljunid, T. Kamigaki, K. Hammad, and H. Oshitani. 2012. Infectious diseases following natural disasters: Prevention and control measures. Expert Review of Anti-infective Therapy 10(1):95–104.

Lee, D. C., V. K. Gupta, B. G. Carr, S. Malik, B. Ferguson, S. P. Wall, S. W. Smith, and L. R. Goldfrank. 2016. Acute post-disaster medical needs of patients with diabetes: Emergency department use in New York City by diabetic adults after Hurricane Sandy. BMJ Open Diabetes Research and Care 4(1):e000248.

Lowe, D., K. L. Ebi, and B. Forsberg. 2013. Factors increasing vulnerability to health effects before, during and after floods. International Journal of Environmental Research and Public Health 10(12):7015–7067.

Mace, S. E., C. J. Doyle, K. Askew, S. Bradin, M. Baker, M. M. Joseph, and A. Sorrentino. 2018. Planning considerations for persons with access and functional needs in a disaster—Part 1: Overview and legal. American Journal of Disaster Medicine 13(2):69–83.

Malilay, J., M. Heumann, D. Perrotta, A. F. Wolkin, A. H. Schnall, M. N. Podgornik, M. A. Cruz, J. A. Horney, D. Zane, R. Roisman, J. R. Greenspan, D. Thoroughman, H. A. Anderson, E. V. Wells, and E. F. Simms. 2014. The role of applied epidemiology methods in the disaster management cycle. American Journal of Public Health 104(11):2092–2102.

McFarlane, A. C., and R. Williams. 2012. Mental health services required after disasters: Learning from the lasting effects of disasters. Depression Research and Treatment 2012:970194.

McKibben, B. 2020. Racism, police violence, and the climate are not separate issues. The New Yorker. https://www.newyorker.com/news/annals-of-a-warming-planet/racism-police-violence-and-the-climate-are-not-separate-issues (accessed September 1, 2020).

Miller, A. C., and B. Arquilla. 2008. Chronic diseases and natural hazards: Impact of disasters on diabetic, renal, and cardiac patients. Prehospital and Disaster Medicine 23(2):185–194.

Mokdad, A. H., G. A. Mensah, S. F. Posner, E. Reed, E. J. Simoes, M. M. Engelgau, and Chronic Diseases and Vulnerable Populations in Natural Disasters Working Group. 2005. When chronic conditions become acute: Prevention and control of chronic diseases and adverse health outcomes during natural disasters. Preventing Chronic Disease 2(Special Issue):A04.

Mulchandani, R., M. Smith, B. Armstrong, C. R. Beck, I. Oliver, and English National Study of Flooding and Health Study Group. 2019. Effect of insurance-related factors on the association between flooding and mental health outcomes. International Journal of Environmental Research and Public Health 16(7):1174.

Nakagawa, I., K. Nakamura, M. Oyama, O. Yamazaki, K. Ishigami, Y. Tsuchiya, and M. Yamamoto. 2009. Long-term effects of the Niigata-Chuetsu earthquake in Japan on acute myocardial infarction mortality: An analysis of death certificate data. Heart 95(24):2009–2013.

NAME (National Association of Medical Examiners). 2002. A guide for manner of death classification, 1st ed. https://name.memberclicks.net/assets/docs/4bd6187f-d329-4948-84dd-3d6fe6b48f4d.pdf (accessed August 1, 2020).

NASEM (National Academies of Sciences, Engineering, and Medicine). 2019. Building and measuring community resilience: Actions for communities and the Gulf Research Program. Washington, DC: The National Academies Press. https://doi.org/10.17226/25383.

NCHS (National Center for Health Statistics). 2017. A reference guide for certification of deaths in the event of a natural, human-induced, or chemical/radiological disaster. Hyattsville, MD: National Center for Health Statistics. https://www.cdc.gov/nchs/data/nvss/vsrg/vsrg01.pdf (accessed June 26, 2020).

Noe, R. 2018. Applications: Disaster-related mortality surveillance: Challenges and considerations for local and state health departments. In J. A. Horney (ed.), Disaster epidemiology: Methods and applications. London: Academic Press. Pp. 55–63.

Norris, F. H., M. J. Friedman, P. J. Watson, C. M. Byrne, E. Diaz, and K. Kaniasty. 2002. 60,000 disaster victims speak: Part I. An empirical review of the empirical literature, 1981–2001. Psychiatry 65(3):207–239.

Peters, M. N., J. C. Moscona, M. J. Katz, K. B. Deandrade, H. C. Quevedo, S. Tiwari, A. R. Burchett, T. A. Turnage, K. Y. Singh, E. N. Fomunung, S. Srivastav, P. Delafontaine, and A. M. Irimpen. 2014. Natural disasters and myocardial infarction: The six years after Hurricane Katrina. Mayo Clinic Proceedings 89(4):472–477.

Purnell, N., and F. Solomon. 2020. Coronavirus doesn’t have to be so deadly. Just look at Hong Kong and Singapore. The Wall Street Journal. https://www.wsj.com/articles/coronavirus-doesnt-have-to-be-so-deadly-just-look-at-hong-kong-and-singapore-11590491418 (accessed October 28, 2020).

RxOpen. 2020. Facilities map. http://rxopen.org (accessed June 26, 2020).

Santos-Burgoa, C., J. Sandberg, E. Suárez, A. Goldman-Hawes, S. Zeger, A. Garcia-Meza, C. M. Pérez, N. Estrada-Merly, U. Cólon-Ramos, C. M. Nazario, E. Andrade, A. Roess, and L. Goldman. 2018. Differential and persistent risk of excess mortality from Hurricane Maria in Puerto Rico: A time-series analysis. Lancet Planet Health 2(11):e478–e488.

Schnall, A. H., A. F. Wolkin, R. Noe, L. B. Hausman, P. Wiersma, K. Soetebier, S. T. Cookson. 2011. Evaluation of a standardized morbidity surveillance form for use during disasters caused by natural hazards. Prehospital and Disaster Medicine 26(2):90–98.

Schnall, A., R. Law, A. Heinzerling, K. Sircar, S. Damon, F. Yip, J. Schier, T. Bayleyegn, and A. Wolkin. 2017. Characterization of carbon monoxide exposure during Hurricane Sandy and subsequent nor’easter. Disaster Medicine and Public Health Preparedness 11(5):562–567.

State of Michigan. 2020. Michigan awarded two federal grants to strengthen behavioral health services during COVID-19 crisis. https://www.michigan.gov/coronavirus/0,9753,7-406-98158-526521--,00.html (accessed June 26, 2020).

Stephens, K. U., Sr., D. Grew, K. Chin, P. Kadetz, P. G. Greenough, F. M. Burkle, Jr., S. L. Robinson, and E. R. Franklin. 2007. Excess mortality in the aftermath of Hurricane Katrina: A preliminary report. Disaster Medicine and Public Health Preparedness 1(1):15–20.

Swerdel, J. N., T. M. Janevic, N. M. Cosgrove, J. B. Kostis, and the Myocardial Infarction Data Acquisition System Study Group. 2014. The effect of Hurricane Sandy on cardiovascular events in New Jersey. Journal of the American Heart Association 3(6):e001354.

Thomas-Henkel, C., and M. Schulman. 2017. Screening for social determinants of health in populations with complex needs: Implementation considerations. Center for Health Care Strategies, Inc. https://www.chcs.org/media/SDOH-Complex-Care-ScreeningBrief-102617.pdf (accessed September 1, 2020).

Tzvetkova, S. 2017. Not all deaths are equal: How many deaths make a natural disaster newsworthy? Our World in Data, July 19. https://ourworldindata.org/how-many-deaths-make-a-natural-disaster-newsworthy (accessed June 28, 2020).

Valderas, J., B. Starfield, B. Sibbald, C. Salisbury, and M. Roland. 2009. Defining comorbidity: Implications for understanding health and health services. Annals of Family Medicine 7(4):357–363. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2713155/pdf/0060357.pdf (accessed September 1, 2020).

Washoe County Regional Medical Examiner’s Office. 2020. What is the difference between manner and cause of death? https://www.washoecounty.us/coroner/faq/difference_cause_and_manner_of_death.php (accessed August 1, 2020).

Watson, J., M. Gayer, and M. Connolly. 2007. Epidemics after natural disasters. Emerging Infectious Diseases 13(1):1–5. doi: 10.3201/eid1301.060779.

Wolkin, A., J. R. Patterson, S. Harris, E. Soler, S. Burrer, M. McGeehin, and S. Greene. 2015. Reducing public health risk during disasters: Identifying social vulnerabilities. Journal of Homeland Security and Emergency Management 12(4):809–822.