Keywords

FormalPara Learning Objectives

This chapter should enable readers to understand and discuss:

  • The stages, outcomes, and potential modification of the natural history of an infectious disease in patients

  • The importance of understanding the natural history

  • The intrinsic determinants of the natural history

  • Common obstacles to conducting natural history studies in outbreaks

  • Strategies to improve the natural history research response in outbreaks

1 Introduction

When a novel or reemerging pathogen begins to infect large numbers of people in a short period, a concerted emergency research response is required. Preclinical and clinical research to characterize and understand the natural history of a disease and its determinants is critical for improving patient care and outcomes, including through the development of medical countermeasures (MCMs). For example, HIV/AIDS almost inevitably resulted in disability and death for many years after the first clinical descriptions (CDC 1981, 1982; Fauci and Lane 2020; Lundgren et al. 2023). After an inexcusably slow start in the eyes of many patients, HIV/AIDS ultimately motivated collaborative, well-resourced research attention to understand the natural history in a broad global effort (Fauci 2021) that led to effective therapeutics. Persons infected with HIV/AIDS now generally enjoy a high quality of life for a near-normal span, provided they receive timely diagnosis and standard treatment. This represents a remarkable redirection of the natural history of an infectious disease once correctly seen as a death sentence.

Four decades later, but far more rapidly, the natural history of COVID-19 after SARS-CoV-2 infection has arguably received more clinical, research, and public health attention than any infectious disease in history. Well-resourced research at unprecedented scale, pace, and high level of resolution has made it possible relatively quickly to provide better care and improve clinical outcomes. A detailed understanding of the natural history of the human-SARS-CoV-2 interaction has been essential to the accelerated development of preventive and therapeutic countermeasures.

Until the COVID-19 pandemic response made it plain, the need to integrate research into infectious disease emergency response was slow to win broad acceptance, even as the scientific tools facilitating an accelerated research response were becoming increasingly powerful and available. Similar efforts were not normative in the past; with few exceptions, understanding and reporting the natural history of clinical disease caused by novel or new-variant pathogens lagged behind public health response and other research efforts. An absent, incomplete, or at best low-resolution picture of the natural history of clinical disease in humans has historically been the rule for most emerging or reemerging infectious diseases. We focus in this chapter on general principles important to understanding and reporting natural history to improve patient outcomes, illustrating these using prototypic examples of infectious diseases that have posed historical and may pose future challenges.

2 Framework

2.1 Defining Terms: The Natural History of an Infectious Disease

As articulated by the U.S. Food and Drug Administration (FDA), natural history refers to the “course a disease takes in the absence of intervention in individuals with the disease, from the disease’s onset until either the disease’s resolution or the individual’s death” (FDA/CDER 2019). Another description adds detail: the “natural course of a disease from the time immediately prior to its inception, progressing through its pre-symptomatic phase and different clinical stages to the point where it has ended and the patient is either cured, chronically disabled, or dead without external intervention” (de la Paz et al. 2010; Jewell 2016). Although typically emphasized in the study of cancer and rare genetic diseases, fundamental concepts of the natural history are crucial to consider in understanding and treating diseases caused by infectious pathogens.

A distinguishing feature of the natural history of an infectious disease is that it emerges from an evolving host–pathogen interaction, including an exposure that leads to an infection (acute or persistent); the onset of a disease syndrome (acute or chronic); and outcomes traditionally captured at either individual (as disability, dysfunction, or death); or population levels (as morbidity or mortality). We consider these stages in greater depth below, focusing on the clinical bedside and patient outcomes, the importance of understanding the natural history, factors determining the natural history, and how the natural history is optimally captured, understood, and reported to improve clinical outcomes.

2.2 Context: Infectious Disease Outbreaks

Effectively understanding and reporting disease natural history presents particular challenges for patients, clinicians, and researchers in infectious disease outbreaks.Footnote 1 This chapter focuses on understanding and reporting natural history in the context of historic, current, and future infectious disease outbreaks of epidemic and pandemic potential, i.e., the infectious diseases that have given rise to Public Health Emergencies of International Concern (PHEIC) since the International Health Regulations were revised in 2005 (WHO 2016) and those considered high risk for future emergencies. Past is prologue, and the research measures that have been most effective should carry forward into response to future outbreaks, with improvements made possible by scientific and technological advances and organizational refinements (Simpson et al. 2020; Van Kerkhove et al. 2021). For example, virus families of concern are included in the prototype pathogen approach (► Chap. 12) (Cassetti et al. 2022; Ford et al. 2023). A few selected examples—COVID-19, Ebola virus disease, Lassa fever, and mpox—will be explored here in more depth, with reference to other infectious diseases where relevant.

2.3 Target: Patient-Centered Care

The interplay between the natural history of the individual’s health and public or community health cannot be disentangled, especially in infectious disease outbreaks. Though perspectives on the relative primacy of the individual or the community vary dramatically among cultures, we focus here primarily on how an infectious disease evolves in individual patients. In the clinical and clinical research setting, natural history may be best characterized at the level of resolution of the individual patient (or larger groups of individual patients) as disease manifests, evolves, and is modulated. There is much to be said about how an infectious disease outbreak evolves at population and community levels, but these themes are covered elsewhere (► Chaps. 21 and 26, In Focus 21.1).

2.4 Approach: Principles and Practice

In considering the importance of understanding and reporting the natural history of an infectious disease with the goal of improving care for patients, each section will first consider general principles broadly applicable across diseases and outbreaks, while recognizing that each infection, disease, and outbreak setting is unique. More concrete examples will follow, drawing predominantly from a few exemplary infectious diseases (COVID-19, Ebola virus disease [EVD], Lassa fever, and mpox). Practical examples may highlight disease-specific successes, but more often than not illustrate knowledge gaps, cautionary notes related to current uncertainty, research attention needed to remedy uncertainty, and suggested strategies to address these gaps in the future.

3 Stages of the Natural History and Outcomes in Outbreaks

3.1 Overview: Through the Lens of the Host–Pathogen–Care Interaction

The clinician working in emergency outbreak response usually interacts with patients presumed to be ill with the disease of concern, either presenting for care or having received some initial care. The provision of even the most basic clinical care is by definition an “intervention.” Despite the natural history definitions above, observation absent any medical intervention would be unethical in human clinical care or research (unlike animal modeling). We will consider the natural history and its outcomes through the host–pathogen–care framework, focusing specifically on medical aspects of clinical care. Effective redirection of the natural history requires consideration of the features of the host–pathogen interaction that determine infection, disease, and the outcomes of that disease in an infected individual. General considerations applicable to most infectious diseases will be illuminated by pathogen- and disease-specific examples.

As seen in ◘ Fig. 1, the natural history can be considered in stages, beginning with an exposure (1A) of a human host to a virus that may lead to an infection. An infection may cause (1B) symptoms or signs of disease that vary across a spectrum of severity; those with significant disease are more likely to come to medical attention and receive clinical care (1C), eventually leading to immediate (acute) and longer-term outcomes (1D). Understanding the determinants of outcomes requires consideration of features intrinsic to the host, the virus, and the exposure. The interplay between these antecedents determines emergent features of the natural history. For example, peak viral load or viral load at admission, often among the strongest predictors of outcome, is not determined solely by the virus, the host, or the exposure, but emerges from their complex interaction, which in turn reflects increasingly complex interactions at multiple levels between a pathogen and human physiological systems, organs, tissues, cells, and so on. This host–pathogen–care heuristic will anchor much of the subsequent discussion which focuses on the clinical bedside and outcomes. Later sections consider how intrinsic host, virus, and exposure characteristics might determine the natural history.

Fig. 1
figure 1

Stages of the natural history and outcomes. (Author, artwork by Jiro Wada)

Full size image

3.1.1 Distinguishing Between Infection and Disease

As defined epidemiologically (as distinguished from a molecular virological definition), “pathogenicity” describes the proportion of infected persons who develop signs or symptoms, while “virulence” refers to the proportion who develop severe disease or death (CDC 2012). Characterization of the natural history is most important in hospitalized patients, who are at risk for the worst outcomes, and on whom we focus in ► Sect. 3.2. However, it is important to recognize the disease spectrum after most viral infections includes subclinical or mild disease in many if not most cases: failure to recognize this might impact the clinician or clinical researcher’s analysis of risk-benefit in decision making and research study design. For most viral infections, the longer-term natural history in infected patients with asymptomatic or only mild disease is unclear, but might reasonably be assumed to be less consequential. Whether the same is true for the public-health risks associated with asymptomatic or subclinical infection, namely, viral persistence that could cause new outbreaks, most likely depends on the specific viral infection.

3.1.2 Practice: Distinction Between the Infection Fatality Rate (IFR) and the Case Fatality Rate (CFR)

The COVID-19 pandemic has demonstrated the importance of distinguishing between outcomes in all those infected vs. those who come to medical attention. With rare exceptions, this holds true for almost all viral pathogens. It has long been recognized, for example, that most serologically confirmed infections with Lassa virus (LASV) do not cause clinical disease, producing a very low IFR. Hospitalization with confirmed Lassa fever, by contrast, is associated with CFRs >20 to 50% (Buba et al. 2018; Grant et al. 2023; Okokhere et al. 2018). Some of these distinctions have only become apparent when large outbreaks prompted careful research into the infection vs. disease spectrum. For example, the CFR of Ebola virus disease (EVD) was presumed for many years to be >70 to 80% and to approximate the IFR. Since the first identification of EVD in 1976, the CFRs from the few larger outbreaks (of >100 patients) corroborated this assumption (Jacob et al. 2020). More recent outcomes from much larger outbreaks in West Africa (2014–2016) suggest a lower overall CFR even in the absence of virus-specific therapeutics (Rojek et al. 2019), and a wider spectrum of disease that includes individuals who were exposed and infected but did not develop, recognize, or recall clinical symptoms (Gayedyu-Dennis et al. 2023; Glynn et al. 2017; Kelly et al. 2022; Timothy et al. 2019).

3.2 The Clinical Bedside and the Evolving Natural History

3.2.1 Overview

For obvious reasons, understanding the natural history of disease in the hospitalized patient is paramount: disease outcomes are typically modified here as the host–pathogen–care interaction evolves. In that regard, effective clinical management redirects the natural history of an infectious disease on three fronts (◘ Fig. 1c):

  1. 1.

    Safe and effective pathogen-targeted therapeutic intervention (e.g., antiviral therapeutics)

  2. 2.

    Safe and effective disease-modifying therapeutic intervention (e.g., immunomodulators)

  3. 3.

    Provision of appropriate supportive care (e.g., intravenous fluids, organ support) across a spectrum of disease severity

The importance of a clear understanding of the natural history to inform tactics on each front will be further explored after outlining key principles that include several cautions. Inadequate characterization of the natural history is the rule early in infectious disease outbreaks and is inescapable with a novel pathogen. Timely capture and reporting of early clinical signals are crucial to urgently informing patient care, optimizing standards of supportive care (► Chap. 20), and setting the stage for well-designed, well-conducted clinical trials to identify safe and effective medical countermeasures (MCMs). Accurate natural history begins to resolve key research questions to be answered in clinical trials:

  • Is a specific intervention safe and effective?

  • For which patient population?

  • At what stage of infection or disease?

  • At which dose and by which route?

  • In which clinical setting?

3.2.2 Capturing Natural History Data at the Clinical Beside

After diagnosis, characterization of clinical disease should routinely capture:

  • Host demographics

  • Medical history (including clinical symptoms of presenting illness, antecedent treatment, and targeted review of systems; exposures; comorbid conditions; chronic medications; and vaccination history)

  • Clinical signs (including vital signs; relevant physical examination, including the presence of intravenous access)

  • Clinical laboratory features (especially of organ dysfunction or disease-related complications)

  • Viral load (in blood and other relevant diagnostic samples)

  • Clinical or postmortem pathology when available

Additional technical capacity, when available, captures more data, often at higher resolution. Electrocardiography and medical imaging, for example, can be very useful but have typically not been available during most outbreaks.

In addition to descriptive characterization of the presenting illness, investigation should also follow disease evolution over time and in response to intervention. Especially with the severe disease seen in outbreaks that become emergencies, a single snapshot at the clinical bedside cannot reliably predict disease course or outcomes. Indeed, accurately describing the natural history requires effectively mapping dynamic interactions between the pathogen (load, location) and the host (response, damage), with the goal of illuminating how the interactions determine outcomes and if, how, and when the disease trajectory may be redirected. As observations accrue, clinicians and researchers are increasingly able to delineate key phases of disease expression to inform therapeutic intervention.

3.2.3 Caution: Beware of Assumptions

Predictors of outcome may exist for diseases caused by some of these pathogens, but their reliability is often uncertain. Knowledge is often very limited; even in the rare cases when a relatively thorough understanding of the natural history is available, outcome predictors generated in under-resourced clinical settings may not be predictive in others. For example, the delivery of advanced supportive care, including extracorporeal support, to critically ill patients with Ebola virus disease had been considered by some clinicians to be futile prior to 2014. Albeit in a limited number of patients, the delivery of advanced care provided proof-of-principle that the natural history of EVD could be redirected even in patients with extremely high viral loads and multisystem organ failure (see ► Sect. 3.2.5) (Uyeki et al. 2016b).

3.2.4 Caution: Beware of Magic-Bullet Thinking

Clinical and research attention in outbreak settings has historically focused first on pathogen-specific interventions; more recent attention during the COVID-19 pandemic has also focused on host-directed disease-modifying therapies, e.g., immunomodulatory approaches. At times historically, supportive care efforts have lagged, and have often been under-emphasized during infectious disease outbreaks. Disease-specific therapeutics may be critical to success at the bedside, but are not “magic bullets” that can be uncoupled from effective supportive care as part of a complete bundle of optimized standard of care (SOC). With optimized SOC, all available strategies are deployed to prevent or mitigate clinical symptoms and signs, organ dysfunction and damage, severe disease and death, and clinical sequelae and pathogen persistence in survivors. Effective use of available treatment strategies depends on efficient understanding and reporting of the natural history and rapid dissemination of findings. Increasingly, “living” clinical care guidelines (► Chap. 20) that are regularly updated as the evidence base evolves provide clinicians in outbreak settings with near real-time access to the most current standard of care (SOC) recommendations.

3.2.5 Caution: Beware of Clinical Operational Gaps

Understanding the natural history and redirecting it effectively are resource-intensive endeavors, whether one relies on pathogen-directed interventions, disease-modifying interventions, or optimized SOC. Many clinical operational challenges arise in outbreak settings, including shortages of requisite medical staff, infrastructure, supplies, systems, and security. Renewed motivation and capacity improvements to provide more advanced clinical care and clinical research support in resource-limited outbreak settings have been increasingly evident in recent years, and further efforts are under active discussion at both national and international levels (GPMB 2023; WHO 2022a, b). Nevertheless, challenges intrinsic to under-resourced settings are likely to hinder optimal SOC and clinical research for years to come. Even in well-resourced settings, inadequate understanding of the natural history of a novel infectious disease, as in the early phase of the COVID-19 pandemic response, can lead to suboptimal clinical care, flawed research study design, and unintended harm to patients.

3.2.6 Practice: The Evolving Role of Optimized SOC in EVD

Over many decades after Ebola virus was first identified in 1976, typical clinical care provided in “Ebola isolation units” consisted only of the most basic case management. Patients received limited supportive care targeting the relief of symptoms (fever, pain), the prevention and treatment of dehydration (usually oral), and basic empiric treatment of possible bacterial or malarial coinfections. Delivery of even this minimal bundle of supportive care was constrained by lack of well-trained staff and supplies, suboptimal care environments, and uncertainty about protecting caregivers from infection; even the use of intravenous fluid replacement was considered controversial. Clinical laboratory testing was often unavailable near treatment units. Recognition of the need to improve care, especially in the face of consistently high case fatality, led to calls from the community to refocus on the clinical bedside and patient (Bausch et al. 2007).

Early in the historically largest West Africa EVD outbreak (2014–2016), calls to improve supportive care continued, but progress was limited by frequent resource-mission mismatches as unprepared healthcare facilities and rapidly erected treatment units were overwhelmed in a rapidly expanding outbreak. During the outbreak, a small number of EVD patients were cared for in well-resourced healthcare settings in the United States and Europe. Based on these few observations, the delivery of advanced supportive care that included extracorporeal organ support provided an important proof-of-principle that optimal SOC could be provided safely and effectively to critically ill EVD patients (Uyeki et al. 2016b). These observations also enabled the first high-resolution descriptions of multi-system organ dysfunction and “critical illness phenotypes” in EVD. The asymmetry in care provided and outcomes led to renewed emphasis on the need to develop the evidence base and improve delivery of optimal SOC in African settings (Lamontagne et al. 2018). Though efforts continued to be limited by resource constraints, exemplary fit-for-purpose EVD treatment units were able to provide advanced SOC to infected healthcare workers in West Africa by the end of the outbreak (Dickson et al. 2018).

Building on these initial steps, the delivery of optimal SOC took a major step forward during the 2018–2020 EVD outbreak in the eastern Democratic Republic of Congo. Significant advances during this outbreak that have become standard of care included key components:

  • Novel care structures optimized to provide safe and effective patient-centered care and improve communications between providers, patients, and families (► In Practice 40.1)

  • Near-treatment unit diagnostic clinical laboratories using standard testing platforms

  • Specific guidelines for standardizing optimal SOC in EVD that were rapidly developed, distributed, and trained during the outbreak (WHO 2019)

  • Provision of well-trained staff, including in key specialty areas

  • Procurement of requisite supplies to operationalize optimal SOC

The commitment to provide this capacity and facilitate the standard delivery of improved SOC was evident throughout this outbreak (Fischer et al. 2019). Importantly, these advances also enabled optimal SOC as part of the PALM RCT that led to the first regulatory approval of two effective Ebola virus-specific therapeutics (► In Practice 17.1, 23.1, and 40.1) (Mulangu et al. 2019).

A continued commitment to maintain and improve these standards will be crucial to improve EVD outcomes in the future, particularly in patients with high viral loads and multi-system organ dysfunction, in whom case fatality remains high despite the receipt of effective, pathogen-specific therapeutics. For example, the presence of acute kidney injury (AKI) predicts poorer outcomes even in EVD patients receiving effective therapeutics (Mulangu et al. 2019). Efforts to optimize prevention and management will require a more detailed understanding of the clinical presentation, evolution, risk factors, and pathogenesis of AKI in EVD. Building on prior lessons learned from the care of severely ill patients in the United States and Europe, characterizing the natural history of differing critical illness phenotypes in African patients will be crucial to future success. As illustrated in ◘ Fig. 2, current SOC guidelines for EVD focus on a number of key features of illness and clinical management. Each of these areas reflects key focus areas for further clinical research to understand the natural history of EVD.

Fig. 2
figure 2

Advancing optimized standard of care for Ebola virus disease: key focus areas. (Author)

Full size image

3.3 Outcomes of the Natural History

Accurately describing and reporting outcomes is crucial for clinicians and clinical researchers alike (◘ Fig. 1, 1d). Indeed, the optimal design of clinical studies to identify safe and effective interventions requires defining appropriate candidate interventions with clinical benefit, namely improving how a patient “feels, functions, or survives” (FDA 2020). Clinical outcomes may be clinician-reported, patient-reported, non-clinician observer-reported, or based on a performance assessment. Assessing both acute and longer-term clinical outcomes is important, as is evaluating viral clearance or persistence.

3.3.1 Acute Clinical and Virologic Outcomes

Historically, the outcomes reported from the clinical bedside in outbreak emergencies have often been limited to death or survival. Caution is needed in extrapolation or comparison of historical or current outbreak case fatality ratios to individual outcomes. Each viral disease comes to be associated with an epidemiologically observed case fatality ratio (CFR); over time and multiple outbreaks, average CFRs are assumed to provide a good indicator for natural history. However, early in outbreaks, the proportion of poor outcomes may seem high, since severe or fatal cases are usually the first to bring a new or re-emerging viral pathogen to attention. These early signals thus tend to exaggerate the severity of the disease based on a subset of the most severe cases. Furthermore, overall outbreak CFRs include all outbreak cases, many of whom may have died in the community rather seeking care at a treatment center.

In addition, accurate understanding of natural history requires characterization of the entire spectrum of clinical outcomes, ranging from full health and well-being to long-term disability or death, along with observed and laboratory-measured indications. Even in patients who fully recover, the individual and public health risks of viral persistence argue for investigation and documentation of pathogen clearance.

3.3.2 Post-acute Clinical and Virologic Outcomes

The potential for acute severe viral infections to leave survivors with fixed (non-evolving) or ongoing (evolving) clinical sequelae has been long recognized, perhaps most famously in the encephalitis lethargica syndromes described after the 1918 influenza pandemic (Berger and Vilensky 2014). Given the emergency response required in infectious disease outbreaks, often in challenging and under-resourced settings, and the difficulty in following large numbers of survivors, it is perhaps not surprising that post-acute clinical sequelae of these diseases have not received a great deal of clinical or research attention until recently. Renewed interest was provoked by follow-up of survivors of the 2014–2016 West Africa EVD outbreak (see ► Sect. 3.3.3), and more recently by “long COVID,” or post-acute sequelae of COVID-19 (PASC). Determining the causes of post-acute sequelae can be challenging. They may be a generic consequence of severe and prolonged critical illness, as in the increasingly recognized “post-intensive care syndrome” (Nakanishi et al. 2021; Quinn et al. 2023). They may be specific to the viral infection and disease of interest. Further considerations include defining whether sequelae are a fixed consequence of organ dysfunction/damage that occurred during acute illness or represent an evolving pathobiologic process—either ongoing infection and/or host immunopathology in the presence or absence of the pathogen. Understanding the natural history and longer-term outcomes accurately likely requires prospective, longitudinal, and well-controlled observational cohorts of survivors.

Recent decades have highlighted the potential for viruses, even those once presumed to cause only acute infections, to persist in tissues or bodily fluids considered to be “immune-privileged”. It is heuristically useful to consider viral persistence in terms of consequences for individual and for public health. As illustrated below for EVD survivors, viral persistence poses risks for the individual patient, including recrudescent organ-specific inflammatory syndromes and potential “relapse” of systemic disease that may be clinically indistinguishable from the primary acute infection syndrome. Persistent virus or viral antigens may also contribute to nonspecific post-acute symptoms and signs (e.g., fatigue, arthralgia, and myalgia) or systemic inflammatory syndromes in survivors. Viral persistence may also pose a risk to public health, potentially reigniting human transmission months or even years after an outbreak has ended. For almost all the pathogens under discussion, the host–virus determinants of persistence and these individual or public health consequences remain underdetermined. Viral clearance from blood and other bodily fluids during and after acute infection has become an important virologic outcome, even in clinically recovered patients.

3.3.3 Practice: Clinical Sequelae and Viral Persistence in EVD Survivors

Though long-lasting effects of EVD had been infrequently described since 1976, usually from patient self-report, no controlled observational studies had been published prior to the West Africa outbreak. In its aftermath, case reports and series initially called attention to the need to understand the natural history in EVD survivors in order to address their urgent care needs, and potentially to protect public health. Rapidly assembled small observational cohorts of EVD survivors contributed to a growing understanding, but conclusions from these studies were often limited by the absence of physical examination and clinical laboratory findings. An array of clinical symptoms and signs have been noted in case reports, case series, and observational studies that (at minimum) included physical examination (see ◘ Fig. 2).

However, these data generally came from studies that did not include closely matched control groups, which are critical to determine whether particular sequelae are truly associated with EVD. Highlighted in ◘ Fig. 2 are the clinical features that were significantly different between EVD survivors and a group of close contact controls at 1 year of the 5-year PREVAIL III longitudinal natural history study of EVD survivors (Sneller et al. 2019). Notably, only two of these post-acute clinical syndromes (red dots in ◘ Fig. 3) have been associated with Ebola virus (EBOV) persistence. This result highlights the need to consider both clinical and virologic outcomes and their rare but potential overlap.

Before the West Africa EVD outbreak, very limited data suggested that EBOV or EBOV RNA could persist in survivors, but individual or public health consequences had not been shown (Thorson et al. 2016). It became clear during and after that outbreak that EBOV persistence in immune-privileged tissues and/or bodily fluids had consequences for both the individual survivor and public health (◘ Fig. 4). Emerging data indicated the longer-term persistence of EBOV RNA and in some studies infectious EBOV in the semen of male EVD survivors (Barnes et al. 2017; Deen et al. 2017; Diallo et al. 2016; Fischer et al. 2017; Sissoko et al. 2017a; Sneller et al. 2019; Subtil et al. 2017; Thorson et al. 2021; Uyeki et al. 2016a) (◘ Fig. 4a). Viral persistence in semen was further associated with rare instances of sexual transmission that resulted in ongoing transmission (Diallo et al. 2016; Mate et al. 2015). Case reports of maternal-fetal transmission in pregnant EVD survivors, after resolution of acute EVD or absent previously recognized EBOV infection, signaled a similar public health risk, albeit a rare one (Bower et al. 2016) (◘ Fig. 4b). Finally, the relative risk of maternal-infant transmission of EBOV via breastmilk remains undetermined, but has been strongly suspected to lead to fatal EVD in at least one infant (◘ Fig. 4c) (Sissoko et al. 2017b). Notably, in several of these reports, the mother was not known to have been infected with EBOV, presumably having survived an unrecognized, mild, or subclinical infection.

Fig. 3
figure 3

Post-acute clinical sequelae in observational studies of EVD survivors that included physical examination. Items labeled a and b are symptoms or signs that were significantly different versus close contact controls in the PREVAIL III Natural History Study (Sneller et al. 2019). Data marked c are from uncontrolled cohorts, case series, or case reports. Details of abnormal exam findings (d–f) may be found in Jacob et al. (2020) from which figure 3 was reprinted by permission

Full size image
Fig. 4
figure 4

Individual and public health implications of Ebola virus persistence in EVD survivors. (Author)

Full size image

Individual consequences for EVD survivors have included well-documented case reports of recrudescent organ-specific inflammatory syndromes (uveitis, meningoencephalitis) associated with infectious EBOV persistence (◘ Fig. 4d–e) (Jacobs et al. 2016; Varkey et al. 2015). Though only rarely reported, the actual prevalence of viral persistence and these inflammatory syndromes in the central nervous system (CNS) and eye is unknown. Uncertainty remains about whether such cases presented a public health risk: it had generally been assumed that survivors with viral persistence in the eye or CNS did not pose a threat outside of direct contact with intraocular fluid, cerebrospinal fluid, or associated tissues. During the 2018–2020 EVD outbreak in the Democratic Republic of the Congo (DRC), however, a previously vaccinated patient who was diagnosed with acute EVD subsequently cleared EBOV RNA in blood after treatment with a monoclonal antibody-based therapeutic and recovered. Six months later, the same patient developed severe systemic “EVD-like” illness with detectable EBOV RNA in blood, rapidly decompensated, and died with what was considered an EVD “relapse” after genetic sequencing confirmed relatedness to his initial infection. Given the setting and the severity of illness, cerebrospinal fluid could not be sampled, and relapse from a central nervous system or similar source could not be ruled out. This case led to ongoing human-to-human transmission and more than 90 subsequent EVD cases over a wide geographic area (Mbala-Kingebeni et al. 2021).

Many open questions remain about the risks posed by EBOV persistence in EVD survivors. Routine outbreak genetic sequencing has demonstrated several outbreaks in the DRC to be related to transmission from an EVD survivor (likely from semen) rather than a new zoonotic spillover (Mbala-Kingebeni 2022; Pratt 2021). In 2021, genetic sequencing suggested a new outbreak in Guinea was related to transmission from a survivor from the earlier West Africa EVD outbreak, though the epidemiology and mechanisms of transmission remain unclear (Keita et al. 2021). Open questions include:

  • What are the determinants and risk factors for sexual (or other modes of) transmission from EVD survivors with viral persistence?

  • Does persistence of virus or viral RNA in semen, which poses a rare but consequential public health risk for sexual transmission, have any health consequences for the individual male survivor?

  • Is EBOV or EBOV antigen persistence associated with very common generalized symptoms (e.g., fatigue, arthralgia/myalgia) seen in many survivors? (Thus far, other tissues or organs in which EBOV persists have not been identified.)

  • Are EVD survivors of subclinical infection at risk for viral persistence?

  • During acute EVD, what role could EVD-specific therapeutics play in preventing, mitigating, or treating EBOV persistence?

  • In EVD survivors, what is the role of EBOV-specific antivirals to clear EBOV RNA (e.g. from the semen of male survivors as one signal suggests) (Higgs et al. 2021)?

Studies to date confirm the need to better understand the host-virus-therapeutic determinants of viral persistence and its recrudescent inflammatory or public health consequences at molecular, cellular, organ/tissue, individual, and population levels. Answering these questions will require well-designed natural history studies of EVD survivors that enable longitudinal long-term follow-up and comparison with well-matched controls.

3.3.4 Viral Persistence in Other Diseases of Interest

Lassa fever virus (LASV) and LASV RNA have recently been shown to persist in the semen of male Lassa fever survivors; thus far, any association with human transmission or any clinical sequelae have not been described (Thielebein et al. 2022). Longer-term viral persistence after acute infection has been documented with SARS-CoV-2, though not in immune-privileged sites in particular. Rather, prolonged detection of virus or viral antigen has been most commonly associated with immunodeficient hosts and with ongoing clinical signs and symptoms. The clinical significance of viral antigen in autopsy tissues from patients who died with COVID-19, even long after the resolution of acute illness, remains unclear (Stein et al. 2022). The relationship of SARS-CoV-2 viral or antigen persistence to post-acute sequelae of COVID-19 (PASC) is also unclear, but it is a hypothetical contributor (Davis et al. 2023; Proal et al. 2023). Similarly, prolonged, severe acute clinical disease associated with mpox (monkeypox) virus (MPXV) has been described in immunodeficient people living with HIV/AIDS, but has not so far been associated with typical immune-privileged sites (Fink et al. 2023; Govind et al. 2023).

4 Why Is Understanding and Reporting the Natural History Important?

4.1 Informing and Optimizing Patient Care and Care Guidelines

Rapid characterization of disease in extremely important to inform the urgent development of standards for clinical care in an outbreak, epidemic, or pandemic. It requires both the experience of clinicians directly caring for patients in the field who are qualified to characterize disease processes as well as expertise capable of gathering and considering evidence as it emerges to develop evidence-based consensus guidelines. In recent infectious disease outbreaks, curation of this development has been managed through globally supported outbreak response networks and rapidly established local, national, and global clinical discussions and guidelines panels. Early results should be published or otherwise disseminated as soon as possible. The widespread use of pre-print servers in medicine since the COVID-19 outbreak began has accelerated access to information before peer review is complete. This has contributed to the ability of emergency clinical guideline panels to gather needed evidence but has also made their expertise essential to weigh that evidence, determine any consensus, and identify outstanding questions. Such expert panels have become a key feature of recent large outbreaks (EVD, mpox, COVID-19) and have produced “living” clinical guidelines that can be updated frequently as new evidence emerges (► Chap. 20). These efforts are just the beginning of more careful research efforts to characterize the natural history. A subsequent section will consider how to best continue characterization in the research environment typical of ongoing outbreaks. Early clinical characterization not only informs SOC guidelines but also supports the design of interventional clinical trials to evaluate medical countermeasures (MCMs).

4.2 Informing Rational Design and Implementation of Clinical Trials

Accurate characterization of the natural history is crucial to the rational design of clinical trials to identify safe and effective interventions. Indeed, insufficient or inaccurate understanding of the natural history may lead to clinical trials that are poorly designed, poorly stratified, over- or under-powered, or cannot be generalized to meaningfully impact outcomes. Though clinical trial design and implementation is a focus elsewhere in this volume, a few rational design implications merit mention in this chapter (► Chaps. 12, 14, and 22).

4.2.1 Who: Selecting Study Populations

Accurate natural history information is important to identify, define, and select study populations most likely to benefit or at higher risk of harm from an intervention. In clinical trials of a candidate dengue virus vaccine, recognition that excess hospitalizations for dengue fever were observed among vaccine recipients 2–5 years of age who had not had prior dengue virus exposure/seropositivity was an instructive lesson (Sridhar et al. 2018). Prior understanding of the natural history may be valuable to risk stratification in designing randomization strategies. For example, previously described associations between EBOV load at admission (as proxied by the RT-PCR cycle threshold value) and survival informed rational allocation of randomization in the context of the PALM randomized trial of EVD therapeutics (Mulangu et al. 2019).

4.2.2 When: Timing Therapeutic Interventions

High-resolution characterization of the natural history can help distinguish key phases of an infectious disease that may have implications for the best timing of an intervention. For example, early recognition an early “virologic” phase of COVID-19 (driven by viral replication), and a later, “hyperinflammatory” phase (driven by host immunopathology) informed the design of therapeutic trials of antivirals and disease-modifying immunomodulation (Horby et al. 2021; RECOVERY Collaborative Group 2021, 2022).

4.2.3 How: Adequate Statistical Power for Meaningful Clinical Trial Results

The natural history informs selection of meaningful primary and secondary outcomes in clinical trial design. Outcomes of the natural history are discussed, but the identification of meaningful measures of how a patient “feels, functions, or survives” help determine the statistical power and generalizability of study findings (FDA 2020). Such information is especially valuable in clinical trials of emergency therapeutics with survival as the primary endpoint. Calculating statistical power is especially fraught early in outbreaks, when the typical disease course is not yet clear and CFRs may be biased toward the severe end of a disease spectrum.

4.2.4 How: Identifying Biomarkers of Disease

As defined by the FDA-NIH Biomarker Working Group, a biomarker is a “defined characteristic that is measured as an indicator of normal biological processes, pathologic processes, or biological responses to an exposure or therapeutic intervention” (FDA 2023; FDA-NIH Biomarker Working Group 2021). Well-designed natural history studies can identify key biomarkers that, when validated, serve as useful endpoints for clinical trials (as proxies for outcomes that may be more difficult to measure). Biomarkers can be categorized based on their utility for understanding and intervening in disease history, and include susceptibility/risk, diagnostic, monitoring, predictive, prognostic, response, and safety biomarkers. Clear criteria have been defined to determine when a biomarker might be a surrogate for clinical benefit (Institute of Medicine 2010).

4.2.5 Practice: The Kole Nat Hx of Mpox Study and Clinical Trials of Tecovirimat

After the first human case of mpox (then monkeypox) was described in the Democratic Republic of the Congo (Ladnyj et al. 1972), endemic disease was described over the next decades (Breman et al. 1980; Ježek et al. 1987) (Breman et al. 1980; Ježek et al. 1987). Two decades later, an increasing incidence of mpox was observed, thought in part to be related to waning cross-reactive immunity after the cessation of smallpox vaccination (Rimoin et al. 2010). Though smaller case series had been published, more detailed studies of the natural history of mpox were lacking. An important prospective observational natural history study conducted in Kole, DRC from 2007 to 2001 provided detailed clinical, virological, and pathologic characterization of the natural history of disease (Pittman et al. 2023). Findings from this pivotal natural history directly informed deliberations around rational design and selection of primary outcomes in an ongoing randomized, placebo-controlled evaluation of the safety and efficacy of tecovirimat in pediatric and adult patients with clade I mpox in DRC (The PALM007 trial; NCT05559099). Furthermore, after the global spread of clade II mpox was detected in 2022, the design of the PALM 007 DRC trial subsequently served as an important protocol template to inform the design of a number of international randomized clinical trials to evaluate tecovirimat in the treatment of clade IIb mpox (Rojek et al. 2023).Footnote 2

4.2.6 Practice: The LASCOPE Study of Lassa Fever to Inform Clinical Trials

Much of the limited natural history data on Lassa Fever are retrospective and preceded the availability of reliable diagnostics and clinical laboratories. LASCOPE (NCT03655561) was a prospective observational cohort study to more fully characterize the natural history of Lassa Fever, standardize case management, and set the stage for clinical trials. In addition to characterizing disease and identifying risk factors for poor outcomes, the study helped define a reference mortality rate to inform future clinical trials; furthermore, results suggested that the need for dialysis should also be considered an important outcome in any evaluation of therapeutics. (Duvignaud et al. 2021).

4.3 Informing Preclinical Development of Novel and Repurposed Pathogen-Directed and Disease-Modifying Therapeutics

Especially in outbreaks of understudied infectious diseases, early description of the natural history in humans can rapidly inform the development of in vitro assays and animal models enabling studies of pathogenesis and the evaluation of novel or repurposed countermeasures. This needs to be considered for both pathogen-targeted therapeutics and host-targeted interventions. Early and accurate characterization of severe disease clinical phenotypes may also provide insight into specific sub-phenotypes within the larger clinical disease presentation.

4.3.1 Practice: Characterizing Pathophysiologic Stages in COVID-19

As the first natural history signals began to be reported early in the COVID-19 pandemic, poor patient outcomes were associated with hyperinflammation, including increased C-reactive protein and interleukin-6 (IL6) levels. Further understanding of the pathophysiologic signatures of this inflammation led to the study of broad (e.g., corticosteroids) and targeted (e.g., tocilizumab and baricitinib) immunomodulation to improve outcomes in particular subsets of COVID-19 patients (Horby et al. 2021; RECOVERY Collaborative Group 2021, 2022).

4.3.2 Practice: Characterizing Severe Illness Phenotypes in EVD

The care of severely ill EVD patients in well-resourced settings provided the first higher-resolution characterization of previously undescribed multi-organ dysfunction syndromes and critical illness phenotypes, including acute kidney injury, acute respiratory failure, circulatory shock, and central nervous system dysfunction (Uyeki et al. 2016b). Though based on a small number of observations, these case reports/series provided important insight into pathogenesis and the first proofs-of-principle that extracorporeal organ support could be safely and effectively provided to EVD patients. Deeper exploration of the mechanisms of organ dysfunction is needed to improve supportive care in the outbreak setting. Further study of severe EVD in some of these patients also suggested shared features between EVD and hyperinflammatory macrophage activation syndromes seen in other infectious diseases and autoimmune syndromes (McElroy et al. 2019), leading to further research in non-human primate models (Liu et al. 2023). Whether this shared “sub-phenotype” is truly a shared “endotype” reflecting common pathobiology and thus shared opportunities to modify disease is not yet clear (Reddy et al. 2020); nonetheless, higher-resolution characterization of the natural history of disease may generate hypotheses useful in the search for effective MCMs, including immunomodulatory approaches.

5 Determinants of the Natural History

As illustrated above (◘ Fig. 1, 1a), the true natural history of any infectious disease, absent intervention, is ultimately determined by intrinsic characteristics of the host, the virus, and the exposure that are antecedent to the development of infection and disease. Though we focus much of the prior discussion on the natural history of clinical disease (and its emergent features), these host, viral, and exposure factors merit consideration in understanding how the natural history is determined.

5.1 Host Determinants

5.1.1 General Principles

Several host factors have been identified or hypothesized to contribute to susceptibility or resistance to infection and disease and to short- and long-term outcomes. Host risk factors (◘ Fig. 1a) include, but are not limited to age; sex; immune status, including prior exposures (i.e., vaccinations, previous infections) and immunodeficiency states; nutritional status; pregnancy; co-morbidities; co-infections; host genetic factors; and host microbiome. Individual host factors may impact the risk and progression of infection and disease with similar effect, or they may influence each stage differently; therefore, a determinant of the risk for infection may or may not be a determinant of disease severity or outcomes. For many of the diseases under consideration, data on host contributory factors have been limited to what is easiest to capture (e.g., host demographics like age and sex, pregnancy, and basic nutritional status). Determining risk related to less easily measured host factors has been more challenging in outbreaks, especially in low-resource areas where more complex clinical or research laboratory capacity may be unavailable. Missing capabilities may include diagnostics for co-infections or comorbidities, the measurement of non-routine biomarkers, DNA characterization, and the absence of large data sets.

5.1.2 Practice: COVID-19

Recent experience with COVID-19 suggests that focused, well-resourced research on large datasets can expeditiously define the complex host determinants of a novel disease. Early in the COVID-19 pandemic, it became evident that age, sex, and certain comorbidities were associated with severe disease and death (Russell et al. 2023). Large, collaborative, multi-center international studies identified inborn errors in Type I interferon (IFN) responses (Zhang et al. 2020) or autoantibodies (Bastard et al. 2020; Reynolds et al. 2006b) against components of the Type I IFN immune response as major genetic or immunologic determinants of COVID-19 severity. As the pandemic progressed, these studies have continued to identify new genetic factors critical to understanding risk, elucidating pathophysiology, and informing immunomodulatory strategies for antiviral therapy (Covid-19 Host Genetics Initiative 2023; Kalil et al. 2021; Pairo-Castineira et al. 2023; Recovery Collaborative Group 2022; Reynolds et al. 2006a). More recently, genetic associations between human leukocyte antigen type and asymptomatic SARS-CoV-2 infection have emerged (Augusto et al. 2023). Finally, genome-wide association studies are beginning to describe host genetic variants in large cohorts of patients with PASC (Vilma et al. 2023). For almost all the other pathogens and diseases under discussion in this chapter, we still have a limited understanding of how host genetic factors influence the risk of infection and disease severity. In the future, similar efforts to define host genetic risks could inform clinical care and improve patient outcomes.

5.2 Viral Determinants of the Natural History

5.2.1 Viral Toolkits, Infection, and Disease

Each virus (as a member of a conceptual virus species) has its own molecular tools enabling infection, replication, local or systemic dissemination, and possible transmission to another host. Viral toolkits evolve to evade both innate and virus-specific adaptive immune responses in an ongoing host–pathogen evolutionary arms race (Crespo-Bellido and Duffy 2023; Ploquin et al. 2018). En route to replication and forward transmission, viral infection may cause dysfunction of or damage to cells, tissues, and organs or organ systems by direct (cytopathic) or indirect (noncytopathic) mechanisms and may provoke immunopathologic responses; clinical symptoms and signs of disease result. Detailed discussion of virus-specific molecular mechanisms of infection and disease is outside the scope of this chapter.

5.2.2 Within-Outbreak Viral Evolution and the Natural History

Sustained human-to-human transmission and replication enables viral evolutionary exploration of the human host. Theoretically, viral evolutionary trajectories should be oriented toward enhanced replication, transmission, and avoidance of host immune responses. The effects of ongoing evolution on pathogenicity are less clear, though they are generally thought to trend toward more transmissible and immune-evasive variants rather than variants that cause more severe disease. Sustained SARS-CoV-2 transmission over time has provided a clear example: early SARS-CoV variants (e.g., Alpha, Delta) appeared to cause more severe disease in humans and animal models than later variants (e.g., Omicron) (Nyberg et al. 2022); whether this trend will continue is unclear. The capacity for rapid genetic sequencing to provide near real-time molecular epidemiologic information during an outbreak is a recent phenomenon, having been deployed for the first time at scale during the 2014–2016 EVD outbreak in West Africa. During that relatively long period of human-to-human transmission, certain mutations became dominant in circulating EBOV, suggesting some fitness benefit to the virus. Efforts to determine any impact on the natural history, however, even in animal models, have not been revealing (Marzi et al. 2018). We do not have such detailed sequence data for most viral pathogens that infect and transmit among humans.

5.2.3 Intra-Species Viral Strains and the Natural History

In general, differentiation of virulence among viral variants (within a viral species) is limited to epidemiologically derived CFRs associated with different variants. Determining whether increased CFRs are caused by a more virulent virus strain (or isolate) typically requires in vitro and animal modeling studies that do not provide a complete picture of human infection and disease even when they can be conducted. For some viruses, animal models have not identified significant differences in virulence between specific outbreak strains; for example, EBOV-Mayinga, EBOV-Kikwit, and EBOV-Makona variants associated with different outbreaks do not appear to differ in severity in animal models (Yamaoka and Ebihara 2021). Subspecies differences in disease severity have been identified for other viral infections. Observations of higher CFR with mpox virus clade I (CFR 5–11%) versus clade II (CFR 1–3%), for example, suggest viral genomic variance contributes to an altered natural history of mpox. However, these observations may be confounded by other variables, including geographic origin, host genetic variation, and the availability and quality of care (Gessain et al. 2022). Comparisons in nonhuman primate models suggest differential lethality, viral loads, and organ-tropism between mpox virus clade I and clade II (Saijo et al. 2009). Similar observations have been made in Lassa fever caused by different clades of the Lassa virus in humans and in animal models (Grant et al. 2023; Stein et al. 2021).

5.3 Exposure Determinants of the Natural History

5.3.1 Dose, Route, Sample Matrix, Environment

Exposure to a larger quantity of infectious material is generally presumed to increase the likelihood of infection. A similar correlation with disease severity is likely, though simple exposure dose–response relationships (to either infection or disease severity) are challenging to demonstrate in either animal-to-human or human-to-human transmission and must rely instead on proxy animal modeling. In general, dose-ranging of viral exposure in animal models suggests that increased exposure increases either the disease severity or its progression. Determining the impact of the route of exposure generally must also rely on animal modeling. It is likely that infectivity and transmission also depends on the exposure matrix (e.g., blood versus bodily fluid versus semen) but this is not well understood beyond epidemiologic observations. It is also likely that the exposure environment affects infectivity and secondary attack rates given the differential stability of viruses in particular environments (e.g., relative humidity and air flow for viruses transmitted by aerosol particles or respiratory droplets) (Thornton et al. 2022).

5.3.2 Practice: Monkeypox Virus (MPXV) Exposure and the Natural History of Mpox

It has been long recognized that the route or site of initial exposure to orthopoxviruses directly impacts the clinical presentation. The cowpox pustules on the hands of infected milkmaids observed by Edward Jenner are perhaps the best-known example (Cowpox and paravaccinia 1967; Jenner 1798). In the context of endemic mpox (caused by Clade I MPXV infection) in the DRC, the route of exposure has been previously associated with more severe disease, though this may be confounded by a dose effect (Pittman et al. 2023; Reynolds et al. 2006b). During the global spread of mpox through human-to-human transmission in 2022–2023 (caused by Clade IIb MPXV infection), atypical clinical presentations were frequently described after high-risk sexual exposures, including localized skin lesions and regional lymphadenopathy, anogenital lesions, and proctitis. Clinical comparison to the disseminated skin lesions more typical of endemic mpox, and more often associated with animal-to-human or human-to-human transmission without high-risk sexual contact (clade I or II MPXV), suggests an important contribution of the exposure route to the natural history (Mitjà et al. 2023). Recently reported descriptions of mpox cases after likely heterosexual transmission of the virus (clade II MPXV) in Nigeria suggest a similar route dependence (Ogoina and James 2023).

5.3.3 Practice: Ebola Virus (EBOV) Exposure and the Natural History of EVD

Human-to-human transmission of EBOV is thought to occur predominantly from mucosal exposure to blood or body fluids from symptomatic or deceased patients (Vetter et al. 2016). Though an exposure dose response has not been confirmed in human studies, animal models suggest dose-dependent differences in the severity, character, or pace of illness after mucosal (oral or conjunctival) versus intramuscular inoculation (Cross et al. 2023; Johnson et al. 2023). However, dose-ranging studies have also demonstrated 100% lethality even after very small inocula after intramuscular or aerosol exposures. Limited observations after accidental laboratory exposures suggest that intravenous or intramuscular exposure (via a needlestick injury, for example) poses a high risk of infection and severe disease in humans (Vetter et al. 2016). Uncertainty remains around the effect of bodily fluid matrix on EBOV transmission, though its influence is very plausible. In comparison to human-to-human transmission of acute EVD, the paucity of cases associated with exposure to the semen of male EVD survivors (with high viral loads in semen, especially early in convalescence) suggests a matrix effect that lowers sexual transmission risk among other potential bottlenecks (Jacob et al. 2020).

6 Natural History Studies: Optimal Design and What Can Be Achieved in Outbreaks

6.1 Outbreak Realities: Challenges to Natural History Studies

Historically, the realities of outbreak settings have limited opportunities for well-designed natural history studies. Outbreaks occur unpredictably, often in remote, under-resourced settings. The immediate urgencies of outbreak response are not well suited to pre-planned, prospective studies, so understanding and reporting the natural history often proceeds as shown in ◘ Fig. 5. It is perhaps understandable that true natural history studies have typically only been established during prolonged outbreaks and are often focused on survivors rather than acute disease.

Fig. 5
figure 5

Understanding the natural history in infectious disease outbreaks: typical progression. (Author)

Full size image

6.1.1 Rapid Communication via Existing Clinical Networks

As described above, the first signals of the natural history typically come from discussions of the clinical disease from providers caring for patients in the outbreak setting. In recent outbreaks (e.g., COVID-19, mpox, and filovirus disease outbreaks), connecting external subject matter experts to in-field clinicians has been invaluable to discuss emerging clinical challenges, begin to inform or update clinical guidelines, identify research questions, and rapidly inform preclinical and clinical research strategies.

6.1.2 Early Observational Studies of Natural History

During an outbreak, observational data is often first published as descriptive case reports, series, or small cohort studies. In these cases, data is derived retrospectively from already available clinical charts and medical records. Typically, these are also cross-sectional (collected during a specified limited time period). While this enables rapid design, analysis, and publication, these studies are susceptible to biases and limitations, including:

  • Missing or inconsistent data or methods of data collection from existing records not designed to reliably capture all relevant data

  • A bias toward a particular patient population may limit generalizability of findings. This is particularly true of case and series reports early in outbreaks, in which unusual or particularly severe clinical presentations may be more likely to be published

  • Limited cross-sectional observation periods may not capture evolution of disease over time

Nonetheless, in most novel outbreaks, early descriptions of clinical disease play an important role in informing care and in shaping clinical research. For example, the first reports of clinical disease from China early in the COVID-19 pandemic critically set the stage to describe phases of disease, begin to identify risk factors for poor outcomes, and suggest strategic approaches for interventional redirection of the natural history (Guan et al. 2020).

6.1.3 Optimal Design of Natural History Studies: Prospective, Longitudinal, and Controlled Observations

Prospective studies evaluate events based on a prespecified study design. Longitudinal studies collect data from all patients in a pre-defined cohort over a defined period and are more likely to yield information about the onset and evolution of disease. Prospective, longitudinal observational cohorts with well-matched controls (where relevant) provide the most comprehensive, reliable, and generalizable information about the natural history of an emerging infectious disease. Optimally, over the outbreak period, these studies would provided an increasingly resolved picture of the natural history. However, the time, resources, and effort required to implement such studies during an emergency have limited their use in infectious disease outbreaks. Conducting effective natural history studies in infectious disease outbreaks likely requires alternative strategies.

6.2 Strategies to Improve Understanding of Natural History During Outbreaks

6.2.1 Early Identification of Research Gaps

The identification of key research gaps for any disease often requires systematic reviews—efforts that require considerable expertise, effort, and time, limiting their utility in emergency research response. For recognized priority pathogens with pandemic potential, formulation of a research agenda should emphasize understanding natural history to inform development of diagnostics and MCMs. A “rapid research needs appraisal” (RRNA) protocol aiming “to identify important gaps in evidence and knowledge in a robust, systematic, and replicable manner to rapidly inform clinical research prioritization” within five days is under development. It was recently piloted using a Lassa Fever outbreak scenario, with online rapid-review software, and evaluated in comparison with an expert Lassa fever panel (Sigfrid et al. 2019). This and analogous efforts to accelerate identification of key research gaps are needed.

6.2.2 Standardized Clinical Characterization Protocols (CCP) for Specific Diseases or Disease Syndromes

Another obstacle to urgently and widely collecting data for understanding natural history is the lack of standardized data collection protocols to enable harmonized collection from multiple sites and independent outbreaks of the same disease. “Prototype pathogen” approaches to help understand viral families should be paralleled by the development of “prototype protocols” that could be pre-positioned to understand the disease caused by those pathogens. Developing disease-specific or syndrome-specific clinical characterization protocols agreed upon by stakeholders in advance would help resolve this issue. Depending on the syndrome, these CCPs are easily adapted to novel diseases presenting with similar syndromes. For example, generic International Severe Acute Respiratory and Emerging Infection (ISARIC) CCPs have proven useful in understanding the natural history of COVID-19 from large international datasets, including in disease characterization (Drake et al. 2021; Millar et al. 2022; Sullivan et al. 2021; Swann et al. 2020), risk assessment (Knight et al. 2022), and evaluation of clinical care. In addition, proposed “perpetual observational studies” could become standard for characterization of the natural history of disease across outbreaks (Hassoun-Kheir et al. 2022).

6.2.3 Prepositioned Clinical Research Networks (and Researchers)

Establishing and strengthening clinical research networks (and researchers) across geographic areas susceptible to the same infectious diseases is a valuable strategy to improve understanding of novel diseases or endemic infectious diseases with significant knowledge gaps. These networks might be initiated in response to specific disease threats but maintained during inter-outbreak periods and able to pivot to new threats. For example, at the national level, the PREVAIL, PREGUI, and PALM collaborative efforts (in Liberia, Guinea, and the DRC, respectively) had origins in research responses to EVD outbreaks but have pivoted to other emergence research needs. Regional examples include The African Coalition for Epidemic Research, Response and Training (ALERRT), and the Pan-African Network for Rapid Research, Response, Relief and Preparedness for Infectious Disease Epidemics (Pandora-ID-Net) as well as focused training networks like the Clinical Research During Outbreaks (CREDO) initiative (ALERRT 2023; Kayem et al. 2019; PANDORA 2022).

6.2.4 Global Clinical Platforms

To enhance the understanding of the natural history in emerging infectious diseases, the WHO has proposed the use of “Global Clinical Platforms” (GCP) in which secure web-based databases could capture individual-level anonymized clinical data from patients around the world, providing large datasets for analysis. Platforms are to be disease or syndrome-adapted and focused on understanding natural history. For example, the aims of the WHO GCP for Viral Hemorrhagic Fever (WHO 2023) are to:

  • Describe the disease course, its natural history, and severity.

  • Identify the association of clinical characteristics of viral hemorrhagic fevers with outcomes.

  • Inform strategies for use of clinical resources to provide high-quality supportive care.

6.2.5 Core Protocols: Integrating Natural History Studies into Clinical Trials of Medical Countermeasures

The challenges of conducting interventional clinical trials in unpredictable outbreak settings have led to calls to implement core (or master) protocols that could make it possible to continue a study across successive independent outbreaks (Dean et al. 2020). In addition to evaluating the safety and efficacy of a candidate therapeutic, the standardized clinical characterization collected during interventional studies could provide important insights into the natural history of disease. For example, platform adaptive randomized core trial protocols under discussion for the evaluation of new and repurposed treatments for filovirus diseases likely could also valuably improve our understanding of natural history.

7 Conclusion

Understanding and reporting the natural history of infectious diseases is an important but challenging component of emergency research response. A clear understanding of natural history not only informs clinical care but also sets the stage for better design of preclinical and clinical research to evaluate medical countermeasures. Recent experience in the COVID-19 pandemic suggests that unprecedented research goals can now be accomplished, albeit with exceptional collaborative effort and resources. Characterization of the evolving host–pathogen–care interaction has been much more limited for diseases caused by most priority pathogens. Any disease caused by a novel pathogen will require new characterization efforts to begin as soon as possible. As we plan for better preparedness for and response to future infectious disease emergencies, we must devise and incorporate strategies to rapidly and effectively characterize, understand, and redirect disease natural history.

Discussion Questions

  1. 1.

    Why is it important to develop an early, detailed understanding of the natural history of a novel disease in patients when there is a novel infectious disease outbreak?

  2. 2.

    What are the intrinsic determinants of the natural history of a disease in a human patient?

  3. 3.

    Name some obstacles to conducting natural history studies during outbreaks.

  4. 4.

    Suggest some ways of better preparing to conduct future natural history studies of novel infectious diseases.