From the frontlines in West Africa, there are a variety of lessons to be learned when evaluating the sociological, political, and anthropological factors that contributed to the evolution of the 2013–2016 Ebola virus disease (EVD) epidemic. Additionally, the outbreak provides its share of lessons for the management of EVD in individual patients, as well as the balance between public health needs and clinical medicine. In this chapter, I discuss the challenges facing health care workers (HCWs) during this 2013–2016 epidemic and what can be learned in order to strengthen future global public health responses.
While most previous EVD outbreaks occurred in rural areas, the geography of the 2013–2016 epidemic created multiple challenges. The virus initially presented itself in a forested region, similar to past outbreaks, likely due to contact with a bat vector or consumption of infected bush meat. This forested region encompasses the porous borders of three previously Ebola-naïve West African countries with mobile populations that share a similar ethnic group and language and have accessible roads between their major urban centers.
EVD quickly demonstrated its disrespect for political boundaries within the region, crossing from Guinea into Sierra Leone, Liberia, Nigeria, and Mali within months. The ease of human migration within the region; a combination of both rural areas with limited health care and densely populated capital cities; cross-border politics; and difficulty with coordination made it challenging to initially investigate case leads and maintain effective contact tracing efforts. Additionally, community resistance to containment efforts, fueled by misunderstandings about EVD, negatively impacted case identification and contact tracing efforts.
Traditional burial practices and funeral rights in the region often involve washing and touching the body of the deceased. Safe burials in the context of EVD require extensive protective equipment to prevent disease transmission to family and those caring for the body as bodily fluids of EVD patients are believed to be the most virulent when their bodies are overcome by EVD at the time of death. In the context of the 2013–2016 outbreak, contact with these very infectious bodily fluids often led to numerous new EVD cases and several deaths arising from one traditional burial.
When cases were eventually suspected and identified, the overburdened health care system was limited in its ability to manage cases and control the epidemic. The lack of material resources, such as appropriate personal protective equipment and diagnostic equipment, presented a significant barrier to management. This was compounded by relatively low numbers of HCWs, especially those with training in relevant infection prevention and control principles for EVD. The death of many HCWs early in the course of the epidemic, along with issues with the availability of relevant equipment and appropriate compensation for local HCWs that considered the risks of the work they were assuming, also presented a significant barrier to care.1
The World Health Organization (WHO) estimates that in the three most impacted countries—Guinea, Sierra Leone, and Liberia—there are only one to two physicians available to treat 100,000 people.2 In addition, most of these physicians are concentrated in urban areas, leaving large parts of the country without access to physician services. These countries also have among the lowest numbers of trained nurses and midwives in the world relative to population density.3 This translates into an unsustainable workload for local health care providers that affects productivity, quality, and job satisfaction, and likely contributes to challenges of emigration or “brain drain” of the trained health care workforce.
The impacted countries had been making improvements in the health sector prior to this EVD outbreak. For example, Liberia had nearly halved its maternal mortality ratio (per 100,000 live births), from 1,200 in 2000 to 640 in 2013.4 Attempts to better train and retain local talent were underway, and Liberia had increased intake and scope of its postgraduate training programs for physicians just months prior to Ebola’s emergence.5 Despite forward progress, the regional health care infrastructure still proved underresourced to manage the emergence of this new virus.
Despite being involved with management of the initial outbreak, miscalculations were made by WHO and US Centers for Disease Control (CDC) experts on the trajectory and complexity of this outbreak in the spring of 2014. Meanwhile, EVD was crossing borders in the Kissi language area and traveling through the region largely unrecognized for some time.6
Early symptoms of Ebola appear similar to regionally endemic illnesses such as malaria, typhoid, and Lassa fever. Despite decades of prior outbreaks elsewhere on the continent, West Africa had never before documented a case of EVD. Dr. Margaret Chan, director general of the WHO, noted “old diseases in new context will deliver surprises,” and these post-conflict countries lacked the trust needed to initially mobilize communities to be part of the solution. Some impacted populations had baseline suspicion of the state, considering prolonged periods during which their basic needs were not provided for after having recently emerged from decades of political coups in Guinea, prolonged civil wars in Sierra Leone and Liberia, and transnational conflicts. Moreover, impacted governments in this fragile post-conflict reconstruction phase had not previously dealt with the unique social and economic disruptions of EVD.
By August 2014, 12 percent of deaths from Ebola in Liberia were among HCWs alone.7 Historically, HCWs have accounted for up to 25 percent of those infected in previous outbreaks.8 HCW exposure to EVD is more often in the general hospital or clinic setting and less often in dedicated Ebola treatment centers. This may be the result of unrecognized or mis-triaged cases, hospital overcrowding, and lack of training in or availability of appropriate personal protective equipment and infection control practices. Rising infections and lack of appropriate resources among HCWs in the hospital context degraded confidence in the already limited health care infrastructure. This led to the functional closure of much of the existing health care system in impacted areas while Ebola cases were arising. When Ebola was appropriately recognized, there was ongoing difficulty with the ability to refer and transport patients to treatment units or centers with the proper training and protective equipment to provide Ebola care.
As EVD spread quickly in mobile and urban environments, the ability to isolate patients in Ebola treatment units (ETUs) was quickly overwhelmed. Médecins Sans Frontières historically managed much of the clinical care in the context of prior EVD outbreaks in conjunction with the WHO, CDC, and local ministries of health. However, the pace, geographic spread, and magnitude of this epidemic quickly called for the involvement of other humanitarian and medical relief groups to assist in case management.
There are several strains within the Ebolavirus genus (for more information, see chapter 1). Deaths from prior outbreaks of the Zaire ebolavirus viral species that appeared in Western Africa were as high as 90 percent.9 This meant that exposure and mortality risk to the extremely limited numbers of HCWs in the region was of paramount concern early in the epidemic. Considering the infectivity and severity of the virus, significant tension existed between the ability to effectively isolate large numbers of patients to prevent community spread of EVD and the ability to provide high-quality patient care to improve chances of individual survival. The public health priority of breaking chains of transmission through isolation of presumed or confirmed cases and safer burials at times took precedence over comprehensive individual case management. For example, when there were overwhelming ratios of patients to trained health care providers, some treatment centers only provided oral medications and did not assume the risks of intravenous therapy (i.e., exposure of HCWs to the blood of patients or to potential needlesticks).
This lack of adequate human and material resources that would have allowed for focus on improvement of individual patient outcomes from EVD arguably undermined the public health response. Some perceived ETUs not as a place to seek treatment, but as a place people went to die, resulting in reluctance to acknowledge symptoms or seek care for EVD. This resistance potentially led to more deaths and unsafe burials in communities while propagating transmission, especially since traditional funerals often involved unsafe burials (by Ebola management standards) and transmission from dead bodies led to many new Ebola cases.10
For the first time in history, an emergency meeting of the United Nations Security Council was called because of the overwhelming nature of the public health crisis. HCWs on the ground in West Africa at the time were relatively few, and those who were present were both overstretched and underresourced in attempting to manage the skyrocketing caseload and infection prevention measures. WHO’s Director General deemed this a public health emergency of international concern (PHEIC) on August 8, 2014, at which point there were already 1,779 cases and 961 deaths.
This was only the third-ever declaration of a PHEIC, the two others being the 2009 H1N1 flu pandemic and the 2014 polio resurgence after its near-eradication. The UN Mission for Ebola Emergency Response was subsequently formed as a coordinating body on September 18, 2014. This was the first-ever UN mission dedicated to public health emergency response. The United States and other members of the international community committed resources, including funding, equipment, and military deployments. The response finally began to scale up more appropriately.
As Ebola patients were repatriated to Europe and the United States for further management, the concern in the international community started to rise in a commensurate fashion. When cases developed in the United States among exposed HCWs, concern about possible domestic outbreaks and modes of transmission escalated. Evolving early guidelines and inconsistent messaging from government officials and the media in August and September 2014 led to skepticism among the public regarding advice from government officials and misgivings about returning HCWs. This distrust resulted in fear superseding science in some cases, such as quarantine restrictions on returning HCWs enacted in New York and New Jersey.11
Concern about a potential Ebola outbreak in the United States served to hamper the response in West Africa, leading to restrictions on deploying HCWs, either by employers unwilling to release staff for participation as humanitarian responders or due to concern about evolving government restrictions on returning HCWs. One international humanitarian organization lost approximately 25 percent of their potential Ebola responders when quarantine policies were established.12 International awareness certainly led to an increase in the material and financial resources mobilized, but the sheer lack of HCWs and delays in resource distribution meant that clinical patient management, even for very basic and commonly treatable medical problems, was limited in most clinical settings. For HCWs who were working to rebuild local systems prior to Ebola’s emergence and those responding with an awareness of health care delivery in a high-resource setting, including myself as a clinician, this extreme lack of access to care was distressing to witness.
In late September 2014, the Centers for Disease Control released their first Ebola response prediction models. This epidemiologic modeling tool was based on Ebola case data and doubling times for the virus in Liberia and Sierra Leone through late August. The model projected that if the EVD epidemic were to continue without intervention, as of January 20, 2015, cases would be estimated to reach 550,000 (or 1.4 million cases if corrections were made for underreporting). The same prediction models indicated that if intervention were scaled up such that 70 percent of patients were in treatment or isolation, and if there were changes in community behavior, particularly decreases in unsafe burial practices, as of January 20, 2015, there would be 14,000–35,000 cases.13
While slow to launch, the support provided to the local health care systems and governments by the international response did allow for further case management, tracing of suspected or confirmed Ebola cases, and social mobilization.
As more treatment centers opened, more staff were trained, and more resources were provided in terms of protective equipment, there was more of an ability to shift from a focus on isolation and minimal treatment to increased intervention and improvement in patient mortality. Reflecting the CDC prediction models, scaled-up intervention helped reduce potential cases by an order of magnitude. Based on the WHO situation reports, cases were at approximately 22,000 as of January 20, 2015. As of September 2, 2015, cases were still within the CDC-predicted range given intervention at approximately 28,073.14
Ideally, we would have better systems in place to immediately mobilize the resources necessary so that future responses will not experience tension between isolation for the public good and the highest possible standard of clinical care for the individual patient. For the remainder of this chapter, I’ll talk about the deficits in knowledge and care, and what we can do to ensure that we do not lose control of future infectious disease outbreaks through inaction, lack of preparedness, and limited clinical knowledge.
Prior to this epidemic, EVD had never been managed in the context of a high-resource health care setting. The opportunity to manage EVD in this new setting, combined with the sheer magnitude of the outbreak in Western Africa, led to a marked increase in knowledge about EVD pathophysiology, clinical management, and the sociologic impact of EVD on affected communities.
Initially, baseline strategies successful at controlling prior Ebola outbreaks were employed. These focused on finding suspected cases as soon as possible and getting them into an ETU. Additional strategies included testing with real-time virus diagnostics in field or regional laboratories, following patient contacts for 21 days in order to identify symptom onset as early as possible, and communication and risk-reduction messaging to impacted communities.
The initial challenges posed by the current epidemic and interactions with communities impacted by EVD unveiled the importance of social mobilization and taking community perceptions into account, especially when addressing EVD in a highly mobile population crossing several national borders in a post-conflict setting.15 Based on my observations in Liberia, attempting to stop rumors and confusion with simple and clearly communicated health promotions messages about Ebola through radio, television, pamphlets, roadside billboards, community and country leaders, health workers, and survivors appears to have eventually been an important factor in prevention and care-seeking behaviors for those with EVD symptoms. A better understanding of community perceptions, reasons for resistance, funeral and burial practices, eating and living styles, and learning best communication channels eventually allowed for more effective messaging and outreach.
Experiences with the recent outbreak should inform best practices in responding to future EVD outbreaks. Regardless of available resources, attention to providing the best available compassionate care to patients, with a focus on minimizing fear and helping maintain hope and dignity, remains paramount in the ETU setting. Anxiety around Ebola is heightened as a result of the personal protective equipment necessary for safe EVD care. When possible, clinicians making themselves known to patients face-to-face across transparent ETU fencing prior to initiating care in protective equipment can help improve care and communication. Surviving patients serve to mobilize others in the community to seek care if they feel they were well cared for in the ETU setting. And for HCWs with limited resuscitation resources, helping afford a dignified death for patients who would otherwise be alone and incapable of caring for themselves appears to be valuable for both patient and provider.
Children represent a particularly vulnerable population and one that is difficult to best care for in the ETU setting. Unaccompanied minors, or those with discordant EVD testing results from their parents or family, present challenges not only in their medical care but also with their safety, supervision, feeding, and so on. EVD mortality for the under five-year-old population is among the highest, and their acute care is particularly difficult for health care workers who can only spend limited time with them due to the constraints of personal protective equipment. This takes both a physical and emotional toll on HCWs. When safe and feasible, cohorting children with relatives or friends, or engaging EVD survivors to assist in their care, may be strategies worth employing.
If children survive to discharge, another set of challenges arise, as they may often be orphaned or stigmatized when attempting to return to their communities. ETUs and partner organizations, with local government as indicated, are charged with coordinating the safe return of children to their families or placement with relatives or protective custody as needed at the time of discharge. Managing the impact of this EVD epidemic on children will need to be a continuing effort considering the long-term regional disruption to schooling, childhood, and family structures in impacted countries.
The ability to engage families of ETU patients, and for patients to interact with visitors, is important not only for the pediatric population, but for all patients receiving ETU care. Transparent fencing, providing a safe area for visitation (at a distance), and allowing families to be present for burial services when possible can aid in this process. Providing religious or psychosocial support to both patients and families is of particular importance. Counselors, pastors, nurses, community leaders, and others should receive focused EVD psychosocial training in order to best support patients, families, and staff around the challenges of seeking Ebola care, providing care in the ETU setting, and discharge back to the community (whether they be negative for EVD or have attained survivor status). All of the above strategies allow for better communication and more dignified care, and appear to improve treatment-seeking behaviors.
Attempts to improve the standard of care for EVD are ongoing. As the epidemic unfolded in real time, frontline HCWs in ETUs on remote Liberian hilltops were searching data from cases managed in high-resource settings. They consulted with organizational or WHO focal points on specific patient-management topics in order to better treat the next ailing patient in their low-resource treatment center with up-to-date information. Coordination with UN Mission for Ebola Emergency Response and WHO experts allowed for knowledge pooling and resources to consult when particular management challenges arose.
EVD management strategies were largely based on expert opinion at the start of this epidemic. There is still much to be learned from the data collected during patient care at the ETUs across West Africa. An important focus moving forward will be thorough analysis of these findings so as to have evidence-based guidelines in place for patient management in the future. Two areas of focus where we are gaining further understanding about pathophysiology and treatment strategies are among pregnant patients and among those with possible bacterial infections secondary to Ebola.
The perspective on pregnant EVD patients evolved during the course of this epidemic, and, again, should inform future clinical practice. Historical data put maternal death rates upward of 90 percent and, at the time of writing, fetal death in the neonatal period is uniform.16 As a result of this heightened mortality among pregnant women and nonviable fetuses, nonintervention during pregnancy and delivery had been the standard of care. Like most Ebola management strategies until this epidemic, guidelines were based on very limited numbers, small case series, and expert opinions.
Pregnant patients are at increased baseline risk of contracting Ebola because of their contact with health care services for both prenatal and delivery care. Transmission in maternity settings has been an issue in previous major outbreaks. Consequently, in the context of an EVD outbreak, it was challenging for pregnant women to gain access to lifesaving maternal and newborn interventions provided at health facilities. This lack of care caused adverse maternal and neonatal outcomes such as infection or death, regardless of the underlying issue. Increased community outreach and distribution programs for clean delivery kits were vital considering the scale of the epidemic and its secondary impact on the obstetric population.
EVD often causes spontaneous abortion (miscarriage) with heavy bleeding during the first or second trimester. However, unrelated to EVD, spontaneous abortions are the most common complication of early pregnancy and happen to 15 percent of otherwise healthy women in early pregnancy.17 Therefore, in the setting of an EVD outbreak, common presentations or complications of pregnancy may become suspect EVD cases. This leads to the challenges of care in the more limited context of an ETU, without surgical or procedural capacity, and most often without a blood bank for transfusions as needed for hemorrhage. Alternatively, women may not even seek care in the hospital setting due to fear of getting infected while there.
The overall number of pregnant women impacted in the course of this epidemic remains unknown. It was previously unclear whether EVD transmission from mother to baby was happening during pregnancy, or via exposure in the birth canal, breastfeeding, or simply close contact after birth. Several cases have now confirmed transmission of EVD across the placenta; therefore, even when maternal virus is cleared from the blood, ongoing fetal EVD infection may still imply unique risks at delivery. EVD case reports have recently been published on maternal assistance during delivery contrary to prior recommendations, including amniocentesis, episiotomy, and other procedures done in the maternal healing phase.18 Good outcomes for these patients have opened the door to more focus on interventions to improve mortality among pregnant EVD patients over the course of the epidemic.
Médecins Sans Frontières opened a dedicated 33-bed maternity unit at one of their Ebola treatment centers in late January 2015 in Sierra Leone in order to provide more specialized care to this impacted population. Since relatively little is known about Ebola and pregnancy, looking at cases and dedicated obstetric management in this unit, along with pooled data from other treatment centers, will help the humanitarian and research communities glean a better understanding of how to treat these vulnerable women and neonates moving forward.
Secondary infections are significant risks for patients with EVD; anti-malarial treatment was given empirically in many ETUs from the beginning of the outbreak due to the endemic nature of malaria and overlapping presenting symptoms. EVD is caused by a virus and is thus not responsive to antibiotics that are used to treat bacterial infections. Therefore, antibiotic administration was variable in the ETU setting and, if given, often varied in type, route of administration, timing of initiation, and duration.
There is, however, increasing recognition that secondary bacterial infections may play a larger role than previously assumed. Data from patients treated in higher-resource settings with careful monitoring of their electrolytes (which impact cardiac, neurologic and multi-organ system function) have given insight and informed protocol development for patients in the lower-resource ETU setting. These data have also confirmed several cases of a particular type of bloodstream infection from gram-negative bacteria, confirmed by blood cultures of patients treated in a high-resource setting, which indicate the need for aggressive intravenous antibiotic therapy.19 20 21 This is thought to originate from severe gastrointestinal tract manifestations in which bacteria that normally live inside the bowel translocate outside it and cause systemic infections.22 As a result, many more ETUs provided empiric intravenous antibiotic therapy aimed at bowel pathogens later in the epidemic than initially.
Changes in medical management such as treatment with intravenous antibiotics for possible secondary bacterial infections, electrolyte repletion protocols, and more aggressive critical care measures are the result of the science and understanding of Ebola’s pathophysiology evolving in real time. This, combined with more human and material resources available to treat patients later in the course of the epidemic, likely contributed to an apparent reduced fatality rate over the course of the epidemic, observed from the ETU. While supportive care currently remains the keystone of EVD treatment, there are several experimental vaccines, treatments, and point-of-care tests being studied.
While the 2013–2016 EVD epidemic has ended, continued vigilance and clear community messaging is necessary in order to facilitate early recognition of potential EVD case flare-ups that may continue. Just as important, the international community needs to ensure the rapid evaluation and dissemination of data collected and knowledge gained about EVD transmission patterns, pathophysiology, and clinical management over the course of this epidemic. If an effective vaccine and curative treatments are still lacking, this information will be incredibly useful not only to end this epidemic, but for whenever and wherever the next outbreak occurs, which should be anticipated in the context of animal reservoirs of EVD.
An ongoing commitment to health systems strengthening is also needed for the region to recover from the cascading impact of Ebola. The challenge of access to basic health care, which was marginal but improving prior to the ongoing Ebola outbreak, has been exacerbated by further losses among the HCW population due to EVD. The limited HCW population further aggravates the baseline challenge of stressful conditions of service, and without commensurate recognition and compensation, may lead to emigration for some of the most talented from the pool of HCWs.23 Therefore, increasing numbers, capacity, diagnostics, treatments, and other material resources needed for the health care workforce to accomplish their jobs is an area that will require significant ongoing investment by the international community, in partnership with local governments and nongovernmental organizations. In addition, assisting severely impacted communities and orphaned children while addressing survivors suffering from stigmatization and post-EVD medical problems—as well as loss of employment, housing, possessions, and family—are all matters of ongoing concern.
The insufficient baseline infrastructure to contain and manage this EVD epidemic in West Africa highlights the role of infectious diseases and epidemics from a national and international security perspective. Ebola has demonstrated that in our interconnected global community, outbreaks can now move faster than the methods historically used to control them. This demands new approaches to the identification, communication, control, and treatment of infectious outbreaks such as EVD in resource-limited settings and offers a new perspective on the importance of health systems strengthening for the global health security agenda.