Managing the Global Health Response to Epidemics

Social science perspectives

Mathilde Bourrier, Nathalie Brender, Claudine Burton-Jeangros

• Publisher: Routledge
• Year: 2019
• Print ISBN: 9781032570181
• eBook ISBN: 9781351263023
• Edition: 1
• Topic: Medicine
• Subtopic: Health Care Delivery

Recent epidemics have prompted large-scale international interventions, aimed at mitigating the spread of disease in a globalized world. During a crisis, however, global health actions – including planning and organizing, communicating about risk, and cost–benefit evaluations – aren't usually part of a single, integrated global response. Arguing that an uncoordinated approach can be challenged by local conditions and expectations, generating a wide range of resistance and difficulties, this volume provides important insights for future outbreak management and global health governance.

Drawing on experiences with A(H1N1) and Ebola virus disease, the book is divided into three parts looking at how responses to global health crises have developed, lessons learned from particular pandemics and the ethical implications of our management of them. Individual chapters focus on, among other issues, financing, cost–benefit analysis, matrix management, risk communication and organizational strategies.

Taking a social science perspective, this valuable book outlines the current state of global health emergency responses and explores ways in which they can be improved. It is a useful read for academics and practitioners interested in global health, the sociology of health and illness, health economics and emergency management.

Part I

Setting the stage

1The challenges of building pandemic response systems based on unique cases

2003 SARS, 2009 A(H1N1) and 2014 Ebola epidemic

Ann Keller

Introduction¹

A growing body of research on crisis response attempts to identify the cognitive and organizational challenges involved in managing a crisis in an effort to prepare organizations for the decisions, processes, coordination and logistical feats necessary to limit its negative effects (e.g., Ansell, Boin, & Keller, 2010; Christensen, Lægreid, & Rykkja, 2016; Comfort 2007; Lagadec, 2009; LaPorte, 2007; Moynihan, 2012; Nohrstedt & Weible, 2010; and ‘t Hart, 2013). Apparent successes in preventing crises – e.g., the low failure rates of high-reliability organizations (LaPorte & Consolini, 1991; LaPorte, 1996) or the use of incident command systems in firefighting (Bigley & Roberts, 2001; Comfort, 2007; Lutz & Lindell, 2008; and Moynihan, 2009) – create both the imperative and the challenge of trying to improve capacity in other domains. Though countries tolerate endemic disease by engaging in incremental, long-term public health campaigns, unpredictable outbreaks of severe diseases prompt resource intensive, concentrated responses. Professionals engage in planning and post hoc evaluation of such events in order to draw lessons that might improve future performance. Two significant challenges emerge from this orientation. First, public health agencies charged with leading outbreak response efforts must transform their organizations from a routine-operations mode into that of emergency response. Second, organization leaders face pressure to develop expertise in responding to short-lived, infrequent events.

The most obvious approach to improve organizational performance, trial and error learning, is not available for health crises that are infrequent and distinct. Because of the lack of opportunity to make use of trial and error learning, organizations with the responsibility to respond to crises must find other means of developing what might stand in for experience. One mechanism is to review a recent crisis in order to draw from it lessons that might help responders improve their performance or, perhaps, limit the number of failures when responding to the next epidemic or pandemic. In this chapter, I examine whether efforts at generating lessons learned from isolated and infrequent events is likely to improve performance in future crises.

This exercise arises from the observation that, when examining three crises – 2003 SARS, 2009 A(H1N1) and 2014 Ebola together – the lessons drawn, respectively, from each begin to look less certain The difficulty in drawing lessons, I argue, stems from the nature of the problems themselves. Sudden outbreaks of deadly diseases create an imperative for rapid and effective response. Yet these health events exhibit properties of “wicked” (Rittel & Webber, 1973) and “unruly” problems (Ansell, 2016) where response efforts can become a source of surprising dynamics. Challenges arise from: (1) the novelty of pathogens and the unpredictability of even known pathogens; (2) surprises arising from geography and scale; and (3) difficulties stemming from the distributed nature of response, which involves predicting the behavior of autonomous actors, including both individuals and organizations. The intersections of these three sources of uncertainty can produce unique dynamics that may not repeat in future crises. By failing to note the deep and persistent sources of uncertainty inherent in responding to acute public health crises, lessons learned tend to presume uncertainties can be resolved quickly in order to roll out appropriately scaled response. An alternate approach would be to build response efforts around persistent uncertainty, surprise and contingency in ways that support responders in poorly understood and dynamic response environments.

Theoretical framework

While learning in the simplest organizational settings can occur through trial and error, many organizations, owing to the nature of their tasks, have difficulty making causal inferences that allow for simple learning (Wilson, 1989). Moreover, as Argyris and Schön point out, trial and error learning represents single-loop learning – i.e., learning directly about the actions and theory used to accomplish an organizational task or goal (1978). A focus on single-loop learning can inhibit double-loop learning where the organization shows a willingness to question values and assumptions related to the goal itself. Learning from isolated events should pose even greater challenges for learning. This can stem from the infrequency of events that allow organizations to drift away from practices deemed necessary following a major organizational failure (Mahler & Cassamayou, 2009). Analyses that point to forces inhibiting the application of learning in organizations over time presume that organization leaders correctly identified sources of organizational failure in the first place.

When it comes to learning from “unruly” events, the challenges for learning may be even more formidable. For instance, even when the same infectious agent is driving an outbreak, outbreaks can display remarkably unique dynamics. While responders may gain valuable experience during a crisis, the learning may not be particularly useful for the next. Even more frustrating, past experience can prove to be misleading in that novel aspects of a current outbreak may be discounted. This occurs when confirmation bias encourages responders to identify expected rather than unexpected patterns during a crisis and fail to note atypical signals. Errors, while clear in hindsight, are incredibly hard to identify during a crisis as responders attempt to characterize events using partial, uncertain and error-prone data to understand their circumstances (Keller et al., 2012).

Drawing lessons from such events is also challenging given that, during any event, officials engage in multiple activities to change the course of an epidemic or pandemic. Moreover, even without intervention, pandemics are often self-limiting; one cannot claim with confidence that public health intervention stemmed the tide of a given outbreak. Even if the outcome is a clear success or failure, quite rare in and of itself, it is impossible, with precision to determine whether one or some combination of interventions made a difference. While there are occasional natural experiments that shed light on the effectiveness of some intervention, even comparative learning can be challenging. For example, countries select intervention strategies not simply for technical reasons, but also for cultural and political reasons (Baekkeskov, 2016; Connor, 2016; MacPhail, 2014). Thus, analysts need to treat differences in outcomes in light of substantial socio-geographical differences across cases. While those who study crisis response might think of response options as discrete and portable entitles, responses can often not be abstracted from the socio-political contexts in which they are used.

Some of the difficulty in learning from epidemic/pandemic crises stems from the nature of the public health emergencies themselves: they express many features of both “wicked” and “unruly” problems. Difficulties addressing wicked problems stem from complexities with problem uniqueness and problem formulation. If problems are unique, performance cannot be improved through trial and error learning. Challenges involved in formulating a problem make it difficult to assess what “state” has occurred and what responses might be appropriate or even when a response has been successful (Rittel & Webber, 1973). Unruly problems highlight how temporal complexities add mounting challenges to response efforts (Ansell, 2016). Unruly problems manifest unevenly over time and space and can produce surprising discontinuities or crescive dynamics where incremental change in the problem fails to generate alarm until the problem has grown past the point of manageable intervention. They can produce vicious cycles where feedbacks aggravate the original problem. Moreover, a response to an unruly problem can, itself, generate new and unexpected dynamics.

Many of these characteristics identified by wicked and unruly problems arise during health crises. Specifically, a disease outbreak can cross geographical and sectoral boundaries, requiring communication and coordination between actors and organizations that may not have a history of routine interactions. Thus, they force experimentation with novel organizational networks and processes (Ansell, Boin, & Keller, 2010). Where these can produce surprising successes, they are also likely to generate unpredictable failures. Health crises can attack the very organizational apparatus that is supposed to contain them in that healthcare workers are often at higher risk than the general population during an outbreak. This can create a vicious cycle in that, once a healthcare system is overwhelmed, it may become a source of transmission rather than a site for outbreak control.

Disease outbreaks can exhibit temporal surprises including sudden shifts in state or sense-making hurdles associated with crescive problems. An example of the former occurred when SARS, though generating cases within China for weeks, suddenly appeared across continents after several travelers contracted the disease from a single doctor staying in a hotel in Hong Kong. Difficulties in recognizing crescive problems dogged the early days of the response to Ebola 2014. Officials, fully aware of ongoing transmission in three countries, assumed existing response efforts would soon reduce the number of new infections. The realization that the opposite was occurring – an outbreak spiraling towards epidemic – came weeks after a stepped-up response could have kept the outbreak within the scope of prior Ebola events. In addition, the very same disease agent can have dramatically different expressions across geographic settings. How cases of Ebola spread in West Africa versus their very limited spread in the United States and Europe demonstrates a quite predictable difference in the socio-geography of an outbreak. Rich countries with vast resources are not likely to experience sustained transmission. However, this is not just an issue of low versus high-income settings. For example, concurrent with the 2014 epidemic in West Africa, the Democratic Republic of Congo experienced an outbreak that spread across several villages and totaled sixty-six cases.² The relatively small size of this outbreak in the Democratic Republic of Congo cannot be attributed to differences in resources across the four countries.³ Thus, one needs more than a level of development to explain why an outbreak of Ebola in similarly poor countries in Africa grew to almost seventy times the size of the next largest epidemic.

The behavior and beliefs of citizens who are at risk of infection produce a significant source of uncertainty and surprise. Another stems from response systems themselves. Responders, who face especially challenging dynamics involved in understanding the nature of the outbreak, are also required to carry out response at incredibly high levels of performance. For example, managing infection control practices for highly infectious and virulent diseases in hospital settings requires that organizations and their workforces transform their performance from routine to high-reliability operations.⁴ These transformations, if they are to be successful, need to take place within hours as an infected patient who arrives in an emergency department can put other patients and the healthcare workforce at risk. Public health workers who conduct contact tracing – finding all potential contacts with someone during a suspected period of contagion – can allow new transmission chains if they fail to identify a s...

Table of contents

• Cover Page
• Half Title Page
• Series Page
• Title Page
• Copyright Page
• Table of Contents
• Notes on contributors
• Acknowledgments
• Introduction
• PART I: Setting the stage
• PART II: Lessons learned from the A(H1N1) pandemic and 2014 Ebola virus disease: a multidisciplinary point of view
• PART III: Complementing views: double standard in ethics and care
• Conclusion: global health revisited
• Index