In December 2013, a little-known viral hemorrhagic fever in West Africa was detected in a forest region of Guinea. It was soon confirmed to be Ebola Virus Disease, an illness previously seen primarily in Central Africa. Within months, Ebola would become a global outbreak. To contain the disease, national and international actors needed access to timely, accurate, and precise data. Yet as the disease spread across Guinea and into the neighboring countries of Liberia and Sierra Leone, critical outbreak data often were missing, unclear, or contradictory. Unexplained peaks and valleys in Ebola case counts, for example, and dramatically differing forecasts of the disease’s potential spread, complicated the relief effort and raised important questions about why it was so difficult to track the disease.
This report details key findings from research focused on the collection, management, analysis, and use of paper-based and digital data and information comprising the Ebola outbreak response. It reveals common sources of the confusing data picture, particularly in the early days of the response, and examines the use of digital technologies to support data and information flows, considering both common barriers and insights from what worked. Building upon a qualitative research approach, including interviews with over 130 individuals engaged in the response, including representatives from NGOs, UN agencies, donor and national governments, and the technology and private sector between November 2014 and February 2016, the research addresses the following questions:
- What contributed to the “fog of information" that characterized much of the early stages of the Ebola outbreak response?
- What can be learned from the use of data, information, and digital technologies during the Ebola outbreak response? How and where were they used effectively?
- What should be done to improve the use of data, information, and digital technologies in emergency contexts, to support long-term recovery, and to build resilience against future shocks?
Recalling the context in which the outbreak flourished and revisiting predictions of Ebola’s spread sets the scene for analyzing the Ebola outbreak and response: three countries with porous borders, interconnected populations, and fragile health systems and infrastructure following years of civil conflict and war. In this context, Ebola quickly spread out of control, catching national and international actors unprepared, with regard both to the rapid rise in prominence of a rarely encountered viral hemorrhagic fever, and to the degree of coordination of actors, systems, and data that a hybrid public health and humanitarian emergency of this scale would entail.
A number of critical factors contributed to an unclear and asynchronous picture of the disease’s spread--the “fog of information” that characterized the effective collection and use of data in the early days. These included: weak infrastructure, such as gaps in reliable electricity and/or digital connectivity; an absence of baseline data, including commonly-used unique citizen identifiers and comprehensive and accessible geographic maps; and the predominance of non-machine-readable data, including the collection of many kinds of disease outbreak and response data on paper, and the publication of summary reports in non-machine-readable format, slowing and narrowing the use of these data. Coordination challenges surfaced in responding to a hybrid health and humanitarian emergency, the impact of non-aligned data standards, and a lack of interoperability between the numerous data systems used to manage data prior to and during the response also contributed to the unclear and asynchronous data picture. Finally, the outbreak underscored the importance of the cultural context and trusted affinity groups in delivering behavior change and other messaging, regardless of delivery channels.
Data about Ebola cases were structured, collected, and used in many complex forms, and frequently involved the time-consuming task of being transferred from paper to digital format. An array of data sets supported the response, including Ebola case data. Their definition, use, and management informed the Ebola response. Case data about individual patients, as well as caseload data composed of aggregated individual case data, were central to understanding the disease’s trajectory and to formulating corresponding aspects of the operational response. Tensions arose regarding the collection and use of data, such as by health professionals treating patients on the one hand, and humanitarian responders tracking and responding to the effects the disease’s spread on the other. Case data, once digitized, were more easily managed and used; however, the proliferation of data platforms negatively impacted on the response.
Nine case studies demonstrate the flow of Ebola response data and information via digital channels and the contributions and challenges of their use. Although digitized data and information flows did not constitute the norm, they did contribute meaningfully to the Ebola outbreak response in specific instances. When used effectively, they introduced both quantitative and qualitative differences in data and information flows. The case studies illustrate the differences the integration of digital technologies enabled, such as increasing the diversity of information flows (e.g. “up” for data collection, as well as horizontally among peer groups, and back “down” through feedback loops) among a greater plurality of actors (e.g., frontline health workers, citizens, governments, and “remote” responders).
The value proposition for integrating digital technologies lies in enabling richer, more diverse, and more rapid data and information exchange, the benefits of which can accrue to health and humanitarian programs, particularly in crises. These include:
- increased accountability, insights, and incentives
- an ability to create feedback loops through the sharing of contextualized data and information back to the point of origin
- the ability to implement continuous learning and adaptive programming, in which activities are modified and, ideally, regularly adapted on the basis of real-time or near real-time data and information and
- the ability to make better-informed decisions at all levels
The use of digital technologies will not, as this research demonstrates, automatically confer these benefits. Indeed, the report finds that the barriers that impede the effective flow of data and information in paper-based environments must also be considered in the integration and use of digital technologies for data and information exchange, and for use in decision-making.
While many of the recommendations in the report require significant investments in human capacity and institutional policies and procedures to reap the benefits of data and digital technologies, certain "quick wins" can be rapidly deployed in the next emergency to support responders and affected communities. These include adopting machine-readable forms, deploying pre-negotiated short codes, ensuring online- and offline-sync capacity for digital technologies, instituting GIS-enabled systems, and providing wifi and satellite-based mobile and broadband connectivity to health facilities and other key institutions.
To unlock the full value of digital technologies, increased attention to and investments in the human, institutional, and policy and regulatory enabling ecosystems are required. Specific recommendations include:
- Invest in physical infrastructure that extends digital connectivity in order to increase resilience in the context of health, humanitarian, and other crises.
- Conduct baseline, country-wide ICT assessments to gauge the reach, quality, and citizen access to mobile and broadband connectivity, and publish findings on shared repositories using machine-readable formats. This allows the prioritization of investments to extend the physical infrastructure that enables digital connectivity.
- In an emergency, develop and implement emergency protocols for rapid updates to baseline country-wide ICT assessments that gauge the reach, quality, and citizen access to mobile and broadband connectivity, and catalogue the effects of an emergency on baseline connectivity infrastructure and access. Doing so enables an understanding of the extent to which mobile and broadband technologies can support the response by rapidly identifying critical connectivity gaps.
- Build staff capacity and data literacy as well as institutional capacity to leverage digital systems and real-time data in support of operations, programs, and decision-making. Effectively using data and digital technologies require more, not fewer, staff to coordinate and manage collection of information across multiple partners, to support use and adaptation of digital platforms, and -- most importantly -- to analyze data in order to inform decision-making.
- Negotiate preparedness protocols with key actors (governments, mobile network operators, and regulatory bodies) to increase telecommunications network access in emergency situations. This will facilitate rapid collaboration with key actors and support the deployment of ICTs during an emergency response.
- Support the development of digital health strategies connected to interoperable emergency preparedness protocols. Where appropriate, linking emergency health data systems with national routine health data systems, such as disease surveillance, will make standing up emergency systems during a crisis easier and faster, and help to improve data quality.
- Advance the ethical and responsible use of data and digital technology. Good data practices include establishing protocols that protect individuals’ privacy and security, including for vulnerable populations.
- Agree upon and support the broad uptake of common data standards to enable effective sharing of data across sectors, systems, and silos.
- Build processes that work toward openness and interoperability. Reduce fragmentation and duplication related to data and ICT to maximize investments and to ensure maximum value of data.
- Encourage coordinated and sustained investments in interoperable data and data systems or platforms. Minimize duplication of efforts and funding and co-invest to achieve scale.
- Consider the use environment, including the digital infrastructure, sociocultural, and psychosocial context in designing and deploying digital technologies. Ensure digital technologies are used in a manner that is relevant, appropriate, ethical, and efficient.
- Insert feedback loops in the full lifecycle of project conceptualization, from design and delivery to monitoring and evaluation. Increase the effectiveness of programming and improve humanitarian and development outcomes.
- Design programs to incorporate digitized data and information flows. Enable faster feedback and iteration, and expand the nodes of connection in order to increase the effectiveness of programming.
Implementing these recommendations will not be easy or quick. Taken together, however, they will help strengthen the effective use of data and information to support health and humanitarian programming--whether in emergencies such as Ebola, in recovery, or to promote long-term resilience.
 "Fog of information” is a variation of the term “fog of war,” first attributed to the Prussian military strategist Carl von Clausewitz and more recently popularized in the documentary film of that title that explored the difficulties of decision-making in the midst of conflict, when full situational awareness is often absent. We adopt this term, which several interviewees used, to describe the lack of timely, accurate, and accessible data, which clouded situational awareness, impeded effective decision-making, and stymied the response.