outbreak). 6 2. Barriers to data sharing during the Ebola outbreak: Some of the most commonly raised barriers to data sharing were related to 1) the relatively undeveloped state of health infrastructure in the three most affected countries; 2) the fractured sources of relevant data; 3) the lack of coordination and clear roles for data generating sources including non-governmental organizational and governmental responders; 4) the development of diagnostics, therapeutics, and vaccines through multiple layers of public-private partnerships; 5) incentives for private sector researchers that penalized or prohibited sharing; 6) incentives for public sector researchers that penalized or prohibited sharing; 7) ethical and legal constraints related to patient confidentiality and informed consent; and 8) community-level barriers to data sharing. a. Standardized and Uniformity in Data Collection and Sharing Before the Ebola outbreak, Guinea, Liberia, and Sierra Leone had suffered from devastating civil wars or internal conflict, which leveled a corresponding effect on the countries’ health system infrastructure. 51 Under WHO assessments, their health infrastructures were among the weakest in the world. 52 These weak infrastructures led to two related problems in the context of data sharing. First, the provision of healthcare and the surveillance and reporting roles often undertaken by public authorities were fractured among dozens of nongovernmental organizations, many of which paid higher salaries or offered employment on more favorable terms than state-administered entities of the healthcare system. These organizations maintained non-uniform systems for collecting, centralizing, analyzing and transferring data. Second, when EVD cases emerged in remote areas of the most affected countries, data about illnesses and deaths was confused with other common causes of morbidity and mortality that occurred at high rates in all three countries. i. Epidemiological Data Diagnosing and treating an infectious disease with epidemic or pandemic potential involves individual-level data, exposure data, and population-level data.19 These data are used to create line-lists and projection models. 2 Population-level statistics such as demographics and geographic information are then used to predict the future spread of the disease. From December 2013 until March 2014, the lack of uniformity in collecting data and the lack of standardization in reporting it explains discrepancies in the initial assessments by NGOs on the ground, particularly MSF, WHO and the Governments of Guinea, Liberia, and Sierra Leone.21 In March 2014, WHO activities in the three most affected countries were extensive and cases in Guinea between April and May had declined under the metrics then used. 53 Although WHO released official case definitions of confirmed, probable and suspected Ebola cases, different 2 Line-lists are tables that list each infected patient and contain demographic details such as age, race, potential exposures and transmissions, etc. Line-lists are used to determine how diseases are spread among populations, as well as contact tracing and control efforts. 7 countries adopted different testing strategies, thereby limiting the opportunity for intercountry comparison. The Government of Sierra Leone used only laboratory confirmed cases in its preliminary response analyses. 54 In Guinea, deceased individuals were not tested for Ebola, meaning these individuals were never classified as confirmed cases, unlike in Liberia and Sierra Leone. In Liberia, “ministries (including port, airport, finance, health and environment) local governments, clinicians, nongovernmental organizations, suppliers, and donors” all collected data related to identifying cases and taking immediate action, but there was “no information sharing” because there was no centralized authority or resource to do so.54 Even within data collected, inconsistences limited usefulness. Dates recorded on a case document might have referred ambiguously to when data was collected, submitted, or edited. Some NGOs providing health services in Guinea and Sierra Leone worked under agreements that authorized the sharing of relevant data only with official health authorities and, in some cases, only with specific administrators.55 Requests by other NGOs, especially for contact lists, were rejected pretextually or actually on this basis. 7 For instance, one district-level Ebola response centre (DERC) in Sierra Leone found it problematic that NGOs engaged in contact tracing spontaneously but in coordination with their role of providing supplies to families in quarantine. The NGOs took it upon themselves to start taking temperatures, recording travel and contact histories. There was already a contact-tracing team from the DERC visiting them over several days to monitor symptoms and gather information on exposure to risk. There were several other organizations visiting or doing similar activities, undermining the role of the centralized, official authority. Inconsistent and haphazardly collected and transmitted data bottlenecked at the hospital, ministry, and international levels. Data quoted by Sierra Leone’s Ministry of Health and Sanitation, for example, were inconsistent with WHO’s which was in turn inconsistent with determinations made by