Key data for outbreak evaluation: building on the Ebola experience

Cori, Anne; Donnelly, Christl A.; Dorigatti, Ilaria; Ferguson, Neil M.; Fraser, Christophe; Garske, Tini; Jombart, Thibaut; Nedjati-Gilani, Gemma; Nouvellet, Pierre; Riley, Steven; Van Kerkhove, Maria D.; Mills, Harriet L.; Blake, Isobel M.

The West African Ebola epidemic has prompted an ongoing public debate about the ethical, practical and scientific implications of wide data access [14, [146] [147] [148] . Ethical considerations require removal of data that might compromise anonymity, but such detailed information might be required to answer important public health questions. Mechanisms also need to be found to appropriately acknowledge those who collected and digitalized the data. From a scientific perspective, having several groups analysing the data is desirable, as independent analyses leading to similar results will reinforce their utility for policymaking [3, 8] . Such parallelized efforts have been formalized through consortiums of highly experienced groups, e.g. for modelling of HIV [149] , influenza [150, 151] and malaria [152] . However, if results differ, understanding what assumptions drive these differences may confuse and delay the process of decision-making. A consequence of data sharing, therefore, needs to be an increased emphasis on evidence synthesis, such as systematic reviews of the different analyses. To enable this process, groups should explicitly state all assumptions underlying their analyses and which data they are based on. Like the original analyses, reviews need to be timely to be useful. These issues have partly been addressed through new data availability policies [153] . Furthermore, this process would ideally be performed by a group independent of those performing the primary analyses. Identifying an effective system, appropriate personnel and appropriate recognition for this important role needs to be planned in advance of future outbreaks.