The recent epidemic of Ebola virus disease in west Africa showed the power of data generated and analysed by the responder and academic communities to help shape and improve the public health response in international health emergencies. The swift response of clinical, public health, and academic professionals from diverse backgrounds was exemplary. Data and insights from many sciences, especially in peer-reviewed papers and online data, were crucial to planning the immediate and medium-term strategy.1, 2, 3, 4 The early response to the Ebola epidemic, however, was also accompanied by several examples of individuals and organisations being unwilling to share data in real time, including data that was vital for public health planning in this difficult, fast-moving emergency. This problem was encountered by the countries most affected, governmental and non-governmental organisations, and even WHO. 5 It would be invidious to highlight particular examples; in our experience, this lack of data sharing was serious in its potential effect on the early response. The Ebola outbreak was not the first public health emergency where this has occurred, even though the response to the 2003 severe acute respiratory syndrome epidemic showed the power of real-time data sharing in rapid public health responses. 6 There are already moves afoot to ensure that key data supporting all publications in major journals are made publicly available. However, the situation of public health emergencies requires particular and urgent attention in this regard. During the recent Ebola outbreak the main reasons given for not sharing data or giving early notification of results to responders were the perceived disincentives to share data, the lack of a mechanism to enable data sharing, and the absence of positive incentives to share data. Three disincentives were frequently mentioned: data sharing would jeopardise subsequent publication; it would allow pre-emptive use of data by others for their own publications; and it would breach confidentiality agreements. None of these reasons showed those involved in a good light, given an unfolding emergency with minimum systematic information on which responders could base early decisions. This unwillingness to share data threatened the lives of both communities and health-care workers, including some of those who provided the data in the first place. Individual journal editors have already made clear that putting data or results into the hands of responders, or indeed public databases, will not threaten subsequent publication. 7 The International Committee of Medical Journal Editors (ICMJE) may well formalise this position at their forthcoming annual meeting, and this decision would be very welcome since it removes one perceived disincentive. There is no standardised best practice framework for data sharing during outbreaks, and no widely available standardised platforms and mechanism for such data sharing. WHO and others are moving to set up public databases that should help improve mechanisms for data sharing, although these would need to be optimised for public health use as well as scientific sharing early in epidemics. Journal editors can, however, go further and provide incentives for those wishing to publish data to share these before publication with responders and public health authorities. Journals could state that they will only publish data-driven research arising from a public health emergency if it is accompanied by an explicit statement from authors that they had shared data and results with authorities and legitimate bodies responding to the emergency at the earliest possible opportunity. Such a recommendation would provide a strong but not an onerous positive incentive. © 2015 John Moore 2015 Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active. The model for this suggestion is the farsighted decision by ICMJE journal editors not to accept trials that have not been prospectively registered. 8 That decision transformed the availability of prospective trial information in the public domain. The simplest method for data sharing in health emergencies would be to require authors to make a statement that they had offered to share data and results and provide supporting evidence, which would be published alongside the paper. As users, funders, and producers of research we are proud of the part the academic and public health community played alongside courageous clinical staff in helping to combat Ebola, as it has in other global health emergencies. Although those who did not share data and information at the earliest opportunity were a minority in this global effort, this failure was damaging. Many groups have responsibilities to improve data sharing in epidemics and health emergencies, but journal editors are in a strong position to provide positive incentives to share data early, and we would encourage them to do so actively.