In August 2014, just days after the World Health Organization (WHO) officially declared that the Ebola virus disease epidemic in West Africa was a public heath emergency of international concern, I received an email from International Medical Corps, a global humanitarian organization, asking me to join their initial emergency response team in Liberia. This was not my first emergency deployment with International Medical Corps; over the past 5 years, I have treated earthquake victims in Haiti, managed a trauma field hospital outside Misurata, Libya, and set up a refugee camp clinic in South Sudan. But launching an Ebola response would require a completely new set of skills. As the first medical coordinator for International Medical Corps’ first Ebola Treatment Unit, one of my tasks was to develop our protocols for clinical care for patients with suspected or confirmed Ebola. In doing so, I scoured the medical literature for clinical trials, diagnostic studies, or even observational studies that could help us develop evidence-based protocols for clinical care. My search came up largely empty. Even prior guidelines developed by WHO and Medecins Sans Frontieres (Doctors without Borders) contained only a handful of primary sources in their list of references. This dearth of research on the clinical management of Ebola virus disease, despite nearly 40 years of international experience responding to epidemics of Ebola in Africa, did not come as a surprise to me. It mirrors the lack of evidence on acute care and public health interventions in all types of humanitarian emergencies, from the management of crush injuries in earthquakes to gender-based violence in conflict settings, from acute malnutrition during famine to cholera outbreaks in refugee camps. Indeed, despite the steady rise in both ‘‘natural’’ and ‘‘man-made’’ emergencies (and the corresponding rise in funding for emergency response), humanitarian healthcare today looks a lot like the rest of medicine did 50 years ago—based largely on anecdote rather than evidence, with few high-quality research studies to guide practice. The model for responding to humanitarian crises in general, and Ebola outbreaks in particular, can be summarized as ‘‘see one, do one, teach one.’’ We respond to each new emergency with the same tools we used to manage the last one, never truly knowing whether any of it worked, or what alternative methods might work better. There certainly have been efforts in recent years to change this dynamic. The SPHERE guidelines (http:// www.sphereproject.org), developed in the 1990s after a cholera epidemic in the refugee camps of Goma in the former Zaire claimed over 30,000 lives, have helped to standardize humanitarian response. But the SPHERE guidelines, similar to those developed for Ebola, are only as good as the research that supports them, of which precious little has been conducted to date. More recent initiatives have been working to build that evidence base. Evidence Aid (http://www.evidenceaid.org), an offshoot of the prestigious Cochrane Collaborative, has done an impressive job of summarizing existing research on humanitarian and disaster response through high-quality systematic reviews and rapid release of pertinent data in response to major crises, including the recent Ebola epidemic. Perhaps even more important, it has done a significant amount of work to highlight the current gaps in the literature and highest priorities for disaster research. In addition, Enhancing Learning and Research for Humanitarian Assistance (ELRHA, http://www.elrha.org) is a UK-based initiative that has begun working in recent years to bridge the gap between research and practice in the humanitarian realm. They have