Research in crises: examples from the Ebola outbreak

Humanitarians are increasingly being asked to deliver more. Many agencies and donors now require them to report on the impact of their work, and to prove that the response they mounted was the best possible option and the most effective and efficient path to recovery – in other words, finding and using ‘what works’. There are two challenges with this: first, measuring impact and showing proof is often extremely difficult in emergency contexts; and second, current systems and mechanisms for generating information and knowledge are often not capable of meeting these additional expectations. It is possible that new or adapted systems, people and concepts will have to be taken up.

A key concept in measuring impact and proving what works is evidence. However, there are many different definitions of humanitarian evidence, and the ways in which different agencies are collecting and using it vary greatly. This article discusses one way of producing evidence – namely research – and discusses how research and evidence is being used in the context of the Ebola outbreak. It draws on experience from two research projects. The first, ‘The Ebola Response Anthropology Platform’, is being led by academics from the London School of Hygiene and Tropical Medicine, Njala University in Sierra Leone, the Institute of Development Studies at the University of Sussex and the University of Exeter in the UK.1 The second, ‘Participatory Behavioural Change To Reinforce Infection Prevention and Control for Ebola Virus Disease in Sierra Leone’, is led by the International Rescue Committee (IRC) in partnership with Charité – Universitätsmedizin in Germany, Durham University in the UK and Njala University. Both were funded by Research for Health in Humanitarian Crises.

How does research build evidence?

AThere are two fundamental characteristics of evidence: it must be methodologically sound, and it should establish causality. D. Dijkzeul, D. Hilhorst and P. Walker, ‘Introduction: Evidence-based Action in Humanitarian Crises’, Disasters, 37(S1), 2013.

In this context, a methodology is the set of principles, pro-cesses and practices used to collect, analyse and synthesise evidence. The words ‘sound’ and ‘robust’ are often used to describe methods that are ethical; contextually appropriate; follow established standards; and are described in detail, so that all stakeholders can access and read the methodology and judge for themselves if it is sound or not.

Methodologies could be unsound or unethical if, for example, a researcher failed to get parents’ permission to interview children; if local or field research teams were not trained to understand what constitutes high-quality, valid data; if field team composition was not ethnically appropriate, gender-sensitive and competent in the most suitable languages; or if they did not talk to enough people, evaluate enough camps or inspect enough infrastructure to have a representative sample on which to base their analysis.

Establishing causality is a big problem in complex situations. In practical terms it means that, if an organisation decides to launch a particular health promotion campaign, then they should be able to prove that it was their campaign that led to measurable changes in behaviour; or that, by supplying certain amounts of water to a care centre, an agency should be able to objectively measure improved health profiles in vulnerable groups and link improvements directly back to their activities. There are two main problems with these scenarios. First, it is very difficult to measure changes and improvement in a particular group of people after each and every humanitarian action. Second, it is difficult to isolate the impact of a single programme and claim that it was the key action that led to observed and measured changes.

Research can help to measure impact and demonstrate ‘what works’. A good humanitarian researcher or academic, who is familiar with the context and region of the crisis and who has experience in a particular field, can make sure that evidence is robust and ethical and can answer specific questions with real proof. They can also analyse rationally how much impact a programme or action might have had, given all the other variables (conditions and actions) in play.

Research and humanitarians

Research is important because it is key to an evidence-based approach. Research in the Ebola outbreak produced evidence and advice that influenced humanitarian action in real time. Some examples of active research from the Ebola Response Anthropology Platform are highlighted in Box 1 (see below).

Researchers in emergencies need to pose clear questions and design a methodology that answers those questions with the minimum of resources. They ultimately require rigorous justification to proceed – it is important to be prudent with research, given competing priorities in a crisis. Perhaps the highest priority will be research that can quickly answer specific, critical questions in a way that could impact that response very quickly. Another high priority may be research that answers critical questions that come up repeatedly in different responses, so that humanitarian response is made more effective in the future.

Linking research to practical needs

Three key ingredients are necessary to link research with need: rapid response funding; knowledge of – and networks between – researchers who are already experts on the region and the context of the emergency; and immediate access to grassroots or ‘field’ researchers, who will often be local, within this network. If these three ingredients are lacking, research can still be important and can contribute to improving responses to future crises. However, it is unlikely to be able to contribute evidence to a current emergency.

The first point can be difficult to address, as there are few rapid response funding mechanisms available. Both of the projects discussed here were funded through Research for Health in Humanitarian Crises (R2HC), which managed a special, fast-track Ebola Emergency Call in late 2014.

In line with the second point, Anthropology Platform managers very quickly brought together researchers who   were already experts in West Africa and public health there. The Anthropology Platform Steering Group comprises researchers mostly from the UK and Sierra Leone. This mix means that field, national and international experts are able to contribute as needed, depending on the questions that arise and the analysis that needs to be done, in a responsive and flexible way. This group could be pulled together quickly because there was an existing partnership between the lead university in Sierra Leone (Njala University) and the UK.

The third ingredient – field research – is critical. It often takes years for a researcher to build up field experience and grassroots knowledge within a region, and in an emergency this can form the foundation for new, tailored research.

Obstacles and challenges

Many of the obstacles and challenges that confront humanitarian responders also affect researchers. In the same way that practitioners often struggle with coordination, a high-profile emergency like the Ebola outbreak can lead to a rush of researchers, with varying levels of experience, who may not be coordinated and may have agendas which compete with each other and with the humanitarian response. The IRC project leaders found that having the University of Sierra Leone as a partner provided good insight into potential overlaps or conflicts in and between research projects. It is also important that researchers are sensitive to humanitarian imperatives and priorities, and work in collaboration with humanitarian staff. IRC project staff in-country kept in close contact with the health cluster and the Ministry of Health, to maintain an overview of interventions being done in the outbreak.

Researchers can also do harm, and it is standard practice that many types of research require ethical clearance from the government and from a university in order to ensure that local and/or international standards are followed. However, getting ethical approval can be difficult. For its project the IRC sought ethical approval from the Ministry of Health and Sanitation in Sierra Leone, but the ministry was overstretched with the response and the IRC found it difficult to get priority attention. In this case, obtaining an additional rapid review from a review board at Durham University was key in ensuring ethical oversight quickly.

The timeframe of methodological planning is also a common obstacle. Methodologies must be detailed and robust, and researchers outside a crisis context will normally spend weeks or months designing and testing their methodologies. The Anthology Platform and the IRC addressed this in part by having established partnerships with experts who already understood the context, the countries and/or viral haemorrhagic fevers. Regional and topical experts are the researchers most likely to be able to create a robust research plan quickly.

Whose role is it?

The responsibility to do high-quality research in crises of all kinds is mutual – humanitarian and academic institutions together must make sure that research is accountable, relevant and useful. Amongst researchers, there is an ethical imperative to share their knowledge in a useful way. For humanitarian practitioners and donors, it is important to recognise how and why research in crises should and could be undertaken, and determine how best to fund it alongside active response. This means that humanitarians and researchers must be willing to work together in partnership.

However, each member of the partnership must also have a clear role to play, and be able to deliver specific results. For example, in the Participatory Behavioural Change project, the IRC was already taking a lead on infection prevention and control in Sierra Leone through the nationwide Ebola Response Consortium. Therefore it was able to identify critical problems that could be addressed through research – in this case, improving health workers’ capacity to adhere to standard precautions in order to prevent infection. Academic partners at Durham University contribute anthropological knowledge and expertise; epidemiology expertise comes from partners at Charité – Universitätsmedizin Berlin; Njala University contributes to the design aspects of the project and provides expertise in infection prevention and control for viral haemorrhagic fevers; and the Kenema District Health Team provides contextual expertise on the ground.

Summary

Gathering evidence through research on what works and what doesn’t is both necessary and possible in the midst of a crisis like the Ebola outbreak. It is most effective when researchers and humanitarians work together in partnership. Humanitarian and academic roles can be brought together through better mutual understanding of the importance of evidence in humanitarian practice. Humanitarians can better request and support research, and researchers can better address critical humanitarian needs and support humanitarians, when there is agreement that research matters.

Lisa Guppy is Senior Research Advisor, Enhancing Learning and Research for Humanitarian Assistance. Dr. Melissa Leach, Director of the Institute of Development Studies (IDS), and Ruwan Ratnayake, an Epidemiology Technical Advisor with the IRC, were both interviewed for this article and contributed substantially. Thanks also to Sian Frost for her help.