Impact, risk-taking and avoiding bias: insights from our R2HC Funding Committee

Our sixth Annual Funding Call for humanitarian health research proposals generated more demand than ever before and came to a culmination at our R2HC Funding Committee meeting held earlier this month.

Whilst we can’t yet reveal which research studies have been given the green light, we can share some of our reflections following two days of in-depth deliberation by our Funding Committee.

This year Committee members were faced with the daunting task of reviewing the greatest number of applications we’ve ever received – almost 200 Expressions of Interest applications were submitted. A rigorous multi-stage review process narrowed this down to a short-list of 30 full proposals which were discussed in depth across the two-day meeting.

Diversity and bias in research teams

Encouragingly, this year saw the greatest diversity of applicants to date. Record numbers of proposals were led by organisations based in the Global South (a quarter), and two thirds of shortlisted studies had at least one partner based in the country or region where the research would be undertaken – local organisations provide important local context to the study design and delivery, and are frequently involved in data collection and oversight activities.

Who is in your research team also matters for other reasons. Commonly, studies are looking to evaluate the effectiveness of an intervention or approach. Humanitarian organisations are, after all, understandably keen to know whether their programmes are effective. In such cases though, research teams need to pay particular attention to ensuring the independence of researchers from the implementation agency. The potential for bias through the research process is a genuine concern, especially if the research team is seen as having a vested interest in demonstrating the success of a particular intervention. Rigorous research needs to demonstrate independence when evaluating the effectiveness (or otherwise) of a programme. If the research team includes members of the operational organisation that champions an approach, there’s a risk that the findings won’t persuade others to adopt it even if shown to be successful.

A related concern is with studies designed predominantly to influence a single NGO delivering the intervention being studied, rather than reach wider audiences. Our Committee regularly highlights the core objective of our programme: to influence the global humanitarian response.

A focus on impact, an appetite for risk

We are increasingly focused on how we can support research teams to consider how the impact of findings can be maximised. With this in mind, our Committee looks to understand the pathways to impact that applicants have mapped out in their proposals; they consider which stakeholders they plan to engage with and influence and they assess whether the research team includes the right partners to help make change happen, including local governments and other national stakeholders. Contrary to applicants’ expectations, studies with higher budgets are regarded as having positive value for money where they will lead to similarly high impact for global humanitarian response. Equally, those studies not addressing key humanitarian evidence gaps with real-world applicability of findings are usually of much less interest.

Our Committee recognises our unique role in supporting research in acute emergency settings, even where this means there is an inevitable risk that data collection might be challenging and, in some cases, not as reliable as in other settings. Where studies outline how these risks will be mitigated – as far as possible given the contexts – the Committee is willing to accept the level of risk.

By contrast, applicants occasionally propose to conduct research in what the Committee see as being ‘uniquely bad settings’ or amongst inappropriate population groups. In such cases the chosen settings might contain characteristics too narrow or specific to enable findings that are generalisable to other contexts. Alternative settings might be both more feasible and more typical of population groups found in other humanitarian settings. In some proposals, the suggestion to undertake research in a humanitarian crisis is premature, and the Committee has recommended that initial research be conducted in more stable low-income settings first. This might allow an intervention to be demonstrated as effective amongst less vulnerable groups before being adapted and tested in emergency contexts with especially at-risk populations.

We encourage applicants to include analysis of cost effectiveness in their studies where possible. This should be more than simply collecting data on costs related to the specific intervention though – the Committee is looking to see an analysis of the relative costs and outcomes/effects of different courses of action.

Conducting research in emergencies

While it is possible to design randomised studies for some humanitarian contexts, our Committee readily acknowledges this is not feasible in all settings, especially those that are the most challenging. Innovative and adapted study designs are welcomed. Studies that include mixed methods often score highest, especially those which use qualitative data to understand the practical impact of how an intervention is implemented.

In some fields of study, the Committee recognises that so little research has been conducted on a particular topic that it isn’t yet clear what methods can produce clear results. In such cases they acknowledge that one of the important impacts of a study might be to develop and learn about appropriate methods, as much as the findings generated themselves.

Although randomisation of control groups is not always possible, the Committee highlights that often more can be done in study design to ensure rigorous comparison. For example, research teams could explore the use of propensity matching approaches to select control groups with similar characteristics to facilitate stronger comparability. They also want to know what the current standard treatment is for the control arm, and want to be confident that the ethical aspects of randomisation have been taken into consideration.

More demand, more funding?

This year, a funding call which started with a huge increase in the number of applications has ended with more studies recommended for funding than ever before. This marks an exciting new phase for us.

With more and more studies demonstrating both their robustness and their relevance to key evidence gaps, we can advocate even louder for more funding to be directed towards health research in humanitarian settings.

By strengthening the evidence base for public health interventions we can improve outcomes for people caught up in crises.

[Photo Credit: Mental Health team supporting a Syrian refugee woman in a primary health centre, Bekaa. Queen Mary University]