Bats, parachutes and bridges

How can epidemiologists improve global health research practice?

Imagine the scenario. A group of epidemiologists is investigating the origins of a new infectious disease in and around the area where “patient zero” is thought to have lived. The epidemiologists hypothesise that a certain type of bat may be responsible for spreading the disease. They plan a household survey, hoping to link the occurrence of disease with people’s contact with these animals. They carry with them pictures of the bats to show, but disappointingly, in interview after interview, people report having never seen the creatures.

That is until during one interview, when a perplexed researcher notices one of the bats flying overhead. Now it is the respondent’s turn to be confused. “What, that one?” they ask. “But it’s so big! The bat in the picture is tiny compared to this one. We don’t have tiny bats here, only big ones. You should have said it was big!”

This is a textbook example of what can go wrong with “parachute studies” – a term used to describe research projects conducted by researchers who are “parachuted in”, lack familiarity with a local situation, and disregard local knowledge systems and expertise. This “tiny bat” story – brief as it is – is emblematic of a much wider set of problems in global health, concerning research integrity and research fairness.

Tiny bat, bigger problems

Integrity and fairness, and the lack thereof, feature prominently in our “tiny bat” story. On the issue of research integrity, we have a photo which poorly represents the bat in question. It is an ill-designed survey tool, probably developed in a hurry and not adequately pre-tested, which will lead to information bias. If a respondent had not pointed out the researchers’ error, the epidemiologists might have unwittingly drawn the wrong conclusions from their investigations but still published their findings anyway, meaning that those study findings are unlikely to be reproducible (unless other studies use a similarly flawed survey tool!). Irreproducibility of findings is a concern as it can fuel mistrust of science. It is also a research integrity issue: irreproducibility may arise from innocent mistakes or poor research methods (as in the “tiny bat” story). At worst, it can be the result of deliberate misconduct and fraud.

As for research fairness, imagine that our epidemiologists are from a European university, travelling to sub-Saharan Africa to conduct their research. This recasts the story as one about a global health study involving researchers from high-income countries working in a low-income setting, which therefore prompts a number of critical questions about study conduct. Was the study conducted in close collaboration with local researchers? Were local communities and local decision makers consulted to develop useful and relevant research questions? If the poor choice of bat photo is anything to go by, the answer is, probably not.

This makes it a typical example of an unfair study, which will ultimately have limited impact at the local level, where the research is conducted. Involving local researchers, communities and decision makers not only helps to ensure better quality research, but also better and more effective communication of research findings to those who need to act on the results. And, in the long term, the involvement of local researchers ensures the development of strong local research capacity to tackle other (perhaps even more) relevant issues.

When we talk about “fairness” and “unfairness” in this context, we refer specifically to power imbalances in global health, resulting from researchers in high-income countries being funded by organisations in high-income countries to conduct research in low- and middle-income countries. Such power imbalances can prevent local stakeholders from shaping the research agenda and competing on a level playing field in international scientific arenas.

Bridging the issues

In order to jointly address the issues of research integrity and research fairness in global health, epidemiologists at KIT Royal Tropical Institute developed a set of good epidemiological practice guidelines in consultation with researchers from Asia, Africa and Latin America.[1] These BRIDGE guidelines draw from existing guidelines focused on research integrity and initiatives to increase research fairness with the aim of helping researchers in international research collaborations to produce technically sound, impactful results. The result is a set of practical tips for funders and researchers at each stage of a study, from conception and planning to the writing up of results.

A number of items from the guideline’s “study preparation” and “data collection” checklists could have helped with the “tiny bat” study. During study preparation, for instance, BRIDGE advocates for the need to “plan and execute research in partnership with local researchers” and to “establish the knowledge gap by searching the literature (peer-reviewed publications and grey literature) as well as by consulting (local) experts, representatives of affected populations and end-users”. So, early on in the study, the BRIDGE guidelines would have helped ensure that the researchers had a reasonably good understanding of the study population’s concerns and viewpoints.

Meanwhile, during data collection, the guidelines emphasise the importance of “using valid and reliable research instruments” and to “pilot test, and if possible, field test all research instruments prior to the start of effective data collection”. It is possible that pre-testing the photograph-based tool to measure the frequency of bat sightings would have revealed its flaws. Elsewhere, the guidelines remind researchers to “ensure that research instruments are locally adapted and culturally appropriate” and to “select data collection staff according to technical as well as cultural criteria”. A reflection on this criterion might have helped the researchers realise that relying on a photo in the context of the “tiny bat” study was culturally inappropriate.

The criteria for study preparation and dissemination/communication have a strong emphasis on redressing existing power imbalances in global health research, by putting local stakeholders and local researchers at the centre of the research endeavour. Indeed, the guidelines aim to ensure that local stakeholders (e.g. national and local representatives of the ministry of health, health facility workers and community members) are engaged early on in the study to ensure that the research deals with their priority problems – not the researchers’. For example, it is realistic to think that local stakeholders might have preferred that the European researchers work on strengthening the disease’s routine surveillance system rather than studying its origin, as this is likely to have more tangible benefits for the local population’s health. The guidelines also offer a blueprint to go beyond tokenistic involvement of local researchers, by ensuring that they are in a position to lead analysis and dissemination efforts. For instance, the guidelines prompt researchers to agree on publication plans, data sharing agreements, and professional development (e.g. training, coaching) early on in the study. These criteria aim to strengthen local research systems, which is paramount in order to enable a comprehensive response to all diseases which affect the local population – not just the one of interest to a particular group of foreign researchers who happen to have a specific interest in bat-spread diseases at this one point in time.

Tiny bats or big bats – making mistakes and learning from them is an integral part of any scientific endeavour. Mishaps along the way can be expected and no set of guidelines will ever be enough to safeguard research from all possible blunders. But the BRIDGE guidelines hopefully can help researchers steer clear of questionable and unfair research practices that may arise, and work towards a long-term positive impact on local research systems and local communities.

The involvement of local researchers ensures the development of strong local research capacity to tackle other (perhaps even more) relevant issues

Acknowledgement

Many thanks to my colleague Ente Rood for the “tiny bat” story, which originally featured neither a bat nor an epidemiological study.

* This article is a reproduction of the article “Bats, parachutes and bridges: How can epidemiologists improve global health research practice?” by Sandra Alba published in Significance Volume18, Issue 4 August 2021. Link to the original article https://rss.onlinelibrary.wiley.com/doi/full/10.1111/1740-9713.01542

**Disclaimer: Reproduction, posting, transmission or other distribution or use of the final Contribution, in whole or in part, in any medium by the Contributor, as permitted by this Agreement. requires a citation to the Journal suitable in form and content as follows: (Title of Article, Contributor, Journal Title and Volume/Issue, Copyright © [year], copyright owner as specified in the Journal, Publisher). Links to the final article on the publisher website are encouraged where appropriate.

Reference

1. Alba, S., Verdonck, K., Lenglet, A., et al. (2020) Bridging research integrity and global health epidemiology (BRIDGE) statement: guidelines for good epidemiological practice. BMJ Global Health, 5:e003236.