Start main page content

Crowdsourcing: Scale of COVID-19 calls for new approaches to research

- Chris William Callaghan

Crowdsourcing is a promising approach to biomedical research and development (R&D) and could produce solutions to pandemics like this one.

COVID-19 is not the world’s first pandemic. Nor is it the only type of big problem that the world faces. Environmental degradation and drug-resistant diseases are other examples.

What’s special about the new coronavirus is the speed with which it has arisen and multiplied.

The current system of scientific and academic research can’t respond fast enough to problems like these, especially when data is still being generated. But there are potentially ways of overcoming this mismatch.

I propose that crowdsourcing is a promising approach to biomedical research and development (R&D) and could produce solutions to pandemics like this one.

The biomedical research and development industry largely responds to private incentives – even if subsidised by governments. Drugs are more likely to be developed for wealthy markets offering opportunities for chronic medicines that need to be taken for the rest of one’s life. This is because drug companies face the challenge of very large initial investments which they might not be able to recoup if a drug is not profitable. There’s less incentive to do R&D that could have wider social benefits.

A large-scale response to the COVID-19 outbreak is under way and there are already vaccines under development. But there is no indication yet that they will be successful. The current R&D response may simply not be large enough to stop the pandemic quickly enough.

How do we stop COVID-19?

To stop this pandemic, it may be necessary to move activities out of already productive (and profitable) research activities. And this may have to happen on a scale that is proportionate to the scale of the cost of the outbreak. A radical restructuring of the incentives of the biomedical research industry may be necessary to shift this activity away from its profitable uses and into (uncertain) vaccine research.

Academic research suggests how this might be done. Probabilistic innovation theory suggests that problems such as COVID-19 need to be exposed to processes that radically increase their probability of success. This may require novel technologies and methods to greatly increase the chances of solving the problem such as biomedical crowdsourcing, machine learning and big data science.

These have already demonstrated their effectiveness in biomedical research, but not yet at the scale required to stop the pandemic. Another useful example of biomedical crowdsourcing is gamification, a process whereby complex biomedical problems are used as the basis for computer games, with the goal of solving them. The site FoldIt is successfully using protein folding games to solve these kinds of problems.

A useful way of thinking about this approach is in terms of a societal benefit ratio. This is the ratio of the research efforts invested in solving a problem to the consequences of the same problem. In other words, many problems with very high human and economic costs don’t receive enough problem solving resources – the scale of the investments in solving these problems should be appropriate to the scale of the problem, or it might not be solved.

Current R&D efforts aimed at tackling the pandemic may produce a societal benefit ratio that is too small. The COVID-19 pandemic potentially affects around 8 billion people. Estimates of necessary interventions suggest that, if not addressed, the damage to the global economy could be in the trillions of dollars. The problem with current approaches seems to be that they are largely rooted in the profit-driven structure of the biomedical industry. Even with academic collaborators, this restricts the size of the investments in solving the problem.

Existing efforts will surely come up with a solution given enough time. But it might be necessary to consider other scientific approaches that have already demonstrated their success in biomedical research, and try them at a large enough scale.

How crowdsourcing works

Advances in technology have made it possible to crowdsource solutions to biomedical problems. Biomedical crowdsourcing is a problem-solving methodology based on putting problems online as an open call for anyone to solve. Sites like InnoCentive provide platforms for the initiation and administration of scientific crowdsourcing, but a large-scale global project might be administered by the United Nations or the World Health Organisation.

The successes of crowdsourcing in biomedical research are well documented. Sites such as InnoCentive have shown that complex scientific problems can often be solved more cheaply and quickly than they would using in-house R&D departments.

It could be argued that the scale of the crowdsourcing efforts to date has been too small to force activity into uncertain avenues of research. If governments across the world were to pledge a portion of their ongoing economic costs of the pandemic, it might be possible to offer a large biomedical crowdsourcing award, for example in excess of a thousand billion dollars. The scale of this award would better match the scale of the consequences of the pandemic. Such countries would not have to pay a cent if a solution were not found. Those seeking to solve the problem (solvers) bear the cost and risk of these efforts. This makes it necessary to offer a very large award. These costs include opportunity costs, such as the costs of not doing other work in the meantime.

If crowdsourcing were to solve this problem, then what of others that we have failed to solve until now? The current pandemic might offer researchers a unique chance to test this methodology at a large scale. If necessity drives invention, then there is no more important time than this to try new ideas.The Conversation

Chris William Callaghan, Professor, University of the Witwatersrand. This article is republished from The Conversation under a Creative Commons license. Read the original article.

Share