Vaccination, hydroxychloroquine, masks and long COVID — all have become politicized, leading to the distortion of medical research during the COVID-19 pandemic and highlighting the severe harm that can accrue from medical misinformation and disinformation.

When biomedical researchers publish, we are not merely sharing our work with colleagues; we are broadcasting it into a broader media and social media ecosystem in which our findings can be misused, misapplied, misinterpreted and misrepresented in the service of any number of causes.

Credit: Kris Tsujikawa

Misinformation harms people by directly misinforming them, as we saw with hydroxychloroquine. It also harms people by aiding and abetting bad policy, as we have seen in Florida where antivaccine misinformation drives state policy. Many of us went into this profession to advance human health and wellbeing, but misinformation has the power to turn our own research against us.

Since February 2020, I have split my time between modeling SARS-CoV-2 epidemiology and studying misinformation. Based on this experience, I have eight key suggestions. First, be aware of the information landscape into which you are releasing your work. Consider the debates and controversies to which your work might pertain, the special interests that might pick up on your findings and spin them, and the biases that your findings might reinforce. Be deliberate in framing and situating your work in this context.

Second, avoid creating hype around your work or making tenuous claims about its significance. Although this sounds obvious, abuses are sadly common. Research in animal models should not be accompanied by incautious claims about human applications; in the absence of causal information, correlations should not be taken to suggest causation; and modeling should never overstate its scope. Be meticulous in specifying sources of uncertainty.

Third, recognize that data visualizations are widely shared on social media and that they can be used for good or ill. Make sure that your figures stand on their own, separated from your manuscript. If you are seeking public impact, consider whether non-specialists would be prone to misinterpreting the figures.

Fourth, where specific abuses of your findings are likely, take steps to head these off. I previously published a paper about how publication bias could allow false claims to become established as scientific knowledge. Our model pertained to small-scale scientific claims, such as those about the affinity of a receptor, not to large-scale conclusions such as the general safety of vaccines or the reality of anthropogenic climate change. Yet it was easy to imagine the paper being misused and so we concluded the paper with a paragraph explicitly explaining why our findings do not cast doubt on scientific consensus around major issues.

Fifth, if you intend to post a preprint, understand how preprints are received by the public and the media. Biomedical papers often have immediate public relevance and preprint servers have been invaluable for rapid communication during the COVID-19 pandemic, but the press and public do not necessarily understand their provisional nature. If your study has health implications, you face an extra duty of care — if you are not ready to see your manuscript appear as is in a leading medical journal, you should think carefully before posting it as a preprint.

Sixth, take direct responsibility for any press release that your institution issues about your work. Exaggerated university press releases are a major driver of inaccurate health reporting. Your press office is charged with promoting your work as broadly as possible and will welcome your help in walking the line between enthusiasm and exaggeration.

Seventh, interact responsibly with traditional media. Understand the most common sources of error in the reporting of medical research and contextualize your work to the press with the same diligence that you use in a scientific paper. Journalists who take the time to reach out to you will be invested in getting the story right, so take their calls, be patient and forthright, and help them author the accurate stories they seek to write.

Eighth, consider engaging on social media. Public engagement on social media is not for everyone, but it can be a powerful tool to combat misinformation. When a paper is misrepresented online, a correction from the authors themselves can be a powerful remedy. At the same time, be realistic and recall Brandolini’s asymmetry principle: “the amount of energy needed to refute bullshit is an order of magnitude larger than to produce it”. Even when you directly repudiate a gross misrepresentation of your work, it may continue to spread virally. In dealing with misinformation, an ounce of prevention is worth a pound of cure.

Working in the biomedical sciences, we are blessed and cursed with societal relevance and public interest. We have an obligation to provide clear and accurate information to a society hungry for such information, while curtailing inadvertent misinformation and stymying deliberate disinformation about our work. This is not an easy charge, but with care, effort and practice we can do a great deal to ensure that our scientific contributions are reflected fairly and accurately in the public discourse.