Read science news critically and look for original studies: An example of misleading headlines related to COVID-19 vaccines in mainstream media.

In July 2021, after a representative of the Serbian health authorities had cited a recent study from Sri Lanka in the media that the Sinopharm antiCOVID-19 vaccine (BBIBPCorV) was very effective against the Delta strain of SARSCoV-2, I tried to locate the original study. The headlines in the Serbian media read: ‘Sinopharm is the most effective vaccine against the delta variant, studies show’ (RTS, 2021). While conducting a simple online search using a combination of keywords, I was surprised to come across several articles in the English language with headlines suggesting that the Sinopharm vaccine is not effective against the Delta strain (‘Sinopharm’s COVID-19 shot induces weaker antibody responses to Delta -study shows’) (Reuters, 2021). Unfortunately, most people will not read the full articles, let alone attempt to locate the original study to locate the actual results. Reading the full text of the original manuscript revealed that the Sri Lankan study (Jeewandara et al., 2021) had reported that the titre of specific antibodies was comparable between Sinopharm-vaccinated individuals and those who survived a natural infection with the Delta variant. Another observation reported in the study was that the titre of antibodies specific to the Delta strain in vaccinated individuals was lower than that of the original Wuhan variant. It was clear from the text of the original study that the authors had not actually compared the different vaccines; thus, the headlines in the Serbian media implying that this vaccine was ‘the most effective’ had not accurately described the cited research. Nevertheless, the Serbian media headlines correctly emphasised the protection aspect, given that the vaccine obviously offers seroconversion comparable to that of a natural infection. In contrast, the English-language media headlines wrongly emphasised that the vaccine offers a ‘weaker response to Delta’; the study actually showed a weaker response compared with the Wuhan variant, but similar to a natural infection with the Delta strain. These authors had not studied the effectiveness of the vaccine in preventing infection, hospitalisation or death; rather, they had focused on the evaluation of specific antibodies and T-cell responses. Of note, this was a manuscript uploaded to a preprint server and not a peer-reviewed article; it should also have been read critically andwill, hopefully, soon go through fair review by experts in the field. In the meantime, this is an exemplary case of how easy it is to misinterpret research findings, and how easily some of the ‘fake news’ develops. Special care is needed to avoid cases such as this, to avoid further erosion of people’s trust in health systems and vaccines protecting against COVID-19. This case is also a beautiful reminder that we should always refer to the primary literature.