Characterisation of social media conversations on syphilis: an unobtrusive observational study

Abstract

Background Conversations around disease conducted through social media provide a means for capturing public perspectives that may be useful in considering public health approaches. Syphilis is a sexually transmitted disease that is re-emerging. We sought to characterise online discourse on syphilis using data collected from the social media platform, Twitter. Methods We extracted English-language tweets containing the word ‘syphilis’ posted on Twitter in 2019. Tweet identification number and URL, date and time of posting, number of retweets and likes, and the author’s screen name, username and biographical statement were included in the dataset. A systematically sampled 10% subset of the data was subjected to qualitative analysis, involving categorisation on content. All tweets assigned to the category of medical resource were assessed for clinical accuracy. The engagement ratio for each category was calculated as (retweets + likes):tweets. Results In 2019, 111,388 tweets mentioning syphilis were posted by 69,921 authors. The most frequent content category – totalling 5370 tweets (48%) – was a joke. Of 1762 tweets (16%) categorised as a medical resource, 1484 (84%) were medically correct and 240 (14%) were medically incorrect; for 38 (2%), medical accuracy could not be judged from the information posted. Tweets categorised as personal experiences had the highest engagement ratio at approximately 19:1. Medical resource tweets had an engagement ratio of approximately 7:1. Conclusions We found medical information about syphilis was limited on Twitter. As tweets about personal experiences generate high engagement, coupling an experience with information may provide opportunity for public health education.