Aims, Objectives and Background The purpose of this study is to unpack the cognitive dimensions of learning and the subjective internalisation that resulted from Twitter activity by investigating the following research questions: 1) What types of information were shared on Twitter pre and post-publication of the CRASH-3 trial? and 2) Did this information signify knowledge translation? This topic is important because translation of medical research into medical practice can take up to 20 years and Twitter-aided knowledge translation has the potential to shorten this. This study is the first to analyse tweets thematically through two methods, including Bloom’s Digital Taxonomy (BDT), bringing the realities of Twitter users to the foreground.

Method and Design Pre-publication tweets (n=92) and post-publication tweets (n=742) were analysed using 1) Braun & Clarke’s six-step thematic analysis framework and 2) BDT. The highest-order thinking skill (HOTS), during BDT analysis, was assigned following a consensus meeting between two independent coders.

Results and Conclusion Eight overarching themes emerging from the pre-publication phase: emotion and feeling (90.21%), hashtagging (40.21%), tagging (26.09%), education-related information (10.87%), research-related information (9.78%), conference (7.61%), statement (3.26%) and poll (2.17%). 16 overarching themes emerged from the post-publication phase: hashtagging (56.06%), tagging (36.79%), article posting (23.05%), emotion and feeling (21.83%), education-related information (19.54%), summarising (14.42%), notification of results (9.57%), media outlook (7.01%), conference (6.74%), commenting (6.06%), open questions (5.26%), judging (4.99%), research-related information (4.04%), recommending (1.75%), quoting (1.48%) and comparing trials (0.40%). Some tweets applied to more than one category of themes.

There was an increase in HOTS during the post-publication„ phase, signifying an increase in the cognitive dimensions of learning and the subjective internalisation of information. This was likely due to the increase in social media activity and information sharing following the release of trial outcomes. These findings support the role of „Twitter, through the social capital model, in facilitating higher-order learning.