Ashley DePriest, LD, MS, RD, CNSC
Clinical Dietitian
Wellstar Health System
Marietta, Georgia
Disclosure information not submitted.
Megan Van Berkel Patel, BCCCP, PharmD, FCCM
Clinical Pharmacy Specialist
n/a
Disclosure information not submitted.
Tabitha Brown, PharmD
PGY-2 Critical Care Pharmacy Resident
Ballad Health Johnson City Medical Center, United States
Disclosure information not submitted.
Title: Factors contributing to a successful Twitter journal club: an analysis
Introduction: Use of Twitter for critical care education has become an increasingly popular tool for healthcare information sharing. Twitter journal clubs (TJC) are one method of disseminating education and facilitating discussion across professions. Despite their widespread use, the optimal factors influencing the reach of a specific journal club are yet to be elucidated.
Methods: TJCs hosted by the Southeast Chapter of the Society of Critical Care Medicine and conducted between January 2020 and June 2021 were included. TJCs were learner-led and open to any profession. The primary outcome was to correlate days between article publication to TJC date and total impressions. Secondary outcomes were to determine if time of day influenced number of impressions, number of contributors and number of professional disciplines involved. Additional outcomes were to correlate total impressions with number of tweets per TJC and number of contributors. Impressions were assessed for one week after each TJC date with Symplur analytics using each chapter's TJC hashtag. Correlation outcomes were assessed using a Pearson Correlation, and other variables with a Mann-Whitney U test.
Results: A total of 4 TJCs were included for analysis. At the time of analysis, the chapter had 936 followers and had been active on Twitter for 7.2 years. The median days from article publication to TJC were 145 with an interquartile range (IQR) of 117-273. Two TJCs were held during midday (1100-1400 EST) and two were held in the afternoon (1401-1700 EST). All article topics were concerning pharmacotherapy. The median impressions per TJC was 40551 (IQR 33183-59236). There was no correlation in days between article publication to TJC and total impressions (r= -0.030, p= 0.970). There were no differences in the time of day and impressions (p=0.333), individual contributors (p=1.0), or number of professional disciplines that participated (p=0.667). Finally, we did not find a correlation with total impressions and either number of tweets (r=0.242, p=0.758), or individual contributors (r=0.504, p=0.496).
Conclusions: In this analysis of a single host TJC, we did not find a particular TJC characteristic that influenced the number of impressions. However, the findings are limited by the small sample size.