JMIR Publications

ABSTRACT

Background:

Early conversations on social media by emergency physicians offer a window into the ongoing response to the SARS-CoV-2 pandemic.

Objective:

This retrospective observational study of emergency physician Twitter use details how the healthcare crisis has influenced emergency physician discourse online, and may have use as a harbinger of ensuing surge.

Methods:

Followers of the three main emergency physician professional organizations were identified using Twitter’s application programming interface. They and their followers were included in the study if identifying explicitly as United States-based emergency physicians. Statuses or ‘tweets’ were obtained between January 4th, 2020, when the new disease was first reported, and December 14th, 2020, when vaccinations first began. Original tweets underwent sentiment analysis using the previously validated Valence Aware Dictionary and sEntiment Reasoner (VADER) tool as well as topic modeling using Latent Dirichlet Allocation unsupervised machine learning. Sentiment and topic trends were then correlated with daily change in new Covid-19 cases and inpatient bed utilization.

Results:

3,463 emergency physicians produced 334,747 unique English tweets during the study period. 910 (26.3%) stated that they were in training, and 446 (51.7%) of those who provided a gender identified as a man. Overall tweet volume went from a pre-March mean of 481.9 ±72.7 daily tweets to 1,065.5 ±257.3 daily thereafter. Parameter and topic number tuning led to 20 tweet topics, with a topic coherence of 0.49. Except for a week in June and four days in November, discourse was dominated by the healthcare system (45,570, 13.6%). Discussion of pandemic response, epidemiology, and clinical care were jointly found to moderately correlate with COVID-19 hospital bed utilization (Pearson’s r = 0.41), as was the occurrence of ‘covid’,‘coronavirus’, or ‘pandemic’ in tweet text (0.47). Momentum in COVID-19 tweets, as demonstrated by a sustained crossing of 7 and 28-day moving averages, was found to have occurred 45.0 ±12.7 days before peak COVID-19 hospital bed utilization across the country and four most contributory states.

Conclusions:

COVID-19 Twitter discussion among emergency physicians correlates with and may precede rising hospital burden. This study therefore begins to depict the extent to which the ongoing pandemic has affected the field of emergency medicine discourse online and suggests a potential avenue for improving understanding of the predictors of surge.