Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : Social media platforms (SMPs) are frequently used by various pharmaceutical companies, public health agencies, and NGOs for communicating health concerns, new advancements, and potential outbreaks. While the benefits of using them as a tool have been extensively discussed, the online activity of various healthcare organizations on SMPs during COVID-19 in terms of engagement and sentiment forecasting has not been thoroughly investigated.

OBJECTIVE : The purpose of this research is to analyze the nature of information shared on Twitter, understand the public engagement generated on it, and forecast the sentiment score for various organizations.

METHODS : Data was collected from the Twitter handles of five pharmaceutical companies, ten U.S. and Canadian public health agencies, and World Health Organization (WHO) between January 01, 2017 - December 31, 2021. A total of 181,469 tweets were divided into two phases for the analysis: before COVID-19 and during COVID-19, based on the confirmation of the first COVID-19 community transmission case in North America on February 26, 2020. We conducted content analysis to generate health-related topics using Natural Language Processing (NLP) based topic modeling techniques, analyzed public engagement on Twitter, and performed sentiment forecasting using 16 univariate moving-average and machine learning (ML) models to understand the correlation between public opinion and tweet contents.

RESULTS : We utilized the topics modeled from the tweets authored by the health organizations chosen for our analysis using Non-Negative Matrix Factorization (NMF) ('c_umass' scores: -3.6530 and -3.7944, before COVID-19 and during COVID-19 respectively). The topics are - 'Chronic Diseases', 'Health Research', 'Community Healthcare', 'Medical Trials', 'COVID-19', 'Vaccination', 'Nutrition and Well-being', and 'Mental Health'. In terms of user impact, WHO (user impact: 4171.24) had the highest impact overall, followed by the public health agencies, CDC (user impact: 2895.87), and NIH (user impact: 891.06). Among pharmaceutical companies, Pfizer's user impact was the highest at 97.79. Furthermore, for sentiment forecasting, ARIMA and SARIMAX models performed best on the majority of the subsets of data (divided as per the health organization and time-period), with Mean Absolute Error (MAE) between 0.027 - 0.084, Mean Squared Error (MSE) between 0.001 - 0.011, and Root Mean Squared Error (RMSE) between 0.031 - 0.105.

CONCLUSIONS : Our findings indicate that people engage more on topics like 'COVID-19' than 'Medical Trials', 'Customer Experience'. Also, there are notable differences in the user engagement levels across organizations. Global organizations, like WHO, show wide variations in engagement levels over time. The sentiment forecasting method discussed presents a way for organizations to structure their future content to ensure maximum user engagement.


Singhal Aditya, Baxi Manmeet Kaur, Mago Vijay