Using Social Media to Help Understand Patient-Reported Health Outcomes of Post-COVID-19 Condition: Natural Language Processing Approach

• Published in: Journal of medical Internet research 2023-09, Vol.25 (2), p.e45767-e45767
• Main Authors: Dolatabadi, Elham, Moyano, Diana, Bales, Michael, Spasojevic, Sofija, Bhambhoria, Rohan, Bhatti, Junaid, Debnath, Shyamolima, Hoell, Nicholas, Li, Xin, Leng, Celine, Nanda, Sasha, Saab, Jad, Sahak, Esmat, Sie, Fanny, Uppal, Sara, Vadlamudi, Nirma Khatri, Vladimirova, Antoaneta, Yakimovich, Artur, Yang, Xiaoxue, Kocak, Sedef Akinli, Cheung, Angela M
• Format: Article
• Language: English
• Subjects: Application programming interface; Bidirectionality; Caregivers; Clinical outcomes; Comorbidity; Computational linguistics; COVID-19; Data collection; Datasets; Evaluation; Fatigue; Headaches; Health care industry; Health status; Humans; Infections; Internet; Language; Language processing; Long COVID; Medical personnel; Medical research; Medicine, Experimental; Multimedia; Natural language interfaces; Natural Language Processing; Normalization; Original Paper; Patient Reported Outcome Measures; Patients; Personal health; Proximity; Public health; Reliability; Reproducibility of Results; Severe acute respiratory syndrome coronavirus 2; Social Media; Social networks; Spatial analysis; Surveillance; Symptoms; Terminology; Uncertainty
• ISSN: 1438-8871; 1439-4456; 1438-8871
• DOI: 10.2196/45767

While scientific knowledge of post-COVID-19 condition (PCC) is growing, there remains significant uncertainty in the definition of the disease, its expected clinical course, and its impact on daily functioning. Social media platforms can generate valuable insights into patient-reported health outcomes as the content is produced at high resolution by patients and caregivers, representing experiences that may be unavailable to most clinicians. In this study, we aimed to determine the validity and effectiveness of advanced natural language processing approaches built to derive insight into PCC-related patient-reported health outcomes from social media platforms Twitter and Reddit. We extracted PCC-related terms, including symptoms and conditions, and measured their occurrence frequency. We compared the outputs with human annotations and clinical outcomes and tracked symptom and condition term occurrences over time and locations to explore the pipeline's potential as a surveillance tool. We used bidirectional encoder representations from transformers (BERT) models to extract and normalize PCC symptom and condition terms from English posts on Twitter and Reddit. We compared 2 named entity recognition models and implemented a 2-step normalization task to map extracted terms to unique concepts in standardized terminology. The normalization steps were done using a semantic search approach with BERT biencoders. We evaluated the effectiveness of BERT models in extracting the terms using a human-annotated corpus and a proximity-based score. We also compared the validity and reliability of the extracted and normalized terms to a web-based survey with more than 3000 participants from several countries. UmlsBERT-Clinical had the highest accuracy in predicting entities closest to those extracted by human annotators. Based on our findings, the top 3 most commonly occurring groups of PCC symptom and condition terms were systemic (such as fatigue), neuropsychiatric (such as anxiety and brain fog), and respiratory (such as shortness of breath). In addition, we also found novel symptom and condition terms that had not been categorized in previous studies, such as infection and pain. Regarding the co-occurring symptoms, the pair of fatigue and headaches was among the most co-occurring term pairs across both platforms. Based on the temporal analysis, the neuropsychiatric terms were the most prevalent, followed by the systemic category, on both social media platforms. Our spatia