Identification of Risk Factors and Symptoms of COVID-19: Analysis of Biomedical Literature and Social Media Data

Jeon, Jouhyun; Baruah, Gaurav; Sarabadani, Sarah; Palanica, Adam

doi:10.2196/20509

Cited by 31 publications

(33 citation statements)

References 19 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors identified 25 novel symptoms through social media posts, specifically Twitter posts. Some of the less common symptoms identified in that work align with the less-documented symptoms reported by participants in this study such as ear and eye problems, weight loss, and memory disorder, however, results in this study also include symptoms that were not identified by Jeon et al [32] such as worsened prior condition (e.g., asthma) and elevated blood pressure.…”

Section: Discussionsupporting

confidence: 69%

See 1 more Smart Citation

Learning from the Experiences of COVID-19 Survivors: A Descriptive Study

Prioleau

2021

Preprint

View full text Add to dashboard Cite

BackgroundThere are still many unanswered questions about the novel coronavirus, however, a largely underutilized source of knowledge are the millions of people who have recovered after contracting the virus. This includes majority of undocumented cases of the COVID-19 which are classified as mild or moderate and received little to no clinical care during the course of illness.ObjectiveTo document and glean insights from the experiences of persons with first-hand experience with coronavirus, especially the so-called mild to moderate cases that self-resolved in isolation.MethodsThis online-based survey study called C19 Insider Scoop recruited adult participants that are 18-years or older who reside in the United States and tested positive for COVID-19 or antibodies. Participants were recruited through various methods including online support groups for COVID-19 survivors, advertisement in local news outlets, and advertisement through professional and other networks. The main outcomes measured include knowledge on contraction/transmission of the virus, symptoms, and personal experiences on road to recovery.ResultsA total of 72 participants (53 females/19 males, ages 18 - 73 yrs. old, mean = 41-yrs.) from 22 U.S. states participated in this study. We found that the top known source of how people contracted the COVID-19 virus was through a family or household member (n=26 or 35%). This was followed by essential workers contracting the virus through the workplace (n=13 or 18%).Participants reported up to 27 less-documented symptoms that they experienced during their illness such as brain/memory fog, palpitations, ear pain/discomfort and neurological problems. In addition, 47 out of 72 participants (65%) reported that their symptoms lasted longer than the commonly cited 2-weeks even for mild cases of COVID-19. In our study, the mean recovery time was 4.5-weeks, and exactly one-half of survivors (50%) still experienced lingering symptoms of COVID-19 after an average of 65-days following illness onset. Additionally, 37 participants (51%) reported that they experienced stigma associated with having COVID-19.ConclusionThis study presents preliminary findings which suggests that emphasis on family/household spread of COVID-19 may be lacking and there is a general underestimation of the recovery time even for mild cases of the virus. Although a larger study is needed to validate these results, it is important to note that as more people experience COVID-19, insights from prior survivors can enable a more informed public, pave the way for others who may be affected, and guide further research.

show abstract

Section: Discussionsupporting

confidence: 69%

“…They identified symptoms and trajectories that were indicative of need for hospitalization. More closely related to this work, Jeon et al [32] used a combination of biomedical literature and social media data to characterize symptoms of COVID-19. The authors identified 25 novel symptoms through social media posts, specifically Twitter posts.…”

Section: Discussionmentioning

confidence: 99%

Learning from the Experiences of COVID-19 Survivors: A Descriptive Study

Prioleau

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…While Twitter data has been used to identify self-reports of symptoms by people who have tested positive for COVID-19 [ 3 , 4 ], the shortage of available testing and the delay of test results in the United States motivated us to assess whether Twitter data could be scaled to identify potential cases of COVID-19 that are not based on testing and, thus, may not have been reported to the CDC. There are studies that have not limited their exploration of COVID-19 symptoms on Twitter to users who have tested positive for COVID-19 [ 5 - 8 ]; however, limiting the detection of potential cases to symptoms may still underutilize the information available on Twitter.…”

Section: Discussionmentioning

confidence: 99%

Toward Using Twitter for Tracking COVID-19: A Natural Language Processing Pipeline and Exploratory Data Set

Klein¹,

Magge²,

O’Connor³

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

Background In the United States, the rapidly evolving COVID-19 outbreak, the shortage of available testing, and the delay of test results present challenges for actively monitoring its spread based on testing alone. Objective The objective of this study was to develop, evaluate, and deploy an automatic natural language processing pipeline to collect user-generated Twitter data as a complementary resource for identifying potential cases of COVID-19 in the United States that are not based on testing and, thus, may not have been reported to the Centers for Disease Control and Prevention. Methods Beginning January 23, 2020, we collected English tweets from the Twitter Streaming application programming interface that mention keywords related to COVID-19. We applied handwritten regular expressions to identify tweets indicating that the user potentially has been exposed to COVID-19. We automatically filtered out “reported speech” (eg, quotations, news headlines) from the tweets that matched the regular expressions, and two annotators annotated a random sample of 8976 tweets that are geo-tagged or have profile location metadata, distinguishing tweets that self-report potential cases of COVID-19 from those that do not. We used the annotated tweets to train and evaluate deep neural network classifiers based on bidirectional encoder representations from transformers (BERT). Finally, we deployed the automatic pipeline on more than 85 million unlabeled tweets that were continuously collected between March 1 and August 21, 2020. Results Interannotator agreement, based on dual annotations for 3644 (41%) of the 8976 tweets, was 0.77 (Cohen κ). A deep neural network classifier, based on a BERT model that was pretrained on tweets related to COVID-19, achieved an F1-score of 0.76 (precision=0.76, recall=0.76) for detecting tweets that self-report potential cases of COVID-19. Upon deploying our automatic pipeline, we identified 13,714 tweets that self-report potential cases of COVID-19 and have US state–level geolocations. Conclusions We have made the 13,714 tweets identified in this study, along with each tweet’s time stamp and US state–level geolocation, publicly available to download. This data set presents the opportunity for future work to assess the utility of Twitter data as a complementary resource for tracking the spread of COVID-19.

show abstract

“…Some of the less-common symptoms identified in their work align with the less-documented symptoms reported by the participants of this study, such as ear and eye problems, weight loss, and memory disorder. However, this study’s findings also include symptoms that were not identified by Jeon et al [ 29 ], such as worsened pre-existing condition (eg, asthma) and elevated blood pressure.…”

Section: Discussionmentioning

confidence: 76%

“…They identified symptoms and trajectories that were indicative of the need for hospitalization. More closely related to this work, Jeon et al [ 29 ] used a combination of biomedical literature and social media data to characterize the symptoms of COVID-19. They identified 25 novel symptoms by analyzing social media posts, specifically Twitter posts.…”

Section: Discussionmentioning

confidence: 99%

Learning From the Experiences of COVID-19 Survivors: Web-Based Survey Study

Prioleau¹

2021

JMIR Form Res

View full text Add to dashboard Cite

Background There are still many unanswered questions about the novel coronavirus; however, a largely underutilized source of knowledge is the millions of people who have recovered after contracting the virus. This includes a majority of undocumented cases of COVID-19, which were classified as mild or moderate and received little to no clinical care during the course of illness. Objective This study aims to document and glean insights from the experiences of individuals with a first-hand experience in dealing with COVID-19, especially the so-called mild-to-moderate cases that self-resolved while in isolation. Methods This web-based survey study called C19 Insider Scoop recruited adult participants aged 18 years or older who reside in the United States and had tested positive for COVID-19 or antibodies. Participants were recruited through various methods, including online support groups for COVID-19 survivors, advertisement in local news outlets, as well as through professional and other networks. The main outcomes measured in this study included knowledge of contraction or transmission of the virus, symptoms, and personal experiences on the road to recovery. Results A total of 72 participants (female, n=53; male, n=19; age range: 18-73 years; mean age: 41 [SD 14] years) from 22 US states were enrolled in this study. The top known source of how people contracted SARS-CoV-2, the virus known to cause COVID-19, was through a family or household member (26/72, 35%). This was followed by essential workers contracting the virus through the workplace (13/72, 18%). Participants reported up to 27 less-documented symptoms that they experienced during their illness, such as brain or memory fog, palpitations, ear pain or discomfort, and neurological problems. In addition, 47 of 72 (65%) participants reported that their symptoms lasted longer than the commonly cited 2-week period even for mild cases of COVID-19. The mean recovery time of the study participants was 4.5 weeks, and exactly one-half of participants (50%) still experienced lingering symptoms of COVID-19 after an average of 65 days following illness onset. Additionally, 37 (51%) participants reported that they experienced stigma associated with contracting COVID-19. Conclusions This study presents preliminary findings suggesting that emphasis on family or household spread of COVID-19 may be lacking and that there is a general underestimation of the recovery time even for mild cases of illness with the virus. Although a larger study is needed to validate these results, it is important to note that as more people experience COVID-19, insights from COVID-19 survivors can enable a more informed public, pave the way for others who may be affected by the virus, and guide further research.

show abstract

Identification of Risk Factors and Symptoms of COVID-19: Analysis of Biomedical Literature and Social Media Data

Cited by 31 publications

References 19 publications

Learning from the Experiences of COVID-19 Survivors: A Descriptive Study

Learning from the Experiences of COVID-19 Survivors: A Descriptive Study

Toward Using Twitter for Tracking COVID-19: A Natural Language Processing Pipeline and Exploratory Data Set

Learning From the Experiences of COVID-19 Survivors: Web-Based Survey Study

Contact Info

Product

Resources

About