Your browser doesn't support javascript.
Using Twitter Data to Estimate Prevalence of Mental Disorder Symptoms in the United States During the COVID-19 Pandemic: Ecological Cohort Study.
Cai, Ruilie; Zhang, Jiajia; Li, Zhenlong; Zeng, Chengbo; Qiao, Shan; Li, Xiaoming.
  • Cai R; Department of Epidemiology and Biostatistics, Arnold School of Public Health, University of South Carolina, 921 Assembly Street, Columbia, US.
  • Zhang J; Department of Epidemiology and Biostatistics, Arnold School of Public Health, University of South Carolina, 921 Assembly Street, Columbia, US.
  • Li Z; South Carolina SmartState Center for Healthcare Quality, Arnold School of Public Health, University of South Carolina, Columbia, US.
  • Zeng C; University of South Carolina Big Data Health Science Center, University of South Carolina Big Data Health Science Center, US.
  • Qiao S; South Carolina SmartState Center for Healthcare Quality, Arnold School of Public Health, University of South Carolina, Columbia, US.
  • Li X; University of South Carolina Big Data Health Science Center, University of South Carolina Big Data Health Science Center, US.
JMIR Form Res ; 2022 Nov 30.
Article in English | MEDLINE | ID: covidwho-2198060
ABSTRACT

BACKGROUND:

Existing research and national surveillance data suggested an increase of the prevalence of mental disorders during the coronavirus disease 2019 (COVID-19) pandemic. Social media, such as Twitter, could be a source of data for estimation due to its real-time nature, high availability, and large geographical coverage. However, there is a dearth of studies validating the accuracy of Twitter-based prevalence for mental disorders through the comparison with CDC-reported prevalence.

OBJECTIVE:

This study aims to verify the feasibility of Twitter-based prevalence for mental disorders symptoms being an instrument for prevalence estimation, where the feasibility is gauged via the correlations between Twitter-based prevalence of mental disorder symptoms (i.e., anxiety and depressive symptoms) and the one based on national surveillance data. In addition, this study aims to identify how the correlations changed over time (i.e., the temporal trend).

METHODS:

State-level prevalence of anxiety and depressive symptoms were retrieved from the National Household Pulse Survey (HPS) through the Centers for Disease Control and Prevention (CDC) from April 2020 to July 2021. Tweets were retrieved from the Twitter streaming API during the same period and used to estimate the prevalence of mental disorder symptoms for each state using keyword analysis. Stratified linear mixed models were employed to evaluate the correlations between the Twitter-based prevalence of mental disorder symptoms and those reported by the CDC. The magnitude and significance of model parameters were used to evaluate the correlations. Temporal trends of correlations were tested after adding the time variable to the model. Geospatial differences were compared based on random effects.

RESULTS:

The Pearson correlations between the overall prevalence based on CDC and Twitter for anxiety and depressive symptoms were 0.587 (P<.001) and 0.368 (P<.001), respectively. Stratified by four phases (i.e., April 2020, August 2020, October 2020, and April 2021) defined by HPS, linear mixed models showed that Twitter-based prevalence for anxiety symptoms had a positive and significant correlation with CDC-reported prevalence in phases 2 and 3 while a significant correlation for depressive symptoms was identified in phases 1 and 3.

CONCLUSIONS:

Positive correlations are identified between Twitter-based and CDC-reported prevalence, and temporal trends of these correlations were found. Geospatial differences in the prevalence of mental disorder symptoms were found between the northern and southern U.S. Findings from this study could inform the future investigation on leveraging social media platforms to estimate mental disorder symptoms and the provision of immediate prevention measures to improve health outcomes.

Full text: Available Collection: International databases Database: MEDLINE Type of study: Cohort study / Experimental Studies / Observational study / Prognostic study / Randomized controlled trials Language: English Year: 2022 Document Type: Article Affiliation country: 37582

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: International databases Database: MEDLINE Type of study: Cohort study / Experimental Studies / Observational study / Prognostic study / Randomized controlled trials Language: English Year: 2022 Document Type: Article Affiliation country: 37582