Media Influence on Anxiety, Health Utility, and Health Beliefs Early in the SARS-CoV-2 Pandemic-a Survey Study

Matthew Greenhawt, Spencer Kimball, Audrey DunnGalvin, Elissa M Abrams, Marcus S Shaker, Giselle Mosnaim, Pasquale Comberiati, Nikita A Nekliudov, Oleg Blyuss, Martin Teufel, Daniel Munblit

Research output: Contribution to journalArticlepeer-review

17 Downloads (Pure)


Background: The psychological effects from the COVID-19 pandemic and response are poorly understood. Objective: To understand the effects of the pandemic and response on anxiety and health utility in a nationally representative sample of US adults. Design: A de-identified, cross-sectional survey was administered at the end of April 2020. Probability weights were assigned using estimates from the 2018 American Community Survey and Integrated Public Use Microdata Series Estimates. Participants: US adults 18–85 years of age with landline, texting-enabled cellphone, or internet access. Intervention: Seven split-half survey blocks of 30 questions, assessing demographics, COVID-19-related health attitudes, and standardized measures of generalized self-efficacy, anxiety, depression, personality, and generic health utility. Main Measures: State/Trait anxiety scores, EQ-5D-3L Visual Analog Scale (VAS) score, and demographic predictors of these scores. Key Results: Among 4855 respondents, 56.7% checked COVID-19-related news several times daily, and 84.4% at least once daily. Only 65.7% desired SARS-CoV-2 vaccination for themselves, and 70.1% for their child. Mean state anxiety (S-anxiety) score was significantly higher than mean trait anxiety (T-anxiety) score (44.9, 95%CI 43.5–46.3 vs. 41.6, 95%CI 38.7–44.5; p = 0.03), with both scores significantly higher than previously published norms. In an adjusted regression model, less frequent news viewing was associated with significantly lower S-anxiety score. Mean EQ-5D-3L VAS score for the population was significantly lower vs. established US normative data (71.4 CI 67.4–75.5, std. error 2 vs. societal mean 80, std. error 0.1; p < 0.001). EQ-5D-3L VAS score was bimodal (highest with hourly and no viewing) and significantly reduced with less media viewership in an adjusted model. Conclusions: Among a nationally representative sample, there were higher S-anxiety and lower EQ-5D-3L VAS scores compared to non-pandemic normative data, indicative of a potential detrimental acute effect of the pandemic. More frequent daily media viewership was significantly associated with higher S-anxiety but also predictive of higher health utility, as measured by EQ-5D-3L VAS scores.

Original languageEnglish
Pages (from-to)1327-1337
Number of pages11
JournalJournal of General Internal Medicine
Issue number5
Publication statusPublished - 24 Feb 2021


  • COVID-19
  • EQ-5D-3L
  • SARS-CoV-2
  • State-Trait Anxiety Inventory
  • anxiety
  • health utility
  • media consumption
  • social media
  • state anxiety
  • trait anxiety
  • vaccine hesitancy


Dive into the research topics of 'Media Influence on Anxiety, Health Utility, and Health Beliefs Early in the SARS-CoV-2 Pandemic-a Survey Study'. Together they form a unique fingerprint.

Cite this