Background: COVID-19 cases first emerged in China in December 2019. An air of fear and anxiety engulfed the globe as it rapidly spread across borders affecting the entire human race. Media outlets started reporting information containing rising death tolls and horrific visuals generating a global hysteria(1). Stress is a well-known trigger for acute coronary syndromes (2) and stroke however, peculiarly enough, by March 2020, hospitals around the world reported a 20-80% drop in admissions for acute myocardial infarction (AMI), urgent coronary angiography (3,4) and stroke (5). This paradox was also observed by physicians in St Louis, Missouri after the first case of COVID-19 on March 7, 2020. Our local “infodemic” also fueled nervousness whereby patients requiring even emergent care for life-threatening conditions like stroke and AMI avoided the hospitals (6). To ascertain any relationship between local media coverage of COVID-19 and hospital admissions of AMI and stroke, we performed a retrospective time-series analysis using hospital records from St Louis metropolitan area.
Methods: All nine acute care hospitals in St Louis City and County (covering a population 1,294,781) were queried for weekly admissions of AMI (ICD-10 codes I21.0-I21.4) and stroke (ICD-10 codes I63.341-I63.543, I66.3, I67.82) from December through May of 2016-2019 (control years) and 2019-2020 (observation year). Total mortality for those time periods were supplied by the Missouri Department of Health and Human Services (Jefferson City, MO). Emergency Medical Services (EMS) calls for in-home cardiac arrest were obtained from St Louis City EMS. The percentage of front-page news articles regarding COVID-19 in the St Louis Post Dispatch were obtained from Newspapers.com, while the percentage of COVID-19 television news coverage was calculated from http://www.KSDK.com/news. Time-series causality analysis was done using the cross-correlation function and Pearson’s correlation utilizing R statistical software (R Foundation for Statistical Computing Vienna, Austria http://www.R-project.org).
Results: AMI admissions abruptly decreased by a maximum of 37% in the second week of April 2020 compared to the average of the same week in 2016-2019 (p< 0.001). This decline is highly correlated with the spike in media coverage (Pearson’s r= -0.91, r2 = 0.95 p< 0.001) at a lag of zero. Similarly, stroke cases in 2019-2020 started declining by mid-March with a maximum of 42% drop by last week of April. This correlated with a rise in media coverage for the same period (Pearson’s r= -0.87, r2 = -0.93, p< 0.001) at a lag of zero. Weekly all-cause mortality increased to a maximum of 49% in April 2020 (p< 0.05). EMS calls for in home cardiac arrest were unchanged.
Conclusions: There was a statistically significant decline in AMI and stroke admissions that highly correlated with an increase in local media coverage suggesting possible association of news-media-induced avoidance of medical facilities, due to fear of COVID-19 exposure.