The first Ebola virus case on American soil was confirmed September 30, 2014, in a 45-year-old man. He had entered the country on September 20, 2014, from Liberia. Feeling ill, he visited a Dallas, Texas, hospital 5 days later where he was released but subsequently returned September 28 gravely ill. These events created unprecedented media attention and exacerbated fears of a widespread Ebola outbreak in the United States. A study using syndromic surveillance data was conducted to examine whether ED visits changed among metropolitan Dallas–Fort Worth (DFW), Texas, residents after the case was reported.
Methods: This study used the Texas Health Service Region 2/3 syndromic surveillance data and associated ESSENCE analytics through the North Texas Syndromic Surveillance System from July 21, 2013, to July 22, 2015. Focusing on the metropolitan DFW area, interrupted time series analyses were conducted. The analyses used segmented regression models with autoregressive errors of daily ED visits, overall and for several chief complaints, including fever and fever with gastrointestinal distress (GI Fever), and date that the first case of Ebola virus was confirmed (the “event”). Visits and visit rates by chief complaint before and after the event were analyzed, and absolute and relative effects of the event were estimated.
Results of analyses: Interrupted time series results (figure 1) indicated the event was highly significant for ED visits overall (p=0.0147) and for the rate of GI Fever visits (p<0.0001). Immediately after September 30, 2014, an increase was observed in total ED visits of 1,023 visits per day (95% CI: 797, 1,253) (table 1). This is an increase of 11.77% (95% CI: 9.22%, 14.36%) in daily ED visits overall. GI Fever visits increased by 40% (95% CI: 34.43%, 46.03%), but this amounted to an increase of about 2 to 3 visits per day.
Total ED visits remained significantly above baseline, even accounting for seasonal surge periods related to the cold and flu season, until January 22, 2015. The total impact of the Ebola Virus Disease (EVD) case on total ED visits from September 30, 2014, through January 22, 2015, was 95,690 (95% CI: 69,185, 116,202) excess visits. The total impact of the EVD case on GI Fever visits, which returned to baseline by March 20, 2015, was 2,151 (95% CI: 2055, 2247) excess visits attributable to the fatal EVD case.
Immediately after the EVD case was confirmed, ED visits in metropolitan DFW increased significantly—both within and outside the symptom profile for EVD. Health officials determined most increases were not due to EVD symptoms because the people seeking care had no fever, had not reported travel, and had not been exposed to the man confirmed to have EVD. ED visits remained elevated for some time even after adjusting for seasonality both within symptom-specific chief complaints and overall.
After the first case of Ebola Virus Disease was confirmed in Dallas–Fort Worth, Texas, syndromic surveillance data showed an increase in emergency department (ED) visits—including visits for symptoms that did not match the disease profile.
- Syndromic data are useful for monitoring upticks in ED visits and characterizing the disease. Syndromic data (similar symptoms grouped into categories) can be used to characterize a disease before labs submit reports that confirm the disease. Early notification can lead to specialized testing among at-risk populations. These data can help set apart the “worried well,” who may unnecessarily overwhelm an ED, from people with symptoms similar to (or vastly different from) symptoms associated with the disease.
- An uptick in ED visits by the “worried well” can inform those who communicate with targeted populations and who provide technical assistance to news media. Accurate, timely reporting can reduce public fears that lead to needless ED visits.
- Syndromic surveillance data improves situational awareness. Situational awareness is the term used to describe a heightened awareness of populations seeking healthcare services, their symptoms, and geographic areas with an uptick in healthcare services. Syndromic data can also be used to inform and monitor the effectiveness of interventions and communication strategies.