Ebola Outbreaks May Often Go Undetected

July 20, 2019 NEW YORK (Reuters Health) - At least half of all Ebola-virus-disease spillover events and small outbreaks go undetected, researchers from the U.K. estimate. "Ebola spillover is extremely rare, and we can't fault clinics with limited resources for misdiagnosing or failing to detect it," Emma E. Glennon of the University of Cambridge told Reuters Health by email. "However, we can still build up infrastructure and ensure strong infection-control practices are universally in place to prepare for when that unlikely event occurs," she said. "When we are consistently controlling nosocomial transmission of all infections, we are better able to prevent even undetected outbreaks." Most reports of Ebola have been large outbreaks, but given the low reported rate of transmission of the disease and the high frequency with which cases infect no one else, most outbreaks could be very small (<5 people). Glennon and colleagues used three distinct data sets from the 2013-2016 outbreak in West Africa and the properties of person-to-person Ebola transmission to estimate the likely true distribution of outbreak sizes and the number of small outbreaks and spillover events (from wildlife to people) that are likely to have gone undetected since 1976. According to their geometric model, the median individual probability of detection was 8.4% per case irrespective of cluster size. Their logistic model predicted that an isolated case has a median 2.4% probability of detection, while outbreaks of sizes greater than 10 approach 100% detection. The overall probabilities of detection since Ebola was first reported were estimated at 26.4% overall, 48.0% for Sierra Leone and 16.8% for Guinea, the researchers write in PLoS Neglected Tropical Diseases, online June 13. The estimated detection probabilities of dead-end zoonotic spillovers causing a single human case ranged from 0.1% to 6.7% overall, from 6.0% to 12.5% for Sierra Leone and from 1.2% to 3.0% for Guinea. These models suggest that there could have been well over 100 cases that went undetected (from a minimum of 16 from the Sierra Leone data to a maximum of 317 from the full outbreak data). "It is essential to build up public health and primary healthcare infrastructure in the regions where Ebola spillover occurs," Glennon said. "No matter how well funded an outbreak response team is or how many vaccines we have stockpiled, a national or international response team can't stop an outbreak no one knows about." "Late detection means these sorts of teams will be trying to catch up with an outbreak from their first day on the ground," she said. "We could save lives, money, and time by noticing outbreaks early enough to contain them locally with relatively simple infection control measures - or by implementing those measures consistently enough to prevent transmission even before identifying the true etiology."