Fleeing Patients Complicate Uganda Ebola Control
Recombinomics Commentary
December 7, 2007
Twenty two people have so far died of the fever.
"Cumulatively, we have got 101 cases of Ebola -- those who fit the case definition," Kabwa said, adding that all were in western Uganda's Bundibugyo district, except for two in Kampala, including a doctor who died.
He said 39 had been admitted to hospital in serious condition and 350 others suspected of having been exposed to the virus were being urged to stay at home.
"They are being observed because they are possible contacts with Ebola cases," Kabwa said. "They are not being confined."
The outbreak, which started in August, has sparked panic amongst officials, health workers and the public.
"A bigger Ebola bomb could explode, claiming many more lives," Bundibugyo district chairman Jackson Bambalira was quoted as saying in the state-owned New Vision daily on Friday.
The above comments provide an update on the Ebola cases in Uganda. However, these numbers are incomplete and easily misinterpreted by the popular press. Some media reports have cited a 22% case fatality rate based on the 22 deaths out of 101 diagnosed patients. However, this calculation is incorrect. It assumes that all cases that have not died will not die, and there is no data to support such an assumption. There is no cure for Ebola, so many of the recently diagnosed patients will not recover.
The above comments cite 39 patients admitted in serious condition. However, the report does not say how many diagnosed patients have fled. As the hospitals empty because of patients requesting discharge and staff absenteeism, the number of diagnosed patients who are admitted also declines because patients fear infections from the hospitals. This fear is based in part on fatal infections in health care workers. Thus, many of the diagnosed patients have no outcome (death or discharge), and have instead chosen not to be admitted. Moreover, new suspect cases are being identified across southern Uganda (see satellite map), which creates additional tracking problems.
Therefore, the case fatality rate will be a very soft number until more outcomes are known, and these outcomes may not include fleeing patients, or contacts of fleeing patients.
Without details of outcomes, publication of case fatality rates is misleading.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings