Six years ago, Boston-based computer whiz-kids and doctors came up with a seemingly smart pandemic-fighting idea. With Ebola spreading in west Africa, they proposed using digital data to monitor and predict the spread of the disease.

A warning system based on mining thousands of web-based data clues, the HealthMap platform was hailed in the media as a fantastic innovation. “How this algorithm detected the Ebola outbreak before humans could,” said one headline at the time.

But according to Susan Erikson, a Canadian aid worker and medical anthropologist who was then based in Sierra Leone, Big Data’s success wasn’t quite so clear-cut. Data scientists based in the west assumed that a disease such as Ebola behaved like malaria and was spread as humans moved; they also presumed that if mobile phones were on the move, their owners were travelling as well.