Are adaptive randomised trials or non-randomised studies the best way to address the Ebola outbreak in west Africa?