Table 1

During the 2014–2015 outbreak, were these conditions consistent (√) or inconsistent (–) with placebo individually randomised controlled trials (placebo iRCT), active iRCT and stepped-wedge/ring-vaccination designs of efficacy studies for candidate Ebola vaccines?

	Placebo iRCT	Active iRCT	Stepped-wedge or ring-vaccination trial
1. General
The study intervention is needed urgently for many people.	√	√	√
b. No study assignment involves a procedure that is expected to actively harm any participant more than it directly benefits her.	√	√	√
No intervention for the condition studied is approved (and available to candidate participants).	√	√	√
2. Participant access to study intervention
All participants receive the study intervention during the trial—except ones who die or get infected before the intervention reaches them.	–	–	√
b. All participants receive the study intervention during the trial or shortly thereafter—except ones who die or get infected before the intervention reaches them.	√	√	√
c. For any prespecified level of statistical power to detect a given effect size, typically the number of participants who die or get infected before the study intervention reaches them is minimal.	√	√	–
3. Impact on participants' prospects
Participation improves medical prospects for each participant (both before and immediately after randomisation) more than any alternative available outside the trial.	√	√	√
b. Participation improves medical prospects for each participant (both before and immediately after randomisation) more than any alternative, including even alternatives that compromise trial efficiency or validity.	–	–	–
c. Participation improves medical prospects for each participant (both before and immediately after randomisation), making their prospects better than those of relevant non-participants.	√	√	√
d. Equipoise (understood here as equal medical prospects in all arms immediately after randomisation) obtains.	–	–	–
e. In no trial arm do the typical participants have grim medical prospects (immediately after randomisation).	√	√	√

For simplicity, the table assumes trials with two equally sized arms.