If the observed effect of an intervention does not rise above the background variance it means that the intervention does not do what you want it to do.
Either (1) it has no effect or (2) the strength and direction of the effect are random. Either of those qualities renders the intervention ineffective. You can’t justify giving a patient something that will have no effect, or will make them worse off half of the time when there are more effective options are on the table.
Of course this is comes with the caveat that the study is adequately powered to detect the strength of the effect you are looking for. That being said, there is plenty historical data for the placebo, so I don’t think that being underpowered because of misestimation of the background variance would be an issue.
Either (1) it has no effect or (2) the strength and direction of the effect are random. Either of those qualities renders the intervention ineffective. You can’t justify giving a patient something that will have no effect, or will make them worse off half of the time when there are more effective options are on the table.
Of course this is comes with the caveat that the study is adequately powered to detect the strength of the effect you are looking for. That being said, there is plenty historical data for the placebo, so I don’t think that being underpowered because of misestimation of the background variance would be an issue.