This does not seem to change the design of the study. Frequency is still the better measurement.

I still can't comment on the "clinical significance" with available info. It will affect whether your protocol will be approved. Regardless, given your years' effort into, this, you should go ahead apply for a trial.

The following questions and statements you may want to think about. If it doesn't make you feel comfortable, just ignore. I have thought over and felt I should pass it to you.

The goal of trial is to validate what you know about the intervention in human subjects. Clincial trial is not a tool to gain new insight or knowledge. When you design a trial, assume you are a doctor and you are treating patients. YOU ARE NOT EXPERIMENTING. This is important. You should have a complete therapeutic protocol, just not certain its efficacy. To your best knowledge, this is how the procedure will be carried out in your practice. for example, antibiotics A to treat urinary tract infection. your description of the intervention will look like:

Antibiotics A 500 mg orally twice a day for a week.

Do you understand vaccination? Most of the vaccinations require a boost. once boosted, it will maintain its effect for a long period of time even lifelong. this is different from repeat where you have to take/get it at regular intervals, e.g. once a day, etc. What is your situation? repeat or boost?

in your case: Who will benefit from the repeat/boost treatment? Those who did not respond to treatment initially(50%) (Will these people ever?), or those whose response subsided later on (30%)? If repeat is needed, have you established the optimal interval? Is there a sustainable effect in some (20%), meaning these group only need one treatment in their life (therefore you need longer follow-up)?

This procedure is effective in preventing 50% patients from having an attack for the duration of xx. For those who have demonstrated effect, 30% will lose the effectiveness in a year, 20% of the patients will have sustained effects beyond one year for xxx. For those who did not respond to the initial treatment will not respond to any future treatment either and deemed non-candidate for this procedure. For those, whose effect has subsided will benenfit from repeat treatment at one-year time....all these details

Or, for those who did respond to the initial treatment, may still respond to repeat treatment and should have a repeat procedure in xx months...

Or, the procedure should be given on a yearly basis to those who have demonstrated the effect.....

Again,I am just all guessing..