Hard News by Russell Brown

I got cross and stopped reading at the "we didn't include women becoz omg periodz" bit. Maybe I'll come back and read the rest later. Maybe I won't, because ffs.
If menstrual variations are a concern, there are plenty of women in the 25-65 age bracket (particularly at the upper end, but also some younger ones for various reasons) who don't menstruate.

Really? It’s the first stage of a study in which they’re not yet sure they’ll see an effect at all. It surely makes sense to exclude any potential confounds. And they haven’t said “omg periodz” – one of the study team has a PhD thesis in relevant aspects of the menstrual cycle. If this first stage produces results, the study group will expand. Fwiw, I'm excluded too, for a reason it would mess things up for them to explain.

Get them back next year! Things moving it seems, and good to know NZ is involved.

Yes really. If they're worried about menstrual cycles, why is the default men rather than non-menstruating women?
I don't have an issue with their logic on the menstrual cycle affecting results. I have an issue with excluding women on the basis of menstruation when plenty of women in their target age-group don't menstruate.
If there is no effect, will they then test on women in some way, or just assume that because there's no effect for men there's none for women either?

I see this as a sort of pre-study proof-of-concept. When you're designing the initial stage of a six week course in which you're not sure you'll even detect measurable results, it makes sense to choose candidates most likely to hopefully register on the scale. Not as any sort of representative sample, but to prove that the results of larger scale testing will be measureable. No?

The point being that in a much larger sample and/or a longer timeframe, hormonal cycle effects should reduce to random noise, but there isn't yet the evidence to justify that sample size?
Even so, this is not ideal. If the effects are generally weaker in men than in women (which is plausible simply on the basis of dose per body mass) a male-only pilot study could well conclude, wrongly, that there is nothing to see.

When you’re designing the initial stage of a six week course in which you’re not sure you’ll even detect measurable results, it makes sense to choose candidates most likely to hopefully register on the scale.

Sure - but they make no mention of any reason to suspect that men would be more likely to register on the scale than women.

If the effects are generally weaker in men than in women (which is plausible simply on the basis of dose per body mass) a male-only pilot study could well conclude, wrongly, that there is nothing to see.

Quite - or even if the dose is done on a per kg basis (which I'd kind of expect), there could be some interaction with testosterone (since their previous research has demonstrated that hormones make a difference) that would be different in women because we have less of it.

I caught up with Suresh last week and he has just finished a funding proposal for a follow-up study involving only women. He was still quite clear that it was going to be too difficult to design the first one for both men and women.

I would think the cohort for that will be similar to the first one: healthy grad students in their late 20s.