I think bhouston is arguing (correctly) you can’t have a 10% of the population drop in dementia prevalence by an intervention that only targets 3% of the popula

I think bhouston is arguing (correctly) you can’t have a 10% of the population drop in dementia prevalence by an intervention that only targets 3% of the population, so even if CPAPs contribute, that does not explain most of the drop. (If everybody who uses CPAPs would get dementia, and they are 100% effective at preventing that, the drop would still be ‘only’ 3% of the population)