Trial adherence in an oral trial with strict administration requirements is genuinely worse than trial-published data often suggests. That is why the exposure variability in oral formulations is mentioned repeatedly in the discussions here.
Take the reasoning and check the arithmetic; I do not always get it right.