3 ms·

> Notably, though, diabetes and high blood pressure were substantially more frequent in the control group How did this happen? Just bad luck with randomization

• zamfi 6y ago
  That's the implication from the paper.
• stjohnswarts 6y ago
  ack those could definitely affect this, isn't that why they normally try to select otherwise healthy patients as a baseline? Bring in the unhealthy people for the big studies and they'll most likely be spread much more evenly