They do mention it in passing, but there is no way to statistically control for that without a randomised controlled trial (or Mendelian Randomisation). There

They do mention it in passing, but there is no way to statistically control for that without a randomised controlled trial (or Mendelian Randomisation). There is no reason to think that working outside hours is the cause of the problem. It could be a consequence of an individual struggling outside of work that then leads to having to try to catch up.