This chart in the article makes me suspicious: https://assets.bwbx.io/images/users/iqjWHBFdfxIU/i_t8UFPOJVkA/v1/800x-1.png https://assets.bwbx.io/images/users/

This chart in the article makes me suspicious: https://assets.bwbx.io/images/users/iqjWHBFdfxIU/i_t8UFPOJVkA/v1/800x-1.png https://assets.bwbx.io/images/users/iqjWHBFdfxIU/i_t8UFPOJVk... The nurses who took 6 hour workdays took roughly the same number of sick days two years later as they did at the start. Technically they took 5% fewer days, and perhaps that's statistically significant, but it's not (for lack of a better word) interesting. The control group is interesting! Why did a group of nurses taking a normal schedule suddenly have such a huge increase in sick days? I get the feeling that something else affected these results that we're not seeing here. I'm skeptical that the 6-hour workday explains this. Can anyone with more experience in the medical industry chime in to say if it's normal for the number of sick days a nurse takes to increase that much over 2 years (for an 8-hour workday)? People on hackernews are always skeptical and I usually hate that. I'd like to start a trend of being appreciative. This study took two years to complete! It's hard to foresee things happening that could make the data look strange like this. Maybe they should have taken more information, so they could see which trait in the control group had the highest correlation with increased sick days. But I can't say with certainty that I would have thought to do that. Interesting study, at the very least it's an indicator that there's something interesting going on here worth investigating :).