[alister]

Looking at the graph in the study, it looks like 0.972 fraction of the very regular sleepers are alive after 7.8 years and 0.945 of the highly irregular sleepers. The difference is 0.027, or in other words 2.7% more of the highly irregular sleepers have died off after 7.8 years. It might be significant in the statistical sense, but it looks like a pretty small difference to me.

I don't know to translate that into a statement like: If you were able to switch from a highly irregular sleep schedule to a very regular schedule, you would live __(x days)__ longer on average. With some hand-wavy reasoning I arrived at something like 10 days longer over a period of 10 years. I.e., a very small amount on average. I'd welcome someone with a statistics background to do a real calculation.