[computator]

Whenever I see a study about something that affects longevity, I want to know how much of a difference it makes expressed as a number I can relate to. If you were able to switch from a highly irregular sleep schedule to a very regular schedule, would you live 18 hours longer on average, or 1.5 months, or 5 years? This would be a way to decide how much attention and effort one should to devote to the numerous studies about things that affect mortality.

[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.

[julianeon]

A confounding factor example would be someone who sleeps regularly and someone who sleeps irregularly, two people who both live the same amount of years... but the irregular sleeper lives their last 10 years with greatly impaired capacity after a stroke. [Note: some of my work involves dealing with elderly people and "impaired after a stroke" is an extremely real, and common, thing.] These 'fuzzy' conclusions may be the best we can do.