Hacker News Reader

Comment by Tangokat | Hacker News Reader

The title is bullshit. They found that people who wake up earlier are less likely to be depressed. NOT that you can just wake up earlier and cure your depression.

"Genetically proxied earlier diurnal preference was associated with a 23% lower risk of depression"

View on HN

We've replaced the title with a sentence from the abstract of the study, shortened to fit HN's 80 char limit.

It's worse than just the title:

> This means that if someone who regularly goes to bed at 1 a.m. instead goes to bed at midnight and sleeps for the same amount of time, they can reduce their risk by 23%...at 11 p.m...40%...early risers may benefit from waking up even earlier. For those in the moderate or evening ranges, however, shifting to an earlier bedtime would likely be helpful.

> “This study definitely shifts the weight of evidence toward supporting a causal effect of sleep timing on depression,” Daghlas added.

> Dr. Vetter gives the following recommendations to individuals who desire to go on an earlier sleep schedule...

They are explicitly recommending that people change their sleep schedule, even though the study did not explore changes in sleep schedule.

This bullshit shouldn't have been published in the first place.

Yup I think they ran straight into a causal reversal.

Or it could just be a correlation. Same thing causes both.

So the article tries to separate this out by using Mendelian randomization which supposedly fixes this since whether or not you inherit a genetic variant from your parent is a coin toss. However it has to make some other assumptions that are pretty hard to justify when it comes to something as complex as sleep schedules. There are some strategies to at least detect when things go wrong, though.

Except the problem with Mendelian randomization is that genes that control circadian rhythm are also likely to affect mood.

Well, indeed, that is what their main finding is. The question is do they affect mood by means other than affecting circadian rhythm?

There have been studies suggesting causality in both directions. The resulting vicious cycles you get from that are common in mental illness.

Thanks for the comment. I came here to check if they'd got cause and effect backwards as usual - depression causes getting up late rather than up late causes depression. It surprises me people find this so hard - at uni even correlation was treated as major stuff and the idea of cause and effect almost beyond human knowledge.

Exactly. Depressed people typically sleep more.

Reply on news.ycombinator.com