Comment by Dumblydorr | Hacker News Reader

Comment by Dumblydorr | Hacker News Reader

This data isn't that convincing, a lot of factors go into this messy correlation between ED visits for girl suicide attempts and covid closure of schools. The title of the link doesn't mention ED visits for attempts, which is the actual outcome of interest here, so IMO the link title is too strongly worded from an epidemiologists viewpoint.

Now conjecturing as to correlation confounders. Maybe the girls need more time at home with pets, family, hobbies, more time in their yard, less time exposed to commuting and classes they don't want to take or people they don't want to see. School is such a diverse set of stressors and the data here isn't clear to begin with...

A more verified takeaway is that girls attempted suicide around 2X more often than boys. This we know has something to do with technology and social media, whereas the schools opening is still potentially spuriously linked.

The higher attempt rate for girls is probably due to boys being more effective at killing themselves (4x as effective, in fact).

https://news.ycombinator.com/item?id=27942628

What does the effectiveness of suicide have to do with the rate at which people attempt it?

A successful attempt results in no further attempts.

That's true. However, do the statistics count the same people reattempting?

I'm not sure, but that is a good question. I tend to think that the data sources could count the same individual multiple times because the report ("2019-methodological-summary-and-definitions") says so:

See: https://www.samhsa.gov/data/report/2019-methodological-summa...

> An important issue for users of NSDUH and TEDS data to consider is that the unit of analysis for TEDS is admissions to substance use treatment, whereas NSDUH estimates are for individuals who received substance use treatment. Consequently, individuals who were admitted to substance use treatment multiple times in the reporting period would be counted more than once in the TEDS admissions data.

Earlier in that same document, the authors state that they only ask these two categories about suicidal thoughts:

- the individual is over 18

- the individual is under or over 18, but has had a major depressive episode (MDE)

The statistics also include multiple studies with variable sampling beyond NSDUH and TEDS. I think the best one can do is to presume mixed data, which would indeed include repeat attempts for the same individual.

The confounding factor that I find most interesting is the fact that parents were likely also home much more during this time.

Reply on news.ycombinator.com