Matthew

The CDC has released provisional suicide statistics for 2022. Here's the big picture: suicides are up almost 3% over 2021. Men still commit about 80% of suicides, but their increase was 2.3% versus female suicides jumping almost 4%. Across races, small numbers lead to big disparities in percent changes, but the same pattern as always was still observed: Whites barely shifted their pole position from 76.13% of suicides to 75.75%. An important part of the race differences in suicide rates is population differences in age. The population is getting older, and the White population is way older than the others. To visualize that, consider this data from Pew: If you split up suicide rates by age more granularly, this is clearer: And if you plot national suicide rates over time, it's clear to see that they're related to increases in population age, and also, to opioid abuse. We have causal information about both of these things based on - Pill mills (strong causal identification) - Doctor prescription habits (moderate-quality causal identification) - Drug raids (strong) - Arrests of prominent drug dealers (strong) - Migration-related variance in county ages above and beyond natural aging (low to moderate) Back on-topic: The CDC's post provided a chart that showed that the suicide rate increased 37% between 2000 and 2018, but it decreased 5% between 2018 and 2020. The rate then returned to peak levels in 2021, and now, it's increased another 3% on top of that: we're back on trend for record suicide rates, and the 2023 provisional data (which is only available for January) shows a further increase of 26 suicides over January 2022/283 over January 2021. Suicide isn't just age, and there are some complexities to its relationship with age. Suicides are still up for people aged 25 to >=65 years of age, but they actually declined for young people, ages 10-24. Why? I think the answer might have to do with schooling. For many young Americans, school is still not in-person, and a recent preprint provided evidence that in-person schooling promotes youth suicides. They had two main pieces of evidence: 1. During the summer months and winter break, when school is out, youth suicides go down, and they rise again when school comes back into session. 2. When the pandemic forced schools to go online, youth suicides went down. When you supplement this with their data showing that kids who have been bullied reported >3x greater rates of suicide attempts, almost 3x higher rates of suicide plans, and roughly 2.7x higher rates of considering suicide, it all makes sense. The authors could have even gone further and shown that anti-bullying crackdowns do reduce rates of student suicide, although maybe it was wise to avoid mentioning these studies, since, from what I've seen — which is probably not comprehensive — they're all underpowered and I have no doubt there's publication bias involved. Here's the gist. When you're online, your classmates aren't going to bully you. And now that most schools are back in-person, large numbers of them also come with offline options, so now we can get the best of both worlds: kids who want to be online frequently have that option; the majority who want to be in-person can also choose to be in-person. There might be worse learning outcomes from going remote (and lots of data indicates this), but if it saves kids lives, the benefits surely outweigh the costs. And who knows: if, as a recent preprint indicates, self-selection into remote learning can make it a benefit for some people (Bruhn, Campos & Chyn, 2023), that group might overlap with bullying victims. So let's take one more look at this. Youth suicides are clearly highly seasonal, but in 2020, they were down in more months, as youth had more time outside of in-person schooling. This may have continued into more recent years, as students are returning to in-person instruction but they're still often able to stay remote if they opt into that. We know the effect of school closures, but allowing students to sort into in-person and online instruction may have an even larger effect because the people who could have reduced suicide rates with in-person instruction because they need to socialize -> they get to go in-person. And the people who would have higher suicide rates because they'll be crushed under the weight of bullying -> they get to stay online. We may be entering a time of increasing age polarization in suicides, and if sorting into different types of instruction can lead to heterogeneous benefits for the youth suicide rate, the CDC's latest report might caution efforts to push everyone back to in-school instruction. Sources: cdc.gov/suicide/suicide-data… pewresearch.org/short-reads/… nber.org/papers/w30795 nber.org/papers/w31542 Earlier discussion of strategies to cut down on opioid-related mortality: teapaw.org/cremieuxrecueil/…