Weekly deaths in the US down by 7-10k for the week ending March 7 | Hacker News Reader

Weekly deaths in the US down by 7-10k for the week ending March 7

twitter.com

20 pointsnstj19 comments

This data is not accurate. I flagged this post because the headline is substantially wrong.

See https://twitter.com/lymanstoneky/status/1244433376741748736 and also the link posted by rjtobin.

It's good that people are fast enough to error-correct bold claims before they get too many eyeballs, but I am still shocked that smart people post these things publicly with such authority without first thinking through how they could be wrong. You're watching in real time what the value of the peer review process is, even if it's slow.

From the twitter thread, it seems like it’s a data error: https://mobile.twitter.com/mersenneary/status/12444399368766...

At first it seems plausible, due to decreased traffic fatalities and so on. But traffic fatalities are about ~3000 a month, this effect is 15-20x greater than that. So it seems like there should be a high burden of proof here.

Someone pointed out that this could be in part due to a backlog in reporting from other causes of death due to the virus. I'm sure fewer accidents (auto, medical, etc), also play a part. Even the OP (of the Tweet) admits this.

Where I live all nonessential medical interventions are on hold to free up space in hospitals. I'd assume that lowers the death rate temporarily.

Perhaps in the short term, but in the long-term a weeks or months long delay in care will likely cause an uptick in deaths. Many major issues are initially caught as patients investigating what seem like minor issues, but turn out to be cancer, etc.

Were the high numbers from early 2018 caused by bad winter weather?

https://en.wikipedia.org/wiki/2017%E2%80%9318_North_American...

edit: Or the flu season.

https://www.cdc.gov/flu/about/burden/index.html https://en.wikipedia.org/wiki/2017%E2%80%932018_United_State...

Flu season. Also check out http://www.euromomo.eu/

@dang - I'm happy to pull this down, is there a correct mechanism for me to do so? Or should I leave it as flagged.

Ok, but with 150k covid infections already the numbers will rise sharply in the next couple of weeks.

Fewer accidents and mass shootings.

You may know this already, but "mass shootings" account for a small number of deaths per year compared to what the media portrays.

[1] In comparison, you're about 3x more likely to be killed by choking on your food, 7x more likely to die in a fire, 91x more likely to die because of fall, and 397x more likely to die in an accident of some kind.

[2]Overall, violence is becoming less common over time despite the [3]increase in privately owned firearms in the US.

[1]https://www.businessinsider.com/us-gun-death-murder-risk-sta...

[2]https://www.themarshallproject.org/2019/09/30/new-fbi-data-v...

[3]https://www.washingtonpost.com/news/wonk/wp/2018/06/19/there...

Since they’re predominantly focused on when it’s about kids, I’ll add those stats since I have them handy.

Causes of death, 2016, ages 0 to 14 (60,975,441 population estimate, 32,709 deaths reported):

- Congenital defects: 5,598 (17.11% of deaths reported)

- "Otherwise unclassified": 3,161 (9.66% of deaths reported)

- Motor vehicle accidents: 1,455 (4.44% of deaths reported)

- Cancer: 1,284 (3.93% of deaths reported)

- Accidental hanging, strangulation, and suffocation: 1,215 (3.71% of deaths reported)

- Major cardiovascular disease: 901 (2.75% of deaths reported)

- Drowning: 713 (2.18% of deaths reported)

- Non-gun murder: 663 (2.03% of deaths reported)

- Accidents not explicitly listed: 657 (2.01% of deaths reported)

- Influenza & pneumonia: 356 (1.09% of deaths reported)

- Gun murder: 238 (0.73% of deaths reported)

- Firearms accidents: 74 (0.23% of deaths reported)

The mortality rate for someone between the ages of 0 and 14 in 2016 was 0.897%. That means the chances of death from a firearm for that age group was 0.0086%.

Sources:

- Centers for Disease Control. National Vital Statistics Reports, Vol. 67, No. 5. "Deaths: Final Data for 2016" Table 6.

- US Census. Annual Estimates of the Resident Population for Selected Age Groups by Sex for the United States, States, Counties, and Puerto Rico Commonwealth and Municipios: April 1, 2010 to July 1, 2018.

- Social Security Administration. Period Life Table, 2016.

Note: CDC numbers are based off reported cases across a variety of age groups. They do not represent an exact number but are instead meant to be treated relatively (e.g., 4.44% of child deaths are due to car accidents)

At least in the USA if you are shot and killed odds are much higher you committed suicide compared to getting shot by someone else.

Even according to 'anti-gun' sources you're about twice as likely to commit suicide with a firearm than to be a victim of a gun-related homicide.

https://lawcenter.giffords.org/facts/gun-violence-statistics...

All deaths are tragic, but given that violence overall has been falling for decades, at the same time that gun ownership is increasing, I find it difficult to attribute the violence to firearms rather than to social factors.

I often think about how many of these deaths would be prevented if we took all the 'anti-gun' money and poured it into our health care system to support people with mental illnesses.

You left out the obvious comparison to general gun deaths for some reason

I was responding specifically to a comment about mass murder.

That said, gun deaths is a murky statistic due to the inclusion of suicides. Whether that is 'fair' or not is a matter of perspective. I don't believe it is, personally. Primarily because other countries without firearms have higher rates of suicide.

The issue is further convoluted by the fact that [1]many of the most armed regions of the US are the safest, in terms of violent crime. My conclusion is that there's other factors at play that have to be addressed before firearms can be isolated as a variable.

[1]https://wallethub.com/edu/safest-cities-in-america/41926/

People have nowhere to go, so much less driving. Perverse.

accidents probably. mass shootings? not significant enough to register.

Reply on news.ycombinator.com