A Perspective on Death in the USA
Note: This post is part of our League Symposium on Guns In America. You can read the introductory post for the Symposium here. To see a list of all posts in the Symposium so far, click here.
by Major Zed
We are all going to die. This is certain. What is uncertain is when and how. The Centers for Disease Control tabulated over two and a half million deaths in the USA in 2010 (the latest year available for detailed mortality statistics). The FBI tabulated more details in certain categories. This article summarizes some of these findings, with a focus on deaths involving firearms. It is not intended to promote particular policies or positions, but rather to provide a factual context for such discussions.
Table 1 below and Figure 1 display deaths by cause and age group. There are several takeaways. Death is mostly for older people. Once you get out of the teenage years, the death rate approximately doubles for each decade of life (not shown in these tables – take my word for it). The biggest causes of death are the usual culprits we all know about – heart disease and cancer. There are two notable exceptions to the “death is for older people” cliche. Infant mortality is reflected in the last category “perinatal and congenital.” And then there are “external causes.” This is the second largest category for the under-18 demographic and the largest category for the 18-54 group.
Table 1: Deaths by Cause and Age Group, USA, 2010
| Age:0-17 | Age:18-54 | Age:55+ | Total | |
|---|---|---|---|---|
| Heart disease | 1,407 | 64,509 | 718,506 | 784,422 |
| Cancer | 2,022 | 68,020 | 519,613 | 589,655 |
| Mental/Nervous | 1,384 | 12,945 | 247,595 | 261,924 |
| Lung disease | 1,294 | 11,788 | 224,345 | 237,427 |
| Metabolic/Digestive | 1,123 | 31,829 | 157,248 | 190,200 |
| External | 12,357 | 98,137 | 72,764 | 183,258 |
| Other | 4,166 | 12,828 | 114,223 | 131,217 |
| Infectious | 1,120 | 13,513 | 52,949 | 67,581 |
| Perinatal/congenital | 18,174 | 2,616 | 1,835 | 22,625 |
External causes are accidents and intentional harm. These are broken out in Table 2 and Figure 2. Accidents predominate, with motor vehicle accidents being the most common, just ahead of poisoning. Older people are more likely to die from a fall than from a motor vehicle accident, but that is much less likely for younger people. Firearms deaths occur in accidents, suicides, and homicides, but more often in suicides. Deaths from firearms accidents are about one sixtieth of motor vehicle deaths. Firearms suicides are slightly more frequent than all other suicides combined because of the older age group. Firearms homicides are more than double all other homicides because of the middle age group. Total deaths from firearms comes just behind motor vehicle deaths and poisonings.
Table 2: “External Cause” Deaths by Cause and Age Group, USA, 2010
| Age:0-17 | Age:18-54 | Age:55+ | Total | ||
|---|---|---|---|---|---|
| Accidents | Motor Vehicle | 3,593 | 20,958 | 10,778 | 35,329 |
| Poisoning | 1,049 | 26,133 | 5,853 | 33,035 | |
| Falls | 125 | 2,223 | 23,660 | 26,008 | |
| Other | 2,977 | 7,968 | 14,920 | 25,865 | |
| Firearms | 106 | 389 | 112 | 606 | |
| Suicide | Firearms | 695 | 11,032 | 7,663 | 19,390 |
| Other | 959 | 13,293 | 4,715 | 18,967 | |
| Homicide | Firearms | 1,386 | 8,823 | 864 | 11,073 |
| Other | 975 | 3,179 | 1,015 | 5,169 | |
| Other/Unspecified | 887 | 7,489 | 4,336 | 12,712 | |
| Total Firearms | 2,186 | 20,244 | 8,639 | 31,069 | |
The FBI’s tabulations of homicides, while of the same magnitude, do not exactly tally with the CDC count. So I used the FBI proportions and applied them to the CDC numbers. Also, some of the “unknown” counts were allocated back to the knowns.
Table 3 and Figure 3 break out the implements of homicide. Table 2 already showed that firearms are preferred, but here we see that handguns are by far the most common with shotguns and rifles trailing far behind.
Table 3: Homicides by Means, USA, 2010
| Handguns | 9,872 |
|---|---|
| Knives | 2,087 |
| Other non-firearms | 1,512 |
| Hands & feet | 909 |
| Blunt instruments | 661 |
| Shotguns | 613 |
| Rifles | 588 |
Table 4 and Figure 4 tabulate the circumstances surrounding homicides. Most common are arguments and brawls (some with alcohol or drug involvement). A significant number occur during the commission of (other) felonies. Gang shootings also contribute.
Table 4: Homicides by Circumstances, USA, 2010
| Brawls & arguments | 6,962 |
|---|---|
| Other non-felony | 3,753 |
| Gangland/juvenile gang killings | 1,653 |
| Felony, Robbery | 1,519 |
| Felony, Other/suspected | 1,453 |
| Felony, Narcotics | 902 |
There is nothing better than having measurables for discussions like this one. Thanks much.
One question I tried (and failed) to answer for myself was what caliber killed the most people. My intuition tells me that it’s the .22 that has killed the most folks, but I wasn’t able to confirm that to my own satisfaction.
In your research, did you stumble across these stats?Report
Sorry, did not.Report
Dude, totes cool. I was just hoping to have numbers when we eventually got to the part of the discussion that asked if anyone really *NEEDED* a Desert Eagle .44 to hunt ducks and if we should really make it legal to stand in front of a mirror yelling “ARE YOU TALKING TO ME” like some Dirty Harry fantasy before derailing into minutia like whether Dirty Harry and Travis Bickle were two different people (and so on)… well, I was hoping to discuss how most folks aren’t killed by Desert Eagles but, actually, Bryco or Jennings handguns (if, in fact, that was the case… but it wouldn’t surprise me).Report
I too have been searching for what caliber of guns account for most gun deaths. Not sure if that is because of the NRA blocking specific reporting or what. On the FBI (I think) I found where most of the suicides were within the first year of gun ownership, making me think that it is more likely that hand guns account for those and that the guns were probably bought for the express purpose of suicide. But I could be totally wrong since I am only making a calculated guess based on available data and my thinking that it is more likely that one would kill themselves with a handgun.Report
I haven’t a clue where to look for such statistics either, and I doubt they exist. Coroner reports and death certificates I’ve read usually indicate something like “GSW” or “shot,” and don’t include any information about the size of the projectile.
My suspicion is that whatever size of projectile is the most popular, that’s the one that’s going to be used the most in lethal shooting incidents (regardless of the criminality or righteousness of the kill). As Mike D. pointed out in another thread, a single .22 caliber bullet can kill a human being. Whatever is out there, that’s what’s doing the killing, in numbers proportional to whatever is out there.Report
My hypothesis is that the majority of non-police shootings are done with little, cheap (non-heirloom quality), inexpensive ($100-$200) .22 handguns. (And, indeed, in a higher proportion than you’d think warranted. I can see Desert Eagles being used in more suicides than homicides.)Report
Which reminds me — I was reading a few days ago that Congress, sometime back, actually blocked any and all funding for what would (otherwise) be fairly routine governmental studies on things like, you know, gun violence and such.Report
A couple points: First, you need to seperate out ‘avoidable’ deaths from ‘unavoidable’. We are, as you note, all going to die.
The important question is: Did we die too early, from something that might have been prevented? (In other words: What on earth does illness have to do with accident and malice?)
Second, you probably need to find a way to control by time or access. I am on the roads, oh, 10 hours a week, say. That’s 10 hours a week exposure to death via automobile. I am around guns, oh, 10 hours a year. (Exposure to guns via crime will, effectively, be minimal in terms of time spent around a gun).
Avid hunters might spend a hundred or two hundred hours around guns a year. (Locked in a gun safe shouldn’t count. What about resting in your nightstand?).
Just to illustrate the point I’m trying to make — if I spend, say, 10,000% more time at risk from automobile deaths than gun deaths, what’s it mean my risk of dying via gun is almost as high as dying via car? Is a gun really safer than a car at that point?Report
(I realize you did break out accidents. I just don’t really get why illness is even related).
Interesting to note how prevalant handguns are in deaths. That pushes handguns more to the “Heavily regulated” category for me, with hunting rifles and shotguns remaining “registered and insured, but available to anyone with a clean record” category.
I’ve really got no problem with guns designed to kill animals. It’s the ones designed to kill people that worry me. Mostly because, well, people tend to use tools as they’re designed to be used. People by hunting rifles to hunt animals. And other guns if they’re planning on shooting people. (or worried they might have to).Report
“Exposure to guns via crime will, effectively, be minimal in terms of time spent around a gun”
All due respect, but would you really see two guys playing with loaded handguns behind the dumpster of a restaurant you were visiting?
… I didn’t. We just got the hell out of dodge.Report
Morat20,
How do you have any idea how much time you spend around guns? Especially if you consider that you don’t need to be in particularly close proximity to a gun to be within its kill range. You might have colleagues that keep guns in their cars. You might walk past people on the street who are concealed carrying. You might visit the home of a friend who has an unlocked firearm somewhere in the house that you don’t know about.
You are right that frequency matters, but I’d be foolish to say I’ve spent less than 5 hours around guns in my life because that is the only time I’ve seen them displayed.Report
Because all my time spent around guns is time I choose to spend around guns. If I’m ever assaulted with a gun, or have a gun aimed at me in the course of a crime, or even with an interaction with a police officer, it amounts to a handful of minutes at most.
Whereas I go target or skeet shooting, or hunting, with my family for a few hours a year.
I’d imagine most people’s actual “time around guns” is measured in minutes per year.
There might be, as you note, some unaccounted for time. But I’d imagine very, very, very few Americans who are not actually employed to carry a gun spend even a fraction as much time around a firearm as a car.
When the median american spends orders of magnitude more time at risk for car accident than gun accident, is it not rather interesting that his risks of injury for both are even in the same ballpark?
It certainly seems to speak volumes about the relative danger of the gun versus the car.Report
Morat:
The important question is: Did we die too early, from something that might have been prevented?
It’s really hard to do that. My grandmother died of pneumonia. She was 97. If she had gone to the doctor earlier, she may have lived through it. But she was 97. Odds are pretty good something was going to kill her within a reasonable delta of the day she died, anyway. Respiratory failure gets the blame, did it really kill her? Probably within the error margin.
Second, you probably need to find a way to control by time or access. I am on the roads, oh, 10 hours a week, say. That’s 10 hours a week exposure to death via automobile. I am around guns, oh, 10 hours a year. (Exposure to guns via crime will, effectively, be minimal in terms of time spent around a gun).
Avid hunters might spend a hundred or two hundred hours around guns a year. (Locked in a gun safe shouldn’t count. What about resting in your nightstand?).
This is a point, but it’s almost impossible to quantify. You have no way of knowing how many guns you’re exposed to in a day.
I’m also fairly certain that this is a bad metric. Here’s why:
Your point (that use/exposure is a factor) is valid, but I’m not convinced by looking at state-by-state CCW stats that it actually is a significant factor in determining likelihood.
States with a lot of CCW permits issued don’t appear to have significantly more deaths from gunshots, per capita (when controlled for number of permits) than states with few CCW permits. (disclaimer: I’ve only eyeballed this)
This would indicate that there is a staggering offset of “exposure” vs. “use”, which is pretty intuitive when you think about it. You’re very very likely to be exposed to a lot of guns without being exposed to the use of a gun, but with cars you get them both at the same time. The converse of cars; most people who drive will be in an accident at some point in their lives, but most gun owners won’t have a gun violence event at some point in their lives.Report
You’re missing the point: I spend, minimum, about 400 hours a year in a car. (That’s just commuting, really). My exposure to guns? (And by guns, anytime I was in the same building as a loaded, unsecured firearm — counting cops, guards, friends who had a gun that wasn’t locked away, using them myself) — maybe 20 hours? (I did have 8 hours of jury duty, where I was around cops).
So I spend twenty times more of my given year at risk for a car accident than a gun accident.
Yet — and I’m being very nice to guns here by excluding suicide — apparently I’m only 3 times as likely to be hurt by a car than a gun! Including suicide, my odds of death by car or death by gun appear equally likely!
Which makes sense — guns are designed to kill things, and that’s what people use them for. Cars are designed for transport. The only safey feature on a gun is — for some — a safety catch. Cars are designed, mandated, and constantly improved to minimize injury upon accident.
You cannot, CANNOT compare guns versus cars without somehow adjusting to the fact that we spend vast chunks of our lives riding around in a car, but don’t spend vast chunks of our lives carrying around a gun.Report
Except you’re making a slew of assumptions you don’t really have any right to make, Morat.
You don’t know how many people around you are carrying guns (in California, this is probably a low number, in Arizona, this might be a lot higher). You don’t know how many people who live next door or in the apartment below or above you might own guns. For all you know, the guy who lives in the apartment above you is an idiot who constantly forgets to clear his gun before he cleans it, and every day the barrel is pointed down at your apartment with a chambered round, and you’re just lucky to be alive right now.
You don’t know how many people have loaded guns in their glovebox, or in their trunk.
In short, you don’t really have anything credible to even remotely assess your exposure to guns, at all.
Which makes sense — guns are designed to kill things, and that’s what people use them for. Cars are designed for transport. The only safey feature on a gun is — for some — a safety catch. Cars are designed, mandated, and constantly improved to minimize injury upon accident.
If guns are designed to kill things, and “that’s what people use them for”, how come with hundreds of millions of them in the country we only have 20k or so deaths from guns? I think you really, really, really don’t understand what people use guns for.
Which is okay, I don’t really claim to understand what people use guns for, either. But you seem to be pushing your inability to understand into a certainty that they must be using them for the only purpose you can think of. That’s terrible logic, dude.
You cannot, CANNOT compare guns versus cars without somehow adjusting to the fact that we spend vast chunks of our lives riding around in a car, but don’t spend vast chunks of our lives carrying around a gun.
You have something of a point, here, but being unable to directly compare two things isn’t the same as being unable to compare them at all.
You do spend vast chunks of your life riding around in a car. Are you more of a danger to yourself – is your piloting of the car your greatest risk factor – or are you in more danger from the other drivers?
My guess would be that you would answer the second (whether or not it’s true, most people think they’re pretty good drivers).
You don’t carry a gun, and so this isn’t a dichotomy for you. Thus, in particular, your perception of guns and danger is going to be heavily weighed in assuming that your danger is from other people who own guns.
You’re right in one sense; the right way to compare the thing is on use and exposure. But “use” and “exposure” are two very different things for cars vs. guns.Report
Morat20: Did we die too early, from something that might have been prevented?… I just don’t really get why illness is even related.
A lot of disease deaths are, if not ultimately preventable, at least “delayable.” Think obesity and type 2 diabetes. Public health policymakers are concerned about all types of death, so I thought the context was relevant.Report
Eh, it is. But in the end, gun deaths are either accidents or deliberate. They should be compared to, well, accidents or deliberate deaths in general.
Throwing in illness is dodgy. Suicides is a bit dodgy too, and should probably be addressed seperately. (Although, really, I understand gunshot to the head is not as certain as many people think. Botched suicides get awful very quickly.)Report
MZ:
Good post. One major bit that is illuminating that you don’t cover here is Homicide by Relation.
About 10% of homicides are “Committed by Intimate” (http://bjs.ojp.usdoj.gov/content/homicide/intimates.cfm). The use of firearms in these murders used to be about 67% of all cases in 1975, it’s been in steady decline and we’re now down to about half.Report
The FBI link I gave has more detailed breakouts if you want to go exploring.Report
Zed,
Thanks for the very informative post. This makes me think of the question often bandied about (sometimes sarcastically, sometimes cynically) about what is the “acceptable” number of gun deaths per year. I have no idea what to consider “acceptable” and how we determine this vis a vis the acceptability of steps taken to achieve that number.
I’m tempted to say the ideal number of gun deaths is zero. But then I think about how Osama Bin Laden was killed with a gun*. Or the non-zero amount of rapists or would-be rapists killed by their victims during the commission of their crime. And I start thinking that maybe zero isn’t ideal.
* I don’t necessarily mean to argue that the shooting of Bin Laden was a wholly good thing. I don’t know enough about the circumstances but the cynic in me wonders if shooting him was the only option available at the time.Report
Some killings are wrongful. Criminal, even.
Other killings are righteous. Self-defense, for example.
The nature of the weapon used is irrelevant to the righteousness of the kill.Report
So some percentage of gun deaths are righteous. Some percentage would happen regardless of the existence of guns. So what we really ought to focus on are wrongful deaths that would not happen if it were not for the existence of guns. How many is that? Hard to know, but probably much lower than 11K. What steps can we justify to prevent what might as little as 1 or 2% of all deaths? I struggle to think of ones that are both reasonable and effective.Report
I said this in my email to you, but I wanted to say it here in the public forum as well: This contribution adds invaluable data to the various discussions at large. Thanks.Report
You did a great job integrating the tables and figures. Thanks for that.Report
I agree with Tod. Excellent data contributing to a discussion that needs perspective.Report
Here is a theory to consider regarding violence.
http://www.motherjones.com/environment/2013/01/lead-crime-link-gasoline
Not directly about guns but it does address rates of violence.Report
That is fantastic! Why haven’t we heard more about this?Report
Hmmm… I just got that link in my inbox… I wonder if we’re on the same email listserv…Report
I saw Alex post about this at Forbes as well.Report
What surprised me the most was the mantra “legalize drugs – end gang violence – reduce gun deaths” doesn’t address a very large portion of homicides.Report
Well, it addresses the overwhelming majority of “narcotics”, many, if not most, “gang killings”, and, I imagine, a small, but measurable, chunk of the homicides committed during robberies.Report
From the FBI link, it sounds like, but isn’t clearly stated that I could find, that these are only immediate circumstances of the homicide.
It’s quite likely that significant portions of the others have a drug-relationship (gangs, robberies, brawls and arguments) – I am not sure if the data is out there but you’d think it was something someone in the Justice Department would be publishing.Report
Good point – a lot of brawls/arguments, the biggest category, could be related.Report
Yes, but bear in mind… some drugs cause paranoia. Crystal Meth is gonna cause deaths, no way around it (except maybe with better manufacturing).Report
Poisoning, for those who are as confused as I was upon seeing this, primarily consists of drug overdoses, not household chemicals.Report