Monday, October 05, 2009

Epidemiology of Gun Violence in the City of Brotherly Love

A recent paper from the American Journal of Public Health takes a look at gun possession and gun assault in Philadelphia. From the abstract:
Objectives. We investigated the possible relationship between being shot in an assault and possession of a gun at the time.

Methods. We enrolled 677 case participants that had been shot in an assault and 684 population-based control participants within Philadelphia, PA, from 2003 to 2006. We adjusted odds ratios for confounding variables.

Results. After adjustment, individuals in possession of a gun were 4.46 (P<.05) times more likely to be shot in an assault than those not in possession. Among gun assaults where the victim had at least some chance to resist, this adjusted odds ratio increased to 5.45 (P<.05).
The paper also had some stats on gun violence that were surprising, though I admit I don't know much about gun statistics. During the study period (2003-2006) there were 3485 shootings in Philly (Population 1.4M, metro 5.8M) or an average of 4.77 shootings per day and an average of 9 shooting free days per year. I wonder how that compares to a city like Toronto (Population 2.5M, metro 5.5M).

The study was done by taking a case-control approach. Case participants were enrolled via the police department and excluded accidental, unintentional, self-inflicted and police-related shootings - the study interest was assault with a firearm. Control participants were matched according to age, race (black or white only), gender, and time of shooting. In other words, the researchers took people who were shot, matched with people who were not then looked at whether each was in possession of a gun or not. After their analysis they conclude:
On average, guns did not protect those who possessed them from being shot in an assault. Although successful defensive gun uses occur each year, the probability of success may be low for civilian gun users in urban areas. Such users should reconsider their possession of guns or, at least, understand that regular possession necessitates careful safety countermeasures.
However, there were some flaws in the study. For example, the authors "assumed that the resident population of Philadelphia risked being shot in an assault at any location and at any time of day or night", nor do they look at the legality of the guns in question. As such, matching wasn't based on location and thus excludes, for example, the possibility of 'bad neighbourhoods', it also excludes the possibility that people with illegal firearms may be involved in other illegal activities that increase the risk of a shooting. This despite their finding that people being shot were more likely to be in areas with lower income and more illicit drug trafficking.

This, of course, doesn't invalidate any results, but it does make it harder to suggest causation as strongly as the authors do in their discussion. From these results it's difficult to say that carrying a firearm increases your risk of assault vs. being in a high-risk group (eg. living in a certain area) increasing the chances of carrying a gun.

Still, the fact remains that gun possession is an indicator of assault risk (even if not necessarily the cause) and offers no guarantee of protection from being shot in an assault.


Rob said...

As you mention ignoring the location is a fatal flaw for trying to suggest causation.
Additionally, instead of trying to blame the victims of gun violence for being shot, which is the tone of this post, because they carry a gun, perhaps the conclusion of this research should be that work needs to be done to eliminate the causes of gun ownership. As you suggest these may include things like neighbourhood concentrations of low income and illicit drug trafficking. Addressing these issues would decrease legitimate gun owner intentions and those who own guns for criminal purposes.
Otherwise I think that you could paraphrase this research by saying,"Research, done in collaboration with the police, say that you shouldn't own a gun, for your own safety." To me that sounds a bit out of touch with the realities of someone who feels that they live in circumstances where owning a gun is desirable.

Bayman said...

I'm kind of confused. The study compares only people who were assaulted in the first place, armed or not?

What about all the people who didn't get messed with because they were packing? Or the people who didn't get shot at because they shot the other bastard first? Isn't that kind of the point of defensive armament?

Maybe that's another question that should have been asked here? Of people who choose to arm themselves, how many avoid violent attack vs those who do not pack heat?

It's the other side of the issue. Would address Rob's point and would probably be of greater interest to people who feel the need to defend themselves from the dangers of their hood.

Kamel said...

No, they looked at people who weren't assaulted and were/weren't armed as well.

The case participants were those who were shot in an assault, the control participants were those who weren't shot at all (from a random dialing sample). Then they looked at who was carrying and who wasn't. Sorry if that wasn't clear.

From there you can look at who was packing and was shot, who was packing and wasn't, who wasn't armed and was shot, and who was unarmed and un-assaulted.

They found when comparing these groups that people who carried firearms were roughly 4.5 times more likely to be the victims of gun violence.

As pointed out in the OP, by rob and now by you, one of the problems with this study (and a problem with the case-control method in general) is that you can't really separate "people who carry guns and get shot" from "people who are likely to be shot and therefore carry a gun".

BUT from the data in the paper, at the very least, it seems that 1) carrying a gun is an indicator of risk even if not the source of said risk and 2) carrying a firearm does not seem to be protective. And again, yes the study is flawed and one better designed to answer the questions that you and Rob bring up would be preferable.

MC Hawking said...

I'm always packin' my Uzzi,
so when I hear about them bitches at MIT utter the words "string theory"
I give them a lesson in bullet probability.

You say possession can make me a victim,
but you haven't heard the MC hawking dictum,
You'll be dead before you can shoot me,
Even if I have to violate temporal causality.

Anonymous said...

"To me that sounds a bit out of touch with the realities of someone who feels that they live in circumstances where owning a gun is desirable."

Or perhaps it's in line with the realities of numerous other studies that show that you're more likely to be shot by a gun that's owned by yourself, a friend or relative; that for every successful self-defense use there are many more accidental shootings; that having a gun in the home triples the likelihood of a family member being killed.

Kamel said...

Forget all that - check out these photos of other damage a gun can do in your house. Won't somebody think of the vegetables?

Anonymous said...

Cool post as for me. It would be great to read more concerning that matter. Thank you for posting that data.
Joan Stepsen
Buy gadget