And the 2009 Junk Science for gun research goes to...

[Will Brink]

Your hard earned tax money at work. Junk science at a level that makes some of the prior politically motivated anti gun "research" look Nobel Worthy:


September 30, 2009

Penn Study Asks, Protection or Peril? Gun Possession of Questionable Value in an Assault

Those possessing gun in assault situation 4.5 times more likely to be shot than those not possessing one

PHILADELPHIA – In a first-of its-kind study, epidemiologists at the University of Pennsylvania School of Medicine found that, on average, guns did not protect those who possessed them from being shot in an assault. The study estimated that people with a gun were 4.5 times more likely to be shot in an assault than those not possessing a gun.

The study was released online this month in the American Journal of Public Health, in advance of print publication in November 2009.

“This study helps resolve the long-standing debate about whether guns are protective or perilous,” notes study author Charles C. Branas, PhD, Associate Professor of Epidemiology. “Will possessing a firearm always safeguard against harm or will it promote a false sense of security?”

What Penn researchers found was alarming – almost five Philadelphians were shot every day over the course of the study and about 1 of these 5 people died. The research team concluded that, although successful defensive gun uses are possible and do occur each year, the chances of success are low. People should rethink their possession of guns or, at least, understand that regular possession necessitates careful safety countermeasures, write the authors. Suggestions to the contrary, especially for urban residents who may see gun possession as a defense against a dangerous environment should be discussed and thoughtfully reconsidered.

A 2005 National Academy of Science report concluded that we continue to know very little about the impact of gun possession on homicide or the utility of guns for self-defense. Past studies had explored the relationship between homicides and having a gun in the home, purchasing a gun, or owning a gun. These studies, unlike the Penn study, did not address the risk or protection that having a gun might create for a person at the time of a shooting.

Penn researchers investigated the link between being shot in an assault and a person’s possession of a gun at the time of the shooting. As identified by police and medical examiners, they randomly selected 677 cases of Philadelphia residents who were shot in an assault from 2003 to 2006. Six percent of these cases were in possession of a gun (such as in a holster, pocket, waistband, or vehicle) when they were shot.

These shooting cases were matched to Philadelphia residents who acted as the study’s controls. To identify the controls, trained phone canvassers called random Philadelphians soon after a reported shooting and asked about their possession of a gun at the time of the shooting. These random Philadelphians had not been shot and had nothing to do with the shooting. This is the same approach that epidemiologists have historically used to establish links between such things as smoking and lung cancer or drinking and car crashes.

“The US has at least one gun for every adult,” notes Branas. “Learning how to live healthy lives alongside guns will require more studies such as this one. This study should be the beginning of a better investment in gun injury research through various government and private agencies such as the Centers for Disease Control, which in the past have not been legally permitted to fund research ‘designed to affect the passage of specific Federal, State, or local legislation intended to restrict or control the purchase or use of firearms.’”

This study was funded by the National Institutes of Health. The authors are also indebted to numerous dedicated individuals at the Philadelphia Police, Public Health, Fire, and Revenue Departments as well as DataStat Inc, who collaborated on the study.

http://www.uphs.upenn.edu/news/News_Releases/2009/09/gun-possession-safety/

 

[JohnFH]

A recent study by gun owners have found that epidemiologists at the University of Pennsylvania School of Medicine are 99 times more likely to make asses of themselves in published studies.

 

[wahsben]

Probably funded by the joyce foundation.

 

[Biggsy]

This study doesn't seem to take into account whether or not the firearms were possessed legally. I would guess that most of that 6% (individuals who had guns and were shot) were involved in a criminal lifestyle. Also this study doesn't appear to take into account anyone who was NOT shot due to their being armed. Additionally, isn't the fact that only 6% of the total number of individuals who were shot evidence that being armed is if anything evidence that an armed person is less likely to be shot? Actually I don't think this proves anything. Someone tell me if I'm missing something. Typical university thinking. This is why I don't hang my sociology degree on my wall. wtf.

 

[notalib]

This study doesn't seem to take into account whether or not the firearms were possessed legally. I would guess that most of that 6% (individuals who had guns and were shot) were involved in a criminal lifestyle. Also this study doesn't appear to take into account anyone who was NOT shot due to their being armed. Additionally, isn't the fact that only 6% of the total number of individuals who were shot evidence that being armed is if anything evidence that an armed person is less likely to be shot? Actually I don't think this proves anything.

Exactly... "Lies, Damn Lies and statistics" give me the same data that these guys had and I can prove just the oposite of their conclusion...
Statistics are only valuable if you ask a question first. It seems that their question is flawed. It appears they asked "What is the liklyhood of being shot if you have a gun" Rather than "What Is the likelyhood that someone not involved in crime and with no histroy of criminal behavior and is carrying a gun legally is more inclined to be wounded than an unarmed person during someone else's commision of a felony assault?".. My question was written in 5 secs and needs a lot more thought to get it right.. BUT thats the way it should work. You dont glean knowledge from stats, you ask a question and get an answer to that question..

 

[william_munny]

Yep, another worthless study for the libtards to quote like gospel. Who cares if its erroneous? It conveys their point, and that's all they care about.

 

[Martlet]

Not in this case. I'll bet she feels differently.

 

[terraformer]

Exactly... "Lies, Damn Lies and statistics" give me the same data that these guys had and I can prove just the oposite of their conclusion...
Statistics are only valuable if you ask a question first. It seems that their question is flawed. It appears they asked "What is the liklyhood of being shot if you have a gun" Rather than "What Is the likelyhood that someone not involved in crime and with no histroy of criminal behavior and is carrying a gun legally is more inclined to be wounded than an unarmed person during someone else's commision of a felony assault?".. My question was written in 5 secs and needs a lot more thought to get it right.. BUT thats the way it should work. You dont glean knowledge from stats, you ask a question and get an answer to that question..

This^.

That is exactly the question that has any public policy relevance. The question they answered was how many criminals die when shot and how many criminals will be shot at when carrying a gun. They didn't even compensate for the fact that the "victims" in this case were probably KNOWN to have a gun by the shooters. Lastly, and so basically, was how did they determine who the victim was? Was the victim just the person with holes in them? God, this is horrible science that should have been caught in peer review but their premise is sought after so someone decided to look the other way.

 

[jhblaze1]

how exactly does epidemiology, the study of health and illness among populations relate in any way to man made inanimate tools? Guns have nothing whatsoever to do with either health or illness, unless your grossly stretch the sh*t out of those terms until they no longer resemble their original definitions.

 

[Biggsy]

Not in this case. I'll bet she feels differently.

What the hell does she know. She's probably not even a phd.

 

[Biggsy]

Exactly... "Lies, Damn Lies and statistics" give me the same data that these guys had and I can prove just the oposite of their conclusion...
Statistics are only valuable if you ask a question first. It seems that their question is flawed. It appears they asked "What is the liklyhood of being shot if you have a gun" Rather than "What Is the likelyhood that someone not involved in crime and with no histroy of criminal behavior and is carrying a gun legally is more inclined to be wounded than an unarmed person during someone else's commision of a felony assault?".. My question was written in 5 secs and needs a lot more thought to get it right.. BUT thats the way it should work. You dont glean knowledge from stats, you ask a question and get an answer to that question..

Yup. I'm pretty sure the question they asked is "How can I get my name published in Libertard monthly, and get lots of pats on the back from all my enlightened colleagues?" Answer: S**t on gun owners.

 

[TomH]

You may be 4.5 times more likely to be shot, but how many more times are you likely to get out alive as compared to dead? Probably many more times than that.

 

[RI_John]

I am always on the lookout for actual studies with actual statistics regarding self-defense, whether it be whether unarmed women should fight off armed rapists or whether fighting off car-jackers is better than just giving up the keys. However, one rarely finds such a study that isn't severely biased from the get go.

In THIS case, it's not only biased, it's nonsensical. I honestly don't even understand their research method or how it even remotely relates to the question they're supposedly addressing. It is so flawed it's hardly worth discussing the methodologies, let alone the conclusions.

 

[Teej]

This just in. People who swim are more likely to drown than those who don't and in other news people who own and drive cars are more likely to get hurt in a car accident than those who don't.

This kind of stuff kills me. I always end up asking what is the point? Taken at face value people are more likely to get shot when armed in an armed confrontation than those who aren't armed. Gee how surprising is that? You mean to tell me that if a guy pulls a gun on me and I pull my gun to defend myself I'm more likely to get shot than if I didn't pull my gun based off of the records you have. No surprise there. Sometimes the DB fights back or poos himself and squeezes off more than the load in his shorts or was simply going to shoot anyways whether I had the gun or not. I expect and accept the possibility of that. It's called making an adult, informed decision. Of course there's the flawed aspect of only working with known data while the unknown data is just as important. How many incidences of armed conflict were there where the aggressor flees upon seeing me point a gun back at him, there by resolving said conflict? These kinds of statistics are used all the time. I'd really like to know the statistics for my sarcastic retorts so I can really use them as facts the next time some SFB gun grabbing idiot whips those little nuggets out.

 

[punx45]

those people were probably shot while unlocking their guns, which are of course, locked for their safety.

 

[hminsky]

The study has a very telling line "Six percent of these cases were in possession of a gun (such as in a holster, pocket, waistband, or vehicle) when they were shot.".

So, umm, a guy with a gun in his waistband gets shot in an assault on the street. My intuition says the guy is a thug, and was probably already doing something that made him "4.5 times" more likely to get shot than someone who isn't a thug with a gun in his waistband.

 

[Beeffajitas]

The study estimated that people with a gun were 4.5 times more likely to be shot in an assault than those not possessing a gun.

As identified by police and medical examiners, they randomly selected 677 cases of Philadelphia residents who were shot in an assault from 2003 to 2006. Six percent of these cases were in possession of a gun (such as in a holster, pocket, waistband, or vehicle) when they were shot.

So if only 6% of the samples were in possesion of a gun when they were shot, how does that equal being 4.5 times more like to be shot? My math, based on those numbers alone, tell me that you're 15.7X more likely to be shot when you don't have a gun on you.

 

[drgrant]

This is typical anti BS....

So they rig the research by using a contaminated sample set.

It's Philadumpia we're talking about here- meaning a lot of the people getting shot are likely criminals themselves, that are probably getting included in the sample set. Even if these criminals did not shoot back or otherwise start the altercation, the activities they are involved in make them more likely to get shot than the average person.

Without removing the criminals from the sample set, any notional of saying that "joe average is more likely to get shot while carrying a gun" goes right out the window.

Their "control" is a bunch of crap, too.

This isn't any different when they whine about the death rate by firearms in the US, and a major, overwhelming chunk of that death rate is suicides.

It's intentional misrepresentation of data to suit some political agenda.


-Mike

 

[In_God_We_Trust]

How about, let's look at the study to see how many of those individuals were in high-risk situations (jewelery store, 7-11, gas station) vs. the average CCW who is just out to protect themselves and family. Every time I read this s*it, I get so angry....



Some say a comet will fall from the sky.
Followed by meteor showers and tidal waves.
Followed by faultlines that cannot sit still.
Followed by millions of dumbfounded dipsh*ts.

Some say the end is near.
Some say we'll see armageddon soon.
I certainly hope we will cuz
I sure could use a vacation from this

STUPID sh*t, silly sh*t, stupid sh*t...

One great big festering neon distraction,
I've a suggestion to keep you all occupied.

Learn to swim.

 

[D-503]

For what it's worth, here is his 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).

Conclusions. 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.

-----

The statement "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." is interesting, as it appears to turn logic on its head - he did not look at individuals with guns to see who was shot, but rather the exact opposite. Then there are those magical words "after adjustment"...

Note also that 6% of 677 means he is basing his conclusions on 41 individuals who were in possession of a gun when they were shot. Presumably the number who actively resisted is somewhat less than this. I will try to acquire the article tomorrow and see if he controlled for things like whether the victims had concealed carry permits or whether they were criminals themselves.

 

[n1oty]

I'll bet they are including all the gangbangers who are shot by the cops or lawfully carrying civilians. A definite load of horse manure. Like my father always said, figures can lie and liars can figure.

 

[RI_John]

I'll bet they are including all the gangbangers who are shot by the cops or lawfully carrying civilians. A definite load of horse manure. Like my father always said, figures can lie and liars can figure.

Yup. I'm also sure if they ASKED they would have found "39% of all gun owners who were assaulted with a firearm have smoked crack in the past year. Therefore, owning a gun increases the likelyhood of smoking crack"

Gotta get the bangers out of the gun studies. Skews them just a teensy bit.

 

[KMaurer]

I want to read tha actual article, but I suspect that I can identify the primary flaws without having seen it. When they "adjust for confounding factors", that usually means simple things such as age, sex, income, race & ethnicity. I can guarantee that it won't include a lot of significant risk factors such as illegal drug use. Aside from their inabilities to control these factors, there's also the simple fact that, even with all other things being equal, people who are most at risk of assault are more likely than others to carry a gun for protection. People are also more likely to carry a gun in circumstances where they're at higher risk (e.g., when working nights or traveling through high-risk areas). I'm guessing it's the sort of things I used to see junior faculty write all the time ("publish or perish"): good enough to get past the referees and get published (particularly easy when the findings agree with the known preferences of the publishers), but nothing they or anybody else would really take seriously.

Ken

 

[Garys]

how exactly does epidemiology, the study of health and illness among populations relate in any way to man made inanimate tools? Guns have nothing whatsoever to do with either health or illness, unless your grossly stretch the sh*t out of those terms until they no longer resemble their original definitions.

Because most of the diseases that public health organizations were founded to combat, have been eradicated or controlled. Just like with the Swine Flu/Bird Flu "pandemic". They need something to justify their phony baloney jobs. Therefore they have to expand the definition of "public health" to include guns, seat belts, child seats, window grates, motorcycle helmets, bicycle helmets, and so on. None of which were contagious diseases which threaten the health of the general public the last time I looked.

Gotta get the bangers out of the gun studies. Skews them just a teensy bit.

Well, if they just included lawful gun owners, they'd have very, very, very, small samples. Like 1-2 a year in the entire state.

I'm guessing it's the sort of things I used to see junior faculty write all the time ("publish or perish"): good enough to get past the referees and get published (particularly easy when the findings agree with the known preferences of the publishers), but nothing they or anybody else would really take seriously.

Ken

Are you saying it's all part of the academia game? We need a "Shocked, Shocked" smiley.

 

[D-503]

I've read the article and would like to share a few thoughts.

First, I had to ask "what's a case-control study and why does it matter?" I found a nice definition at the website of the CA Department of Public Health (http://www.ehib.org/faq.jsp?faq_key=34):

A case-control study is an analytical study which compares individuals who have a specific disease ("cases") with a group of individuals without the disease ("controls"). The proportion of each group having a history of a particular exposure or characteristic of interest is then compared. An association between the hypothesized exposure and the disease being studied will be reflected in a greater proportion of the cases being exposed. It is advantageous for the controls to come from the same population from which the cases were derived, to reduce the chance that some other difference between the groups is accounting for the difference in the exposure that is under investigation. A case-control study generally depends on the collection of retrospective data, thus introducing the possibility of recall bias. Recall bias is the tendency of subjects to report events in a manner that is different between the two groups studied. People who have a disease may be more likely to remember exposures more readily than those without the disease. The opportunities to effectively use case-control studies may expand as new ways of characterizing exposure through the use of biologic markers of exposure are developed, which would reduce the problem of recall bias.

I like this description because it includes a summary of the drawbacks that are are present in great measure in this study. Of course there is another one that isn't mentioned here - the case-control study is a retrospective design that provides no evidence regarding causality. With more reading (there's even a decent article on Wikipedia), it quickly became clear that it's pretty much universally accepted that the fundamental limitations of the method are (a) population differences between cases and controls (leading to confounding), (b) different measurement biases between cases and controls, and (c) the inability to make causal inferences.

A good place to look for more critical description of case-control research is Chapter 7 of the National Research Council's report, Firearms and Violence: A Critical Review (http://www.nap.edu/catalog.php?record_id=10881#toc).

Many have already commented on the obvious presence of the first two limitations in the Branas et al. paper, so I am just going to make a comment about the third one. If time permits I will try to go through the issues of confounding and bias. But that is in some ways a distraction, because the main issue is that the report provides no evidence for the conclusion that carrying a firearm will either increase or decrease anyone's likelihood of being shot in an assault, regardless of whether they are a criminal.

In their introduction, Branas et al. cite the National Research Council report's conclusion that a variety of previous case-control studies have done little to reveal the impact of guns on homicide or the utility of guns for self-defense:

Several case-control studies have explored the relationship between homicide and having a gun in the home, purchasing a gun, or owning a gun. These prior stdies were not designed to determine the risk or protection that possession of a gun might create for an individual at the time of a shooting and have only considered fatal outcomes. This led a recent National Research Council committee to conclude that, although the observed associations in these case-control studies may be of interest, they do little to reveal the impact of guns on homicide or the utility of guns for self-defense.

However, the recent National Research Council committee also concluded that additional individual-level studies of the association between gun ownership and violence were the most important priority for the future. With this in mind, we conducted a population-based case-control study in Philadelphia to investigate the relationship between being injured with a gun in an assault and an individual's possession of a gun at the time.

Now, if you actually read the report, it is plainly evident that its research recommendations were based on a through and critical review of the weaknesses of both ecological and case-control methodologies. Here are a couple of paragraphs from different parts of the executive summary (pages 5 and 6):

Case-control studies show that violence is positively associated with firearms ownership, but they have not determined whether these associations reflect causal mechanisms. Two main problems hinder inference on these questions. First and foremost, these studies fail to address the primary inferential problems that arise because ownership is not a random decision. For example, suicidal persons may, in the absence of a firearm, use other means of committing suicide. Homicide victims may possess firearms precisely because they are likely to be victimized. Second, reporting errors regarding firearms ownership may systemically bias the results of estimated associations between ownership and violence.

In summary, the committee concludes that existing research studies and data include a wealth of descriptive information on homicide, suicide, and firearms, but, because of the limitations of existing data and methods, do not credibly demonstrate a causal relationship between the ownership of firearms and the causes or prevention of criminal violence or suicide. The issue of substitution (of the means of committing homicide or suicide) has been almost entirely ignored in the literature. What sort of data and what sort of studies and improved models would be needed in order to advance understanding of the association between firearms and suicide? Although some knowledge may be gained from further ecological studies, the most important priority appears to the committee to be individual-level studies of the association between gun ownership and violence. Currently, no national surveys on ownership designed to examine the relationship exist. The committee recommends support of further individual-level studies of the link between firearms and both lethal and nonlethal suicidal behavior.

To cite this report as the primary justification for conducting another case-control study because it had called for more research with individual-level data is flagrantly dishonest. Their whole point was to encourage research that can better identify causal relationships - which they had just pointed out is not not possible with case-control designs. What the committee was saying was that the last thing that anybody needs is more case-control research. Given that the authors are willing to play games with the fundamental premise of the research question, I am not very interested in digging deeper. But if I can hold my nose long enough I will try.

 

[WILDCATT]

educated idiots

you want case studies come on down here.drive by shootings,drug dealer shootings,drinking shootings.a lot of shootings nowadays are sudden.no warning. a robber shoots you first and then robs you.not much chance to defend your self if your dead.a lot of crook to crook shootings.seems the medical profession has to find some victims to cover up their mistakes.called medical misadventures.like 600,000 a yr.some one of you might get ahold of the NRA about this BS.if they have not allready looked at it.I asume that Locke would have something to say on this subject.

 

[06LemansC6]

This clinches it. police should no longer carry guns, it will make their jobs safer. After all they are 4.5 times more likely to be shot while carrying.

 

[notalib]

For what it's worth, here is his abstract:

-----

Note also that 6% of 677 means he is basing his conclusions on 41 individuals who were in possession of a gun when they were shot. Presumably the number who actively resisted is somewhat less than this. I will try to acquire the article tomorrow and see if he controlled for things like whether the victims had concealed carry permits or whether they were criminals themselves.

How many were legally armed and minding their own business. ie, how many were Gang bangers?

 

[D-503]

The closest they get is whether or not a person had prior arrests - from police records for people who were shot ("cases") and by asking those who were interviewed by phone ("controls"). Of those who were shot, 53% had prior arrests. Of those who were shot and in possession of a gun, 37% had prior arrests.

While I'm looking at the table, I can see that 83% of those shot were outdoors at the time, while only 8% of those who weren't shot were outdoors at the same time.

With differences of this magnitude between the case and control populations, it is a little hard to swallow the idea that statistical adjustment can make them equivalent.

 

[D-503]

Another note - many seem to misunderstand what these authors are claiming. They do not conclude that carrying a gun makes you more likely to be shot in a criminal assault. They say "On average, guns did not protect those who possessed them from being shot in an assault." At first this may seem to be a nice conservative conclusion (they are not making any more of the 4.5 odds ratio than saying it is nowhere near the range (that is, less than one) you would expect if gun possession is protective.

Of course, this conclusion is just as incorrect as saying those with guns are more likely to be shot. Aside from the obvious confounding (you can adjust all you want but you can't statistically turn law abiding citizens into criminal gang members, even in Philadelphia), the case-control design cannot tell you anything about causal relationships. The most they can say is that if you have been shot while you were carrying a gun, your gun did not protect you from being shot. Stop the presses!