Wednesday, August 26, 2020

More Medical Propaganda disquised as a "study" about Firearms and Risk

Meme from stgolinsky.com, cropped and scaled by Dean Weingarten

The American Journal of Medicine recently published a paper about deaths associated with firearms. The paper misleads in the first sentence of the abstract. The lie is in the unstated, false assumption. It is a subtle but important shift in causation.

The first sentence in the abstract is: 

News media and policy makers frequently discuss deaths from firearms, drug overdoses, and motor vehicle accidents.

The shift in causation is done when the author inserts the word from before firearms and associates firearms with drug overdoses and motor vehicle accidents. The correct word would be with.

The correct usage would differ for each statistic. A corrected sentence would read thus: 

News media and policy makers frequently discuss deaths committed with firearms, from drug overdoses, and from motor vehicle accidents. 

Firearms are objects. Drug overdoses and motor vehicle accidents are actions. The author repeats the error by listing firearms as a "cause of death".

The Center for Disease Control  (CDC) does not list "Firearms" as a cause of death. They list homicide, suicide, and unintentional Injury. 

Inserting the word "from" defines firearms as the cause. It comes with the assumption: if there were no firearms, none of the deaths listed would have occurred. 

That is a false assumption. 

The rate of total suicides is little affected by a reduction of firearms; similarly, a reduction of firearms has little, if any, effect on homicides.

There may be an effect on fatal accidents. The numbers are so small, it is difficult to know if reducing the number of firearms would reduce the total rate of accidental deaths. It might, or it might not. 

The rate of accidental deaths with firearms has been reduced a remarkable 94% in the last 85 years, which is why the current number is so small. We do not have a good number for how many privately owned firearms there were 85 years ago. The first reasonable estimate is from 1945. 91 of the 94% reduction in the rate of fatal firearm accidents has occurred since 1945.  

The reduction has occurred while the per capita number of firearms has risen from .35 in 1945 to 1.37 in 2019. In 75 years, the number of firearms per person in the United States has increased 390%.  There is little correlation with the number of guns and the number of fatal gun accidents.

By the end of 2020, there will be four times as many privately owned guns per person in the United States as there was in 1945. The homicide rate is lower today than it was in 1945. 

In 1945 the homicide rate was 5.7 per 100,000. In 1933 it was 9.7. In 1957, it was 4.0. In  1974, it was  9.8. In  1980, it was 10.2. In 1984, it was 7.9. In 1993, it was 9.8. By 2013, it was back down to 4.5. 

The homicide rate has gone through wild swings, as the per capita number of firearms has steadily increased. 

There is no correlation between the number of guns and the homicide rate.

An absolutist might claim if there were no guns, there would be no gun accidents. No rational person believes the number of firearms in society can be reduced to zero. Even with the draconian gun controls put in place in Australia in 1997, the number of private guns is now more than it was before the controls were put in place. To reduce guns to zero, you have to eliminate the potential of making them. Making guns at home is getting easier every year.

There were 444 million private firearms in the United States at the end of 2019. The population was 328.2 million

In the results section of the abstract, the error is repeated. This is not a surprise. The error is the heart of the paper. Without the error, the reason for the article evaporates. From the results: 

The lifetime risk of death from firearms, drug overdoses, and motor vehicle accidents was 0.93% (95% confidence interval [CI], 0.92%-0.94%), 1.52% (95% CI, 1.51%-1.53%), and 0.92% (95% CI, 0.91%-0.93%), respectively.

To check the numbers in the paper, I performed much simpler mathematics. 

I took the number of deaths associated with firearms from the CDC in 2018 (39,740) x the life expectancy from the closest year, 2017 (78.54). That yielded the deaths over the time span of the average life expectancy, or 3,119,590. As a percentage of the population, that is .95%, very close to what the study says. But the study only counted people below the age of 85. When you correct for that (in 2016, the percent over 85 was 1.9%), 1.019 x .93 is .95%, so close as to be no difference. 

This simple check shows the mathematics of the study are credible. 

It is the conclusions of the study which are suspect. From the abstract: 

CONCLUSIONS:The lifetime risk of death from firearms, drug overdoses, and motor vehicle accidents is substantial and varies greatly across demographic subgroups and states

The false assumption about causality is in the conclusion, as expected. There are two component messages which are implied. 

One, the reduction of firearms in society reduces the risk of death. That is false. 

Two, the idea that promoting the narrative that firearms cause death is desirable, for reasons of policy.

The two assumptions make the paper into a political statement, instead of a scientific one. 

These assumptions only make sense if the author assumes firearms have no positive uses or contributions to society.  A hundred million Americans show they believe firearms have positive uses. They do so by owing firearms instead of selling them, giving them away, or destroying them.

In 2020, millions of Americans are buying firearms. It appears a new record will be set, with over 21 million firearms sold in 2020. 

It is the heart of Progressive doctrine that the elite know better than the population, what is best for members of the public.  Restricting guns from those who value them, is a core progressive belief.

©2020 by Dean Weingarten: Permission to share is granted when this notice and link are included.

Gun Watch



The false assumptions which are not stated.

False assumption: No firearms, no deaths.

False causality: Firearms cause these deaths. 

False, assumed prescription: Reduce firearms, reduce deaths.

Poor methodology: Only look at risks, ignore benefits.

Percent of population 85 or older: 1.8% in 2010, 1.9% in 2016

 If corrected, the number for lifetime risk becomes 95%  (.93 x 1.019)

 

From amjmed.com:

ABSTRACTBACKGROUND:News media and policy makers frequently discuss deaths from firearms, drug overdoses, and motor vehicle accidents. However, this information is generally presented as absolute numbers or annual rates. Cumulative lifetime risk may be an additional useful metric for understanding the impact of these causes of death.

METHODS:Data on all-cause firearm, drug overdose, and motor vehicle accident deaths were obtained from the US Centers for Disease Control and Prevention (CDC) for the year 2018. Age-specific death rates were used to estimate the cumulative risk of firearm, drug overdose, and motor vehicle accident deaths from birth to age 85 after accounting for other causes of death.

RESULTS:The lifetime risk of death from firearms, drug overdoses, and motor vehicle accidents was 0.93%(95% confidence interval [CI], 0.92%-0.94%), 1.52% (95% CI, 1.51%-1.53%), and 0.92% (95% CI,0.91%-0.93%), respectively. Black males had a 2.61% (95% CI, 2.55%-2.66%) lifetime risk of firearm death, indicating that 1 out of 38 black males will die from firearms if current death rates persist. Residents of West Virginia had a 3.54% lifetime risk of drug overdose death, equivalent to 1 out of every 28 residents dying from overdoses.

CONCLUSIONS:The lifetime risk of death from firearms, drug overdoses, and motor vehicle accidents is substantial and varies greatly across demographic subgroups and states

1 comment:

One of Many said...

HUMM Peter Principle? Doctors do cause the death of how many each year by screw up's?