Durham County

Gunshot survival rate data flawed, says Duke prof. Here’s why there’s been no dramatic improvement.

Surgeons at Harbor-UCLA Medical Center operate on a gunshot victim. A new study by a Duke professor rebuts the belief that more gunshot victims are surviving due to advances in medical care.
Surgeons at Harbor-UCLA Medical Center operate on a gunshot victim. A new study by a Duke professor rebuts the belief that more gunshot victims are surviving due to advances in medical care. TNS

A Duke University professor may have just sent disruptive ripples through academia’s world of public health.

Philip Cook, ITT/Terry Sanford Professor of Public Policy Studies, says the survival rate of U.S. gunshot victims did not dramatically improve over a decade-long period in the 21st century.

Cook is the lead author of “Constant Lethality of Gunshot Injuries From Firearm Assault: United States, 2003-2012,” a study published online June 22 in the American Journal of Public Health.

Cook and co-authors Garen Wintemute, Ariadne E. Rivera-Aguirre and Magdalena Cerdá say new research, based on recalculations and new analysis of how data was collected and coded in a public health survey, runs counter to and corrects a previously wide-held supposition.

A purported increase in gunshot survival rates between 2003 and 2012 has been credited to improvements in emergency and medical treatment of critically injured patients.

This survival rate was based on data indicating more people were getting shot, transported to a hospital and receiving treatment for gunshots at the hospitals, correlated with data indicating homicide numbers simultaneously remained steady.

The notion that more people were getting shot without dying led to a perception that medical practitioners were getting better at saving the lives of shooting victims.

But, Cook said, “Our analysis shows that neither the estimate of nonfatal shootings nor the fatality rate changed much in the decade after 2003.”

Cook and his co-authors examined statistics collected between 2003 and 2012, from the NEISS Firearms Injury Surveillance Study.

The data in the new study came from a sample of around 90 hospitals with six or more beds that provided 24-hour emergency care.

After re-analyzing the numbers and adjusting for the data reporting problems, Cook and his team said the purported increase in nonfatal gun assaults vanished because of two primary reasons.

Firstly, two hospitals that participated in the NEISS survey at its beginning, dropped out of the survey before its completion and were replaced by facilities that provided disproportionate data.

Secondly, Cook and his team found that in the earliest years surveyed in the study, gunshot cases were incorrectly classified as “unknowns,” rather than being correctly classified as what they were – “gun assaults.” Over the years, the longer the survey went on, gun assaults were more often classified correctly.

This means, more and more cases were classified correctly as “gun assaults” – rather than “unknowns” – but, without fixing the earlier coding, the data falsely appeared to show that more and more people were getting shot.

“As a result, it created an illusion that there were many more nonfatal gunshot cases than before,” Cook said.

Previous analyses interpreted the same data to mean nonfatal assaults had increased by 49 percent between 2003 and 2012.

Such a substantial increase is tremendously important in terms of researchers’, writers’ and policy makers’ understandings of violent crime in America and the nation’s power to care for trauma patients.

In 2015, the last year for which a number was readily available, 198 people in Durham were shot or suffered shooting related injuries, more than double the 95 shooting victims in 2014.

Study co-author Wintemute is a practicing emergency physician, professor of emergency medicine at UC-Davis and the director of the UC-Davis Violence Prevention Research Program.

“The idea that there was a sharp reduction in the death rate failed a basic test of plausibility,” Wintemute said. “Absolutely, there have been major improvements in emergency medical care and delivery, but they have taken place over decades. In the short run, they simply could not account for the large change.”

The study produced “readjusted” data estimating there were nationally 41,874 nonfatal gun assaults in 2003 and estimated there were 41,996 nonfatal gunshot assaults nationally in 2012.

With 11,920 homicides reportedly having occurred in 2003 and 11,622 homicides in 2012, that equals a national fatality rate which held steady at around 22 percent over the 10-year span.

The Duke professor’s new study is a response to articles in the Wall Street Journal and in the public health literature that talk about the hidden epidemic of gun violence, Cook said, “there is no hidden epidemic during that period.”

Taking into account that U.S. population increased by 8.3 percent during surveyed decade, Cook added, the rate at which Americans were shot annually declined by 7.8 percent with 18.54 Americans per 100,000 Americans shot in 2003 and 17.07 per 100,000 in 2012.

Colin Warren-Hicks: 919-419-6636, @CWarrenHicks