Nonfatal Firearm Injury Costs

Welcome back. If you follow or occasionally visit this blog, you may be aware that I’ve written about firearms--four times to pitch the need for gun research, not gun control.

As I described in the first of my four posts (see Gun Research), Congress essentially shut down gun research funding by the Centers for Disease Control and Prevention in 1996 and later by the National Institutes of Health. Although research continued without federal support, it was estimated that gun violence research funding and publications were less than 5% of what would have expected based on the statistics for other leading causes of death.

I am happy to note that, in December 2019, Congress changed its collective mind. After more than 20 years, Congress approved funding of $25 million for fiscal 2022 to be split between the CDC and NIH to examine gun violence from a public health perspective. While experts concede the amount is small in comparison to the scope of the issue, they celebrated.

Congress approved fiscal year 2022 funding for the CDC and NIH to examine gun violence (graphic from abcnews.go.com/Politics/decades-long-gap-gun-violence-research-funding-lasting/story?id=80646946).

Which brings me to an example of the gun research that NIH funded.

Long-Term, Often Hidden Costs of Firearm Injuries
Some 45,000 people are killed in the U.S. by firearms each year; more than twice as many sustain firearm injuries and survive.

A team of researchers set out to measure the first year’s changes in clinical and economic outcomes for survivors and their family members after nonfatal firearm injury. These typically include worse mental health, substance use disorders and higher health care spending.

The researchers were affiliated with Harvard Medical School, Massachusetts General Hospital, Harvard T.H. Chan School of Public Health, Harvard University, Michigan University and Boston Health Care for the Homeless Program.

Data for Analysis
The researchers analyzed 10 years of data, 2008-2018, from IBM MarketScan commercial and Medicare claims databases--nationwide samples of employer-sponsored commercial insurance enrollees and Medicare beneficiaries with employer-sponsored Medicare supplemental coverage. The databases contain detailed claims data for more than 40 million persons annually.

They identified two cohorts: survivors of firearm injury and the survivors’ family members, all of whom had been enrolled in the insurance coverage for at least 1 year before through 1 year after the firearm injury. For control, each cohort member was matched to five unexposed participants.

The final dataset comprised 6,498 survivors and 32,490 matched control participants, along with 12,489 family members and 62,445 matched control participants.

About 71% of nonfatal firearm injuries were unintentional; 21% were assaults and 4% self-harm.

Characteristics of firearm injury survivors and their matched control participants. DxCG – Diagnostic Cost Groups risk score is a measure of health status or expected spending; larger values denote greater expected spending. Consumer directed health plans -- preferred-provider organization plans with health reimbursement funded by employer (from Table 1 of www.acpjournals.org/doi/10.7326/M21-2812).

Study Findings
In the first year after injury, the survivors’ medical spending increased 402% per person per month and cost sharing (co-pays and deductibles) increased 176% per person per month over the control participants’ spending. The increase was driven by the first month’s spending of 4,122%  and 1,917% in cost sharing per survivor.

The cost of all categories of the survivors’ health care increased relative to the control participants--40% increase in pain diagnoses, 51% increase in psychiatric disorders and 85% increase in substance-use disorders, in addition to increased pain and psychiatric medications. Family members had a 12% increase in psychiatric disorders relative to their control participants.

These clinical and economic changes were driven by intentional and more severe firearm injuries.

Changes in study outcomes for 1 year after unintentional and intentional nonfatal firearm injuries; intentional firearm injuries include assault, self-harm and law-enforcement shootings (from Table 3 of www.acpjournals.org/doi/10.7326/M21-2812).

Wrap Up
Nonfatal firearm injuries cause survivors, their families, employers, insurers and society to spend thousands of dollars more each month on healthcare. Considering the annual number of gunshot survivors in the U.S., the researchers estimate direct health care spending to be well over $2 billion for just the first year.

The researchers suggest that it might be wise for doctors to screen gunshot survivors and their family members for signs of mental health problems and to be mindful of an increased risk of substance use disorders when treating pain in the wake of a shooting.

Interesting stuff when Congress allows gun research. Thanks for stopping by.

P.S.
CDC tally of firearm deaths: www.cdc.gov/nchs/fastats/injury.htm
Congressional turnabout on funding CDC and NIH gun violence research:
www.apa.org/monitor/2021/04/news-funding-gun-research
www.usatoday.com/story/news/nation/2020/02/09/gun-violence-how-researchers-spend-25-m-gun-safety-funding/4464121002/
Study of nonfatal firearm injury outcomes in Annals of Internal Medicine journal: www.acpjournals.org/doi/10.7326/M21-2812
Article on study on EurekAlert! website: www.eurekalert.org/news-releases/948335

Fatal Police Shootings

Welcome back. Living in the D.C. area for more than 20 years, I was a regular reader of The Washington Post. I still see selected articles, but most of the content goes by. I wasn’t aware, for example, that the newspaper started logging every fatal shooting in the U.S. by an on-duty police officer.

The Post began that project after its investigation of the 2014 shooting of Michael Brown, Jr., in Ferguson, Mo., found the FBI undercounted fatal police shootings by more than half. Reporting by police departments is voluntary and many don’t.

The Post’s “Fatal Force” database begins on 1 January 2015; relies primarily on news accounts, social media postings and police reports; is updated as new information is obtained; and is searchable by state, gender, race, age, mental illness, weapon, body camera, fleeing the scene, year, as well as name.

The Washington Post’s database of fatal shootings by on-duty police officers (from www.washingtonpost.com/graphics/investigations/police-shootings-database/).

Although The Washington Post makes the database available publicly, a team of medical researchers affiliated with Penn, Yale and Drexel universities judged it was critical that fatal police shootings of Black, Indigenous and People of Color (BIPOC) be recognized and treated as a public health emergency. Toward that end, they decided to enter the data into the scientific literature and present it using methods that are accepted by science as rigorous and robust.

Relative Rates of Fatal Police Shootings
For their recently published study, the researchers conducted a longitudinal analysis of 4,653 of the 5,367 fatal police shootings listed through May 2020, omitting those lacking race/ethnicity or age details.

Using generalized linear-mixed models to capture trends with time and rates relative to respective populations, they found shootings of BIPOC, whether armed or unarmed, were essentially constant from 2015 to 2020 and significantly higher than that of Whites. 

Fatal U.S. police shootings by ethnicity, 1 Jan 2015–14 July 2020; graphic’s end date is several weeks after that of study (The Washington Post data; www.statista.com/chart/21857/people-killed-in-police-shootings-in-the-us/).

For armed victims, the rates that Native Americans, Blacks and Hispanics were killed were, respectively, 3 times, 2.6 times and 1.3 times higher than the rate Whites were killed. For unarmed victims, the rates that Blacks and Hispanics were killed were, respectively, 3 times and 1.4 times the rate of Whites. The average age of Whites killed was 38; Native Americans, Blacks and Hispanics were younger, 31, 30 and 33, respectively.

Years of Life Lost
The researchers also calculated the estimated years of life lost by race/ethnic group. Basing the estimates on national historical life expectancy data for U.S. citizens in the victim's birth year vs. age at death, they found an average 31,960 years of life lost annually due to police shootings. Relative to Whites, the years of life lost were 4 times higher for Native Americans, 3.3 and 3.5 times higher for Blacks overall and unarmed, and 1.6 times higher for Hispanics overall and unarmed.

Another, more difficult factor to quantify is that Blacks report worse mental health in areas where there are police killings.

A center-of-the-road memorial to Michael Brown, Jr., for the 5th anniversary of his fatal shooting by a police officer, 9 Aug 2014; the neighborhood memorial was in place on 8 Aug and will be guarded overnight by the man in the folding chair (photo by Robert Cohen for St. Louis Dispatch, rcohen@post-dispatch.com).

Wrap Up
Why should the number of fatal police shootings be almost the same, nearly 1,000, every year since the Post started its tally? The database website points to probability theory for a possible explanation. In essence, the quantity of rare events in huge populations tends to remain stable absent major societal changes. For police shootings, that change could involve a fundamental shift in police culture or restrictions on gun ownership.

Treating the shootings as a public health emergency, the researchers note that what has been done at the local level--body cameras and independent investigations--is insufficient. It must be raised to the state and national level and codified into law.

So, should it be more Law and Order, more Black Lives Matter or…well, what do you suggest? Thanks for stopping by.

P.S.
The Washington Post “Fatal Force” website: www.washingtonpost.com/graphics/investigations/police-shootings-database/
Study of fatal police shootings in Jour. of Epidemiology & Community Health: jech.bmj.com/content/early/2020/10/20/jech-2020-215097
Articles on study on EurekAlert! and Yale University websites:
www.eurekalert.org/pub_releases/2020-10/b-fps102320.php
news.yale.edu/2020/10/27/racial-disparity-police-shootings-unchanged-over-5-years
CDC’s definitions of years of potential life lost: www.cdc.gov/injury/wisqars/fatal_help/definitions_ypll.html