LEXINGTON, Ky. – When gunshots make national news, Mark Bryant's phone rings.

Bryant, 62, is neither a law enforcement officer nor a trauma specialist. He runs a private website, Gun Violence Archive, that updates hourly, with street-level details, most of the gun-related incidents that have occurred in the United States since 2013.

Want to know how many people have been killed by guns so far this year? In your state? In your city? Last year? The year before that? The number of people wounded? How many victims were children? How many mass shootings occurred? Police-related shootings? How many times were guns used in self-defense?

Operated out of a small home just off Richmond Road, Bryant's archive answers such questions for journalists, policymakers, even law enforcement. And despite the public safety menace of gun violence in this country, few others do this kind of work.

Typically, the FBI undercounts shootings and the U.S. Centers for Disease Control and Prevention undercounts gun deaths because they rely on incomplete reports passed along by local officials and extrapolated surveys. Academic study of gun violence slowed nearly to a halt in 1996 once Congress, at the behest of the National Rifle Association, prohibited using federal funding for research that could be used to advocate for gun control.

"For firearms, we have rotten, absolutely rotten data," said Jon Vernick, co-director of the Center for Gun Policy and Research at Johns Hopkins University in Baltimore.

"Compare that to car crashes," Vernick said. "We collect all sorts of information about car crashes. I can tell you not only how many people died in your state last year in car crashes; I can break it down for you by the make and model of the vehicle, the speed it was traveling, the road conditions, the weather, the age and experience of the driver, on and on. But shootings? No. Nothing like that with shootings."

GVA (gunviolencearchive.org) comes closer than most. It has 19 researchers around the country to sweep information about gun-related incidents from the websites of more than 2,000 news organizations and police departments. (One researcher is devoted entirely to Chicago, which can see more than 100 people shot over a single holiday weekend.) Researchers follow up with phone calls and open records requests when necessary.

Incidents are promptly reviewed, categorized and posted, with one or more links to original sources to confirm their authenticity. There is no commentary; GVA is nonpartisan and takes no position on gun ownership or gun control. It simply provides the numbers.

Last year, according to GVA, there were 384 mass shootings in the United States, adopting the federal government's definition of "four or more people shot and/or killed in a single event." There were 671 children up to age 11 killed or wounded by guns, and 3,124 teenagers up to age 17 killed or wounded. There were 1,971 verified defensive uses of a firearm, which can include either brandishing a gun or shooting it. There were 2,198 unintentional shootings.

Overall, there were 15,063 fatal shootings, continuing an upward trend since GVA began counting, and 30,613 gun-related injuries. GVA doesn't track gun suicides.

On any given day, GVA gets about 20,000 page views. A major shooting — say, the Orlando nightclub rampage in June 2016 that left 49 people dead — can spike that to 1.2 million page views a day.

"I know a lot of guys … who are very smart, rational people on most issues. But when it comes to any kind of legislation about gun safety, all that goes out the window," Bryant said. "They have this visceral reaction that people are trying to take away their guns. Nobody is trying to take away their guns. They're just trying to make things a little bit safer for the rest of us."