The .45-caliber pistol that killed Lucas Heagren, 3, on Memorial Day last year at his Ohio home had been temporarily hidden under the couch by his father. But Lucas found it and shot himself through the right eye. "It's bad," his mother told the 911 dispatcher. "It's really bad."
A few days later in Georgia, Cassie Culpepper, 11, was riding in the back of a pickup with her 12-year-old brother and two other children. Her brother started playing with a pistol his father had lent him to scare coyotes. Believing he had removed all the bullets, he pointed the pistol at his sister and squeezed the trigger. It fired, and blood poured from Cassie's mouth.
Just a few weeks earlier, in Houston, a group of youths found a Glock pistol in an apartment closet while searching for snack money. A 15-year-old boy was handling the gun when it went off. Alex Whitfield, who had just turned 11, was struck. A relative found the bullet in his ashes from the funeral home.
Cases like these are among the most gut-wrenching of gun deaths. Children shot accidentally — usually by other children — are collateral casualties of the accessibility of guns in America, their deaths all the more devastating for being eminently preventable.
They die in the households of police officers and drug dealers, in broken homes and close-knit families, on rural farms and in city apartments. Some adults whose guns were used had tried to store them safely; others were grossly negligent. Still others pulled the trigger themselves, accidentally fracturing their own families while cleaning a pistol or hunting.
And there are far more of these innocent victims than official records show.
A New York Times review of hundreds of child firearm deaths found that accidental shootings occurred roughly twice as often as the records indicate, because of idiosyncrasies in how such deaths are classified by the authorities. The killings of Lucas, Cassie and Alex, for instance, were not recorded as accidents. Nor were more than half of the 259 accidental firearm deaths of children younger than 15 identified by the New York Times in eight states where records were available.
As a result, scores of accidental killings are not reflected in the official statistics that have framed the debate over how to protect children from guns. The National Rifle Association cited the lower official numbers this year in a fact sheet opposing "safe storage" laws, saying children were more likely to be killed by falls, poisoning or environmental factors — an incorrect assertion if the actual number of accidental firearm deaths is significantly higher.
In all, fewer than 20 states have enacted laws to hold adults criminally liable if they fail to store guns safely. Legislative and other efforts to promote the development of childproof weapons using "smart gun" technology have similarly stalled. Technical issues have been an obstacle, but so have NRA arguments that the problem is relatively insignificant and the technology unneeded.
Compiling a complete census of accidental gun deaths of children is difficult, because most states do not consider death certificate data a matter of public record. In a handful of states, however, the information is publicly available. Using these death records as a guide, along with hundreds of medical examiner and coroner reports and police files, the New York Times sought to identify every accidental firearm death of a child age 14 and younger in Georgia, Minnesota, North Carolina and Ohio dating to 1999, and in California to 2007. Records were also obtained from several county medical examiners' offices in Florida, Illinois and Texas.
The goal, in the end, was an in-depth portrait of accidental firearm deaths of children, one that would shed light on how such killings occur and might be prevented. In all, the newspaper cataloged 259 gun accidents that killed children ages 14 and younger. The youngest was just 9 months old, shot in his crib.
In four of the five states — California, Georgia, North Carolina and Ohio — the newspaper identified roughly twice as many accidental killings as were tallied in the federal data. In the fifth, Minnesota, there were 50 percent more accidental gun deaths.
The undercount stems from the peculiarities by which medical examiners and coroners make their "manner of death" rulings. These pronouncements, along with other information entered on death certificates, are the basis for the nation's mortality statistics, which are assembled by the National Center for Health Statistics, a division of the Centers for Disease Control and Prevention. Choosing among five options — homicide, accidental, suicide, natural or undetermined — most medical examiners and coroners call any death in which one person shoots another a homicide.
"A homicide just means they died at the hands of another," said Dr. Randy Hanzlick, the medical examiner for Fulton County, Ga. "It doesn't really connote there's an intent to kill."
To get more accurate information about firearm deaths, researchers have pushed for the expansion of the National Violent Death Reporting System. The effort first started in the 1990s at the CDC but was shut down shortly afterward when Congress, at the urging of the NRA, blocked firearms-related research at the centers. The project was revived in 2002 after researchers decided to expand its scope beyond guns, but it is up and running in only 18 states. President Barack Obama has called for increased financing for the program, part of a package of gun-related proposals made after the school shooting in Newtown, Conn., in December.
Another important aspect of firearm accidents is that a vast majority of victims do not die. Tracking these injuries nationally, however, is arguably just as problematic as tallying fatalities, according to public health researchers.
The largest number of deaths came at the upper end of the age range, with ages 13 and 14 being most common — not necessarily surprising, given that parents generally allow adolescents greater access to guns. But the third-most common age was 3 (tied with 12), a particularly vulnerable age, when children are curious and old enough to manipulate a firearm but ignorant of the dangers.
About a quarter of the victims shot themselves, with younger children again especially susceptible. More than half of the self-inflicted shootings involved children 5 or younger; the most common age was 3.