ON A THURSDAY in late November, the entire city of New Orleans recoiled in shared horror at security video of a young medical student collapsed on the sidewalk just off Magazine Street, clutching his bleeding torso as a hooded assailant stood over him with a gun aimed at his head. The film's silence only amplified the menace as the gunman apparently tried to squeeze the trigger, twice, to finish off his already-incapacitated victim, giving up only when a mechanical mercy intervened and the gun refused to fire.

  Two nights later, Bunny Friend Park in the 9th Ward — its almost comically benign name a memorial to a teen who died in an accident in the 1920s — became the scene of the city's next headline-grabbing gun battle. A block party and planned music-video shoot were rent apart by a hail of gunfire, leaving 17 people wounded. At least a half dozen people have been named as suspects as investigators try to piece together how the celebration turned to chaos.

  The bloodshed continued the following weekend, when more young men were killed at some of the city's best-known places: 26-year-old Brandon Robinson on Bourbon Street, 19-year-old Richard Dowell on Canal Street and 19-year-old Devin Johnson near the newly opened Lafitte Greenway.

  Despite those incidents and others, city officials continue to insist they've made significant strides in the struggle against violent crime in New Orleans in recent years. Many measurements — as well as newly published academic studies — back up those statements. But if things are getting better, why does carnage in New Orleans still intrude onto playgrounds, green spaces and tourist thoroughfares? If the violence is the work of a relatively small number of people, why are they so hard to stop?

  The answer to all those questions, crime experts say, may lie in a relatively new understanding of violence as a disease — not only in a rhetorical sense, but in the mechanism by which it festers and spreads through a community like an infection moves through the body. Those experts say the only way to treat the terrifying symptoms being played out in public view is to treat the disease at its origins.

For many New Orleanians, violence is a health threat that claims more lives than most diseases. Homicide is the fourth-highest cause of death in this city each year, behind cancer, heart disease and accidents, according to the Centers for Disease Control and Prevention (CDC). In the U.S. as a whole, homicide as a cause of death doesn't even rank in the top 15. Homicide deaths in New Orleans largely comprise black victims, and for African-Americans between the ages of 5 and 44 — 45 percent of New Orleans' population — homicide is by far the leading cause of death, CDC statistics showed.

  Those killings are concentrated in Central City, Hollygrove, Treme, the 7th Ward, New Orleans East and parts of Algiers. This pattern, a handful of neighborhoods with homicide problems that far outpace the rest of the city, is repeated in urban areas across the country, said Charles West, who leads the new anti-violence initiatives as the head of Mayor Mitch Landrieu's Innovation Delivery Team.

  "There are about three or four neighborhoods in New Orleans that we refer to as our 'hotspots' that have significantly higher incidents of violence," West said. "But if you look at every major city, even those that have much lower murder rates, there are about three or four neighborhoods that have incidents of murders that are about four times that city's average."

  Those neighborhoods in New Orleans all share the same factors: an unemployment rate that is twice as high as the rest of the city, educational attainment that's half the city's average, a high number of babies with low birth weights and high infant mortality rates. "All of the same things you look at from a public health standpoint," West said. "You overlay them, and they exactly overlap with these neighborhoods."

  When such boundaries can be drawn, it becomes easy for the rest of the city's residents to view violence as a neighborhood's problem rather than a citywide issue. Criminals, however are stepping beyond the boundaries. In late October — midway between several brazen Uptown restaurant robberies and the latest outbreak of headline-grabbing bloodshed — two businesses on S. Claiborne Avenue, a barbershop and daiquiri bar, were robbed at gunpoint by masked men. Even after the suspects were arrested, workers at those businesses declined to discuss their ordeals for fear of retribution from the perpetrators.

  Although the crimes were strikingly similar — the businesses were robbed during operating hours and patrons were forced to the floor — there were no mayoral news conferences or New Orleans City Council hearings in response to the barbershop and daiquiri bar robberies like there were for the restaurant holdups. No mounted patrols were assigned to S. Claiborne Avenue and no news vans or cameras lined the street. The disparity in public attention based on a crime's location is a "medieval" approach to the problem, said state Sen. J.P. Morrell, comparing it to public views that prompted leper colonies: People in "safe" areas might pity those near the source of infection, but their primary hope is that the disease doesn't spread to them.

  "For communities that are struggling with violence and a violent culture, the lack of police presence and a lack of resources make it difficult to overcome what already exists," Morrell said. "When you talk to people from these communities, the common thing you always hear is 'The city has forgotten about us.' That hopelessness breeds that perpetual ongoing resurgence of crime."

  That lack of shared investment in all crimes can perpetuate a system of street justice and retaliation that fuels continuing violence and creates victims who don't trust the "system" to protect them. Researchers describe that fear as an expression of a lack of "police legitimacy" — the law-enforcement system has less sway in some communities than does the constant cycle of street justice.

  In 2013, New Orleans made a significant dent in violence, dropping from just under 200 homicides in 2012 to 155 in 2013 — a figure that held steady in 2014 and is roughly on pace to do so in 2015. According to research published by two University of Cincinnati criminologists in the August issue of Criminology & Public Policy, that 20 percent decrease has a clear and statistically significant cause: adoption of the Group Violence Reduction Strategy, a new form of targeted intervention specifically against the individuals involved in a crime. The strategy that has been successful in a number of urban areas.

  The new concept begins with intelligence gathering. New Orleans Police Department (NOPD) officers identified 59 street-level neighborhood gangs totaling 600 or 700 members. Previous reviews showed 2 percent or fewer homicides were gang-related, but the new concept identified 55 percent of homicides as involving gangs. Police detectives then compiled information about people involved in violent crimes — their relatives, companions, social media interactions, what cars they use, if they are connected to any guns, and what shootings they've witnesses. Officers then construct maps of social networks among people involved in violence.

  Authorities summon individuals invol- ved in the most active gangs to a courtroom — usually 10 to 15 already in jail and about 30 on parole or probation — and tell them how much the police know about their networks.

  Representatives from the law enforcement community — police brass and detectives, state and federal prosecutors, the mayor — sit on one side of the courtroom; representatives from nonprofits and education agencies sit on the other. They tell offenders that if they or their associates are involved in any additional violence, their entire group could become the focus of organized crime investigations that, they are told, have been successful in putting some of the city's most violent offenders behind bars. If they want out of the gang lifestyle and into the city's job-training programs, they are told, they can be moved to the top of the list.

  "The message we deliver is that we know who you are. We know who you hang out with. We know that your group has committed acts of violence and we're here today to offer you an opportunity to put the guns down," explained NOPD Commander Frank Young, who heads the department's anti-gang units. "We will help you if you let us. We will stop you if you don't."

  The concept is based on research that shows the most effective deterrent to crime is the perceived probability of being caught, the Cincinnati study said, and when gang members learn how much the police know about them, the effect tends to be immediate.

  During the 18 months be tween October 2012 and March 2014, NOPD held five such call-ins with 158 offenders from 54 gangs. (Nearly 60 of them signed up for job training services, though only 25 followed through.) Meanwhile, gang unit investigations have resulted in more than 120 organized crime indictments against street gang members.

  The Cincinnati study concluded that as a direct result of that pressure, "the City of New Orleans experienced a statistically significant homicide rate decline above the average homicide rate change for the 14 highly comparable cities ... a 23 percent homicide rate decline that was specific and unique to New Orleans." The reduction was concentrated in gang member-involved (GMI) homicides, which fell 30 percent during the study period. Non-GMI homicides such as domestic cases or random acts by otherwise nonviolent people) fell only 10 percent. Likewise, "homicides involving black male victims between the ages of 20 and 29 years old experienced a statistically significant decline of 26.7 percent," according to the study. The decline in homicides also corresponded to a similar decrease in nonlethal gun violence the same year, researchers found.

  Young, whose 20-year career includes working with investigation units and stints with the FBI and the Drug Enforcement Agency (DEA), said the study's conclusions matched his own observations about 2013 as a turning point. When the gang units he now leads were created, Young was still assigned to investigations in the 6th District, where many of the most violent gangs were located and could see immediate effects from the units' work.

  "The same people getting shot were the perpetrators," he said. "The same witnesses would then become suspects in other crimes. It was a very small number of people that were responsible for the majority of the crime. I got to watch as these big groups got indicted, and I was the guy that watched all these numbers spiral downward."

  In most years, changes in homicide numbers in New Orleans are similar to variations in overall violent crime and property crime. Therefore, researchers say, increased scrutiny of the most violent offenders seems to be working.

If progress was so clear in 2013, why aren't murder numbers continuing to fall?

  Some law enforcement officers blame a decline in manpower: Only 77 new officers joined the NOPD between 2011 and 2014, while more than 100 left the force each of those years. The study, however, said factors contributing to violence in New Orleans make the city "the most challenging of contexts" for reducing crime, because of its "extremely high murder rates, political and police corruption and a local culture seemingly more tolerant of violence." Police and city officials agree that law enforcement's ability to prevent violence is limited.

  "Children that grow up in a household with family violence themselves often end up as perpetrators later in life," West said. "It is a learned behavior. ... These children, when faced with conflict, behave in the way they have seen."

  Much of the long-term effort to correct problems that incubate violence has focused on the dismal unemployment rate for African-American men, pegged at 52 percent in a 2013 study by Loyola University New Orleans. The City Council requires developers to hire local workers, and the city has a job training and placement support program for people with criminal records. West said more than 100 people have successfully completed that program.

  The large income disparity among the city's residents and neighborhoods makes finding solutions problematic, Morrell said. The wealthy can afford to buy additional security by hiring private guards funded by taxing districts, he says, while the middle class, which used to live adjacent to the crime and was empowered enough to demand officials pay attention to it — no longer exists.

  "We have a problem in this city with opportunity," Morrell said. "There is no black middle class in New Orleans anymore at all. You're either affluent or you're poor. That middle class that would be the cauldron of outrage has dwindled."

Will the public health approach be the key to winning New Orleans' interminable struggle with violent crime?

  Dr. Andre Perry, who has documented striking differences in the life expectancies among people living in different New Orleans ZIP codes, said the multiplicity of factors that give rise to violence in a community bear a resemblance to "some organic system working in concert." A public health approach, he said, has the benefit of a long-term focus on root causes of violence rather than reducing it to a policing or jailing issue.

  "If you give people low opportunities, low education, this is going to happen," Perry said. "The real consequence of not providing a just and equitable system is a violent system."

  The success of the public health approach ultimately will be measured less by what law enforcement does and more by what structural changes can be orchestrated, Perry said, adding that individual programs for job training, diploma acquisition and the like are only a start.

  "It's needed, but it's not sufficient," Perry said. "In New Orleans, it's a much deeper structural problem around employment. We need to further reduce the prison system. Education needs to continue to improve over a longer period of time. We need greater structural change in order to see the kind of reduction that will serve residents well."