And Then You're Dead

The Autopsy / Case File · Violence & Homicide

Interrupted, Then Defunded

Community violence intervention programs measurably reduced homicides in the cities that ran them long enough to matter. Then the grants that paid for them got canceled.

Community violence intervention sends trained outreach workers — often people with their own history in the neighborhoods they now work, sometimes formerly incarcerated themselves — to mediate conflicts before they turn into shootings. It is not a metaphor and not a slogan; it's a specific, evaluated intervention with a measurable effect on the homicide count in the cities that funded it consistently. In April 2025, the Department of Justice canceled roughly $820 million in public-safety grants nationwide, including the largest single funding stream keeping those programs running.

And Then You're Dead · September 2026 · 8 min read

Join the discussion →
Boarded-up rowhouses on East Preston Street in East Baltimore, Maryland
East Preston Street, East Baltimore — the kind of block community violence intervention programs like Safe Streets are built to work, block by block, conflict by conflict.

Community violence intervention, or CVI, is built on a specific theory: that a large share of urban gun violence follows predictable social patterns — retaliation cycles, ongoing disputes, status conflicts — that can be interrupted by a credible messenger before the next shooting happens. That messenger is usually someone the neighborhood already trusts, someone who has often lived through the same violence they're now trying to stop, deployed to mediate a specific beef before it escalates and to connect people at the highest risk of shooting or being shot with services that might change their trajectory. It is not policing. It is, by design, the opposite of policing — outreach workers explicitly don't carry badges and don't report to law enforcement, because the entire model depends on being trusted by people who would never talk to a police officer.

What The Baltimore Numbers Actually Show

The longest-running American test of this model is Safe Streets Baltimore, launched in 2007 and modeled directly on Chicago's CeaseFire program. A team of researchers at the Johns Hopkins Center for Gun Violence Solutions — Carla Tilchin, Elizabeth Wager, Mitchell Doucette, Bijan Niknam, Shani Buggs, Michael Desjardins, and Daniel Webster — published the most comprehensive evaluation of it to date, examining outcomes across 11 Safe Streets sites operating between 2007 and 2023.

0%
reduction in homicides involving youth ages 15–24 associated with Safe Streets at the neighborhood level, per the Johns Hopkins Center for Gun Violence Solutions evaluation
0%
reduction in youth nonfatal shootings associated with the same program, over the same neighborhood-level analysis

The researchers were careful about the limits of their own findings: results varied considerably by site, and at several individual locations the reductions weren't statistically significant on their own — a common outcome in evaluations built on relatively small numbers of shootings per neighborhood per year. But the aggregate, city-wide pattern held across a period that included Baltimore's worst homicide years on record, and it tracks with what happened afterward: the city's homicide count fell 20% in 2023 and another 23% in 2024, a two-year decline city officials and outside researchers have credited in part to sustained investment in Safe Streets alongside other violence-reduction strategies.

Lexington, Kentucky, offers a smaller but similarly documented pattern: homicides in the city fell from 44 in 2022 to 22 in 2024, a decline city leaders have tied to expanded community violence intervention outreach launched in the interim.

Where The Model Came From

Safe Streets was built on the framework established a decade earlier by CeaseFire Chicago, now known nationally as Cure Violence. The program's first rigorous outside evaluation, led by Northwestern University's Wesley Skogan and commissioned by the National Institute of Justice, examined seven program sites using three separate methods — time-series analysis, hot-spot crime mapping, and social network analysis of gang-related killings.

The 2008 evaluation found that violence declined noticeably in six of the seven CeaseFire sites studied, and that in four of them, researchers could specifically link the drop in shooting hot-spot intensity to the program's introduction rather than to chance or to broader citywide crime trends. In the remaining two sites, shootings declined too, but the evaluators couldn't establish with confidence that CeaseFire itself was the cause — an honest methodological limit that has shaped how CVI research has been designed ever since, with an emphasis on neighborhood-level comparison groups rather than simple before-and-after counts.

The Skogan evaluation and the Safe Streets evaluation, published fifteen years apart, reach the same structural finding: the effect is real, it is measurable, and it is not uniform — a program that works in one neighborhood for one set of reasons doesn't automatically work the same way three blocks over.

The Cost-Benefit Case Researchers Have Made

Beyond the mortality data, researchers at Johns Hopkins have separately estimated the economic return on this kind of intervention, putting the benefit at roughly seven to nineteen dollars in avoided costs — medical spending, lost productivity, criminal justice processing — for every dollar spent on community violence intervention programming. That range is wide because it depends heavily on which categories of downstream cost get counted and which city's data the estimate is built from, but even the low end of the range describes an intervention that pays for itself several times over if the underlying homicide-reduction effect holds.

The Funding Cliff

None of that evidence base stopped the Department of Justice from canceling the grants that funded much of this work. In April 2025, DOJ terminated roughly $820 million in previously awarded grants, affecting more than 550 organizations across 48 states and territories. Community-based safety grants — the funding category that includes most CVI programs specifically — accounted for about $169 million of those cuts.

$0M
total DOJ grant funding canceled nationwide in April 2025, spanning more than 550 organizations in 48 states and territories
$0M
portion of that cut specifically targeting community-based safety grants, the funding category covering most CVI outreach programs

The cuts hit specific, named organizations immediately. In Newark, New Jersey, a $5 million reduction across four grants forced the Newark Community Street Team — the city's primary violence-interruption outreach organization — to lay off roughly a dozen staff members, according to Brennan Center for Justice reporting on the cuts. The Department of Justice's proposed fiscal year 2026 budget goes further still, proposing the outright elimination of the community violence intervention funding line altogether, rather than a reduction within it.

The programs that survived the April 2025 cuts are now operating the way syringe service programs and drug-checking labs often do — patched together from state opioid-settlement-adjacent public safety dollars, philanthropic grants, and city general funds, none of which offer the multi-year funding stability that lets an outreach worker build the years-long trust relationships the model depends on. A violence interrupter's effectiveness rests substantially on being a known, consistent presence in a specific neighborhood over time; a program that has to rebuild its staff and its relationships every time a grant cycle lapses is not running the same intervention the Baltimore and Chicago evaluations actually measured.

And Then You're Dead

A 42% reduction in youth homicide isn't a hypothesis anymore — it's a measured outcome, published, peer-reviewed, sitting next to a homicide count that kept falling for two more years after the study period ended. None of that made the program durable. The grant got canceled the same spring the evidence for it was strongest, and the next shooting that outreach worker might have stopped is now, in a lot of cities, nobody's job to prevent.

That's it. That's the whole thing.

Comments

Loading…

Comments are anonymous and moderated.

Sources

Photo: "Rowhouses, 2504–2510 E. Preston Street, Baltimore, MD 21213" by Eli Pousson/Baltimore Heritage, via Wikimedia Commons (CC0 1.0), cropped from the original.