Chicago’s per capita police spending has, officially, more than tripled since 1964. The city now employs about twice as many police officers per capita as the national average—markedly more than any other large city except Washington, D.C. The Chicago Police Department has attempted nearly every possible police intervention and reform. Meanwhile, many of Chicago’s segregated Black and brown neighborhoods continue to suffer from high rates of poverty and violence. In recent years, this violence has begun to spill over into the downtown business core and, as a result, to increasingly concern the city’s wealthy donor class.

Chicago’s mayoral race between Brandon Johnson and Paul Vallas is placing this reality in the national spotlight. Johnson, a progressive, has been calling for change by implementing a public health approach to safety. Vallas, who has often identified himself as a Republican and represents the most conservative edge of the Democratic Party, has—in contrast to Johnson—been calling for the expansion of existing police-centric safety paradigms.

The police model around which Vallas has built his campaign, with backing from Republican donors and the city’s incendiary police union, considers public safety as first and foremost a matter of “crime.” Crime has spun out of control, this framework tells us, because we have not yet spent enough money on police nor hired enough officers to patrol the city’s schools and streets. Consistent with this approach, Vallas’ central electoral strategy has been to cast Johnson as a delusional proponent of “defunding the police” for his suggestions that investing in a public health model for safety would yield better results than repeating the police-first policies that have ruled the city for decades while social services have been defunded and privatized.

I am a physician and public health and safety researcher. I have also spent a decade living and working as an ethnographer on Chicago’s South and West Sides. Like many of my friends and patients, I’ve been violently assaulted at gunpoint. I’ve spent nights in emergency departments with families devastated by shootings. I’ve been at bloodstained street corners the morning after. It is very clear to me, as it is to most voters in Chicago, that we need a mayor who will implement safety policy that works. So what does the evidence show works to actually make communities safe?

The Police Model of Safety

To answer that, we need to first acknowledge what doesn’t. Data indicate that additional police do not make a substantial impact on crime. A Washington Post analysis, for example, found no correlation between spending on police and crime rates. Studies have repeatedly shown that large increases in police funding over the last several decades have not meaningfully reduced crime nor have they increased the rate at which police solve serious crimes. Researchers found that, in the 1990s, funding from the COPS program, which provided $7.6 billion in federal grants to hire thousands of additional police officers, had “little to no effect on crime.” The United States Government Accountability Office own estimates indicate that additional police officers hired from 1993 to 2000 drove only 1.3 percent of the large decline in overall crime during that period.

Hiring more police has been shown to increase reports of police brutality. The US Department of Justice’s own statistics show that use-of-force complaints increase with the size of the police force. These complaints rarely lead to any meaningful consequence for officers involved, let alone systematic changes to address abuse perpetrated by police or cuts to police budgets. Prior to murdering George Floyd and then finally—in an exception to the norm—being held accountable by the law, Derick Chauvin had faced 17 misconduct claims, for example.

More police leads to more arrests, regardless of underlying conditions. A study of NYPD officers found that if you send more officers to a given neighborhood, they end up arresting more people—even if crime is falling. Recent research found that each marginal police officer makes an additional 7.3 arrests for nonviolent “quality of life” offenses. This is especially concerning given the criminogenic effects of jailing—that is, the way that being subjected to time in pretrial detention in jails increases the future likelihood that someone will commit crimes and be involved in violence. In light of this, more arrests for “quality of life” offenses increases the risk of violence in communities down the road.

The police model of safety that Vallas has made the centerpiece of his campaign is full of destructive feedback loops like this. More policing drives more incarceration, regardless of underlying conditions. Incarceration then causes and intensifies poverty. This worsens violence. Incarceration disrupts families and employment, worsening poverty for individuals, families, and whole communities. Research indicates that without mass incarceration, the number of people in poverty in the US would fall by as much as 20 percent.

This reactive paradigm for safety policy has been given trillions in US federal and state dollars since the 1970s. It has been the longest, biggest, and most expensive social experiment in the history of the United States.

So, what has it given us? The world’s largest system of incarceration. Far more violence than peer nations. Dramatically worse life expectancy, racial inequality, and quality of life compared to peer nations. A public trained to feel always fearful and unsafe, and to falsely believe crime is always rising, even when crime is in fact at historic lows. This fear and the public distrust—along with devotion to guns—it breeds makes everyone less safe.

Given these realities, are voters in Chicago finally ready for a change? If so, Brandon Johnson represents a chance to finally invest in people through programs that work rather than repeating the current police model that has already repeatedly failed for over 40 years to produce safety. Johnson has embraced a public health model for public safety backed up by actual evidence.

The Public Health Model of Safety

The public health approach to shared safety for communities begins from several key redefinitions to detach us from distortionary police narratives and to reattach us to obvious but long-neglected truths.

First, violence is not just a matter of interpersonal violence. To effectively stop criminal violence, we must also account for structural violence: poverty, unaffordable housing, unemployment, police violence, barriers to health care access, mass incarceration, etc—all of which kill and disable far more people each year than violent crime, and that also ultimately drive ongoing high rates of criminal violence.

Second, we must redefine “the public” of “public safety.” The police model of public safety has prioritized white middle- and upper-class groups while disregarding others. For reasons not only of justice but also just basic effectiveness at a population level, actual public safety policy must protect everyone—including poor, Black, Latiné, and other criminalized communities.

Third, safety can’t be just about crime. The biggest threats to safety are not in fact violent crime but instead what we can call “the social determinants of safety”: lack of health care and housing, overdose risk, economic insecurity, hopelessness and suicide, lack of consumer and environmental regulations, etc. We must address all of these factors as interrelated conditions if we are to build functional safety systems.

So, what defines the public health model for public safety? It follows basic epidemiological principles, such as the observation that, as in infectious disease epidemics, you can’t protect the public at large unless you protect each individual and community—starting from the most vulnerable. It subscribes to a “trickle-up principle”—that is, that safety must be built from the bottom-up, not top-down. When the poorest and most vulnerable benefit from focused support systems, everyone—including the wealthy—ultimately benefits from improved safety that has ripple effects throughout the whole of society, from education and workforce quality to economic and health-systems efficiency.

To implement a public health approach to safety requires prioritizing supportive investments in the communities suffering from the highest rates of insecurity and violence. In the context of US histories of racial segregation and exclusion that carries long-term consequences, public safety depends upon “intensive care”—not more police and punishment––for our most dispossessed communities to repair the harm that the ongoing legacies of slavery, policing, redlining, and mass incarceration have inflicted.

The public health model takes this into account with a holistic approach. It considers how a wide array of economic, medical, and social factors intersect and what in fact constitutes the biggest and most fundamental threats to public safety. Interpersonal violence can’t be stopped if we focus on it in isolation.

A key change that we need to make to build more effective violence prevention systems is to change the way we measure safety. Why does this matter? Because policy is responsive to the metrics by which it is measured—by officials, researchers, journalists, and the public. Crime rates are clearly a woefully misleading metric for measuring safety and guiding effective safety investments. So what should the metrics for public safety be?

Public health experts argue we should measure the success of violence prevention policies via outcomes that connect more closely to root causes of unsafety: rates of homelessness, poverty, employment, hospitalization, emergency room visits, overdose, child abuse and neglect, educational attainment, infectious disease rates (e.g., HIV, STIs, hepatitis C, Covid-19, influenza, etc.), diabetes, heart disease, and cancer, along with trends in overall life expectancy. And to maximize returns on investment and to generate maximum safety for everyone, we use public health principles to prioritize addressing the populations in the bottom quartile in each of these categories.

The Evidence for Investing in Neighborhoods

One study showed that emergency financial assistance substantially reduced total arrests—including a 51 percent reduction in arrests for violent crimes. Another showed that increasing youth employment (e.g., summer jobs programs) reduces violent crime by up to 43 percent, with long-lasting effects. Additional research shows that increasing access to affordable housing by building more low-income housing units results in significant reductions in violent crime.

Given that economic inequality predicts homicide rates “better than any other variable,”it should be unsurprising that data repeatedly show that addressing economic inequality would make our society dramatically safer. Data from 39 countries indicate that even small decreases in inequality would reduce homicides by 20 percent and cause a 23 percent long-term reduction in robberies.

Addressing inequality and economic insecurity may sound utopian to some, but we already have abundant policies that do this; they should be scaled up. Unemployment insurance and other forms of guaranteed income (i.e., Social Security insurance), for example, have been shown many times to reduce crime and violence. An important study showed that removing youth from Social Security insurance at age 18 led to increases in crime by 20 percent over the next 20 years and increased risk of incarceration by 60 percent. And, considering the overlap of crime and violence risk with substance use disorders and overdose, it’s important to consider these findings alongside another recent study that showed unemployment benefits brought major reductions in opioid overdose deaths.

What about the data for built environment and neighborhood design as violence prevention? Recent studies in Philadelphia, Baltimore, and Youngstown, for example, have all found that maintaining green space and repairing damaged houses dramatically reduces crime. In Philadelphia, a project to repair abandoned homes was associated with a 39 percent reduction in firearm assaults. The project returned hundreds of dollars for every dollar invested in the program. In New York City, research showed that adding streetlights can reduce “index crimes”—including murder, robbery, aggravated assault, and some property crimes—by more than a third. Another study found restoring vacant land in cities significantly improves local residents’ both perception of their safety and their actual physical safety. Restoration projects reduced firearm violence by 30 percent.

What about the data for non-police community safety interventions? Research shows Safe Passage to School—which places civilian guards along specified routes to keep students safe on their way to and from school—reduced violent crime by 14 percent. Chicago’s own Cure Violence model, which takes a public health approach to circuit-breaking violence interruption via street outreach and providing employment to communities at high risk of violent crime, has been associated with significant reductions in violence. This violence-interruption model has been replicated in other cities with noted improvements in safety. Similarly, the violence interruption model introduced by Advance Peace has had success reducing firearm violence in cities like Sacramento, Stockton, Richmond, and Fresno. And research on hospital-based peer interventions for youth has found that such programs significantly reduced the survivor’s risk of experiencing future violence and of being arrested themselves.

What about the data for expanding health care access as a crime and violence prevention measure? ​​In studies released last year alone, one found that men who lose access to Medicaid eligibility are 14 percent more likely to be incarcerated over the subsequent two years. That goes up to 21 percent for people with histories of mental health diagnoses. A second showed that increased access to health care and through Medicaid expansion for formerly incarcerated people led to a 16 percent reduction in violent crime. A third found police arrests significantly declined after expansion of Medicaid access, including a 25 to 41 percent drop in drug arrests and a 19 to 29 percent decrease in violence-related arrests.

As I’ve summarized elsewhere, multiple additional studies have convincingly demonstrated that expanding health care and supportive social services is far more effective for reducing crime and protecting community safety than doubling down, once again, on the police model. Beyond numbers alone, the public health model for safety also provides an ethical basis for a government grounded in a commitment to caring for its people rather than simply punishing them instead.

From my vantage as a policy researcher and a doctor who treats the ramifications of violence every day, it’s clear that building safety for Chicago will require following the evidence to focus supportive resources—not more policing—on communities in greatest need. Johnson has a plan to do this. Alderperson Rossana Rodriguez’s work to build the Treatment Not Trauma platform, which Johnson has embraced as a key piece of his holistic community safety plan, reflects just the tip of the iceberg of what we—city officials, public health and safety experts, and our most affected communities—can do together with someone in the mayor’s office who finally has the integrity to fight for what works rather than what’s politically convenient. ​