From public safety to public health: Re-envisioning the goals and methods of policing.

A MOMENT OF URGENCY

There is an acute need for major reform in policing in North America, and especially in the United States. The murder of George Floyd and the movement it energized emphasized the urgency of this long-term, ongoing need. As the Movement for Black Lives expanded into possibly the largest social movement in U.S. history (Buchanan et al., 2020), many of its protests were policed in ways that only served to heighten tension and mistrust between the police and the public seeking change in their practices (Brunsden et al., 2020).

The political and policy responses to this movement have ranged from proposals to abolish the police and prisons, to proposals to “defund” (i.e., reduce) police budgets and shift resources elsewhere, to specific, limited policy reforms, such as rewriting use-of-force policies to ban choke holds, to mandating additional training (e.g., anti-implicit bias and de-escalation curricula). The breadth of proposals reflects the breadth of opinion and the lack of consensus on how to reform policing in a democracy or, indeed, what the roles and goals of the police should be in American society.

As the nation grapples with defining the proper roles and limits of police generally, and particularly in Black, Brown, and other communities that have borne disproportionate harms from police (as well as from many other institutions), we propose an approach that we believe would be both realistic and effective: adopting the goals, metrics, and lenses of public health. A call for such a union has been voiced for several years now (Burris & Koester, 2013; Krupanski et al., 2020; van Dijk et al., 2019), but we have yet to operationalize it using a discrete and scalable approach. We begin to take up that task here. By replacing current performance metrics with public health metrics and flawed conceptions with ones that are based upon evidence, and by demanding agility and accountability in changing practices and policies when they are shown to cause harm, we can improve the health, safety, and well-being of communities across the United States.

This will require collaboration between the fields of public safety and public health. It will require the introduction of public health terminology into policing in order to establish a common language for setting goals and measuring results, a step that may empower communities to hold police officials accountable in ways that have so far proven elusive (Table I). In theory, this is not a radical recommendation. Police officials already name public health-related goals as their ultimate objectives: to reduce morbidity and mortality related to violence, non-violent victimization, roadway safety, and emergency responses to risk behaviors associated with drug use and mental illness. While this should be a cause for optimism, to transform these stated objectives into actual policy and practice will require that these goals be more explicitly internalized within police departments.

Table I

The reform effects of a paradigm shift from public safety to public health with key concepts of medicine

	Policing/Public Safety Paradigm	Public Health/Medicine Paradigm Shift	Reform Effect
Objectaive	Lowering community violent crime	Population-level morbidity and mortality reduction aggregated from individual interventions	Provides a common language with implications for goals, methods and metrics
Primary focus	Deterrence and identifying offenders	Prevention	Shifts from tertiary prevention through policing to primary and secondary means that address structural determinants
Measures	Productivity measures: arrests, tickets, contraband seized	Surrogate vs. true endpoints	Holds interventionists accountable for their stated goals
Authority	Law enforcement as an end in itself	Law as empowering an agent to pursue an end	Focuses on discrete outcomes rather than assuming the means can achieve them
Negative effects	Collateral consequences	Iatrogenesis	Explicitly calls for reduction; acknowledges self-perception of police as interventionists
Specialization	Generalist response by officers to calls for service	Preliminary diagnosis and referral to specialists in behavioural health as needed	Promotes evidence-based outcomes; realigns municipal budgets as necessary
Minimizing impact	Reduce overpolicing	Compute Number Needed to Treat	Asks prospective question rather than making post-hoc observation; nests with iatrogenesis

In limited ways, this work has begun. A body of research has conceptualized promising police linkages with the pursuit of public health and the opportunity for complementary work (van Dijk et al., 2019). The nascent development of diversion and deflection programs for people suffering substance use disorders reflects some growing acknowledgement among police departments of the value of public health approaches. But we have yet to lay out the basis of a common language or to operationalize concepts in a way that make commensurate the goals and work of policing and public health. This article sketches out the way forward and provides some illustrative examples.

MOVING AWAY FROM TODAY’S INEFFECTIVE AND HARMFUL PERFORMANCE MEASURES

At present, the performance measures commonly used in policing include metrics such as the number of arrests made, tickets issued, guns seized, and pounds of illegal drugs seized. These metrics are rooted in a belief that, through deterrence and incapacitation, these basic law enforcement actions will lead to a jurisdiction-wide reduction in crime (as measured by crime rates). Unfortunately, the evidence base consistently demonstrates little if any connection between the acts measured—and thereby encouraged—through these performance metrics and the ebb and flow of crime rates. This is true even when additional resources are poured in or punishments ratcheted up.

ADOPTING GOOD GOALS AND METRICS

To provide evidence of effective policing, surrogate endpoints should be linked to actual ones we can accept as independently valuable and worth striving for (Beletsky, 2018). That police should endeavour to detect and arrest the perpetrators of serious crimes is a widely accepted goal, but this comprises a small proportion of the daily work of most officers. We recommend adopting the most obvious endpoints as lodestars: the basic public health goals of healthier, longer lives for all people. How policing contributes to or detracts from these endpoints should be measured. The ways police can operationally contribute to these endpoints of healthier, longer lives—such as by engaging with upstream preventive efforts—should be identified, quantified, and then operationalized through performance metrics and incentives.

Some of these will concern direct police interactions: do they save lives or cause trauma? Others will be indirect, such as aspects of life that evidence suggests contribute to a community’s health, resilience, and economic solvency. Police should be explicitly tasked with fostering them. While police are not and should not be primary service providers delivering the key elements of strong social determinants of health, they can facilitate them. Examples include the ability of a community to use public spaces constructively and without fear, or to safely use roadways and public transportation.

IMPROVING PUBLIC HEALTH IMPROVES PUBLIC SAFETY

There is a basic congruence between public safety and public health. The most obvious example is violence: it undermines both. Focusing on Black Americans, the Centers for Disease Control and Prevention ranks homicide as the leading cause of death for Black individuals between the ages of 15 and 29, and specifically men between the ages of 15 and 34 (Centers for Disease Control and Prevention, 2018). This disparity extends to nonfatal shootings: Black Americans are shot at the rate of 113.8 per 100,000 people, which is ten times higher than their white counterparts (Everytown for Gun Safety, 2020). If police can effectively reduce this violence, they will be improving both public safety and public health.

RECKONING WITH IATROGENESIS

Can police effectively reduce this violence without adding to public health harms? A public health-informed approach to policing means adopting—but not co-opting—the concepts and language of public health. One of the critical shifts will be to seriously reckon with iatrogenesis (Anderson & Burris, 2017). Policing has often been impervious to evidence demonstrating that some practices are ineffective or harmful, or that interventions with proximate positive effects foster negative ones that manifest in the long term. The public health-informed approach will require building a willingness to change based on evidence.

Continuing to highlight the experience of Black Americans, they are also the group at greatest risk of being killed during a police encounter, with the lifetime odds of being killed by police approximately 1 in 1,000 (Edwards et al., 2019). The cumulative causes of violent death, injury, and debilitation are acute, recurring public health and safety crises for not only Black Americans but several other marginalized constituencies, such as the LGBTQ+ and Latinx communities, and cultural and religious minorities. These crises provide tragic support for the increasingly widespread acknowledgement that racism is a public health emergency and the growing contention that policing is one, too. Aligning policing with public health therefore not only requires recognizing the proper endpoints of policing but also recognizing when interventions are a source of great harm and have few redeemable features.

MOVING AWAY FROM “HEALTHWASHING”

A reason to be skeptical of current statements by police and political leaders purportedly embracing public health goals is that policing already includes many iatrogenic policies that are promoted specifically in the name of public health. Some see this as a way to take troubling police practices and “healthwash” them in order to make them more palatable. Take the example of America’s failed “war on drugs,” the source of extraordinary iatrogenesis (Alexander, 2012). Not only have its collateral consequences been harmful, but the interventions themselves continue to be largely ineffective. Drug possession and the crimes associated with use and addiction are prime examples of harmful practices justified in the name of health, in continued defiance of the evidence. Indeed, in most jurisdictions, police have consistently undermined the scaling up of evidence-based health services shown to reduce drug crime, chaotic drug use, and overdose mortality—such as harm-reduction agencies and medications for opioid use disorder—through opposition at the political level and through targeting enforcement activities at or near harm-reduction agencies. A comprehensive reform of the United States’ failed drug policies is long overdue (del Pozo & Beletsky, 2020).

A handful of forward-thinking jurisdictions have made substantive changes such as the de facto legalization of the unprescribed possession of addiction treatment medications (del Pozo, Krasner, & George, 2020) or the decriminalization of the simple possession of controlled substances (Levin, 2020). Similarly, the onset of the COVID-19 pandemic led to significant police reforms in the name of preserving health and preventing the spread of infection, from reduced drug arrests to fewer pretrial detentions. These changes present an opportunity to re-center the entire approach to policing addiction, but their results need to be evaluated to make the strongest case possible (del Pozo, Beletsky, & Rich, 2020). The results of these evaluations should help determine which policing practices should be incentivized in order to change police culture, which is often entrenched and difficult to shift when it comes to deflection and diversion in lieu of drug arrests (Barberi & Taxman, 2019).

INCORPORATING IATROGENESIS-AVOIDANCE INTO POLICE CULTURE

Framing negative outcomes as iatrogenic effects would encourage police to think like their colleagues in fields such as mental health and addiction treatment. There is another advantage to invoking this concept that should not be underestimated: it would reconcile the harms caused by policing with the belief held by police leaders that their interventions are meant to protect citizens, help communities, and avert even greater harms.

While the actions police take can be simply harmful, or even criminal or unconstitutional, most are well-intended interventions developed to address dangerous situations. Policing is viewed by most officers as a type of treatment meant to secure or preserve public safety, but taking this idea seriously means acknowledging that, like chemotherapy, even the best-intentioned treatments can cause harms. Talking about the iatrogenic effects of policework requires that collateral harms be both measured and reduced, the most critical among them being unnecessary or excessive force. It also means foreclosing the use of interventions that produce more harm than good, even if lawful or supported by a vocal or politically powerful constituency. For example, arrest, incarceration, and their long-term consequences are deeply harmful to an individual’s well-being (as well as that of their family and community). These should be considered impermissible iatrogenic effects unless they can be unambiguously justified by a safer, healthier community.

THE NUMBER POLICE NEED TO TREAT

The idea of iatrogenesis allows policing to leverage another useful public health concept: the number needed to treat (NNT). This measures the effectiveness of an intervention by measuring how many people it must be applied to, who may need it or not, to prevent one undesirable outcome (Cook & Sackett, 1995). Many police interventions, from traffic enforcement to investigating “suspicious” people, are broadly distributed and intend to prevent undesirable outcomes downstream, be they accidents or violent crime. The police need to consider they may be treating many more citizens than necessary to regulate behaviour, especially if the downstream effects on roadway safety or violent crime are not clearly spelled out by research and may actually reduce a community’s quality of life.

Stopping and frisking is a paradigm case. In New York City, before a court declared the practice unlawful and discriminatory in its effects, many hundreds of people were stopped and frisked for each illegal firearm actually found; over 90% of people stopped were found to have committed no infractions (Goldstein, 2013); and these fruitless stops have been associated with elevated risk of psychological distress in minority communities (Sewell et al., 2016). The constitutionality of these police practices aside, in the same way medicine should minimize the number of patients it needs to treat to preserve health, police should minimize the number of interventions necessary to promote safety. This would commit police to acting based on evidence rather than the simple prerogatives afforded by law, and to use caution when the endpoints being pursued are minor, or the NNT and its accompanying iatrogenic effects are disproportionately high. This way of thinking would preclude the use of many overbearing enforcement techniques before constitutionality—and potential payouts—even comes into question.

LETTING EVIDENCE DIVIDE LABOUR

A police commitment to public health would require four actions: (1) basing procedures and policies on behavioural health and science, (2) collaborating with health professionals such as social workers, harm reductionists, and mental health counselors when situations call for co-response, (3) ceding or referring work to health professionals when a police response would not be particularly effective or requires specialized knowledge or skills that exceed the police remit, and (4) supporting rather than undermining the scaling up of evidence-based health services and practices such as harm reduction in the budgeting and legislative processes and in police practices (Burris & Koester, 2013; van Dijk et al., 2019).

A considerable amount of routine police work involves responding to calls involving three categories of risk behaviour and vulnerability that are especially well-suited to such a division of labour: substance use, mental health crises, and homelessness. Police have traditionally been the default response to these issues, but there is compelling evidence that other actors would be more effective at addressing the problems they pose. The challenge is operationalizing such a shift without seeing an interruption in critical services because the police have been removed from certain roles before effective alternatives have been implemented (del Pozo, 2020).

Work would need to be done to draw lines between items (2) and (3) above. Calls involving drug use and addiction should almost always be routed to specialists in harm reduction and addiction treatment; the law enforcement approach to drug use has been a dismal failure (del Pozo & Beletsky, 2020). However, mental health calls with an element of volatility would require co-response; other such calls could be routed to health professionals. Indeed, promising models in Oregon have demonstrated that properly funded and trained specialists can handle mental health calls in lieu of police officers (Butler & Sheriff, 2020), and cities such as Rochester, San Francisco, New York, and Chicago have taken steps to implement this approach (Westervelt, 2020).

While such programs are piloted and scaled up, police officers can be trained and incentivized to outsource through linkages with the health and harm-reduction sectors. Like general practitioners referring patients to specialists, police can vector people to interventionists or hubs that will provide the proper services, with the proper training and resources. We deliver a training program that helps create these linkages for people who use drugs, and it has been shown to increase police willingness to refer people to treatment in lieu of arrest and to engage in other harm-reduction measures (Beletsky, 2020; Davis & Beletsky, 2009; Rocha-Jiménez et al., 2019). Another approach is to give police officers some of the basic skills of interventionists through crisis intervention team training, which has been shown to increase police officers’ reliance on de-escalation and referrals to psychiatric treatment rather than arrests in responding to mental health crises (Watson & Compton, 2019).

MEDICINE AND PUBLIC HEALTH AS PARTNERS AND MODELS IN POLICE REFORM

Though typically conceived as totally different professions, there are some key conceptual similarities between police and the health fields that, if recognized, may help build partnerships between them. Police interventions, like those in medicine, typically happen at the individual level, in response to proximal threats. In the aggregate, these have the potential to lead to population-level effects; indeed, police departments have population-level responsibilities, namely safe communities. While medicine typically lacks such responsibilities, practitioners often have public health in mind, and they work with public health agencies towards population-level effects. Thinking of police work this way—as a series of individual-level interventions undertaken by a system responsible for achieving population-level effects—may help police leaders understand the congruence of the missions of policing and health. Encouraging police leaders to understand this and accept its implications would provide a means to effectively operationalize a critical step in police reform.

This conception may enhance efforts to build linkages between law enforcement and health actors. The sense of a common mission may help develop alliances among players that operate in different lanes but have many intersections. Indeed, this conception may help police see how to reckon with iatrogenesis. It is the connection between individual health practices and population-level health results that has caused the medical fields to engage with the iatrogenic effects of their own practices, such as the origins of the opioid crisis (Frakt & Monkovic, 2019) and the disparate outcomes of medical practice by race and class (Paul et al., 2020). Medical and public health efforts to come to grips with their own legacies of racism, through the careful use of evidence-based methods and metrics, may be a model for how police agencies can own their failures while working to reform themselves. If Americans are anxious for police leaders to reframe their work in a sweeping way, health care providers and public health officials can supply the language, methods and metrics that would help them do so, while affirming the core police responsibilities of protecting life and helping communities thrive.

The medical experience also highlights that there is much to be gained by police officers and leaders in this approach. We predict greater job satisfaction, lower burnout, improved retention, improved community relations, and fewer lawsuits.

GETTING THERE

To turn these stated objectives into true objectives that are deeply internalized in police organizations and their policies, practices, and cultures will take great effort. The United States has over 18,000 law enforcement agencies. The nation will be on the road to real progress in police reform when police budgets are tied to public health metrics, when there are employment and budgetary consequences for iatrogenesis that goes unfixed, and when officers’ performance and promotional paths are tied to public health outcomes. These metrics should be developed with the input and support of the communities they serve and the people most impacted by policing, and diverting and deflecting to health services should become the default practice. When police leaders and associations stop undermining and instead support the scaling up of those health services, and when political leaders ask police chiefs and candidates for police leadership positions what they will do to ensure their work will advance the jurisdiction’s public health goals, the police profession will be on a path of meaningful reform that delivers not only public safety but public health and well-being.