There are some things that should never happen in medicine.
A surgeon should never operate on the wrong body part. A patient should never get a lethal dose of a medication because a doctor’s handwriting is illegible. We call these mistakes “never events” because they should never happen—regardless of how busy, complicated, or chaotic medicine can be.
The landmark 1999 Institute of Medicine report “To Err Is Human” highlighted the sheer prevalence of never events in medicine—leaving an instrument inside a patient after surgery, giving a patient the wrong type of blood—as well as the less egregious but preventable mistakes such as hospital-acquired infections and surgical complications. All told, the report estimated that between 44,000 and 98,000 medical errors, including never events, were occurring in the U.S. each year.
These numbers helped spawn the modern patient safety movement in health care. Since then, quality improvement initiatives have sought to design ways to provide safer, higher quality care—often yielding clear and measurable improvements. Hospital mortality, for instance, has dropped significantly over time.
In the same way, there should be never events in law enforcement. The most obvious example is a police confrontation that ends in the death of unarmed civilians. The frequency of these events—as well as greater awareness about them—has increased significantly in recent years. A recent FBI report found that police killings are at their highest level in two decades, and poor and black communities are disproportionately affected. These incidents have shaken the country and should now serve as a catalyst for law enforcement reform the way preventable medical mistakes did in health care.
Many advocates argue we need a cultural shift in policing. But that’s only part of the solution. We need a shift in process and approach as well—a clear framework to evaluate quality and demand accountability. As physicians and health policy researchers, we believe it’s worth considering what law enforcement can learn from health care’s efforts to improve the quality and safety of care that Americans receive, given notable but often unrecognized similarities between the professions.
Both clinicians and law enforcement officers are highly motivated professionals who work in high-pressure environments. They make decisions under conditions of great uncertainty with monumental consequences for affected parties. Quick intervention for heart attacks saves lives in medicine; in policing, rapid response to 911 calls can do the same. Both professions serve others, and community engagement is increasingly recognized as vital for success. The public places trust in both professions—granting them special privileges—with the understanding that they exist primarily to protect the public good. The lives of those they serve often hang in the balance, and the public is rightfully outraged when errors are committed.
In both health care and law enforcement, never events—and other errors—occur more frequently in situations in which the speed of response to problems is critical and uncertainty about the right course of action is large. But errors aren’t inevitable. We must construct systems to allow professionals to work with greater safety and efficiency.
We believe this can be achieved by borrowing from health care’s most widely recognized quality improvement framework, developed by the physician Avedis Donabedian. The framework conceptualizes quality into three domains: structure, as in, does a hospital have the necessary structure to provide high-quality care, e.g. appropriate nurse-to-patient staffing ratios or the presence of an ICU; process, as in, does a hospital engage in processes of care that are high quality, e.g. providing antibiotics to patients with pneumonia within one hour of hospital arrival; and outcomes, as in, does a hospital have a higher or lower mortality rate compared with others.
Using this framework to assess police stations as we assess hospitals could help us pinpoint where things are going wrong. And developing standard metrics could facilitate sharing of best practices, rigorous performance assessment, and safer environments for both officers and civilians.
Consider the growing use of body cameras—a structural component of policing—and the mounting evidence of their benefit. A randomized trial found that after the introduction of body cameras in Rialto, California, use of force by police officers dropped by 60 percent and citizen complaints declined by 88 percent. In San Diego, the initiation of body cameras was associated with a 46.5 percent reduction in use of force and 40.5 percent fall in complaints.
But according to a report from the Department of Justice, only a quarter of police departments used body cameras in 2013. While most departments are now considering the technology, there is little consensus on how and when to use them. Though early evidence suggests body cameras help, many concerns remain about cost, privacy, and storage of collected video.
Other police equipment also deserves greater scrutiny in terms of the effect it has on outcomes. Since the 1990s, law enforcement agencies have acquired billions of dollars of military equipment, much of which was originally designed and purchased for war zones. But this practice is controversial and may benefit from more civilian input, greater standardization of permissible equipment, and mandated training for its appropriate use. We should collect evidence on how this equipment affects outcomes to determine whether its continued use is beneficial.
The workforce’s racial and gender composition is another vital aspect of structure. Police departments, like health systems, should aim to be representative of the populations they serve. Research suggests that racially diverse police departments are less likely to conduct potentially discriminatory traffic stop searches. Other work has found that female officers are less likely to use force and that victims of domestic abuse are more likely to report it to female officers. A recent report, however, found that minorities remain underrepresented in most police departments. Fixing this structural problem could help improve the safety, efficiency, and efficacy of police departments.
The process quality domain focuses on the interactions between professionals and those they serve. Process metrics reflect the procedural steps that lead to the outcomes we want and help health systems understand how their actions ultimately affect patients. In health care, carefully selected process measures such as same-day clinic appointments, screening for diabetes-related complications, and immunization rates can help focus doctors’ attention and improve care. For example, by focusing on increasing vaccination rates in a particular community—a process measure—we can ultimately achieve the desired outcome of fewer patients infected with vaccine-preventable diseases.
In law enforcement, a typical process measure might be how quickly police respond to 911 calls. But process measures could also evaluate how effectively police interact with communities by focusing on how and when officers engage in processes of relationship-building, such as youth education, CPR training, and serving meals at shelters. Some departments might choose to encourage foot and bicycle patrols; others could train officers to connect homeless or substance-dependent individuals with services. Evidence suggests that these strategies can lower crime rates and recidivism. By designing and adopting process best practices, police departments can take the necessary steps to allow the desired end results to follow.
Finally, outcome measures are vital for evaluating new initiatives and policies. In health care, the “quadruple aim”—enhancing patient experience, advancing population health, reducing costs, and improving employee work experience—helps doctors work toward the right goals and track progress along the way. This framework has obvious corollaries in law enforcement: enhancing community relations and maintaining public safety with judicious use of taxpayer dollars, while promoting officer respect and morale. Just as health care has increasingly recognized preventive care as a pillar of good medicine, so too law enforcement should focus on preventive policing. A department might choose to reward not the number of arrests made, but the number of adolescents diverted from jail. Selecting the appropriate outcome metrics—measuring what matters to officers and the communities they serve—is vital for advancing toward a law enforcement system that provides safer and higher-quality policing.
There are many potential outcome measures in law enforcement that could become the focus of a department: violent crime rates, false arrests, traffic citations, among others. But health care’s experience should also serve as a cautionary tale: excessive documentation, meddling metrics, and check boxes can seriously impair moral and service quality. We should strive to create supportive environments and nudge decision-making, but we should also trust professionals—be they officers or clinicians—with the freedom to exercise their judgment and do their jobs.
As medical professionals, we can and should identify ways in which knowledge gained from our own experiences can help address structural violence and racial bias. There are many ways to join a burgeoning movement that demands justice and reform: research, advocacy, writing, introspection. But perhaps the most effective and expansive contribution we can make is sharing our approach to improving safety and quality.