John G. Peters, Jr., Ph.D., and Darrell L. Ross, Ph.D.
©2025. A.R.R.
In September of last year, the Police Executive Research Forum (PERF) unveiled its report, 15Principles for Reducing the Risk of Restraint-Related Deaths, sparking critical discussions within the law enforcement community. While PERF’s standing as a trusted advocacy organization lends weight to its recommendations, many of these principles demand scrutiny before adoption. As seen with the International Association of Chiefs of Police’s Model Policy on electronic control weapons – highlighted in Police and Security News just last year – such guidance can contain hidden “land mines.” Some of PERF’s principles offer practical benefits, but others raise concerns about catering to anti-police narratives.
PERF 15 Restraint-Related Death Principles
If you have not read the 2024 PERF report, here are the 15 principles which PERF Executive Director Chuck Wexler stated, “Nearly all guidance in this report can be applied to any situation” (p. 1). Few things can truly be said to apply universally to “any situation,” so it’s ultimately up to you, the reader, to evaluate the validity of this claim. Many of the following principles, while not controversial, cannot be implemented in “any situation” and that requires critical reading of the explanations and recommendations following each principle.
1. Identify Warning Signs of a Medical-Behavioral Emergency.
2. Plan and Develop Protocols for a Coordinated Medical-Behavioral Emergency Response.
3. ICAP Principles Apply to Medical-Behavioral Emergencies.
4. De-escalate Wherever Possible, but at a Minimum – Don’t Escalate.
5. Evaluate the Need to Immediately Restrain.
6. Multiple Electronic Control Weapon (ECW) Applications May Increase the Risks Associated with Restraint.
7. A Supervisor or Other Leader Needs to Take Charge.
8. Designate a Patient Safety Officer.
9. Prone Restraint Carries Potential Risks and Should be Limited.
10. Not Following Directions Does Not Always Mean Willful Non-Compliance.
11. The Goal is Control, NOT Complete Immobilization.
12. The Ability to Speak Does Not Mean a Person Can Breathe.
13. On-scene Coordination and Collaboration Between EMS and Law Enforcement is Crucial During MBEs.
14. Keep Emotions in Check and Be Ready to Step Up and Step In.
15. Commit to Learning from Every Incident. (pp. 3-21; policeforum.org).
Redirecting an actively resistive person into the prone position for control and likely restraint happens daily and seldom is there a serious injury or an even more rare fatal outcome. This is supported by 23 published scientific clinical studies and four epidemiologic (field) studies. These studies have concluded there is no inherent risk of danger to the subject in the prone restraint position. Admittedly, there is no totally risk-free force option, but the likelihood of a subject dying in the prone restraint position is statistically rare and most likely for reasons beyond the officer’s control.
We commend PERF for tackling the critical issue of non-firearm officer-associated deaths. However, before adopting its proposed principles, agency leaders, managers and trainers should carefully evaluate their potential impact. While space constraints prevent a comprehensive analysis, this article highlights key concerns which must be addressed to ensure these recommendations do not inadvertently lead to harmful changes in policy and training.
Scientific Studies vs. Theory
PERF’s failure to integrate published scientific findings into its 15 recommendations is contrary to PERF’s 2021 prior recommendation of using science to inform decision-making (What Police Chiefs and Sheriffs Need to Know about Collecting and Analyzing Use-of-Force Data; policeforum.org). The published scientific literature on prone restraint is not referenced in the most recent PERF report. Of the 32 footnotes analyzed, only one theoretical case study was cited (fn. 21) along with many Associated Press articles. Case studies do not test hypotheses and the cited case study assessed two autopsy deaths which is an unreliable unit of analysis.
The collective scientific literature about redirecting an actively resisting/combative subject into the prone and restrained position finds no evidence of clinically significant effects on cardiac output, ventilation, hypoxia, or hypercarbia. Approximately ten percent of people sleep on their stomachs with no adverse effects and prone positioning has been shown to improve patient breathing during some medical surgeries and treatments.
The four epidemiological field research studies on prone positioning collectively showed that no one died in that position and that serious injuries were rare. Various force options were applied including a conducted energy weapon. Most of the subjects exhibited symptoms associated with illicit drugs and/or mental illness. Collectively, the scientific research shows the prone restrained position is a safe and often preferred position for controlling and restraining actively resisting/combative individuals and did not pose a deleterious risk of harm to them.
The PERF report introduces the concept of Prone Restraint Cardiac Arrest as a potential explanation for sudden deaths among restrained individuals. While this theory has not been scientifically validated, its proponent, Alon Steinberg, M.D., suggests that individuals who struggle with officers – either before being taken to the ground or during and after restraint – may experience cardiac arrest. This hypothesis, though unproven, challenges the notion of “positional asphyxia” and aligns with common sense, as underlying cardiac issues may go unnoticed by both the individual and the officer(s) involved. However, Dr. Steinberg’s assertion that “the best way to avoid cardiac arrest is to remove individuals from prone restraint as soon as possible” warrants scrutiny. Cardiac arrest can occur regardless of body position – whether standing, running, lying supine, or sitting. Notably, the PERF report provides no scientific evidence to substantiate this specific claim.
De-escalation
PERF Principles 4 and 5 recommend “don’t escalate” and “delay restraint until Emergency Medical Services” (EMS) arrive. In contrast, we suggest caution, as this could potentially compromise the officer’s safety. De-escalation is a commendable intervention strategy to attempt, but it usually takes two individuals for it to work (subject and officer). Frequently, officers cannot script their responses and often must decide to use a force option based on the perceived behaviors, noncompliance and the resistance of the subject. The report conveys mixed messages to the reader: that the incidents are high risk critical incidents; subjects may not comply with directions and may exhibit erratic/irrational behaviors, as well as paranoia and delusional behaviors, and be severely agitated; to wait on EMS; use restraint for persons at imminent risk of self-harm or harm to others and if the subject is at significant risk of flight (which lacks description).
Principle 5 of the PERF report says, “If a person is reasonably contained . . .” but failed to define or clarify what “reasonably contained” means. Restricting an officer’s response to these factors fails to consider the totality of the circumstances an officer may confront which may simply involve a dangerous person without medical issues. It also ignores the Fourth Amendment’s standard of objective reasonableness established in Graham v. Connor (1989), its progeny, and intervening based on the perception of the officers on scene. These incidents are dynamic, unpredictable and contain innumerable variables. An officer may not have the luxury of continuing with crisis intervention strategies and may have to immediately respond to protect him/herself and others, as well as control and protect the subject. The application of de-escalation strategies is not a substitute for the safety of the involved officer(s).
Small Agencies Reign
Per the Bureau of Justice Statistics (BJS), a majority of law enforcement agencies employ about ten officers. It is common for only one or two officers to be assigned patrol duties per shift and if EMS response time is slowed due to a priority call, the officer may not have time to perform a full assessment, continue with de-escalation techniques or await EMS. Waiting for backup may not be possible and EMS may not be forthcoming particularly in smaller communities. Recommending that an officer delay a response or restrict the use of certain force options places the officer at risk and usurps the ability of the officer’s decision-making based on his/her perception and the totality of circumstances.
Limit ECW Applications
It is clear from reading the PERF report that the TASER® 10 was not considered in the recommendation of Principle 6: “Multiple Electronic Control Weapon (ECW) Applications May Increase the Risks Associated with Restraint.” While no one recommends multiple ECW applications without reasonable justification, the PERF report appears to have ignored the scientific ECW study findings which show that using an ECW with limited, repeated applications did not produce an adverse impact on the subject which is consistent with other controlled laboratory studies. ECWs do not work on everyone and, therefore, more than one application may be required. The TASER 10 has multiple probes which can be applied to a person until the best connection is identified by the device. The number of ECW applications must be communicated to EMS and/or other medical responders.
No Supervisor
PERF Principle 7 calls for a supervisor or other leader to take charge. What if no one is available to “take charge”? Again, the PERF report appears to ignore the many smaller agencies which may not have a supervisor on duty or, in some larger agencies, this, too, may be impracticable. If a supervisor is unavailable or is across town or the county, the default contact is the chief of police or the sheriff. While noteworthy, this recommendation may be problematic to implement for many agencies.
What about Psychological or Simply Dangerous Suspects?
Medical-Behavioral Emergencies (MBE) are a primary focus of the PERF report which Wexler describes as “generally involv[ing] a person experiencing a medical issue (often drug-induced or drug-enhanced) that presents to police as a behavioral issue” (p.2). Per this definition of MBE, the behavioral signs fail to include those associated with mental illness, lack of oxygen, people who are simply aggressive and will fight the police to escape, and do not include the numerous mortality enhancers associated with a non-firearm, arrest-related deaths. Very few officers are medically trained beyond first aid and are not trained psychologists or psychiatrists. There are plenty of violent encounters when the violator does not have medical issues.
Force Options
Other than cautioning the misuse of an ECW, there are no report recommendations about selecting force options, including control techniques and force devices; the application of restraints; the use of a restraint chair; the use of spit restraint devices; the use of aerosols; or using other force equipment. While the recommendation to monitor a restrained person is appropriate, the report does not direct officers on how or what to monitor for, or direct officers to be prepared to provide medical assistance as warranted. There is no recommendation that the involved officers submit an incident report or that involved officers or a supervisor follow clearly established law, agency policy and their training.
Disputable Best Practices
Referring to the PERF 15 principles as “best practices,” Wexler confuses the reader by saying they “can be employed during nearly any incident, not only MBEs” (p.2). This contradicts his earlier claim that “Nearly all guidance in this report can be applied to ANY situation” [emphasis added] (p.1). Regardless, the principles do not meet the criteria of “best practices.”
According to Kerzner (2013), “[a] best practice begins with an idea that there is a technique, process, method or activity that is more effective at delivering a desired outcome than any other approach and with fewer problems and unforeseen complications.” A “best practice” implies there is one and only one way of accomplishing a task; that an ideal has been achieved; and that, in this case, law enforcement officers and EMS providers have been performing some activities incorrectly.
It is strongly recommended to use the terms “proven” or “evidence-based” practices. Per Garfield (2017), “Proven practices are methods that have been demonstrated to be effective and lend themselves to replication to other groups, organizations and contexts.” When independent and scientifically rigorous evaluations are needed, “evidence-based” practices are used.
“Evidence-based practices” have the following components: (1) They have been studied using appropriate scientific methodology; (2) they have been replicated in more than one geographic or practice setting with consistent results; (3) they have been recognized in scientific journals by one or more published articles; (4) there is an implementation manual to follow; and (5) specific outcomes are produced (Campos, 2011).
These definitions suggest the PERF 15 Principles are nothing more than non-scientific recommendations. As previously discussed, none of the scientific findings on prone restraint and officer-associated deaths is included in the PERF report. An unproven theory and two case studies are the only pillars for MBE recommendations.
Summary
PERF is applauded for tackling this issue, but its myopic focus on MBEs, lack of scientific support and claiming these are “best practices” when they do not meet the definition requires critical reading of the report before adopting any of the recommendations. Few will disagree that unnecessarily applying an ECW is unacceptable and there is needed guidance on this and other important issues. PERF, like other advocacy groups, has no ministerial authority or legislative power to “mandate” law enforcement agencies adopt its recommendations. Plaintiff counsel will conflate them into “standards” in an attempt to hold municipalities and officers accountable.
John G. Peters, Jr., Ph.D., serves as president and chief learning officer of the Institute for the Prevention of In-Custody Deaths, Inc. and Executive Director of the Americans for Effective Law Enforcement, Inc. A judicially qualified expert witness, he is also a graduate of the MIT Sloan School and Computer Science and Artificial Intelligence Laboratory program on Artificial Intelligence. He is a frequent contributor to Police and Security News.
Darrell L. Ross, Ph.D., is professor emeritus at Valdosta State University. An experienced expert witness and former correctional officer, Dr. Ross is a prolific author and speaker. His research interests focus on prone restraint, perceived unusual behaviors and sudden in-custody deaths.
