An exaggeration of the sudden in-custody death problem is generating “persecution and prosecution” of LEOs and their agencies and is resulting in “reactionary changes in policy and procedure that may well be based in conjecture rather than fact,” according to new findings by a Canadian research team.
In particular, the study group challenges the widely held belief that simply placing a subject face-down in a prone position after a use of force creates a substantial threat to life. In fact, the only subject who died in more than 1,200 consecutive force encounters that the team meticulously analyzed was lying on his side, a position commonly advocated for its presumed safety.
“[S]udden in-custody death,” the study concludes, “has more to do with the features of the individual” than with his or her positioning. Even among suspects supposed to be at high risk--the drug and alcohol intoxicated and those with mental illness--there appears to be no scientific basis for believing that prone positioning is as dangerous as its reputation suggests, the researchers report.
“The true understanding of use-of-force events will come from comprehensive, scientific protocols such as this one, which was carried out under the rigorous conditions of a scientific medical study and supervised by the Institutional Review Boards of the universities involved,” Hall told Force Science News.
“Rather than focusing solely on one type of force or another, the goal of the study is to evaluate all use-of-force events to better understand which subjects and situations may represent particular risk.”
The investigative team was headed by Dr. Christine Hall, an emergency room physician and a medical faculty member at the University of British Columbia and the University of Calgary. An internationally recognized expert on excited delirium syndrome, Hall is also on the faculty for the certification course in Force Science Analysis conducted by the Force Science Institute.
Origin of Controversy
Concern about the relationship between suspect positioning and death first arose in the early 1980s after a medical examiner in Washington State analyzed three cases in which prisoners were transported while hogtied and concluded that such positioning was associated with sudden in-custody death because of the suffocating effect on breathing and the inability of the subjects to shift to a different position.
More articles warning about “positional asphyxia” and “restraint asphyxia” followed in medical journals and law enforcement publications and positioning became a widespread training and policy issue.
Hall points out that the medical examiner’s original work on the effects of positioning was shown to be “fraught with methodological errors,” and that other, more sophisticated studies disputed alleged adverse findings. Nonetheless, the debate about suspect positioning “gradually translated into the unsupported idea that any and all prone positioning for any length of time is immediately dangerous.”
Moreover, prone positioning came to be regarded “without scientific evidence” as “particularly dangerous” to restrained suspects who are “under the influence of intoxicants such as alcohol and/or stimulants, experiencing mental distress, or in the agitated, incoherent state known as excited delirium,” as well as those who have been TASERed.
“When a cause of death cannot be found otherwise, positional asphyxia is often suggested to have [played] a role...,” the researchers note.
As the bias against any form of prone positioning intensified, Hall perceived that what was lacking in the debate was a broad-based, detailed examination of real-life incidents to determine exactly what variety of physical positions restrained suspects end up in after force encounters with police and what association there might be between each of these positions and ultimate unfavorable outcomes.
With the cooperation of police executives in an unnamed Canadian city with more than 1,100,000 population and nearly 2,000 street cops, Hall and her team collected a vast statistical database across a recent three-year period. (The city is anonymous in their report to protect privacy, an ethical consideration required by the universities involved.)
Without officers being aware of the researchers’ interests, certain study-related entries were embedded in the department’s standard electronic form that must be filled out after any use of force above soft-hand control. Among other things, officers had to specify the apparent cognitive condition of the subjects they encountered and their “final resting position...once physical control had been achieved...and while awaiting further disposition”: i.e., either prone (face-down) or not prone (face-up, side-lying, sitting, kneeling, or standing). The department involved does not specify a favored position in policy, so there was no motive for officers to shade their responses, Hall points out.
Forms involving suspects at least 18 years old and all “general police duty” uses of force beyond merely “a bent wrist or straightened elbow to gain compliance” were then parsed for particulars. “We did not pick and choose between subjects, events, or outcomes,” Hall says, “but included all uses of force over three consecutive years.”
The pool the researchers analyzed ended up being 88 percent males. Some 40 percent appeared drunk at the time of police contact; 10 percent were judged to be drug-intoxicated; 9 percent seemed mentally distressed; and 28 percent demonstrated some combinations of those states.
In all, out of 1,566,908 interactions between police and subjects across the study’s 3 years, force was used in only 1,269 contacts.”[C]ontrary to current suggestion that police use of force is rampant, 99.92 percent of all police-public interactions [recorded by the agency studied] did not involve police use of force,” the researchers report. “This finding did not vary across three years of study.”
In terms of positioning, nearly 43 percent of subjects ended up prone in their final resting pose after their force encounter. About 57 percent were not prone. All were handcuffed, but none was leg- or ankle-restrained or hogtied. “[E]ven though prone positioning was very common,” the researchers noted, none of the “hundreds of subjects in the prone position” died.
Indeed, with only one death [a not-prone subject] discovered in the entire study, statistically “there was no [significant] difference in the death rate between prone and not-prone positions.”
Suspects with drug/alcohol intoxication, mental illness, or some combination thereof were essentially evenly distributed between prone and not-prone positioning. Despite the prevalent suspicion that such individuals are at greater risk, “no subject died in the prone position even with a large number of abnormally behaving individuals” in the force pool.
Of the large number of subjects against whom a TASER was used, about 29 percent ended up prone, 25 percent not prone. Again, positioning did not determine how these suspects fared, the researchers emphasize.
The one subject who died during the study “was an individual who was assessed by officers on the scene as having both drug intoxication and mental distress, had undergone a single contact stun exposure with a CEW, and had many abnormal characteristics before and after police involvement at the scene,” the study report says. “This subject was placed in the side-lying position at the conclusion of the use-of-force event, prior to cardiopulmonary collapse.
“[He] displayed multiple features of excited delirium. The details...are strikingly similar to other in-custody death occurrences both in and out of the prone position and...with and without CEW application.”
Hall points out that this is believed to be the “only study to document the incidence of sudden in-custody death across all use-of-force modalities” in a major urban police agency.
Besides dispelling some persistent myths about the risk of prone positioning, she feels the findings are important for confirming the overall rarity of sudden in-custody death. It’s “profoundly low in the real-world environment of police use of force which includes many abnormally behaving individuals in varied circumstances.”
Media publicity of in-custody deaths without an appreciation of their context “results in the presumption of a falsely elevated prevalence” of these occurrences, she writes in the study report. “This, in turn, generates persecution and prosecution of individual police officers and police agencies at great personal and societal cost, fear and mistrust for the police in the public eye, and results in reactionary changes in policy and procedure that may well be based in conjecture rather than fact.”
Hall acknowledges that all research studies have some limitations. Because hogtying was not used in any of the encounters her team studied, the researchers could not assess its specific danger to prone subjects, if any, for example. Also, while they were aware that “weight force” was very likely applied to the back of some of the proned subjects, they were not able to determine how often that occurred. That variable will be evaluated in the future.
“[R]egardless of the low incidence of sudden in-custody death, further study is required,” the team reports. Hall’s objective is to “determine what, if anything, can predict sudden in-custody death in order to improve the outcomes for people undergoing police use of force.”
Meanwhile, the researchers caution that suspects restrained in a prone position--any position, actually--should not be abandoned “for protracted lengths of time.” Suspects always bear close observation.
“[T]he best way for non-medically trained pre-hospital personnel to monitor the [well-being] status of any human being is through observation of the face...,” the researchers suggest. Therefore it is “advisable to assign 1 officer to monitor the restrained subject’s face for signs of distress/difficulty.”
This study was funded by the Canadian Police Research Center. The results appear in the latest issue the Journal of Forensic and Legal Medicine, under the title “Incidence and outcome of prone positioning following police use of force in a prospective, consecutive cohort of subjects.” Click here for a brief abstract that’s is available free of charge. The full seven-page study paper can be ordered for a fee at that site.