Causes of Death in the Presence of Law Enforcement in Johnson County, Iowa, 2011–2020

Abstract

Objectives. To examine sudden and unexpected or trauma-related deaths that occurred in the presence of law enforcement in Johnson County, Iowa, between 2011 and 2020.

Methods. We identified deaths in the presence of law enforcement using definitions from the National Association of Medical Examiners. We obtained data, including demographics, cause and manner of death, toxicology results, and circumstances and location of event leading to death, from comprehensive medical examiner investigative reports.

Results. There were 165 deaths that occurred in the presence of law enforcement: 114 were from a known disease, and 51 were either trauma related or the sudden, unexpected initial presentation of a previously unrecognized disease. Three deaths occurred in the context of physical restraint by law enforcement. Suicide was the leading manner of death among trauma-related deaths; the means of suicide was predictable based on in-custody (hanging) or precustody (firearm) circumstances.

Conclusions. Our findings highlight the potential role of medical examiners and coroners in improving completeness of data on reporting death in the presence of law enforcement to public health agencies. (Am J Public Health. 2024;114(6):642–650. https://doi.org/10.2105/AJPH.2024.307616)

Medicolegal death investigation (MDI) systems, which involve medical examiners, coroners, or justices of the peace, have a unique public service position that bridges the medical, public health, and legal communities. Recognizing the societal value of understanding how people die, US states and territories have created MDI systems by statute to investigate all sudden, unexpected, unexplained, or trauma-related deaths.

A common example of sudden, unexpected death is a death caused by previously undiagnosed heart disease. Common examples of injury-related deaths are blunt trauma from a motor vehicle collision, asphyxia from hanging, illicit drug intoxication, and gunshot wounds. In general, MDI jurisdiction also includes some specific death types (e.g., of those younger than 18 years, from a novel infectious disease, of an unidentified decedent, maternal death, and in-custody death). The outcome of every investigated case is a medical opinion of the cause of death (e.g., a specific disease or injury) and the manner of death (e.g., natural, accident, suicide, homicide, or undetermined).

The work of a death investigation system is centered on public health: learning from the deceased to improve population health and reduce mortality. By investigating sudden, unexpected deaths, trauma deaths, and additional specific types of death as required by law (e.g., in-custody deaths), MDI systems conduct surveillance and generate epidemiological data. Surveillance examples are (1) the recognition and early description of hantavirus pulmonary syndrome in 1993 by the New Mexico Office of the Medical Investigator,¹ and (2) real-time recognition of illicit drug deaths.² Examples of epidemiological data generated by death investigation systems are (1) common characteristics of COVID-19 deaths in Cook County, Illinois,³ and (2) common characteristics of the sleep environment in sudden unexplained infant deaths.⁴

Recognition of MDI as a public health activity is complicated by the incorrect stereotypes of focusing solely on homicide and of conducting an autopsy on every investigated death. Homicides (i.e., injury-related deaths caused by the volitional act of another person) are always medical examiner or coroner cases but represent only a small fraction of overall caseload. Furthermore, medical examiner and coroner death investigations may, but do not always, necessitate an autopsy, which is not the only tool available. The high profile of homicide deaths also encourages a false impression of an MDI office as working for law enforcement. Medical examiner and coroner independence is a core professional principle no matter an agency’s administrative placement in government.⁵ Death investigation systems do not investigate crimes. Rather, they investigate certain types of death as outlined by state or territorial laws (i.e., sudden, unexpected, unexplained, or traumatic). This independence is particularly important during the evaluation of deaths that occur in the presence of law enforcement. This is an uncommon but not rare case type, and each investigation demands a thoughtful, unhurried approach focused on correct identification of cause and manner of death.

Despite the legal requirement for medical examiners and coroners to investigate in-custody deaths, deaths that occur in the presence of law enforcement are undercounted in the US public health record. This occurs because there is no single comprehensive reporting agency, compulsory reporting mechanism, or uniform definition of law enforcement presence.^6,7 Currently, public health professionals seeking a data resource must rely on a blend of federal- and state-funded programs, including the Death in Custody Reporting Act of 2013 and its derivative, the Federal Law Enforcement Agency Deaths in Custody Reporting Program; the Mortality in Correctional Institutions program; the National Vital Statistics System; and the National Violent Death Reporting System.⁸

Public health professionals seeking to study deaths in the presence of law enforcement are further limited by the lack of a uniform definition in the existing body of literature. Definitions include deaths that occurred in a correctional facility only,^9,10 a specific manner of death or scenario of death,^11–13 and an expanded definition of in custody that includes deaths occurring during law enforcement personnel pursuit or apprehension.^14–16 Relatively few studies have examined county-level data on deaths that occurred in the presence of law enforcement,^10,17–19 and none, to our knowledge, are specific to the state of Iowa.

We performed a descriptive analysis of deaths that occurred in the presence of law enforcement and were investigated by the Johnson County, Iowa Medical Examiner Department (JCME) between 2011 and 2020. We aimed to demonstrate that medical examiners and coroners are key sources of local, case-specific data on deaths that occur in the presence of law enforcement.

METHODS

In a 2017 position paper, the National Association of Medical Examiners (NAME) put forth an encompassing definition of “in custody” that delineates the breadth of precustody and in-custody scenarios of death in the presence of law enforcement.²⁰ We employed the NAME-defined categories “precustody” and “in custody.” We use the term “law enforcement presence” to broadly refer to both precustody and in-custody scenarios.

NAME defined precustody deaths as deaths that occur

under the perceived or physical control or restraint of a law enforcement officer, a corrections officer (including a private corrections officer), or an authorized employee or agent of a district juvenile secure facility or youth residential facility.^20(p607)

The NAME position paper emphasized the perception of limiting an individual’s “freedom of movement” during the precustody phase and included scenarios in which an individual was in the process of being detained or arrested. Examples were motor vehicle pursuits by law enforcement with the intention of detention or arrest and law enforcement surrounding a house during a hostage or barricade situation. Both examples illustrate that one need not be physically restrained to perceive a limitation of freedom of movement by law enforcement. NAME defined in-custody deaths as deaths that occur while the decedent is “in actual police custody, corrections custody (both pre-trial and sentenced), and when the individual is in legal custody but not in the custody of a correctional agency”^(p607) and defined the start of this phase as the moment of physical arrest.²⁰

Medical Examiner Jurisdiction

JCME conducts medicolegal death investigations in Johnson County, Iowa. Although it encompasses only 614 square miles and hosts an estimated population of only 152 854 as of 2020,²¹ Johnson County is home to both the University of Iowa Hospitals and Clinics (UIHC), a tertiary care academic medical center that receives patient transfers from much of Iowa and adjacent areas of border states, and the Iowa Medical Classification Center, a correctional facility with medical and psychiatric beds and the only hospice unit for the Iowa Department of Corrections.

JCME has jurisdiction over deaths in Johnson County that are sudden and unexpected or trauma related, as well as over any “death of a person confined in a prison, jail, or correctional institution.”^22(p1) By convention, JCME also investigates all deaths in the presence of law enforcement that occur before the individual is taken into custody. JCME jurisdiction is also determined by the location of death (i.e., in Johnson County) and not by the location of an ultimately fatal incident and, therefore, includes many deaths that occurred at UIHC for which the event leading to death occurred in a location other than Johnson County (e.g., an individual with an arrest-related injury sustained in a location other than Johnson County may be transferred to UIHC for a higher level of medical care).

Data Collection and Analyses

We identified deaths in the presence of law enforcement by reviewing all JCME case files for 2011 to 2020. We extracted data from deaths in the presence of law enforcement from medical examiner investigative narratives, medical records, and autopsy reports, where applicable, to obtain age, sex, cause and manner of death, toxicology results, circumstances and location of events leading to death, location of death pronouncement, and year of death. We obtained the decedent’s race through review of death certificates generated using the Iowa Electronic Death Registration System. Funeral homes collect racial identity via interviewing the decedent’s next of kin.

We report descriptive statistics as frequencies and percentages. We used χ² analysis to compare categorical variables, with statistical significance set at α = 0.05. We conducted statistical analysis using SAS version 9.4 (SAS Institute, Cary, NC).

RESULTS

There were 165 deaths that occurred in the presence of law enforcement: 114 (69.1%) were attributable to a previously known disease process, 37 (22.4%) were trauma related, and 14 (8.5%) were the sudden and unexpected initial presentation of an ultimately fatal disease process.

A mean of 16.5 deaths in the presence of law enforcement (median = 15.5; range = 9–28) occurred annually during our 10-year study period. Of the 28 deaths in 2020, 13 (46.4%) were attributed to known COVID-19 disease as either the primary cause of death or a significant infection that contributed to death from another condition (e.g., COVID-19 infection hastening the death of a hospice cancer patient; Figure 1).

FIGURE 1—

Deaths in the Presence of Law Enforcement: Johnson County, IA, 2011–2020

Note. A total of 165 deaths occurred in the presence of law enforcement during this period. Deaths in the presence of law enforcement can be further categorized as sudden and unexpected or trauma-related deaths (n = 51, depicted in blue) and deaths occurring as the result of a known disease (n = 114, depicted in orange).

Previously Known Disease Process

There were 114 individuals who died in the presence of law enforcement from a known disease process, most of whom were enrolled in hospice (Table 1). These decedents were primarily male (98.2%), White (81.6%), or Black/African American (17.5%) and had a mean age of 62.4 years (median = 65; range = 21–89). These decedents were all in custody at the time of death, manner of death was natural, and none required an autopsy examination because death resulted from a known fatal disease process.

TABLE 1—

Demographic Characteristics of Deaths in the Presence of Law Enforcement: Johnson County, IA, 2011–2020

	Total (n = 165), No. (%)	Sudden and Unexpected or Trauma-Related Deaths (n = 51), No. (%)	Deaths From Previously Known Disease (n = 114), No. (%)	P a
Sex				.002
Male	156 (94.5)	44 (86.3)	112 (98.2)
Female	9 (5.5)	7 (13.7)	2 (1.8)
Race				.38
White	138 (83.6)	45 (88.2)	93 (81.6)
Black/African American	25 (15.2)	5 (9.8)	20 (17.5)
Asian	2 (1.2)	1 (2.0)	1 (0.9)
Age, y				< .001
≤ 24	10 (6.0)	8 (15.7)	2 (1.8)
25–34	15 (9.1)	13 (25.5)	2 (1.8)
35–44	12 (7.3)	9 (17.7)	3 (2.6)
45–54	28 (17.0)	15 (29.4)	13 (11.4)
55–64	39 (23.6)	4 (7.8)	35 (30.7)
≥ 65	61 (37.0)	2 (3.9)	59 (51.7)

^a We used the χ² test to compare categorical variables. Statistical significance is set at α = 0.05.

Sudden and Unexpected or Trauma Related

Of the 51 deaths in the presence of law enforcement that were sudden and unexpected or trauma related, decedents were primarily male (86.3%) and White (88.2%). The mean age was 39.4 years (median = 40; range = 13–79; Table 1). All but 1 received an autopsy examination. Relative to those for whom death in the presence of law enforcement was from a known disease process, decedents whose deaths were sudden and unexpected or trauma related were significantly more likely to be younger (P < .001) and female (P = .002).

The most common manner of death among sudden and unexpected or trauma-related deaths was suicide (n = 22; 43.1%; Table 2). Suicide deaths included 11 (50%) in-custody hangings in which the ligature was a bedsheet (n = 8), telephone cord (n = 2), or jail-issued pants (n = 1). All 9 self-inflicted gunshot wound deaths occurred before the individual was taken into custody and all involved a handgun. The majority of sudden and unexpected disease-related deaths (i.e., natural manner deaths) were attributed to hypertensive or atherosclerotic heart disease or both (n = 8; 57.1%). The majority of accident deaths occurred during motor vehicle collisions while being pursued by law enforcement (n = 5; 55.6%). Among the 5 deaths classified as a homicide, 2 (40.0%) were from a law enforcement shooting (precustody) and 2 (40.0%) were from an incarcerated person altercation (in custody).

TABLE 2—

Manner and Cause of Death Among Sudden and Unexpected or Trauma-Related Deaths: Johnson County, IA, 2011–2020

	No. (%)
Suicide	22 (43.1)
Hanging	11 (50.0)
Self-inflicted gunshot wound	9 (40.9)
Exsanguination from intentional manipulation of a medical device	1 (4.5)
Jump from height	1 (4.5)
Natural	14 (27.5)
Cardiovascular disease	8 (57.1)
Infection	3 (21.4)
Pulmonary thromboembolism	2 (14.3)
Neoplasm	1 (7.1)
Accident	9 (17.7)
Motor vehicle collision during pursuit	5 (55.6)
Fall from standing height	2 (22.2)
Complications of ethylene glycol intoxication	1 (11.1)
Complications of methamphetamine intoxication with law enforcement  restraint	1 (11.1)
Homicide	5 (9.8)
Inmate altercation	2 (40.0)
Law enforcement shooting	2 (40.0)
Law enforcement restraint	1 (20.0)
Undetermined	1 (2.0)
Complications of methamphetamine intoxication and excited delirium  with law enforcement restraint	1 (100.0)

Note. Sample size was n = 51.

More of the 51 sudden and unexpected or trauma-related deaths occurred in custody (n = 30; 58.8%) than during the precustody phase (n = 21; 41.2%; Table 3). The manner of death differed significantly by custody status (P = .002). In-custody deaths were primarily natural manner (n = 14; 46.6%), followed by suicide (n = 12; 40.0%), homicide (n = 2; 6.7%), and accident (n = 2; 6.7%). Precustody deaths were primarily suicide (n = 10; 47.6%), followed by accident (n = 7; 33.3%), homicide (n = 3; 14.3%), and undetermined (n = 1; 4.8%). Among Black/African American individuals, there was a larger percentage of precustody (19.0%) than in-custody (3.3%) deaths. In addition, 3 precustody deaths occurred while law enforcement physically restrained the individual.

TABLE 3—

Sudden and Unexpected or Trauma-Related Deaths by Custody Status: Johnson County, IA, 2011–2020

	Total (n = 51), No. (%)	In Custodya (n = 30), No. (%)	Precustodyb (n = 21), No. (%)	P c
Manner				.002
Suicide	22 (43.1)	12 (40)	10 (47.6)
Natural	14 (27.5)	14 (46.6)	0 (0.0)
Accident	9 (17.6)	2 (6.7)	7 (33.3)
Homicide	5 (9.8)	2 (6.7)	3 (14.3)
Undetermined	1 (2.0)	0 (0.0)	1 (4.8)
Race				.13
White	45 (88.2)	28 (93.4)	17 (81)
Black/African American	5 (9.8)	1 (3.3)	4 (19)
Asian	1 (2.0)	1 (3.3)	0 (0.0)

^a We defined in-custody deaths as any death that occurred while (1) in police custody (arrested or detained), (2) in corrections custody, or (3) in legal custody but not in custody of a correctional agency.

^b We defined precustody deaths as any death that occurred while (1) in the process of being detained, (2) in the process of being arrested, (3) in the process of a law enforcement pursuit, or (4) there was perceived limitation of one’s freedom of movement by law enforcement.

^c We used the χ² test to compare categorical variables. Statistical significance was set at α = 0.05.

For most of the 51 decedents, the event ultimately leading to death (i.e., injury or initial presentation of a previously unknown disease) was not immediately fatal and death was pronounced at UIHC (78.4%). The location of the event ultimately leading to death was most often the Iowa Medical Classification Center (31.4%) or another correctional facility (25.5%), followed by a place of residence (17.6%), in a vehicle either parked or on a roadway (15.7%), outdoors while on foot (5.9%), in a restaurant (2.0%), or at a mental health facility (2.0%).

Postmortem toxicology examination was performed on blood or urine specimens or both collected from 48 decedents in our study and was positive in 21 (44.0%) cases, nearly half of which (47.6%) were positive for multiple substances. Most decedents with positive postmortem toxicology test results died in the precustody phase (76.2%) and were White (76.2%) or Black/African American (23.8%).

All 5 decedents who died of motor vehicle collision injuries during law enforcement pursuit had positive postmortem toxicology test results. Two of the 3 decedents driving a car, truck, or SUV during the event leading to death were not wearing seatbelts. Neither decedent riding a motorcycle during the event leading to death was helmeted.

DISCUSSION

This study provides a descriptive analysis of 165 deaths in the presence of law enforcement that occurred in Johnson County, Iowa, between January 1, 2011 and December 31, 2020. JCME had jurisdiction over every death; it is presumed that JCME captured every death in the presence of law enforcement in Johnson County over the 10-year period. One of our goals was to demonstrate the role of medical examiners and coroners as key sources of data on deaths in the presence of law enforcement. Improving the completeness of data available to public health professionals may improve population health and—possibly—reduce precustody and in-custody mortality.

Most decedents in our study were male, which is comparable to federal data collected by the Bureau of Justice Statistics and corresponds to the gender distribution of incarcerated individuals in the United States.^14,23 Black/African American individuals constituted 15.2% of all deaths in the presence of law enforcement and 19% of precustody deaths, although only 4.1% of Iowa’s population identifies as Black/African American.^21,24 Individuals of color are overrepresented in Iowa’s correctional facilities.²⁵ Approximately one quarter (25.4%) of individuals incarcerated in Iowa are Black/African American.^21,24 Likewise, 23.1% of arrests that occurred in Iowa in 2020 were of individuals who identified as Black/African American.²⁶

The overall leading manner of death was natural from previously known disease and, less frequently, the sudden, unexpected presentation of a previously unrecognized disease. These findings parallel those reported by the Bureau of Justice Statistics in a large study of federal in-custody and arrest-related deaths using similar definitions and methodology.¹⁴ It is worth noting that, among all deaths in the United States during 2019 (the most recent pre–COVID-19 year), more than 90% were from natural causes.²⁷ By contrast, 77.6% of deaths in the presence of law enforcement during our 10-year study were from natural causes. During 2020, the number of deaths in the presence of law enforcement in Johnson County, Iowa, markedly increased because of the COVID-19 pandemic. Throughout the pandemic, incarcerated populations proved to be particularly vulnerable to COVID-19 outbreaks across the United States, primarily because of high transmission rates in prison and jail facilities and the inability to adequately test, monitor, or quarantine individuals in custody.²⁸

More than one half of the trauma-related deaths were suicides. Our findings reveal predictable self-injury means among in-custody (hanging) and precustody (gunshot wound) individuals, suggesting the need for future studies regarding improved in-custody suicide risk mitigation in correctional facilities and de-escalation techniques when the subject may have access to a firearm.

Many decedents had positive toxicology test results for at least 1 substance, and most of these were before the individual was taken into custody. All 3 decedents who died while being restrained had positive toxicology test results for methamphetamine. Future studies should consider examining the interplay between intoxicant use and death in the presence of law enforcement.

Limitations

Our study was not without limitations. In data extraction, we used an inefficient manual review of case files (paper and, later, electronic). The proportion of deaths of natural cause was increased by COVID-19–related deaths because of the inclusion of the year 2020. Additionally, Iowa’s population has little racial diversity. Given that our findings are specific to JCME, the generalizability of our results beyond southeast Iowa is uncertain.

Public Health Implications

MDI systems are uniquely positioned to recognize, characterize, and report all deaths of those in custody (per the NAME definition) and all deaths occurring in the presence of law enforcement (our terminology). This is because (1) in the United States, medical examiners and coroners generally have jurisdiction over in-custody death cases; and (2) medical examiners and coroners generate a death certificate on every death within their jurisdiction. The death certificate is a well-established tool to collect epidemiological data, and there is a process for implementing updates to identify case characteristics of public health significance (e.g., a new checkbox item to indicate whether a decedent died in the presence of law enforcement).

To an epidemiologist, the death certificate is a survey instrument and data collection device with standard fields and format. Two sections (i.e., “cause of death” and “how injury occurred”) are free text and encourage the certifier to use medical terminology (e.g., cause of death) and a simple narrative (e.g., how an injury occurred if not a disease-related death) that together best reflect the certifier’s opinion. Other sections are checkboxes or fill-in-the-blank format (e.g., manner of death and date of injury, respectively). Death certificates are state- and territory-centered documents, but all are patterned after a standard US certificate of death.²⁹

The standard certificate of death may be updated at regular intervals, with states and territories following suit. The most recent updates of the standard US certificate of death were issued in 2003. Among many changes, the 2003 revision added 3 public health–centered topics: contribution of tobacco use to death, pregnancy status, and general information about transportation scenario deaths.²⁹ A query centered on deaths in the presence of law enforcement is not among the 2003 revisions.

Medical examiners and coroners should know of every death that occurs in the presence of law enforcement in their jurisdiction and can apply consistent practice to description and reporting. Even though medical examiners and coroners certify deaths in the presence of law enforcement, the in-custody characteristics of such deaths may not be captured on the death certificate for 3 reasons: (1) a specific in-custody query is lacking, (2) the “how injury occurred” section is limited to nonnatural manner deaths, and (3) the free text format of the “how injury occurred” section permits omission of in-custody identifiers (e.g., entering “shot by another person” instead of “shot by on-duty law enforcement officer”). Furthermore, the dynamics and relationship between law enforcement and medical examiners and coroners should be an area of further research, especially when deaths in the presence of law enforcement remain undercounted.

It is likely that an explicit designation of medical examiners and coroners as the source for information about deaths in the presence of law enforcement would yield more accurate data collection for public health purposes. For a parallel success story of medical examiner and coroner reporting, one may look to information gleaned from death certificates during the opioid death epidemic. Medical examiners and coroners have jurisdiction over drug-related deaths and issue a death certificate for each. Position papers have described optimal strategies for investigating, interpreting, and certifying drug-related deaths.^30,31 Following standardized recommendations for how medical examiners and coroners certify the cause of death, there has been a marked national improvement in drug-related death data quality.³²

There are at least 2 potential mechanisms for medical examiners’ and coroners’ rapid reporting of deaths in the presence of law enforcement to a public health agency. The first is via death certificates, as described earlier—a strategy that is limited by how the certifier enters data on the death certificate and the absence of a checkbox for deaths in the presence of law enforcement. The second is via a standardized medical examiner or coroner workflow that includes a nondeath certificate mechanism to identify this death type. Optimally, this mechanism would be a component of data collection of every investigated case, whether through a death investigation system’s electronic case management system or a state-level electronic death registration system. Either data record system should be programmable for a mandatory query about law enforcement presence. Such a system was added to Iowa’s electronic death registration system in 2024 (Dennis Klein, MD, Iowa Chief Medical Examiner, personal oral communication, September 18, 2023). The designation of death as occurring in the presence of law enforcement should be standard practice, as standard as reporting age and sex of every decedent.

Our results underscore the medical examiner and coroner as a key source of data regarding deaths in the presence of law enforcement and highlight the importance of generating a more streamlined process for documenting deaths in the presence of law enforcement. If medical examiners and coroners are tasked with the mandatory reporting of precustody and in-custody deaths in their jurisdictions (via a tailored death certificate or other mechanism), a more complete picture of deaths in the presence of law enforcement in the United States will emerge and provide invaluable public health information.