Abstract
One in five fatal police shooting victims may have been experiencing a mental health crisis (MHC) at the time of their death [1]. We use data on fatal police shootings from the National Violent Death Reporting System (2014–2015) to (a) identify incidents where the victim is reported to have experienced an MHC at the time of their death, (b) describe the characteristics of these incidents, and (c) compare the characteristics of MHC to fatal police shootings where the victim was not experiencing an MHC at the time of their death. We systematically coded 633 fatal police shootings from 27 states. Descriptive statistics characterized fatal police shootings, including victim characteristics; their mental health status; and contextual information regarding the police encounter (e.g., reason for police call). Overall, 203 of 633 fatal police encounters (32%) involved victims who showed signs of an MHC at the time of their death. Victims were predominantly white, male, and in possession of a firearm. In 3 of 4 cases, the MHC manifested as suicidal ideation despite any relevant documented history among most victims. Among half of suicidal victims, suicidal ideation was expressed verbally and in-person to a family member/intimate partner who subsequently called the police. Dispatch was aware of the MHC in 1 of 4 of total police calls. Overall, fatal police encounters involving those experiencing an MHC accounted for 1 in 3 of our caseloads. Approximately, 3 of 4 mental health calls involved a suicidal person who mainly expressed intent to a loved one in-person.
Keywords: Firearm, Homicide, Police, Policing, Mental health, Suicide
Introduction
More than 1000 people are shot and killed annually by police officers acting in the line of duty in the USA [2–4]. Between 2012 and 2018, fatal police shootings accounted for approximately 8% of all adult male homicides [5]—with the year 2022 being the deadliest on record since 2013 [6]. Studies using open-source data, such as the Washington Post’s Fatal Encounters database, have found that approximately one in five people who are shot and killed by police may have been experiencing a mental health crisis at the time of their death. Nonetheless, literature on fatal police encounters involving victims with mental health crises remains scant. The National Violent Death Reporting System (NVDRS), a data surveillance system for violent deaths coordinated by The Centers for Disease Control, may provide a rich source of information to better characterize police homicide data, particularly those involving individual with mental health illnesses. The NVDRS counts more violent deaths, including legal intervention (or police) homicides, than government databases like The National Vital Statistics System (NVSS). It also provides contextual and mental health information for police homicide victims, which may serve complimentary to already established open-source data like Fatal Encounters.
Given these advantages, we use data from the National Violent Death Reporting System to describe the characteristics of police homicides in 2014–2015—differentiating between police encounters that did and did not involve a victim who showed signs of suffering from a mental health crisis at the time of their death.
Methods
Study Population
This study used data from The National Violent Death Reporting System (NVDRS) for the years 2014 and 2015. The NVDRS is a database that is coordinated by the Centers for Disease Control and Prevention, which collects data on violent deaths in the USA including homicides, suicides, and unintentional shootings—these include legal intervention homicides (LIHs) [7]. In 2014 and 2015, 27 states participated in NVDRS (see Appendix, Fig. 1). In each state, abstractors use various documents, including coroner and medical examiner (CME) reports, law enforcement (LE) reports and death certificates, to code for standardized variables about the incident. Two incident narratives are also written by the abstractors that summarize the CME and LE reports.
Fig. 1.
Number of legal intervention homicides (state, year)
The NVDRS dataset includes standard variables on demographics (such as age, race, and sex of victim), injury and death (e.g., cause and place of death), and other circumstances (e.g., use of weapon and any mental health diagnoses). An extensive list of the standard NVDRS variables is published online [7]. Among variables routinely coded about victims, NVDRS provides information on the victim’s prior and current mental health condition and treatment. More specifically, NVDRS includes variables for: any prior treatment for any mental health disorders; any prior history of suicidal ideation; current mental health diagnoses; and whether the victim was receiving any mental health treatment at the time of the incident. Among mental health-related illnesses, we include substance dependence for alcohol or non-alcohol related substances.
Case Definitions and Data Source
As part of an earlier study [8], all legal intervention homicides (LIH) in NVDRS 2014–2015 were identified. Eligible cases met the following criteria: (1) the manner of death was described as a homicide; (2) the suspect was a law enforcement officer (local, state, or federal level); and (3) the incident occurred in the line of duty—this included off-duty officers who are acting in the line of duty, but excludes officers who commit homicide outside the scope of their professional role, such as murdering an intimate partner. Through a process described elsewhere [8], all LIH cases underwent a second round of coding as part of which 13 additional legal intervention homicide-specific data elements were coded from the incident narratives (see Appendix, Fig. 2). This second round of coding aimed to include data elements that reflected situational and individual characteristics previously identified in the LIH literature as relevant to a comprehensive understanding of these events (e.g., force used by victims against law enforcement). Of the 13 variables, 2 were directly relevant to mental health illness; specifically: whether the victim was suicidal during the police encounter; and whether, at the time of the incident, the victim’s behavior was impaired by drugs, alcohol, mental health illness, or any other reason.
Fig. 2.
Additional variables used to code narratives
Variable Coding
For the purposes of the current paper, the dataset underwent a third round of coding. We used the LE and ME narratives to code supplementary variables, which further contextualized each police encounter involving mental health crises. These variables were the following: (1) the reason for the police call relayed to dispatch (e.g., mental health crisis, domestic disturbance, suspected crime) and (2) whether there was evidence that the victim was experiencing a mental health crisis immediately preceding and/or at the time of the police encounter, and whether, according to the narratives, dispatch was aware of the MHC. Cases were then grouped into two groups (group 1, group 2) depending on whether a mental health illness influenced the victim’s behavior prior to and/or during the incident (group 1), or whether mental health disturbance was not relevant to the incident (group 2).
Defining a Mental Health Crisis
We define a mental health crisis as an incident where an individual was at immediate risk to themselves or others due to a perceived or known mental health illness or was otherwise disoriented due to a cognitive impairment. Of note, victims who only expressed suicidal intent at the prospect of being apprehended or detained by the police during an active crime—with otherwise no history of suicidal intent or mental health illness prior to or during the incident—were not considered suicidal or suffering from a mental health disturbance.
Statistics Analysis
Descriptive characteristics of incidents in group 1 and group 2 are presented. Between-group comparisons use Pearson’s chi-square tests. If cases involved missing data, they were excluded from the analysis.
Results
In 2014 and 2015, there were 633 cases that met the definition of a “legal intervention homicide” (LIH). Of all cases, 32.1% (n = 203) of victims showed signs of a mental health crisis immediately preceding and/or during their police encounter (group 1). Among all cases (group 1 and group 2), victims were disproportionately male (95.6%), non-Hispanic white (53.2%). Victims in group 1 were older on average than victims in group 2 (mean age 40.9 years vs 35.8 years) and more likely to be white, non-Hispanic (67.5% vs. 46.5%)—see Table 1.
Table 1.
Case characteristics
| Group 1 – Mental health disturbance (n = 203) | Group 2 – No mental health disturbance (n = 430) | Total (n = 633) | P values | |
|---|---|---|---|---|
| Demographics (%) | ||||
| Age | < 0.01 | |||
| 16–35 | 83 (40.9) | 244 (56.7) | 327 (51.7) | |
| 36–55 | 91 (44.8) | 145 (33.7) | 236 (37.3) | |
| 56–75 | 26 (12.3) | 39 (9.1) | 65 (10.3) | |
| ≥ 76 | 3 (1.5) | 2 (0.5) | 5 (0.1) | |
| Mean | 40.9 | 35.8 | ||
| Biological sex | 0.16 | |||
| Female | 12 (5.9) | 15 (3.5) | 27 (4.3) | |
| Male | 191 (94) | 415 (96.5) | 606 (95.7) | |
| Race | < 0.01 | |||
| Black (non-Hispanic) | 25 (12.3) | 135 (31.4) | 160 (25.3) | |
| Hispanic | 27 (13.3) | 68 (15.8) | 95 (15.0) | |
| White (non-Hispanic) | 137 (67.5) | 200 (46.5) | 337 (53.2) | |
| Other | 14 (6.9) | 27 (6.3%) | 41 (6.5) | |
Overall, 91% (n = 576) of victims were reported to have used force against law enforcement. Of those who used force, victims in group 1 were less likely to assault law enforcement (45.4% vs 51.2%), but more likely to threaten LE (51.9% vs 31.9%). Lastly, 73.1% of victims (n = 463) used a weapon during the incident—victims in group 1 used a larger proportion of knives and a smaller proportion of firearms compared with group 2 (Table 2).
Table 2.
Use of force characteristics
| Group 1 – Mental health disturbance (n = 203) | Group 2 – No mental health disturbance (n = 430) | P values | |
|---|---|---|---|
| Did victim use force against law enforcement? (%) | 0.93 | ||
| Yes | 185 (91.1) | 391 (90.9) | |
| No | 18 (8.9) | 39 (9.1) | |
| *Highest level of forced used against LE? (%) | < 0.01 | ||
| Assaulted | 84 (45.4) | 220 (51.2) | |
| Threatened | 96 (51.9) | 137 (31.9) | |
| Other/unspecified | 5 (2.7) | 34 (7.9) | |
| Did victim use a weapon during the incident? (%) | 0.04 | ||
| Yes | 159 (78.3) | 304 (70.7) | |
| No | 44 (21.7) | 126 (29.3) | |
| *Type of weapon used (%) | 0.1 | ||
| Firearm | 96 (60.4) | 206 (67.8) | |
| Knife | 41 (25.8) | 58 (19.1) | |
| Motor vehicle | 5 (3.1) | 17 (5.6) | |
| Other/unspecified | 17 (10.7) | 23 (7.6) | |
Among all LIH cases, LE were most commonly responding to calls for suspected crimes (38.4%), domestic disturbances (23.1%), and behavioral/mental health crises (11.8%) (not shown). Group 1 cases were less likely to occur in response to suspected crimes, with 36.9% being primarily for a suspected mental health disturbance. Domestic disturbance cases were the second most common police call in group 1—accounting for 27.1% of victims (n = 55). Approximately, 1 in 5 domestic disturbance cases in group 1 involved a suicidal person (data not shown). Excluding police responses specifically for mental health disturbances, dispatch was aware of a mental health problem in 1 out of 4 of group 1 cases—see Table 3.
Table 3.
Contextual information on police encounters
| Group 1 – Mental health disturbance (n = 203) | Group 2 – No mental health disturbance (n = 430) | P values | |
|---|---|---|---|
| Reason for police contact (%) | ** < 0.01 | ||
| Domestic disturbance | 55 (27.1) | 91 (21.2) | |
| Mental health crisis | 75 (36.9) | 0 (0.0) | |
| Suspected crime | 38 (18.7) | 205 (47.7) | |
| Other | 35 (17.2) | 134 (31.2) | |
| *Dispatch aware of MH problem (%) | |||
| *34 (26.6) | |||
*Cases for mental health crises were excluded from both numerator and denominator. Apart from police calls for mental health crises specifically, of all other group 1 cases where the police were summoned (n = 128), dispatch was aware of the relevant mental health problem 27% of the time (n = 34)
**Comparing group 1 (all mental health cases) and group 2
The most common reason for a mental health-related police call involved a suicidal person, which accounted for 77.3% (n = 58) of mental health calls—see Table 4. Over half involved a verbal or written expression of suicide; these ranged from the victims verbally informing their loved ones of their suicidal intent during in-person verbal communication to victims sending messages via cell phone or social media to others (data not shown). Three cases described a case of “suicide by cop” where victims, all with a known history of suicidal illness accompanied by acute??mental health deterioration, expressed the intention to call the police with hopes that they [LE] would fatally shoot the victim. Approximately 1 in 5 cases (18.7%) involved a non-suicidal mental health crisis. These ranged from victims suffering from visual hallucinations to the police responding to calls from the public of an individual walking on a highway with disturbed behavior. Two victims died during a police call for an emergency custody order (ECO). These are temporary detention orders that are issued by a magistrate for an individual who poses a risk to themselves or others, due to their mental health, and needs hospitalization for treatment. An ECO is issued by the courts when someone is unable or unwilling to volunteer themselves for either psychiatric hospitalization or treatment. In over half of mental health cases (54.6%), the police were called by the victim’s family members, their partner, or the victim themselves—see Table 5.
Table 4.
Reason for police contact (group 1 – mental health calls)
| Reason for mental health call | Total cases (n = 75) |
|---|---|
| Suicidal ideation | 58 (77.3) |
| Verbal/written intent (e.g., informed partner) | 39 |
| Physical self-harm (e.g., slashing neck) | 8 |
| Overdose or poisoning | 5 |
| Other/unknown | 6 |
| Other behavioral disturbance | 14 (18.7) |
| Person on highway/amidst traffic | 4 |
| Delusions and/or hallucinations | 3 |
| Other (e.g., fecal smearing, self-barricading) | 7 |
| Other/unknown (e.g., emergency custody order) | 3 (4.0) |
| Total | 75 (100.0) |
Table 5.
Mental health calls: who called the police?
| Individual calling 911 | Total cases |
|---|---|
| Parents/other family member | 17 (22.7) |
| Partner | 15 (20.0) |
| Victim | 9 (12.0) |
| Healthcare professional | 4 (5.3) |
| Other/unknown | 30 (40.0) |
| Total | 75 (100.0) |
Of all group 1 cases, 42.4% of victims (n = 86) were noted to have wat least 1 mental health diagnosis, most commonly a depressive disorder (14.8%). Only 25.1% (n = 51) were known to be in active treatment and only 33.9% (n = 78) were known to have ever been treated for a mental health condition. Approximately 1 in 4 victims were known to have had an alcohol problem and 1 in 3 a non-alcohol substance problem (see Table 6). Only a 3.4% (n = 7) of victims were known to have ever reported suicidal ideation prior to their fatal police encounter (data not shown).
Table 6.
Mental health characteristics
| Group 1 – Mental health disturbance (n = 203) | Group 2 – No mental health disturbance (n = 430) | P values | |
|---|---|---|---|
| At least 1 current mental health diagnosis (%) | |||
| Yes | *86 (42.4) | 42 (9.8) | < 0.01 |
| No | 117 (57.6) | 388 (90.2) | |
| Mental health diagnosis (%) 0.79 | |||
| Anxiety disorder | 10 (4.9) | 7 (1.6) | |
| Bipolar disorder | 16 (7.9) | 9 (2.1) | |
| Depression/dysthymia | 30 (14.8) | 12 (2.8) | |
| Post-traumatic stress disorder | 10 (4.9) | 3 (0.7) | |
| Schizophrenia | 18 (8.9) | 9 (2.1) | |
| Other/unspecified | 27 (13.3) | 18 (4.2) | |
| Currently in treatment for mental health problem (%) | < 0.01 | ||
| Yes | 51 (25.1) | 20 (4.7) | |
| No | 152 (74.9) | 410 (95.3) | |
| History of ever being treated for a mental health illness (incl substance abuse) (%) | < 0.01 | ||
| Yes | 78 (33.9) | 34 (7.9) | |
| No | 125 (54.3) | 396 (92.1) | |
| History of alcohol dependence problem (%) | < 0.01 | ||
| Yes | 55 (27.1) | 5 (1.2) | |
| No | 148 (72.9) | 425 (98.8) | |
| History of dependence problem (non-alcohol) (%) | < 0.01 | ||
| Yes | 67 (33.0) | 79 (18.4) | |
| No | 136 (67.0) | 351 (81.6) | |
Discussion
In our dataset, one in three police homicides in the line of duty involved victims with a mental health crisis. In 64% of cases, these crises occurred during police calls for a mental health crisis or a domestic disturbance. Among all LIHs, those that involved a mental health disturbance disproportionately involved a middle-aged, non-Hispanic white men with a background of mental illness who used force against law enforcement in the moments prior to their fatal events—48% of whom were in possession of firearms.
Police calls for mental health crises involved victims who were suicidal, those with non-suicidal MHCs (e.g., signs of acute psychosis, e.g., visual hallucinations and delusions) and other specialist cases, such as those who were sought for involuntary psychiatric admission under an emergency court order. A previous report on mentally unwell victims of police homicides describes a largely untreated population who are not linked to care [9]. This could align with our findings, as only half of those with a mental health diagnosis had received some type of treatment for mental health illness in the past, and only 5% of suicidal victims had a reported history of suicidal ideation. Despite this, caution is warranted in interpreting NVDRS data as a reliable source of prior mental health history given its lack of validation.
Police are commonly called as first responders to mental health crisis situations [10]. This is reflected in the evolution of policing practices in the USA. Some state-based protocols ensure that the dispatcher asks whether anyone at the scene appears to be mentally unwell during the initial police call [11]. According to our narratives, in at least 1 in 4 of our study’s police calls, the dispatcher was aware that the victim was mentally unwell. When available, knowledge of a mental health problem prior to the incident might allow LE agencies to dispatch personnel who have undergone crisis intervention training (CIT)—a type of mental health training designed for police response that has been introduced in over 2700 communities nationwide [12]. The extent to which knowledge of a mental health problem prior to the incident, or deployment of CIT-trained LE, influences the likelihood of a violent or fatal police encounter is unknown. Although research highlights the positive impact of mental health training on self-reported police outcomes—for example, CIT-trained personnel perceive themselves as less likely to use excessive force in hypothetical mental health encounters [13]—further research is required to assess the effectiveness of police-led mental health interventions like mental health-focused triage at dispatch and police response thereafter.
Many theories have been posed to explain the high volume of police homicides of mentally ill victims in the USA. These range from implicit biases in policing against racial minorities (despite making up only 13% of the population, the rate of Black people being by the police is almost double that of white people [3] and/or those with precarious shelter—a significant proportion of whom harbor more severe comorbid mental health disorders that are further compounded by the effects of community policing [14]—to the widespread availability of firearms in the USA, which render the prospect of successful de-escalation more difficult to imagine than encounters without firearms.
One phenomenon that seeks to explain the volume of suicidal persons being fatally shot by the police is “suicide by cop,” which appeared in 3 (4%) of our mental health-related police calls. This term describes an incident where a suicidal individual—typically an intoxicated younger adult white male with a criminal and mental health background [15]—appears to coerce law enforcement to respond with lethal force to end their own life, a manifestation of their suicidal intent. Although excluded from the suicide cases for the purposes of our study, “suicide by cop” may also describe an individual—with otherwise no background of suicidal ideation and no prior intention to end their life—who is apprehended by the police after committing a crime and expresses the preferring of dying at the hands of police to the prospect of being apprehended by the police, charged for the crime and detained.
In our paper, most police calls involving suicidal persons described individuals with a history of mental illness who verbally reported suicidal ideation to an individual in their personal network (e.g., partner, parent). Over half (51%) of our study’s mental health calls were reported to the police by the victim’s partner, family member, healthcare professional, or through an emergency court order due to concerns of the victim’s deteriorating mental health. Thus, most victims did not anticipate their prospective fatal police encounter. In addition, given the well-documented history of deadly forced used by US police against communities with severe mental illness [2, 16–18] alternate theories that explain fatal police encounters involving suicidal persons should be investigated. Studying fatal police encounters involving suicidal persons outside the USA, particularly among police forces who do not traditionally harbor firearms (e.g., police in the UK), may advance understanding of the complexities of this type of fatal police event involving a suicidal person.
In order to reduce fatal police encounters among victims with mental health crises, there has been a growing emphasis on the funding of alternate programs to 911 that deploy mental health practitioners to eligible 911 calls involving mental health emergencies, instead of the police. Examples of such programs include the CARE project in Chicago [18], B-HEARD in New York City [19] and CAHOOTS in Oregon [20]. The introduction of these initiatives also attempts to tackle the role of implicit biases when engaging with individuals with mental health illness, such as the ill-evidenced notion that links mental health illness with a propensity for physical violence. Mental health illness does not predict violent behavior [21] and such stereotypes have the potential to heighten the likelihood of police escalation, including use of force. Despite initial success, there does however remain a strong concern that alternate programs that coordinate with police departments are still victim to social ills, such as pervasive structural racism, that will undoubtedly impact the quality of mental healthcare received by individuals. Keeping community mental health programs separate from police involvement may serve as a method to reduce encounters between the police and highly policed communities, and therefore arrests and subsequent incarceration, from historically excluded populations whose mental health illnesses (e.g., substance abuse) have been historically criminalized [3]. Nonetheless, given increased interactions between the police and those with mental health illness in the community—exacerbated by newfound health policies, such as New York City’s Mental Health Plan, which will use police to involuntarily admit those with severe mental health illness from the city’s streets [22]—there is a dire need to explore further community-based solutions that reduce these fatal police interactions.
Our paper has several limitations. The NDVRS data for 2014 and 2015 does not include all states; hence, the data may not be generalizable to the whole country. Given that more states have now enrolled into NVDRS, it would be important to re-examine the data when data from the remaining states becomes publicly accessible. The narratives used in this study were also of variable quality, particularly in terms of length, detail, and information regarding a timeline of events. Our sample also exclusively focuses on police homicides and therefore does not address the vastly more common incident of nonfatal police violence and policing-related morbidity in the USA. Our focus on police homicides—without a comparator group of non-fatal police encounters involving individuals experiencing a mental health crisis—renders it difficult to conclude whether any victim or incident characteristics that we have identified are risk factors for a fatal police encounter. In addition, NVDRS does not provide much information about the law enforcement officers who killed the victims. Improving access to this type of data is critical to better understanding—and thus preventing—patterns in LIHs among those with mental health illness given the current burden of police violence in the USA and the extent to which this impacts public mental health.
Conclusion
Approximately one-third of fatal police encounters in our study involved victims with a mental health crisis. Among the many ways that an MHC can manifest during fatal police encounters, the modal presentation involved a suicidal adult with a known mental health history who verbally expressed suicidal intent to a family member or intimate partner, who, in turn, called 911 out of concern. Given the extent to which individuals with an MHC die by police homicide, there is a significant need to evaluate alternative or complementary programs to 911 as a potential solution to manage the multitude of ways mental health disturbances can present in police calls.
Appendix
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Fatal police shootings of mentally ill people more likely in small and midsized areas - the Washington Post. https://www.washingtonpost.com/national/police-mentally-ill-deaths/2020/10/17/8dd5bcf6-0245-11eb-b7ed-141dd88560ea_story.html Accessed 24 May 2023.
- 2.Sharara F, Wool EE, Bertolacci GJ, et al. Fatal police violence by race and state in the USA, 1980–2019: a network meta-regression. The Lancet. 2021;398(10307):1239–1255. doi: 10.1016/S0140-6736(21)01609-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Stewart HLN, Jackman M, Agarwal S, et al. Decriminalising being Black with mental illness. Lancet Psychiatry. 2021;8(1):8–9. http://www.thelancet.com/article/S2215036620305198/fulltext Accessed 14 Nov 2022. [DOI] [PubMed]
- 4.Alang S, McAlpine D, McCreedy E, Hardeman R. Police brutality and black health: setting the agenda for public health scholars. Am J Public Health. 2017;107(5):662–665. doi: 10.2105/AJPH.2017.303691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Edwards F, Esposito MH, Lee H. Risk of police-involved death by race/ethnicity and place, United States, 2012–2018. Public Health. 2018;108:1241–1248. doi: 10.2105/AJPH.2018.304559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mapping Police Violence. Mapping police violence. Published October 24, 2022. https://mappingpoliceviolence.us/ Accessed 23 Oct 2022.
- 7.Centers for Disease Control and Prevention. National Violent Death Reporting System - Web Coding Manual Version 5.4.1. National Center for Injury Prevention and Control; 2022 https://www.cdc.gov/violenceprevention/pdf/nvdrs/nvdrsCodingManual.pdf. Accssed 13 Feb 2024
- 8.Wertz J, Azrael D, Berrigan J, et al. A typology of civilians shot and killed by us police: a latent class analysis of firearm legal intervention homicide in the 2014–2015 National Violent Death Reporting System. J Urban Health. 2020;97(3):317. doi: 10.1007/S11524-020-00430-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fuller D, Lamb R, Biasotti M, Snook J. Overlooked in the undercounted: the role of mental illness in fatal law enforcement encounters. 2015. https://www.treatmentadvocacycenter.org/storage/documents/overlooked-in-the-undercounted.pdf Accessed 29 Oct 2022.
- 10.Chidgey K, Procter N, Baker A, Grech C. Police response to individuals displaying suicidal or self-harming behaviours: an integrative review. Health Soc Care Community. 2019;27(3):e112–e124. doi: 10.1111/HSC.12668. [DOI] [PubMed] [Google Scholar]
- 11.Baker CD, Polito KE, Bennett Daniel. Domestic Law Enforcement Guidelines; 2017. https://www.mass.gov/doc/2017-domestic-violence-law-enforcement-guidelines-1. Accessed 13 Feb 2024
- 12.Watson AC, Compton MT, Draine JN. The crisis intervention team (CIT) model: an evidence-based policing practice? Behav Sci Law. 2017;35(5–6):431–441. doi: 10.1002/BSL.2304. [DOI] [PubMed] [Google Scholar]
- 13.Rogers MS, McNiel DE, Binder RL. Effectiveness of police crisis intervention training programs. J Am Acad Psychiatry Law. 2019;47(4):414–421 . doi: 10.29158/JAAPL.003863-19. [DOI] [PubMed] [Google Scholar]
- 14.Devylder JE, Anglin DM, Bowleg L, Fedina L, Link BG. Police violence and public health. Annu Rev Clin Psychol. 2022;18:527–552. doi: 10.1146/ANNUREV-CLINPSY-072720-020644. [DOI] [PubMed] [Google Scholar]
- 15.Patton CL, Fremouw WJ. Examining “suicide by cop”: a critical review of the literature. Aggress Violent Behav. 2016;27:107–120. doi: 10.1016/J.AVB.2016.03.003. [DOI] [Google Scholar]
- 16.2022 Police Violence Report. https://policeviolencereport.org/ Accessed 2 Feb 2023.
- 17.Treatment Advocacy Center Reports - Treatment Advocacy Center. https://www.treatmentadvocacycenter.org/evidence-and-research/studies Accessed 30 Jan 2023
- 18.City of Chicago Mayor’s Office of Violence Reduction. Crisis Assistance Response and Engagement (CARE) Dashboard. https://www.chicago.gov/city/en/sites/public-safety-and-violence-reduction/home/CARE-Dashboard.html Accessed 14 Nov 2022.
- 19.City of New York Mayor’s Office of Community Mental Health. Re-imagining New York City’s mental health emergency response - Mayor’s Office of Community Mental Health. https://mentalhealth.cityofnewyork.us/b-heardAccessed 14 Nov 2022.
- 20.Eugene Police Department. CAHOOTS | Eugene, OR Website. https://www.eugene-or.gov/4508/CAHOOTS Accessed 14 Nov 2022
- 21.Varshney M, Mahapatra A, Krishnan V, Gupta R, Deb KS. Violence and mental illness: what is the true story? J Epidemiol Community Health. 2016;70(3):223–225. doi: 10.1136/JECH-2015-205546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mayor Adams unveils ambitious mental health agenda focused on improving family and child mental heal | City of New York. https://www.nyc.gov/office-of-the-mayor/news/140-23/mayor-adams-ambitious-mental-health-agenda-focused-improving-family-child-mental#/0 Accessed 24 May 2023.