International Journal of Law and Psychiatry xxx (2014) xxx–xxx

IJLP-01019; No of Pages 7

Contents lists available at ScienceDirect

International Journal of Law and Psychiatry

The frequency and nature of resolution of potential police provokedshooting encounters

Louise C. McLeod, Stuart D.M. Thomas ⁎, Dragana KesicCentre for Forensic Behavioural Science, School of Psychology & Psychiatry, Monash University, AustraliaVictorian Institute of Forensic Mental Health, Australia

⁎ Corresponding author at: Centre for Forensic BehavioClifton Hill, Melbourne, Victoria 3068, Australia. Tel.: +612650.

E-mail address: stuart.thomas@monash.edu (S.D.M. Th

http://dx.doi.org/10.1016/j.ijlp.2014.02.0090160-2527/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article as: McLeod, L.C., et al.,tional Journal of Law and Psychiatry (2014), h

a b s t r a c t

a r t i c l e i n f o

Available online xxxx

Keywords:Psychiatric crisisProvoked shootingPolice

Helping people in acute psychiatric crisis has become an increasingly common part of modern community polic-ing. In certain extreme cases, policemay be facedwith a suicidal individual who intends to intentionally provokepolice to shoot them. While fatalities are fortunately rare, anecdotal reports from frontline police suggest thatthese kinds of encounters are occurring on a regular basis. This paper explores 2350 psychiatric crisis incidentsover an eight-month period in Victoria, Australia, and assesses the frequency and nature of potential police-provoked shootings resolved through non-fatal means. Contextual factors relating to the person's behaviourand police responses, and the person's psychiatric and criminal histories were considered to elucidate character-istics common to these incidents. Results suggest that police are potentially encountering a personwho is suicidaland trying to provoke police to shoot them more than twice a week. These individuals share a number of com-mon characteristics with those who have been fatally shot in similar circumstances and are quite differentfrom those who attempt self-inflicted suicide. Results are discussed in relation to the impact of previous criminalcontact from both the suspect and police perspectives.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Some twenty years ago Ruiz (1993) commented that “the job of apolice officer is said to be the one of tedium punctuated by momentsof terror” (p. 157). While many police will go through their entirecareers without drawing their gun, they will commonly encounterpeople who are resisting arrest or who are in crisis situations, therebysometimes necessitating the use of physical force (Bayley & Bitner,1989). While force is most commonly used in relation to resolvingencounters with suspects and offenders (Hickman, Piquero, & Garner,2008), police continue to raise particular concerns that they areresorting to the use of excessive force when attempting to resolveencounters with people experiencing mental illness (Godfredson,Thomas, Ogloff, & Luebbers, 2011). It has been argued that the tradition-al authoritarian ‘must resolve quickly’ approach style of police(Brouwer, 2009), coupled with the complexities of crisis encounterscommonly involving people experiencing mental illness (Clifford,2010), can exacerbate the dynamics of the encounter. This may bedue to perceptions of the policing intruding as opposed to intervening

ural Science, 505 Hoddle Street,3 9947 2600; fax: +61 3 9947

omas).

The frequency and nature ofttp://dx.doi.org/10.1016/j.ijlp

or helping (Coffey, 1990), and can lead to the need for increased forceresponses from police. On some occasions this will inevitably lead tothe need for fatal force (Kesic, Thomas, & Ogloff, 2010); the impact ofsuch on police officers should not be understated (Carroll, 2005).

2. Police provoked shootings

2.1. Definitions

While the broad phenomenon of victim-precipitated homicidehas been recognised for quite some time (Wolfgang, 1959), the re-search literature really only began specifically focusing on victim-precipitated homicide by police in the last 20 years, with (Geberth,1993) analysing “incidents in which individuals bent on self-destruction, engage in life-threatening and criminal behaviour inorder to force the police to kill them”. The subsequent research intothese encounters variously labelled them as suicide-by-cop (Homant &Kennedy, 2000a; Kennedy, Homant, & Hupp, 1998; Lindsay & Lester,2008;Mohandie,Meloy, & Collins, 2009), law-enforcement-assisted suicide(Lord, 2000), law-enforcement-officer-forced-assisted suicide (Hutson et al.,1998), police-involved–victim-provoked shooting (McKenzie, 2006),victim-precipitated homicide (Parent & Verdun-Jones, 1998) and suicideby police (Homant & Kennedy, 2000b; Kesic, Thomas, & Ogloff, 2012).For consistency, and in line with operational definitions in the state of

resolution of potential police provoked shooting encounters, Interna-.2014.02.009

2 L.C. McLeod et al. / International Journal of Law and Psychiatry xxx (2014) xxx–xxx

Victoria, Australia, where this research was conducted, the current paperadopts the term police-provoked shooting.

As our collective understanding of the estimated prevalence of police-provoked shootings, and commonalities found between different inci-dents has developed, attention has shifted to refining the criteria bywhich cases are identified. Lord (2000) used specific elements to identifycases of police provoked shootings: “those individuals who, whenconfronted by enforcement officers, either verbalised their desire to bekilled by law enforcement officers and/ormade gestures such as pointingweapons at officers or hostages, running at officers with weapons, orthrowing weapons at officers” (p. 403). Comparatively, Hutson et al.(1998) specified a number of inter-related elements, specifically includ-ing the ability to attribute suicidal intent to the individual involved;“[when] a suicidal individual intentionally engages in life-threateningand criminal behaviour with a lethal weapon or what appears to be alethalweapon towards lawenforcement officers or civilians to specificallyprovoke officers to shoot the suicidal individual in self-defence or toprotect civilians” (p. 665). More recently, the literature has emphasisedan assessment of police officers' actions in the context of the entire behav-ioural chain of events in order tomost fully consider a person'smotivationand intent (McKenzie, 2006; Pinizzotto, Davis, & Miller, 2005).

2.2. Further delineating types of police-provoked shooting incident

Prominent authors in the area have posited that it is possible to un-derstand the nature of provoked shooting incidents by considering theperson's motivation, either instrumental or expressive (Lord & Sloop,2010; Mohandie & Meloy, 2010; Van Zandt, 1993). An instrumental mo-tivation describes instanceswhere the person purposefully engages in vi-olent behaviour or other premeditated criminal acts and seeks tointentionally use police as a means/method of suicide due to them notbeing able to follow through with self-inflicted suicide for fear of socialor religious reprimand. Alternately, people can have expressive motiva-tions that are more spontaneous and impulsive in nature; these are acti-vated when the person feels frustrated or threatened and are used toexpress hopelessness or anger (Lord & Sloop, 2010; Mohandie & Meloy,2010; Van Zandt, 1993). Lord and Sloop (2010) note that these motiva-tions can co-occur in certain circumstances/situations.

Homant and Kennedy (200b) have suggested that provoked shoot-ing incidents can be distinguished by a more nuanced consideration ofthe chain of events that lead to the police encounter and what unfolds;they propose three types of situation: (1) incidents of ‘direct confronta-tion’ involving the person choreographing a sequence of events, initiat-ing an incident to attract police attention and on arrival then escalatingthe encounter to provoke them to shoot; (2) ‘disturbed interventions’whereby an individual's irrational behaviours attract police and, whilesuicide may not have been planned, intent develops and escalates dur-ing the encounter; and (3) ‘criminal intervention’, where the individualis apprehended in relation to commission of a criminal act and decidesthey would rather die than face the criminal justice consequences.

2.3. Common characteristics of people involved in police provokedshooting incidents

Despite the available literature being variable in quality, methodologyand case ascertainment of provoked shooting incidents, post-incidentcase analyses have highlighted a number of potential commonalities inthe individual characteristics, short-term precipitants and behaviouralelements associated with provoked shootings. Characteristically, theevidence suggests that individuals involved are most commonly middleaged and male, with a notable psychiatric history including a record ofsubstance abuse, along with a history of contact with police as suspectsor offenders (Lindsay & Lester, 2004; Mohandie & Meloy, 2000; Wilson,Davis, Bloom, Batten, & Kamara, 1998). Common short-term catalystswhich may contribute as antecedents to the incident include the personbeing involved in a domestic dispute or being alcohol intoxicated (Best,

Please cite this article as: McLeod, L.C., et al., The frequency and nature oftional Journal of Law and Psychiatry (2014), http://dx.doi.org/10.1016/j.ijlp

Quigley, & Bailey, 2004; Hutson et al., 1998; Mohandie & Meloy, 2010;Wilson et al., 1998). Common behavioural components encompassthose inherent in the classification of a provoked shooting incident: thepossession of a potentially deadly weapon, asking police to shoot them,non-compliance with police commands, and advancing on police(Hutson et al., 1998; Kesic et al., 2012; Lindsay & Lester, 2004).

2.4. Using police as a means of suicide versus self-inflicted suicide

Klinger (2001) suggested that the extant literature lacked theory asto elucidating why people may choose to select the police as a means/method of suicide as opposed to other self-inflicted means. However,Mohandie and Meloy (2000) proposed motivations behind these inci-dents including avoiding the consequences of criminal conduct, goingdown in a blaze of glory, and having a volatile history with police.These findings are interesting in light of recent suggestions that thereappear to be a number of commonalities in the characteristics of thosewho attempt police provoked suicides and those who use self-inflictedmethods. Risk factors common to both means/methods of suicide in-cludemale gender, middle age, temperamental dysfunction, psychiatricand psychological disorders, experience of trauma and substance abuse(Lord & Sloop, 2010). The situational characteristics between the twomeans of suicide are really quite different: the very behaviours thatdefine police provoked shootings (e.g., not responding to police com-mands, advancing on policewith a deadlyweapon, threatening the pub-lic with harm) are not apparent in self-inflicted incidents. Of note, theprovoked shooting group has a more entrenched history of suicideattempts, suggesting more serious psychopathology with this group.

2.5. Potential provoked shootings that are resolved through nonfatal means

A slowly accumulating body of research, mainly from the UnitedStates and the United Kingdom, has considered the prevalence of pro-voked shooting incidents among those who are fatally shot by police,reporting that between 10% and 50% of the fatal incidents meet variouslydefined criteria (Hutson et al., 1998; Kennedy et al., 1998; Kesic et al.,2012; Parent & Verdun-Jones, 1998;Wilson et al., 1998). However, anec-dotally speaking, police report that they encounter people who attemptto provoke the police into shooting them on a regular basis; nevertheless,the police are able to resolve the encounters throughnonfatalmeans. Thisperception, while not well quantified or supported empirically, has somelimited support. Indeed,McKenzie (2006) emphasised that not all peoplewho engage in police provocation to shoot them will be shot and/orkilled, and as such there is a need to focus on these successfully resolvedcases to isolate factors which may impact on the outcome. Lord (2000)suggested that an important line of enquiry would be the comparisonof historical, personality, behavioural and situational factors in incidentswhich are successfully resolved (e.g., de-escalated or abated with lessthan lethal force) compared to incidents involving the use of fatal force.

While fewof these factors to date have statistically discriminated be-tween a fatal and a non-fatal outcome, one of the key factors appears tobe that the possession of a firearm by the individual significantly in-creases the chances of fatal outcome. This supports theories that policeuse of force increases with the officers' perception of threat and dangerto themselves, determined in great part by the limited other force-basedoptions available at a distance (Best&Quigley, 2003; Pinizzotto, Davis, &Miller, 2007).

Mohandie et al. (2009) suggest that suicidal intent can often occurspontaneously during a police encounter, a finding common to pro-voked shooting incidents which have commenced in the context of apsychiatric crisis or criminal act and then escalated through somemeans into a provoked shooting (Homant & Kennedy, 2000b). Such aposition supports the interactional theory of the use of force thatspecifies that it is the cumulative and complex relationship betweenall of the elements of the use of force incident (e.g., characteristics ofparticipants, environmental circumstances, suspect behaviour,

resolution of potential police provoked shooting encounters, Interna-.2014.02.009

3L.C. McLeod et al. / International Journal of Law and Psychiatry xxx (2014) xxx–xxx

individual and officer perceptions and assumptions) which may esca-late into the need for police to use fatal force (Parent, 1996). This isakin to what Pinizzotto et al. (2007) referred to as the “deadly mix” ofindividual, situational and police response variables being present at apoint in time which then leads to escalation and a fatality. While insome situations an absolute resolve to die on the part of the suspectmay negate any other elements in the encounter, including the possibil-ity of negotiation (Mohandie & Meloy, 2010), other situations includeperiods where suicidal intent can fluctuate or the point of crisis startsto subside; these can potentially be a window of opportunity for inter-vention (Deisenhammer et al., 2009).

Given the limited literature in this critically important area for bothmental health and community policing, there remains a distinct need tounderstand the complex interplay between the behaviours of people incrisis and optimal police responses that leads both to nonfatal resolutionand to providing access to health, social and welfare providers.

2.6. Aims

Against this background, the aims of the study were: (1) to gain anestimated prevalence of potential provoked shootings resolved throughnon-fatal means in a sample of psychiatric crisis cases; (2) to character-ise those incidents which meet the criteria for being a potential pro-voked shooting; and (3) to compare the details of the potentialprovoked shooting incidents and individuals involved in those whocomplete police provoked shooting and those who are intent on self-inflicted suicide.

3. Method

3.1. Databases

The current study involved data linkage between three databases:The Victoria Police mental health transfers database, the Victoria PoliceLaw Enforcement Assistance Programme (LEAP), and the Redevelop-ment of Acute and Psychiatric Information Directions (RAPID) database.

The source database was the Victoria Police mental health transfersdatabase. This (L42) database is an electronic record of police incidentsfor which reporting procedures require the completion of a MentalHealth Transfer Form (MHTF). This form captures both categorical infor-mation and narrative information about situations in which police havecontact with a person that they suspect to be mentally disordered whorequires transfer to another agency or service (e.g., ambulance, hospitalor mental health facility) (Short, MacDonald, Luebbers, Ogloff, &Thomas, 2012).

The Victoria Police Law Enforcement Assistance Programme (LEAP)database, in operation since March 1993, records all contacts betweenpolice and members of the public as witnesses of crime, victims ofcrime, and as alleged or confirmed offenders.

The Redevelopment of Acute and Psychiatric Information Directions(RAPID) database, formerly known as the Victorian Psychiatric CaseRegister (VPCR) is managed by the Department of Health and recordsall contacts with the public mental health system (e.g., hospitals, com-munity services and in-patient services) across the state of Victoria.This database does not capture mental health related contacts with pri-vate hospitals or general practitioners; as such, it is likely to provide anunderestimate of the more common mental disorders (including anxi-ety disorders, mood disorders and substance use disorders) which aremost commonly determined and treated outside of the public mentalhealth system (Short, Thomas, Luebbers, Ogloff, & Mullen, 2010).

3.2. Procedure

Categorical information and narrative information about police en-counters were obtained from the L42 database. The initial sample com-prised 2611 Victorian incidents recorded over an eight month period

Please cite this article as: McLeod, L.C., et al., The frequency and nature oftional Journal of Law and Psychiatry (2014), http://dx.doi.org/10.1016/j.ijlp

between 1/12/2009 and 31/07/2010. The current study limited analysesto psychiatric crisis incidents only; as such the other mental healthtransfer contacts (including welfare checks and return of various per-sons to different institutions, e.g., return of involuntary or security psy-chiatric patient) were excluded, leaving a final sample of 2350incidents. Proportionally more of the incidents involved males (n =1336, 56.9% vs. females n = 1001, 42.6%), and the average age of thesample was 35.12 years (SD = 12.98, range = 9–89 years).

Individual identifiers (i.e., full name, gender and date of birth) wereextracted from the L42 database. This was used to create amaster list byassigning each person a unique study code. This master list was thenlinked with the LEAP database to establish the frequency and natureof the individual's prior contacts with police in different contexts; andseparately with the VPCR to ascertain who in the sample had a historyof contacts with the public mental health system, what the nature ofthese contacts were and any diagnoses received. L42 incident leveldata were also extracted in order to obtain a narrative account of the in-cident and to establish a chain of events from the attending police offi-cers' perspective. Data linkage procedures followed contemporaryguidelines regarding epidemiological research using data banks(NH&MRC, 2007).

Criminal history variables extracted from LEAP included: if the indi-vidual had a previous charge, (i) the number of criminal charges; (ii) theage of first charge; (iii) history of violent offending, defined as chargedor convicted with any violent crime; and (iv) history on non-violentoffending, defined as having been charged with, or convicted of, anynon-violent crime. In addition, any instances of criminal victimisationwere also considered. Mental health-related variables extracted fromthe VPCR were age of first contact with public mental health servicesand specific Axis I diagnoses recorded.

4. Materials

The research team extracted information from the L42 databasepertaining to: (a) criteria for the classification of potential provokedshootings, adapted from the Hutson et al. (1998) criteria (detailsbelow); (b) the typology proposed by Homant and Kennedy (2000b)of planned, disturbed, or criminal intervention types of incident; (c) po-lice contact level variables (i.e., subject, family/partner/ex-partner,friend, caseworker/psychologist/psychiatrist, hospital/ambulance/nurse/doctor, lifeline/suicide line, not available); (d) other contextualvariables (i.e., intoxication, domestic dispute); and (e) outcomevariables (i.e., police presence, negotiated, arrested, use of force, notavailable).

4.1. Procedure

The narrative recorded for eachMHTFwas read and coded accordingto elements defined in the coding sheet developed specifically for use inthis study. The length of the narratives available ranged from a shortsentence to a 300-word paragraph; some provided a depth of informa-tion, including all elements required for the current analysis, whileothers were more limited in scope and detail. Criteria employed byKesic et al. (2012), an adaptation of the original Hutson et al.'s (1998)criteria in the assessment of suicide-by-police cases, were used as thebasis from which to screen for incidents of potential provoked shoot-ings. The criteria considered the following: (i) communication of suicid-al intent; (ii) gestures of suicidal intent; (iii) evidence that the personstated that theywanted police to shoot them; (iv) possession of a dead-ly (orwhat looked like a deadly)weapon; and (v) evidence that the per-son deliberately escalated the encounter in order to provoke police intoshooting them.

While these criteria provided a foundation for narrative characteri-zation, the particulars of each criterion component were adjusted toallow for amore inclusive identification of potential provoked shootingsbased on all available information (McKenzie, 2006). The coding of the

resolution of potential police provoked shooting encounters, Interna-.2014.02.009

4 L.C. McLeod et al. / International Journal of Law and Psychiatry xxx (2014) xxx–xxx

data with reference to these criteria was therefore a dynamic process,with the initial revised Hutson criteria being more fully operationalisedto reflect the nature and broader scope of the current study and sourcepopulation under consideration. Specifically, the suicidal gesture criteri-on was operationalised to incorporate the presence of a suicidal gestureat any point in the narrative and included arming oneself with a weap-on, preparingmeans by which the person could commit suicide or pur-posefully placing themselves in a potentially life threatening situation.Additionally, criterion five was adapted to exclude the component ofpolice shooting, with the rest of it (i.e. evidence that the individualwas resisting the police and escalating the encounter through violentand threatening behaviour) remaining the same.

According to Lord (2000), it is the death of the individual whichgives greatest weight to the inferences of suicidal intent being madefrom the specific behaviours detailed in the classification criteria. In po-tential cases, the individual's intent to suicide is commonly less defini-tive, as intent will inevitably fluctuate and perhaps dissipate, allowingfor de-escalation and/or resolution by nonfatal means. In addition tothese complexities, narrative information available in the currentstudywas limited in nature. Only incident level informationwas consid-ered for the purposes of identifying potential provoked shooting cases,but decisionswere ultimately determined on the basis of all of the infor-mation available, on the balance of probabilities, based on consensusviews of the authors.

For the self-inflicted suicide comparison group, cases were excludedif any of the following were present: (i) the available narrative did notidentify both intent and gesture at some point during the incident; (ii)there was no continued gesture or verbalisation upon police arrival;(iii) intent or gesture was communicated or recorded as an act of self-harm or desire to self-harm, rather than specific intent to kill oneself;or (iv) police arrived during an active or after a failed suicide attemptand therewas no further gesture or verbalisation relating continued sui-cidal intent.

4.2. Data analysis

Simple descriptive statistics were used to characterise the sample.Within sample comparisons were considered, comparing potential pro-voked shooting cases to individuals with intent to self-inflict suicide,and the remainder of the psychiatric crisis sample. Data from a recentstudy of suicide by police fatalities (n = 15) was sourced (Kesic et al.,2012); this study had used the same LEAP and VPCR databases toenable valid comparisons on key variables. Continuous variables werecompared using independent t-tests, while categorical variables werecross-tabulated and Chi Squared tests of association calculated. Datawere analysed using SPSS (version 19.0) and Stata (version 10.0).

5. Results

5.1. Sample characterisation

Ninety two cases met the full or partial criteria for being classified asa potential police provoked shooting (PPPS) incident. The majority ofindividuals involved in the PPPS cases were male (n = 67, 72.83%)and their average age at the time of the incident was 35.09 years old(SD = 12.39, range = 14–63 years old). Fifty-seven (62%) subjectshad an Axis I diagnosis recorded. Of these 20 (35.1%) had a depressivedisorder, 13 (22.8%) had schizophrenia, 6 (10.5%) had a substance usedisorder, 5 (8.8%) had bipolar disorder, 5 (8.8%) had an anxiety disorder,and 8 (14%) had other diagnoses. Roughly two thirds (59, 64.1%) hadprevious criminal histories recorded and over half (n = 52, 56.5%) ofthe PPS cases had been a victim of crime.

Of the 92 cases, a fifth (n = 19, 20.7%) met all five of the revisedHutson criteria. The remaining 73 cases met a minimum of three ofthe five criteria; and in each of these cases there was sufficient confi-dence in its inclusion in the study groupwhen all of the details included

Please cite this article as: McLeod, L.C., et al., The frequency and nature oftional Journal of Law and Psychiatry (2014), http://dx.doi.org/10.1016/j.ijlp

in the chain of eventswere considered. At its broadest interpretation, 77(83.7%) of the incidents included detail that the individual specificallysaid that they wanted to be shot by police. Eighty-nine (96.7%) narra-tives identified a verbalisation of suicidal intent; just under half (n =45, 48.9%) included a suicidal gesture; a smaller proportion (n = 37,40.2%) detailed that the person had a deadly weapon; and in over half(n=49, 53%) of the cases the individual intentionally escalated the sit-uation in such a way as to provoke police.

In the 15 cases that did not include specific details that the personstated that they wanted to be shot by police, there was always anindication of suicidal intent (either by communication or gesture), thepresence of a deadly weapon, and/or intentional escalation. For exam-ple, 12 of these 15 cases included details about suicidal intent (commu-nication or gesture), and they had a weapon and escalated theencounter; in two of the cases there was communication and a suicidalgesture and escalation (one was a planned attack and in the other theperson threatened to shoot civilians). In the one remaining case includ-ed as a PPPS the person had suicidal intent (communication andgesture) and a deadly weapon, and there was evidence that he hadplanned the incident, however, there was no information about escala-tion or asking to be shot by police.

The followingnarrative exemplifies someof the complexities of suchincidents: “[The] individual contacted police stating he had a firearm aswas going to shoot himself. He showed police to the rear yard (nilfirearm) where he became aggressive and charged at police. [OleoresinCapsicum] foam deployed to overpower the individual. Individual wasconveyed to hospital for [mental health] assessment. Individual contin-ually made reference to committing suicide”.

Sixty-five (70.7%) incidents occurred in the context of a ‘disturbedintervention’. Twenty-two cases (23.9%) were ‘planned confrontations’and 5 (5.4%) incidents occurred during ‘criminal intervention’. In 17narratives (18.5%) it was noted that the individual was intoxicated atthe time of the incident, while a larger proportion (n = 28, 30.4%) oc-curred in the context of a domestic dispute. For the majority of cases itwas not clear who had made initial contact with police (n = 58, 63%).

For 34 cases with this information recorded, the individual in crisismost commonly initiated contact (n = 13, 38%); less commonly familymembers (n = 7, 21%), the person's case worker, psychologist or psy-chiatrist (n=3, 9%), or a mixture of ambulance or hospital staff, a com-munity bystander, a suicide prevention line or a friend (n = 7, 21%)made initial contact with police. Most L42 narratives did not providedetails on how the incident was resolved (n = 49, 53%). Of those withan outcome description available, 24 involved the use of force, 6 wereformally negotiated, 6 dissipated in police presence and 5 resulted inarrest. Only 1 case indicated that specialised crisis negotiators (CrisisIncident Response Team; CIRT) were used.

5.2. Comparison of potential police provoked shootings (PPPSs) and generalpsychiatric crisis population

Those in the PPPS sample were significantly more likely to be male(χ2 = 9.961, p = 0.002) than the broader psychiatric crisis group.Theywere also significantlymore likely to have a formal history of crim-inal charges (χ2 = 7.194, p = 0.007) and a higher number of previouscharges (t = 2.532, p = 0.011) than the broader psychiatric crisisgroup. This was the case for both violent offending (χ2 = 8.832, p =0.003) and non-violent offending (χ2 = 8.825, p = 0.003) (Table 1).

5.3. Comparison of PPPS cases and self-inflicted suicide cases

Compared to the actively suicidal group, the PPPS group wassignificantly more likely to be male (χ2 = 8.012, p = 0.005). Theywere also more likely to have an Axis 1 disorder recorded on thepsychiatric case register (χ2 = 6.843, p = 0.009) and were signifi-cantly older at their age of first contact with public mental healthservices (t = 2.210, p = 0.03). Those in the PPPS sample were more

resolution of potential police provoked shooting encounters, Interna-.2014.02.009

Table 1Comparisons between the PPPS sample, general psychiatric crisis cases, fatality sample and attempted self-inflicted suicide sample.

PPPS (N = 92) Psychiatric crisis (N = 2258) Fatality sample (N = 15) Attempted self-inflicted suicide (N = 362)

Male gender 67 (72.8%) 1269 (56.5%)⁎⁎ 14 (93.3%) 205 (56.3%)⁎⁎

Age at incident 34.63 (12.51) 34.69 (13.06) 33.94 (SD 9.43) 32.70 (12.20)Age at first mental health contactMean (SD)

29.56 (12.11) 32.35 (27.40) – 28.71 (SD 12.35)⁎

Presence of Axis 1 disorder 57 (62.0%) 1212(53.8%) 8 (53.3%) 169 (46.4%)⁎⁎

Had previous criminal charges 59 (64.1%) 1126 (50%)⁎⁎ 12 (80%) 171 (47%)⁎⁎

Number of charges 42.19 (50.47) 29.41 (47.34)⁎ – 33.04 (SD 51.34)Age at first chargeMean (SD)

30.29 (12.30) 30.81 (29.44) – 26.47 (SD 9.96)⁎⁎

History of violent offending 27 (29.3%) 390 (17.3%)⁎⁎ 11 (73.3%)⁎⁎ 58 (15.9%)⁎⁎

History of non-violent offending 40 (43.5%) 656 (29.1%)⁎⁎ – 97 (26.6%)⁎⁎

⁎ p b 0.05.⁎⁎ p b 0.01.

5L.C. McLeod et al. / International Journal of Law and Psychiatry xxx (2014) xxx–xxx

likely to have a formal history of criminal charges (χ2 = 8.375, p =0.004), and to have been older at the time of their first criminal charge(t = 3.124, p = 0.002). Similar to the previous comparisons, those inthe PPPS sample were significantly more likely to have a history of vio-lent (χ2 = 8.560, p= 0.003) and non-violent (χ2 = 9.690, p= 0.002)offending (Table 1).

5.4. Comparison of PPPS cases and fatal police-provoked shooting incidents

When compared to a fatality sample classified as ‘suicide-by-police’incidents (Kesic et al., 2012), the PPPS sample had a higher proportionof females; they were slightly older at the time of the incident, and hada less entrenched criminal history. Significantly fewer of the PPPS samplehad a history of violent offending (χ2 = 10.90, p= 0.001) (Table 1).

5.5. Estimated number of PPPS incidents attended by police in a year

If the 92 cases are extrapolated to get a prevalence estimate, thesedatasuggest that therewould be 138 potential provoked shooting cases a year.This equates to one incident every 2.65 days, or between 2 and 3 of theseincidents per week that police are resolving through nonfatal means. Amore highly conservative estimate, based on the 19 cases which met allfive of the criteria, would suggest that there are 29 PPPS cases per year,equating to 2.42 incidents a month.

6. Discussion

This study sought to determine how frequently and in what contextpolice encounter situationswhere a personwho is suicidal intentionallyattempts to provoke the police into shooting them. The results suggestthat police are encountering these potential police-provoked shootingincidents between 2 and 3 times aweek, but are resolving them throughnonfatal means.

Consistent with previous research the majority of individuals in-volved in these encounters were male in their mid-thirties (Hutsonet al., 1998; Wilson et al., 1998). The majority of those in the PPPSgroup were known to the public mental health services in some capac-ity: they had their first mental health contact at a younger age thanbeing first charged with a criminal offence. Additionally, they hadtheir first mental health system contact on average five years prior tothe psychiatric crisis encounter captured here. Since that first policecontact, the PPPS group had been in frequent contact with police as sus-pects/offenders, albeit more commonly for nonviolent offences.

Despite all of the encounters considered here being labelled as ‘psy-chiatric crisis’ encounters by attending police, it was interesting to findthat less than two thirds of the PPPS cases had a formal Axis 1 diagnosisrecorded on the public mental health service case register. This may beindicative of the known underreporting of the high prevalence disor-ders according to publicmental health systemrecords and the increasedfocus of treatment and the coordination of care for people with these

Please cite this article as: McLeod, L.C., et al., The frequency and nature oftional Journal of Law and Psychiatry (2014), http://dx.doi.org/10.1016/j.ijlp

disorders in the primary care sector (Ansseau et al., 2004; Short et al.,2010; Staiger, Thomas, Ricciardelli, & McCabe, 2011). At the sametime, it may also point to either more inclusive definitions of what psy-chiatric crisis constitutes in the community and for frontline police, orperhaps a lack of more specialised training in signs and symptoms ofmental illness (Godfredson et al., 2011). Given the commonality of do-mestic crises and alcohol intoxication as precipitants in these incidents,it is likely that police are including people who, while falling short ofthresholds for formalised diagnoses, still present as distressed, confusedand disorientated.

6.1. What are the similarities between potential police-provoked shootingincidents and completed suicide-by-police incidents?

A comparison of cases identified as meeting criteria for being policeprovoked shootings in a study that used the same databases and meth-odology (Kesic, Thomas, & Ogloff, 2012) with the PPPS cases consideredhere suggests that the only statistically significant difference betweenthose who were fatally shot by police was that they had a moreentrenched history of violent offending than the PPPS cases. A potentialpartial explanation for this is that the fatality sample considered policeshootings over a protracted period of time and included a number ofshootings of high profile armed robberswith particularly established vi-olent criminal histories. Interestingly, and linked to this point, therewere proportionally more males in the completed suicide-by-policegroup, while over a quarter of those in the PPPS sample were female.These two issues bring into question the decision-making processes in-herent in the nature of these encounters and the impact of situationaland suspect-level factors (e.g., previous violence, sex, physical size/strength) on the risk assessments made by police.

From the suspect/offender perspective, Van Zandt (1993) suggeststhat the majority of crimes are committed by individuals who exist in asubculture of violence, forwhomaggressive orientedbehaviours are com-monly used as a method of problem-solving. Using this reasoning, indi-viduals with more entrenched violent offending histories may react in amore aggressive manner, especially towards police, exhibiting many ofthe risk factors indicative of a provoked shooting (e.g., noncompliant, es-calating behaviour, moving towards police in a threatening manner).While they may not have specific suicidal intent, they may be more am-bivalent about, or indifferent towards, life and death. In turn, they maybe less likely to respond favourably to police negotiation and want to ex-press their rage or punish society by dying defiantly at the hands of thepolice (Lindsay & Lester, 2004; Mohandie & Meloy, 2010; Van Zandt,1993). Furthermore, the use of less severe force options by police can infact be counterproductive and lead suspects/offenderswhowere perhapsambivalent about the encounter to escalate their behaviour and force amore severe or lethal police response (Scoville, 1998). However, to ex-plore this possibility these cases need to be compared in terms of policeresponses and management.

resolution of potential police provoked shooting encounters, Interna-.2014.02.009

6 L.C. McLeod et al. / International Journal of Law and Psychiatry xxx (2014) xxx–xxx

These issueswere perhaps the casewith some of the cases consideredfor inclusion in the PPPS grouphere.While theymay not have specificallycommunicated or gestured an intent to suicide, a broader analysis of theentire scenario, as recommended by McKenzie (2006) and used by Lord(2000), suggests that it was plausible that their courses of action couldnaturally lead to provoking the police to shoot at them. This kind of rea-sonable foresight standard, common to criminal law as a test of causationto assess whether the conduct/behaviour led to the result (Bronitt &McSherry, 2005, p. 165), could be applied in these situations. For exam-ple, that it would be ‘common sense’ to consider that advancing on policewith a potentially deadly weapon and not responding to police requests/demands to desist would lead them, potentially fearing for their safety, todrawing their firearm. While this approach may be helpful in identifyingpotential cases where the data available are limited, the question still re-mains as to whether the ‘common sense’ type of argument applies insuch highly charged situations.

6.2. Is there something different about those who choose the police as ameans of suicide?

Lord and Sloop (2010) recently reported a number of core similari-ties between those who attempt police provoked shooting and thosewho choose self-inflicted means. These data suggest some interestingdifferences, with lower proportions of females in the PPPS group.

Cases which met the criteria for being potential police-provokedshootings differed from other psychiatric crisis incidents by beingmore likely to be male and with respect to the nature and extent oftheir previous contact with the police. What really sets the PPPS casesapart from people involved in other crisis situations is that the PPPScases involved people who were significantly more likely to have a his-tory of both violent offending and non-violent offending, more previouscontacts with the police as suspects and/or offenders and higher preva-lence of Axis I disorders. They are therefore similar in characteristics tofatal PPS cases. While there are some similarities between the PPPScases and the self-inflicted suicidal group as well, the self-inflictedgroup was found to be coming to police attention as suspects/offendersat a much younger age. More needs to be known about the characteris-tics of the interactions between the self-inflicted group and the police inorder to make any firm comments about these findings. Of particularnote was the lower rate of recorded Axis 1 disorders found among theself-inflicted suicide group as compared to the PPPS group. This findingis perhaps reflective again of the under-reporting of the high prevalenceAxis I disorders and of personality disorders in the sample generally.

6.3. Limitations

When considering the generalizability of these findings, there areseveral issues to consider. Firstly, the analyses relied on existing datathat was collected for purposes other than research in the context ofthe roll-out of a new data recording protocol for police. The narrativesavailable considerably varied in both length and detail and thereforedid not always include all of the elements required to make a completeassessment of the inclusion of a potential case, or to characterise thebroader context in a consistent manner. As such, a more inclusive crite-rion of PPPS was used so as not to unduly underestimate the prevalenceof these incidents. A further limitationwas that therewas commonly lit-tle by way of detail on the specific police strategies that were used togain resolution of the crisis situation,with outcomes generally classifiedas just ‘resolved/negotiated/arrested’ without the necessary chain ofevents that led to this conclusion. As such, the degree of detail availableon successful police tactics was often quite limited. An additional com-plication relates to the ascertainment of suicidal intent. In general sui-cide studies, there is an inherent difficulty in assessing and attributingintent; in provoked shooting studies researchers must infer the addi-tional element of intent to suicide by forcing the police to shoot themand attribute the wish to die, sometimes in the absence of the telling

Please cite this article as: McLeod, L.C., et al., The frequency and nature oftional Journal of Law and Psychiatry (2014), http://dx.doi.org/10.1016/j.ijlp

component of self-infliction. Further to this, in investigating potentialprovoked shootings there is no definitive death to anchor analysis ofthe preceding chain of events, thus a certain level of detail is assumedin the absence of this telling endpoint (Lord & Sloop, 2010). Concernshave also been raised over the attribution of active intent as comparedto instances of an ‘indifference’ to life or death (McKenzie, 2006). Futureresearch should attempt to disentangle these subtle differences andhow different motivations may manifest differently in behaviour. Not-withstanding these potential limitations, the size of the source samplewas large and the method adopted for classifying cases was complete,comprehensive and transparent, thus contributing to increased confi-dence in the estimated prevalence and characterisations providedhere. The studyprovides a useful and timely examinationof the success-fully resolved PPPS cases.

6.4. Implications, future directions and conclusions

The very nature of contemporary policing requires police to be able tothink on their feet andhaveonehundreddifferent responses to a hundreddifferent situations ‘up their sleeves’ at their disposal. While the resultspresented here suggest that police are successfully (non-fatally) resolvingpotentially deadly encounters with people in acute psychiatric crisis reg-ularly, a more nuanced understanding of what ultimately leads to this asan outcome was not possible here. Future research needs to adopt moredetailed, in-depth analyses of police decision-making in these highlystressful scenarios to help lay the foundations for a research-informed,multifaceted multi-layered approach which provides a broader array oftactical optionswhich emphasise the importance of adaptive and respon-sive communication skills (e.g., Parent, 2011).

Issues of thefluctuation in suicidal intent and the influence of this onsuspect demeanour and impulsive behaviours remain unclear and high-ly subjective. Further research should consider the differences betweensuicidal intent and an indifference to life or death as they relate to op-portunities for intervention. The findings reported by Deisenhammeret al. (2009) suggest that there arewindows of opportunity to intervenewith a person who is acutely suicidal; the extent to which these couldapply in a psychiatric crisis encounter involving frontline police remainsunclear at this time. The results do suggest some increase in our under-standing of potential influences on the decision to use police as ameansof suicide over other methods but the results, especially in relation tothe impact of previous criminal histories, need to be reflected elsewherebeforewe can start to develop a theory (Klinger, 2001). Proposals to fur-ther standardise police reporting practices around potential provokedshootings and the adoption of a consistent language by police inreporting protocols (e.g., Reuland, Schwarzfeld, & Draper, 2009)would further enhance our ability to more accurately gauge both thefrequency and nature of these incidents and effective police responses.These recommendations could be readily implemented across differentjurisdictions and would meaningfully contribute to the development ofa revised suite of responses to people in acute crisis who are living in thecommunity but in desperate need of timely interventions and access toultimately what need to be a network of interagency supports fromhealth and welfare agencies.

References

Ansseau, M., Dierick, M., Buntinkz, F., Cnockaert, P., De Smedt, J., Van Den Haute, M., et al.(2004). High prevalence of mental disorders in primary care. Journal of AffectiveDisorders, 78, 49–55.

Bayley, D., & Bitner, E. (1989). Learning the skills of policing. In R. Durham, & G. Alpert(Eds.), Critical issues in policing: Contemporary readings. Prospect Heights, IL:Waveland Press.

Best, D., & Quigley, A. (2003). Shootings by the police: What predicts when a firearms of-ficer in England or Wales will pull the trigger? Policing and Society, 13, 349–364.

Best, D., Quigley, A., & Bailey, A. (2004). Police shooting as a method of self-harming: Areview of the evidence for ‘suicide by cop’ in England and Wales between 1998and 2001. International Journal of the Sociology of Law, 32, 349–361.

Bronitt, S., & McSherry, B. (2005). Principles of criminal law (Second ed.). Sydney: Thom-son Lawbook Co, 147–206.

resolution of potential police provoked shooting encounters, Interna-.2014.02.009

7L.C. McLeod et al. / International Journal of Law and Psychiatry xxx (2014) xxx–xxx

Brouwer, G. (2009). Review of the use of force by and against Victorian police. Melbourne:Office of Police Integrity.

Carroll, M. (2005). Mental health system overburdening police. Police Journal, 18–22.Clifford, K. (2010). The thin blue line of mental health in Australia. Police Policies and

Research, 11, 355–370.Coffey, A. (1990). Law enforcement: A human relations approach. Englewood Cliffs, NJ:

Prentice-Hall.Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., & Weiss, E. M.

(2009). The duration of the suicidal process: How much time is left for interventionbetween consideration and accomplishment of a suicide attempt? Journal of ClinicalPsychiatry, 70, 19–24.

Geberth, V. (1993). Suicide-by-cop. Law and Order, 41, 105–109.Godfredson, J.W., Thomas, S. D.M., Ogloff, J. R. P., & Luebbers, S. (2011). Police perceptions

of their encounters with individuals experiencing mental illness: A Victorian survey.Australian and New Zealand Journal of Criminology, 44, 180–195.

Hickman, M. J., Piquero, A. R., & Garner, J. H. (2008). Toward a national estimate of policeuse of nonlethal force. Criminology & Public Policy, 7, 563–604.

Homant, R. J., & Kennedy, D. B. (2000). Effectiveness of less than lethal force in suicide-by-cop incidents. Police Quarterly, 3, 153–171.

Homant, R. J., & Kennedy, D. B. (2000). Suicide by police: A proposed typology of law en-forcement officer-assisted suicide. Policing: An International Journal of Police Strategies& Management, 23, 339–355.

Hutson, H., Anglin, D., Yarbrough, J., Russell, M., Strote, J., Canter, M., et al. (1998). Suicideby cop. Annals of Emergency Medicine, 32, 665–669.

Kennedy, D. B., Homant, R. K., & Hupp, R. T. (1998). Suicide by cop. FBI Law EnforcementBulletin, 67(8), 21–27.

Kesic, D., Thomas, S. D. M., & Ogloff, J. R. P. (2010). Mental illness among police fatalities inVictoria 1982–2007: Case linkage study. Australian and New Zealand Journal of Psychi-atry, 44, 463–468.

Kesic, D., Thomas, S. D. M., & Ogloff, J. R. P. (2012). Analysis of fatal police shootings: Time,space and suicide by police. Criminal Justice and Behavior. http://dx.doi.org/10.1177/0093854812440084.

Klinger, D. A. (2001). Suicidal intent in victim-precipitated homicide: Insights from thestudy of “suicide by cop”. Homicide Studies, 5, 206–226.

Lindsay, M., & Lester, D. (2004). Suicide by cop: Committing suicide by provoking police toshoot you. Amityville, New York: Baywood Publishing Company, Inc.

Lindsay, M., & Lester, D. (2008). Criteria for suicide-by-cop-incidents 1. PsychologicalReports, 102, 603–605.

Lord, V. B. (2000). Law enforcement-assisted suicide. Criminal Justice and Behavior, 27,401–419.

Lord, V. B., & Sloop, M. W. (2010). Suicide by cop: Police shootings as a method of self-harming. Journal of Criminal Justice, 38, 889–895.

McKenzie, I. K. (2006). Forcing the police to open fire: A cross-cultural/international ex-amination of police-involved, victim-provoked shooting. Journal of Police CrisisNegotiations, 6, 5–25.

Please cite this article as: McLeod, L.C., et al., The frequency and nature oftional Journal of Law and Psychiatry (2014), http://dx.doi.org/10.1016/j.ijlp

Mohandie, K., & Meloy, J. R. (2000). Clinical and forensic indicators of “suicide by cop”.Journal of Forensic Sciences, 45, 384–389.

Mohandie, K., & Meloy, J. R. (2010). Hostage and barricade incidents within an officer-involved-shooting sample: Suicide by cop, intervention efficacy, and descriptive char-acteristics. Journal of Police Crisis Negotiations, 10, 101–115.

Mohandie, K., Meloy, J. R., & Collins, P. I. (2009). Suicide by cop among officer involvedshooting cases. Journal of Forensic Sciences, 54, 456–462.

National Health and Medical Research Council (2007). National statement on ethical con-duct in human research. Canberra: Australian Government.

Parent, R. B. (1996). Aspects of police use of deadly force in British Columbia: The phenome-non of victim-precipitated homicide. (Master's thesis). Burnaby, BC: Simon FraserUniversity.

Parent, R. B. (2011). The police use of deadly force in British Columbia: Mental illness andcrisis intervention. Journal of Police Crisis Negotiations, 11, 57–71.

Parent, R. B., & Verdun-Jones, S. (1998). Victim-precipitated homicide: Police use of dead-ly in British Columbia. Policing: An international journal of police strategies &management, 21, 432–448.

Pinizzotto, A. J., Davis, E. F., & Miller, C. E. (2005). Suicide by cop: Defining a devastatingdilemma. FBI Law Enforcement Bulletin, 74, 8–20.

Pinizzotto, A. J., Davis, E. F., & Miller, C. E. (2007). Deadly mix: Officers, offenders and thecircumstances that bring them together. FBI Law Enforcement Bulletin, 76, 1–10.

Reuland, M., Schwarzfeld, M., & Draper, L. (2009). Law enforcement responses to peoplewith mental illnesses: A guide to research informed policy and practice.NewYork: Coun-cil of State Governments Justice Center.

Ruiz, J. (1993). An interactive analysis between uniformed law enforcement officers andthe mentally ill. American Journal of Police, 12, 149–168.

Scoville, D. (1998). FYI: Getting you to pull the trigger. Police, 11, 36–44.Short, T. B. R., MacDonald, C., Luebbers, S., Ogloff, J. R. P., & Thomas, S. D. M. (2012). The

nature of police involvement in mental health transfers. Police Practice & Research:An International Journal. http://dx.doi.org/10.1080/15614263.2012.736717.

Short, T. B. R., Thomas, S. D. M., Luebbers, S., Ogloff, J. R. P., & Mullen, P. E. (2010). The ep-idemiology of psychosis: Prevalence and use of mental health services in a randomcommunity sample. Australian and New Zealand Journal of Psychiatry, 44, 475–481.

Staiger, P. K., Thomas, A. C., Ricciardelli, L. A., & McCabe, M. P. (2011). Identifying depres-sion and anxiety disorders in people presenting for substance use treatment. MedicalJournal of Australia, 195, S60–S63.

Van Zandt, C. R. (1993). Suicide by cop. Police Chief, 60, 24–30.Wilson, E. F., Davis, J. H., Bloom, J. D., Batten, P. J., & Kamara, S. G. (1998). Homicide or sui-

cide: The killing of suicidal persons by law enforcement officers. Journal of ForensicSciences, 43, 46–52.

Wolfgang, M. E. (1959). Suicide by means of victim-precipitated homicide. Journal ofClinical & Experimental Psychopathology, 20, 335–349.

resolution of potential police provoked shooting encounters, Interna-.2014.02.009