Adm Policy Ment Health DOI 10.1007/s10488-014-0594-9
ORIGINAL ARTICLE
Co-responding Police-Mental Health Programs: A Review G. K. Shapiro • A. Cusi • M. Kirst • P. O’Campo • A. Nakhost • V. Stergiopoulos
Springer Science+Business Media New York 2014
Abstract Co-responding police-mental health programs are increasingly used to respond to ‘Emotionally Disturbed Persons’ in the community; however, there is limited understanding of program effectiveness and the mechanisms that promote program success. The academic and gray literature on co-responding police-mental health programs was reviewed. This review synthesized evidence of outcomes along seven dimensions, and the available evidence was further reviewed to identify potential mechanisms of program success. Co-responding police-mental health programs were found to have strong linkages with community services and reduce pressure on the justice system, but there is limited evidence on other impacts. The relevance of these findings for practitioners and the major challenges of this program model are discussed, and future research directions are identified.
Preliminary findings were presented at the Provincial Human Services & Justice Coordinating Committee: 2013 Educational & Training Provincial Conference. November 25–27, 2013, Toronto, Canada. G. K. Shapiro A. Cusi M. Kirst P. O’Campo V. Stergiopoulos Centre for Research on Inner City Health, Keenan Research Centre in the Li Ka Shing Knowledge Institute, St. Michael’s Hospital, 209 Victoria Street, Toronto M5B 1T8, Canada G. K. Shapiro (&) Department of Psychology, McGill University, 1205 Dr. Penfield Avenue, Montreal H3A 1B1, Canada e-mail:
[email protected]
Keywords Co-responding police-mental health programs Community outreach Mobile crisis response Psychiatric service organization and delivery
Introduction The interaction between police officers and individuals who experience a mental health crisis is high and increasing in North America (Coleman and Cotton 2010). Between 7 and 31 % of police calls involve a person with mental illness (Abbot 2011; Baess 2005; Wilson-Bates 2008). Not surprisingly, police have been described as the ‘‘de facto mental health providers’’ (Cotton & Coleman 2006, p. 2), ‘‘key front-line responders in mental health emergencies’’ (Steadman et al. 2000, p. 645), ‘‘streetcorner psychiatrists’’ (Teplin and Pruett 1992, p. 139), and ‘‘society’s de facto 24/7 mental health workers’’ (Thompson 2010, p. 3). The increase in police involvement with mental illness is reported to be due to a number of factors including deinstitutionalization (i.e. more individuals with psychiatric issues residing within the community), fewer psychiatric hospitals and hospital beds, decreased hospitalization, and changes M. Kirst P. O’Campo Dalla Lana School of Public Health, University of Toronto, 6th floor, 155 College St, Toronto M5T 3M7, Canada A. Nakhost V. Stergiopoulos Department of Psychiatry, University of Toronto, 250 College Street, 8th floor, Toronto M5T 1R8, Canada A. Nakhost Mental Health Services, St. Michael’s Hospital, 30 Bond Street, Toronto M5B 1W8, Canada
G. K. Shapiro Institute of Community and Family Psychiatry (ICFP), Lady Davis Institute, Jewish General Hospital, 4333 Coˆte St-Catherine Road, Montreal H3T 1E4, Canada
123
Adm Policy Ment Health
in mental health laws (Cotton and Coleman 2006; Fisher et al. 2006; Lamb et al. 2002). Increased police interaction with, what some police units call, ‘Emotionally Disturbed Persons’ (EDP) in crisis situations can be costly in both human and economic terms. EDPs consume a significant amount of policing resources (Mental Health Commission 2009; Thompson 2010). In addition, high profile incidents of injury or deaths have longstanding implications for persons with mental illness, their families, and the larger community (Adelman 2003; Hails and Borum 2003). A lesser focus of the media, but equally problematic, is the criminalization of a large number of persons with mental illness who have committed minor crimes and are then taken to jail rather than to hospitals or community psychiatric facilities (Lamb et al. 1995). Individuals experiencing a serious mental illness are at times arrested because police officers cannot see any other option, or hope to avoid the loss of time required to locate and escort an EDP to a more appropriate facility (Finn and Sullivan 1989; Fry et al. 2002). Furthermore, at least in the Canadian context, some officers will occasionally charge a person with an apparent mental illness with a minor offense because, through the justice system, EDPs may have better access to mental health support services, a practice called ‘mercy booking’ (Wood et al. 2011). Notably, it is becoming increasingly clear that individuals with mental illnesses are often arrested not because of symptomatic behavior or crimes related to mental health problems (Peterson et al. 2014), but rather as a result of an individual’s opportunities, activities, and life circumstances (e.g. housing or employment) (Fisher et al. 2006). Nevertheless, diverting individuals with mental illness—or those experiencing a mental health crisis—from the criminal justice system ‘‘is on numerous counts a worthy endeavor’’ (Fisher et al. 2006, p. 548), as it holds promise of connecting these individuals to services that address the social determinants of health. Improving police interactions with persons with mental illness who are in crisis has not surprisingly become a priority for police services and communities. In order to respond appropriately to EDPs, different configurations of police–mental health partnerships have emerged (Laing et al. 2009; Lamb et al. 2002). These partnerships are highly heterogeneous but generally follow one of three strategies to assist EDPs (Adelman 2003; Deane et al. 1999). One program model promotes mental health expertise for police officers so that they are better prepared to provide crisis intervention services (Watson and Fulambarker 2013). This Crisis Intervention Team (CIT) approach, which originated in Memphis in 1988, is a popular model that emphasizes mental health training for officers (for a recent overview of the CIT program see Cross et al. 2014). In a less popular program model, mental health consultants are hired by police departments to provide phone-based or on-site assistance to officers in the
123
field (Saunders and Marchik 2007; Zealberg et al. 1992). In contrast, a third program model has established coresponding police-mental health programs that include both local community mental health workers and police officers that collaboratively respond to EDPs crises. Despite a significant amount of research on the CIT model, there is a paucity of literature on the co-responding police-mental health program model (Boscarato et al. 2014). This is particularly unfortunate as ‘‘[i]n Canada, the predominant mobile response is in fact a joint mobile response, which is exemplified by programs such as the COAST Program in Hamilton, and the similar Car 87 program in Vancouver’’ (Kean et al. 2012, p. 20; Kisely et al. 2010). Similarly, co-responding police-mental health programs have been implemented in the United States with teams that have been evaluated in DeKalb County (Georgia), Knoxville (Tennessee), and Los Angeles (California), among other jurisdictions. The co-responding police-mental health program involves a collaboration of specially trained police officers and mental healthcare workers that provide on-site services to EDPs in the community. The theory underlying these programs is that a joint response is preferable as police are specialists in handling situations that involve violence and potential injury while mental health professionals are specialists in providing mental health consultation to officers and mental health care to individuals in crisis (Forchuk et al. 2010; Lamb et al. 1995). As Rosenbaum (2010) explains: ‘‘[t]hese teams are based on the idea that the more police and mental health workers collaborate, the better the two systems can serve consumers and each other effectively’’ (p. 176). The co-responding police-mental health programs have gone by many names including the Integrated Mobile Crisis Response Team in Victoria (British Columbia), Mobile Crisis Intervention Team in Toronto (Ontario), Crisis Outreach and Support Team in Hamilton (Ontario), Mental Health Mobile Crisis Team in Halifax (Nova Scotia), Mobile Mental Health Crisis Unit in Knoxville (Tennessee), Mobile Crisis Program in DeKalb County (Georgia), the Systemwide Mental Assessment Response Team in Los Angeles (California), and the Police, Ambulance and Clinical Early Response in Melbourne (Victoria) (Allen Consulting Group 2012; Canadian Crisis Response Services List 2008; Rosenbaum 2010). In this review, these programs will be called co-responding police-mental health programs. There are multiple objectives for the co-responding police-mental health program, including: deescalating crises, preventing injuries to individuals in crisis and the response team, linking individuals who are experiencing psychiatric emergencies to appropriate services in the community, and reducing pressure on both the justice system (e.g. by decreasing arrests and officer’s time involved with handling psychiatric emergency situations) as well as the health care system (e.g. by decreasing
Adm Policy Ment Health
unnecessary visits to the emergency department) (Borum 2000; Matheson et al. 2005; Scott 2000). Furthermore, these programs often aim to be accountable and cost effective. A coordinated approach is needed in order to optimally achieve these goals (Wood et al. 2011). Notably, co-responding police-mental health programs differ greatly in the populations they serve, funding levels, program guidelines, type-of response, hours of operation, staff expertise, equipment, and training (Kean et al. 2012; Ligon 1997). Furthermore, though the programs share a common goal of on-scene mobile crisis intervention and outreach, the central activities of each program may differ (Finn and Sullivan 1989; Patterson 2010). For example, some of the varying program activities involve program capacity building, maintaining a presence in vulnerable areas, staff preparedness and training, partnership building, following-up with consumers, short-term counseling, referral, and evaluating program outcomes. There is a considerable need for high-quality evidence to inform policy regarding how and when to implement coresponding police-mental health programs (Abbott 2011; Forchuk et al. 2010; Wood et al. 2011). As Forchuk et al. (2010) explain, ‘‘[s]urprisingly, there is limited understanding of the essential components, processes, and outcomes for crisis services to date’’ (p. 74). A review of coresponding police-mental health programs was therefore conducted to synthesize the published literature, identify promising practices, and determine gaps in the literature to guide future studies.
Method Search Strategy The search included data in the English language, published from 1970 to 2013. Scholarly peer-reviewed literature was accessed through examining the databases Pubmed, Web of Science, JSTOR, and OVID (i.e. Medline, PscyINFO, Cochrane databases, and Embase). The search terms that were used included: ‘mobile crisis intervention team’, ‘mobile crisis unit’, ‘mobile crisis program’, ‘coresponding police-mental health team’; as well as ‘police or law enforcement’ combined with ‘psychiatric nurses’, ‘mental health clinician’, ‘psychiatric nurse’, ‘psychiatrist’, ‘psychologist’, ‘social worker’, or ‘crisis intervention’. Other sources, including Google Scholar, Canadian Mental Health Association (CMHA), National Institute of Mental Health (NIMH), National Mental Health Development Unit (NMHDU), National Alliance on Mental Illness (NAMI), and The Substance Abuse and Mental Health Services Administration (SAMHSA), were accessed (using the same search terms as above) to identify research reports
that have been published in the gray literature. Furthermore, experts were contacted and the reference lists of relevant articles were scanned in order to locate additional relevant literature. Inclusion Criteria Studies involving co-responding police-mental health program were included in the review. Literature on interventions that do not include both police and mental healthcare workers in their response were excluded as they fall outside the scope of our analysis. Data Extraction Eleven peer-reviewed papers, seven reports, and three dissertations were identified and were reviewed in detail (Table 1). Data were extracted that enabled the authors to identify the components and contexts of co-responding police-mental health program, their activities, their critical ingredients, and their evaluated outcomes.
Results A number of outcomes have been used in evaluations of coresponding police-mental health program, including: (1) averting crisis escalation and injury, (2) linking EDPs with community services, (3) reducing pressure on the justice system (i.e. through reducing the number of arrests and police officers’ handling time), (4) improving officers’ perception of individuals who have a mental illness, (5) reducing the number of hospital admissions, (6) cost effectiveness, and (7) program perception. This review conducts a narrative synthesis of the evidence along the seven aforementioned outcome domains, and identifies and proposes potential mechanisms that may underlie program success. A Synthesis of the Evidence Averting Crisis Escalation and Injury The combination of professional mental health staff and law enforcement officers attending crisis situations has been thought to avert crisis escalation and injury (Baess 2005). Although this is an outcome that is difficult to measure, it is believed that through police training and ‘ride-alongs’ there is an opportunity to enhance officer understanding of mental illness and offer alternative in-themoment tactics for deescalating situations without resorting to the use of violence (Abbott 2011). Indeed, Baess (2005) found that clinical staff report feeling safer when attending community crisis calls alongside police officers.
123
Adm Policy Ment Health Table 1 List of included studies Study
Publication type
Methods
Outcomes
Abbott (2011)
Dissertation
Quantitative analysis comparing police attitudes in four police departments at a single point in time
• Averting crisis escalation and injury (1)
Allen Consulting Group (2012)
Report
Mixed-methods approach that included a quantitative comparison of an intervention (with a control), a qualitative consultations with relevant stakeholders, and a cost analysis
• Improving officers’ perception of individuals who have a mental illness (4) • Program perception (7) • Averting crisis escalation and injury (1) • Reducing pressure on the justice system (3) • Reducing the number of hospital admissions (5) • Cost effectiveness (6) program perception (7)
Anderson and Taylor (2013)
Report
Descriptive case report and gap analysis
• Reducing pressure on the justice system (3)
Baess (2005)
Report
Mixed-methods approach that included quantitative analysis, and a qualitative consultations with police officers, professionals, and the public
• Averting crisis escalation and injury (1) • Reducing pressure on the justice system (3) • Improving officers’ perception of individuals who have a mental illness (4) • Reducing the number of hospital admissions (5) • Cost effectiveness (6) • Program perception (7)
Bar-On (1995)
Peer-reviewed publication
Descriptive research
• Reducing pressure on the justice system (3)
Borum et al. (1998)
Peer-reviewed publication
A quantitative analysis surveying police in three crisis programs, a record review of representative cases, and key informant interviews
• Reducing pressure on the justice system (3) • Program perception (7)
Brown et al. (2009)
Report
Descriptive case report
• Reducing pressure on the justice system (3)
City of Toronto Mobile Crisis Intervention Team Coordination Steering Committee (2013)
Report
Descriptive case report, gap analysis, and consultation with stakeholders and consumers/survivors
• Reducing pressure on the justice system (3)
Cobb (1997) as cited in Rosenbaum (2010)
Report cited in a peerreviewed publication
Descriptive case report
• Cost effectiveness (6)
Dean et al. (2000)
Report
Descriptive case report
• Reducing pressure on the justice system (3)
Finn and Sullivan (1989)
Peer-reviewed publication
Descriptive analysis of eight crisis service models
• Linking of EDPs with community services (2) • Reducing pressure on the justice system (3) • Cost effectiveness
Forchuk et al. (2010)
123
Peer-reviewed publication
Ethnographic case-study that compared three crisis service models (by consulting consumer, family member, and service provider focus groups), and a naturalistic observation of crisis services
program perception (7)
Adm Policy Ment Health Table 1 continued Study
Publication type
Methods
Outcomes
Kisely et al. (2010)
Peer-reviewed publication
Mixed-methods approach that included a controlled before-after quantitative study of an intervention area (with a control area), and a qualitative assessment of the views of service recipients, families, police officers, and health staff at baseline and 2 years afterwards
• Linking of EDPs with community services (2)
Lamb et al. (1995)
Peer-reviewed publication
Retrospective design that examined the records of consecutive referrals to a crisis program, and the status of subjects at a 6 month follow-up
• Reducing pressure on the justice system (3) • Program perception (7)
• Reducing pressure on the justice system (3) • Reducing the number of hospital admissions (5)
Landeen et al. (2004)
Peer-reviewed publication
Retrospective design that examined data from four time periods using a qualitative, text-based approach
• Reducing the number of hospital admissions (5)
Ligon (1997)
Dissertation
Quantitative analysis of consumer and family’s surveyed satisfaction with a crisis service
• Program perception (7)
Ligon and Thyer (2000)
Peer-reviewed publication
Quantitative analysis of consumer and family’s surveyed satisfaction with a crisis service
Program perception (7)
Ratansi (2004)
Dissertation
Quantitative analysis of survey data that was completed by the police and the co-responding program, and consultations with members of the coresponding program
• Averting crisis escalation and injury (1)
Saunders and Marchik (2007)
Scott (2000)
Peer-reviewed publication
Peer-reviewed publication
Mixed-methods approach that included a descriptive case report, qualitative interviews with key informants, and quantitative evaluation of surveys completed by law enforcement patrol officers Retrospective evaluation of case records of a coresponding program (and a control group), a quantitative analysis of consumer and police’s surveyed satisfaction, and a cost analysis
• Reducing pressure on the justice system (3) • Program perception (7)
• Improving officers’ perception of individuals who have a mental illness (4) • Program perception (7)
• Linking of EDPs with community services (2) • Reducing pressure on the justice system (3) • Reducing the number of hospital admissions (5) • Cost effectiveness (6)
123
Adm Policy Ment Health Table 1 continued Study
Publication type
Methods
Outcomes
Steadman et al. (2000)
Peer-reviewed publication
Comparative cross-site design examining records of three different police response programs
• Linking of EDPs with community services (2) • Reducing pressure on the justice system (3) • Program perception (7)
On the other hand, Ratansi (2004) examined open-ended questions and found that ‘‘most officers believed that working with MCT [i.e. in the co-responding police-mental health program] had no impact or did not improve the time spent on calls, identification, danger, or the effectiveness of the call’’ (p. 156). Interestingly, the Allen Consulting Group (2012) investigated the number of times a use of force form was completed by the co-responding police-mental health program compared to a control ‘usual service’. They found that less use of force forms were filled out in the coresponding police mental health program catchment area compared to the usual service provision catchment area. However, the difference between the number of use of force forms filled out in the two areas was small and this difference narrowed during the period of investigation (January 2010 to March 2011). Unfortunately, this analysis does not statistically analyze this difference. Furthermore, this report does not consider whether the type of force that is reported differs between the two areas. Nevertheless, the Allen Consulting Group (2012) further reports that the coresponding police-mental health program provided deescalation advice on 10 % of occasions and undertook other activities that might assist in preventing force, crisis escalation, and injury. Linking EDPs with Community Services Connecting individuals in crisis with community services, rather than the justice system or acute care services, is thought to be the most appropriate way to support consumers and prevent reoccurrence of a crisis and ‘revolvingdoor’ recidivism (Finn and Sullivan 1989). As Scott (2000) explains, ‘‘[a]n anticipated benefit of the program was increased access for consumers to community-based emergency services in the least restrictive environment in order to avoid the trauma of psychiatric hospitalization whenever possible’’ (p. 1156). Some research has evaluated this claim. Steadman et al. (2000) examined three sites’ dispatch calls and found that in situations where a specialized response was present, the co-responding policemental health program in Knoxville had the largest
123
proportion of referrals to treatment services compared to the other models (36 %, compared to 0 % in Memphis’s CIT model, and 3 % in Birmingham’s program whereby civilian officers assist police officers in mental health crises). In other words, when the co-responding police-mental health program in Knoxville responded to a mental health crisis, they were more likely to refer an individual to case managers, mental health centers, or outpatient treatment, compared to the other two specialized models. In contrast, Memphis’s CIT model was most likely to take individuals in crisis to a psychiatric emergency room, a general hospital emergency room, a detoxification unit, or another psychiatric facility (75 % of the time, compared to 42 % of the time in Knoxville, and 20 % of the time in Birmingham); while Birmingham’s specialized team was most likely to resolve the situation on scene (64 % of the time compared to 23 % of the time in Memphis and 17 % of the time in Knoxville). Seemingly, linking individuals in crisis with community mental health services is particularly emphasized in the co-responding police-mental health program. This is not an isolated finding. In 2010, Kisely et al. found that individuals who had been in contact with the co-responding police-mental health program in Halifax showed greater service engagement than control subjects, as demonstrated by increased outpatient contacts. Moreover, according to Finn and Sullivan (1989), in Fairfax County, Virginia, 71 % of consumers followed through with a treatment recommendation and were actively engaged in a voluntary outpatient program within 4 weeks of the crisis intervention. Finally, although only 2 % of individuals to whom deputies gave wallet cards listing outpatient mental health services in Washtenaw County sought help, the number seeking community services rose to 18 % when consumers were telephoned to encourage aftercare within 48 h (Finn and Sullivan 1989). Reducing Pressure on the Justice System EDP crises create a significant pressure on the justice system. Co-responding police-mental health programs are believed to mitigate this pressure by reducing the number of EDP arrests (and the related processing time) and
Adm Policy Ment Health
decreasing police officers’ on-site handling time. A number of studies have demonstrated low EDP arrest rates associated with the co-responding program. For example, Steadman et al. (2000) found the co-responding Knoxville program had a low arrest rate of 5 % (compared to 6 % in Memphis and 13 % in Birmingham). Similarly, Brown et al. (2009) report that only 2 % of 445 calls resulted in jail placement during the first nine months of the coresponding police-mental health program in Polk County. Unfortunately, Brown et al. (2009) did not compare the arrest rate in Polk County to a comparable location that does not have a co-responding police-mental health program, or to Polk County arrest rates before implementing the co-responding police-mental health program (in a before-after design). Furthermore, a study by Lamb et al. (1995) found that only 2 of 101 consumers were arrested in Los Angeles’ coresponding police-mental health program, despite 70 of the 101 consumers exhibiting psychiatric symptoms at referral, 20 among them being overtly violent and 29 others who exhibited threatening behaviour. Lamb et al. (1995) thereby concluded that ‘‘because of the use of well-trained teams consisting of a mental health professional and a police officer, these subjects were not criminalized, even though they came from a population at high risk for criminalization’’ (pp. 1269–1270). However, notably, at a 7-month follow-up, 22 % of consumers had been arrested (12 % for crimes of violence) (Lamb et al. 1995). This potentially indicates that despite the immediate success of the crisis team in diverting EDPs from the criminal justice system, consumers may not be successfully diverted in the long-term. This finding underlines that it is especially important that evaluative studies investigating the coresponding police-mental health program examine both the proximal and distal outcomes of the programs. Another study that investigated arrest rates in DeKalb County, Georgia, did not find a significant difference in arrest rates between the co-responding police-mental health program and a control group in a retrospective examination design (Scott 2000). However, a descriptive report by Dean et al. (2000) found that the co-responding police-mental health program in Lumberton (North Carolina) greatly reduced repeat calls from chronic calling problem homes. Other research has demonstrated that co-responding police-mental health programs decrease the on-site handling time of officers, thereby reducing the burden on police (Finn and Sullivan 1989). For example, there were reported reductions in time spent on scene by police officers in the co-responding police-mental health program in Halifax, Canada (136 min), compared to the control area (165 min) (Kisely et al. 2010). Similarly, the Allen Consulting Group (2012) reported that on average the police first responder unit was released a third of the time when
the co-responding police-mental health program was involved (52 min) compared to the control area (nearly 3 h). Baess (2005) also reported that clinician crisis responders were able to attend to more than double the number of high acuity calls when plain-clothed police officers were on shift (Baess 2005). However, the methodology of Baess’s (2005) study is not presented and it is unclear what factors underlie this improvement. Further, while patrol officers waited on average for 121 min when they attended the emergency department on their own, they only waited for 45 min when assisted in the co-responding police-mental health program (Baess 2005). Anderson and Taylor (2013) similarly argue that the co-responding police-mental health program in Surrey, British Columbia, reduces the overall time that police spend in hospital wait times and diversions from hospital (Anderson 2013). Evidently, the co-responding police-mental health program can only decrease arrests and reduce on-site handling time of police officers if the teams are able to reach the scene, and do so within an acceptable amount of time. Some research has evaluated the reach of the co-responding police-mental health program in several jurisdictions. For example, the Allen Consulting Group (2012) found that in 90 % of cases the co-responding police-mental health program responded in 30 min or less from the time that they were contacted. Furthermore, Kisely et al. (2010) found that after implementing the co-responding policemental health program (that offered telephone support 24 h a day backed by a co-responding team) in Halifax, Nova Scotia, the number of crises to which the team was able to respond increased. However, Steadman et al. (2000) found that the co-responding police-mental health program only responded to 40 % of mental disturbance calls in Knoxville (compared to 92 % in Memphis and 28 % in Birmingham). They argue that ‘‘one of the key concerns expressed in this study about the Knoxville mobile crisis unit [i.e. Knoxville’s co-responding police-mental health program] was that response times were excessive and impractical. The delayed response led officers not to use the unit’s services as often as they otherwise might have and forced them to consider alternative dispositions’’ (Steadman et al. 2000, pp. 648–649). The delayed response of some co-responding police-mental health programs has been linked to transportation in geographically spread out locales and a lack of capacity (i.e. staffing concerns) (Borum et al. 1998; Ratansi 2004). As Ratansi (2004) explains, ‘‘the team [in Hamilton County, Ohio] was not effective because manpower motivation, availability, response time, and feasibility were not adequately addressed. For example, having a collaboration in which one of the stakeholders (MCT) is only available 9–5 during the week is a problem when the other stakeholder is available 24/7’’ (p. 179). Indeed, availability and access of mental healthcare workers is particularly
123
Adm Policy Ment Health
problematic at night (Bar-On 1995; City of Toronto Mobile Crisis Intervention Team Coordination Steering Committee 2013). Improving Officers’ Perception of Individuals Who have a Mental Illness Some research has also examined the proposition that the co-responding police-mental health programs improve officers’ understanding and perception of mental illness. Indeed, Baess (2005) found that feedback from two police officers working on the co-responding police-mental health program in Vancouver Island revealed that they developed a better understanding of mental health issues (i.e. diagnostic criteria and behavioural interventions). Furthermore, using a questionnaire that examines community attitudes towards people with mental illness, Abbott (2011) found that surveyed officers working in departments with jail diversion programs in two Massachusetts communities (i.e. Framingham and Quincy) reported greater tolerance for responding to individuals with a mental illness and more strongly endorse their role in supporting persons with mental illness than officers working in departments without jail diversion programs (i.e. Lynn and Peabody). However, there was no significant difference between departments with and without a co-responding police-mental health program regarding levels of acceptance of individuals with a mental illness living in community-based settings. In another study, Saunders and Marchik (2007) demonstrated a statistically significant increase in knowledge of mental health problems before and after officers received training from the co-responding police-mental health programs in Polk County, Iowa. Interestingly however, they found that 76 % of the 210 surveyed officers reported learning about mental health issues directly through observing the coresponding police-mental health program staff in the field or through discussion with the co-responding police-mental health program staff (Saunders and Marchik 2007). Reducing the Number of Hospital Admissions Hospital admissions are costly and present a burden on limited resources. There is mixed evidence that the coresponding police-mental health programs reduce hospitalizations. In one study, 55 % of emergencies handled by the co-responding police-mental health programs were managed without psychiatric hospitalization of the person in crisis, compared with 28 % in a control group of regular police (Scott 2000). Scott (2000) therefore concluded that ‘‘mobile crisis programs can decrease hospitalization rates for persons in crisis and can provide cost-effective psychiatric emergency services that are favorably perceived by consumers and police officers’’ (p. 1153). Similarly, Baess
123
(2005) reported that fewer than 15 % of 1,200 referrals from the co-responding police-mental health program were directed to the hospital’s emergency room (a 7.8 % decrease compared with historical data). Furthermore, the Allen Consulting Group (2012) found that the coresponding police-mental health program they evaluated had fewer referrals to hospital emergency departments (52 % of cases) compared to a control area (82 % of cases). In addition, they found that the length of stay in hospital emergency departments for mental health patients referred by the co-responding police-mental health program was reduced by approximately 2 h compared to the control area (Allen Consulting Group 2012). However, other studies did not find that the coresponding police-mental health program model reduces hospital admissions. For example, in one co-responding police-mental health program individuals are admitted to the hospital in approximately 75 % of apprehensions under the Mental Health Act (Landeen et al. 2004). Similarly, Lamb et al. (1995) examined consecutive referrals and found that of the initial 101 subjects, 80 were initially taken to hospital and 73 were hospitalized. Further, in a sixmonth follow-up of these individuals, 42 % of subjects were re-hospitalized (Lamb et al. 1995). Cost Effectiveness The goals of the crisis teams are presented as humanitarian rather than economic; however, there is some evidence that the co-responding police-mental health programs are also cost effective (Scott 2000). In 1989, Finn and Sullivan reviewed eight law enforcement agencies and social service systems arrangements and concluded that though funding is an obstacle for these programs in the outset, ‘‘those holding the purse strings feel the arrangements are sufficiently beneficial to have funded them without serious hesitation for many years’’ (p. 13). The perceived effectiveness discussed in this study has since been supplemented by more compelling evidence. According to the supervisor in San Diego County’s second district, their coresponding police-mental health program potentially saves more than $2,000,000 a year in the western division of the San Diego Police Department through decreasing jail costs and officer savings by as much as $2,200 per contact (Cobb 1997 as cited in Rosenbaum 2010). In addition, another study found the co-responding police-mental health program to be less costly compared to a ‘service as usual’ control; however, the authors note that this finding must be interpreted in the context of the assumptions that were made to address data limitations (Allen Consulting Group 2012). A further study, by Scott (2000), found that on average each person served by the co-responding policemental health program costs 23 % less, as measured
Adm Policy Ment Health
through psychiatric hospitalization savings. However, as Hubbeling and Bertram (2012) point out, this study did not report statistical testing or confidence intervals. Furthermore, a cost analysis conducted by Baess (2005) found that a police call costs $300 per call while the co-responding police-mental health program call costs $190 due to a reduction of hospital wait times and admissions. Program Perception A number of researchers have investigated the perception of, and satisfaction with, the co-responding police-mental health programs by staff, consumers, and community stakeholders. For example, stakeholder consultations of a co-responding police-mental health program in Victoria (Australia) reported that ‘‘among the stakeholders consulted in this project [i.e. the Area Mental Health Service, Victoria Police, and Ambulance Victoria], there is widespread support for improved inter-agency collaboration in responding and managing mental health crises in the community’’ (Allen Consulting Group 2012, p.vii). More specifically, Saunders and Marchik (2007) surveyed 210 law enforcement patrol officers to assess their experiences with, and benefit from, the co-responding police-mental health program. They found that 86 % rated the program as either ‘very good’ or ‘superior’. In particular, officers’ valued ‘being able to return to the street faster’ and the ‘benefit to clients assessed for MH [mental health] issues’. Officers did suggest, however, that improvements to the service be made by increasing availability and efficiency, providing county officers with greater training, and reducing response times (Saunders and Marchik 2007). Similarly, officers in Framingham and Quincy (Massachusetts) reportedly valued their co-responding policemental health program (Abbott 2011). Other studies, however, have found low satisfaction of the co-responding police-mental health program among officers. In Knoxville, officers rated their preparedness for EDP calls to be high (78.1 %); however, only 52.7 % of officers rated their program as ‘moderately’ or ‘very effective’ in meeting the needs of people with mental illness in crisis, only 51.9 % of officers rated their program as effective in maintaining community safety, and only 7.3 % of officers rated their program effective in minimizing the amount of time patrol officers spent on mental health calls (Borum et al. 1998). Furthermore, though Steadman et al. (2000) found the co-responding police-mental health program in Knoxville to have a low arrest rate (5 %), the program was not perceived to be as effective as other programs in reducing arrests. Accordingly, while 41.8 % of officers at Knoxville reported their program was effective in reducing arrests, 47.9 % of officers in Birmingham and 70.1 % of officers in Memphis believed their program was
effective at reducing arrests (Borum et al. 1998). Similarly, Ratansi (2004) found that all officer groups reported that responding to calls with mental health workers only slightly improved the effectiveness of EDP response. However, Ratansi (2004) also found that officers with the highest number of training hours had more positive attitudes towards being prepared and able to handle EDPs. Consumer and community member perceptions of the co-responding police-mental health program have also been examined. Baess (2005) reported positive feedback from service providers and families. Similarly, Ligon and Thyer (2000) found that both consumers and family members rated the co-responding police-mental health program in DeKalb County, Georgia, favourably. Interestingly, in one study, family members of consumers reported higher level of satisfaction with the service on a Client Satisfaction Questionnaire compared to consumers (Ligon and Thyer 2000). However, no demographic information of the family member and consumer groups was collected, and, as the authors themselves note, the sample size in their study was small (29 family members and 15 consumers), and it is therefore difficult to know if their sample is representative of clients and families in the service area (Ligon 1997; Ligon and Thyer 2000). Furthermore, in an ethnographic case-study that included group interviews and participant observation, Forchuk et al. (2010) contrasted crisis service models in three Ontario communities. They found that all three communities valued their crisis services. However, a number of modifications were suggested including: wanting peer support as part of crisis care; a ‘warm line’ that provides support in addition to a ‘hotline’ that provides immediate attention; including consumers and their families in the collaborative framework of the crisis response; and enabling family members to receive crisis support (Forchuk et al. 2010). In Halifax, Kisely et al. (2010) found participants report their co-responding police-mental health program to be helpful in providing someone to talk to, obtaining advice, and facilitating referral. As an example, one participant wrote that ‘‘they [the co-responding Halifax crisis team] got me through the winter of 2007’’ (Kisely et al. 2010, p. 665). However, participants suggested that the co-responding police-mental health program could be improved by an expansion of service, greater availability, and a more timely response (Kisely et al. 2010). Hypothetical Mechanisms Underlying Program Outcomes We initially sought to conduct a realist synthesis of the coresponding model literature (Kirst and O’Campo, 2012; Pawson 2006), that is to review how, why, for whom, and in what circumstances the programs work. However, the
123
Adm Policy Ment Health
Fig. 1 Mechanisms involved in reducing the number of arrests
evidence concerning program mechanisms—that is the critical ingredients of the program that is responsible for the success or in some cases for the failure of the program—contained in the existing literature was reported in too few articles to undertake a complete realist review. However, in reviewing the literature we were able to generate some initial propositions about the mechanisms of successful co-responding programs. Based on the literature reviewed above, we identified two potential mechanisms underlying key intended outcomes of co-responding policemental health programs, namely diversion of EDPs from the criminal justice and hospital systems, and connecting EDPs to community resources (Lamb et al. 1995; Steadman et al. 2000). Focusing on reducing the number of arrests, it is hypothesized that providing adequate mental health training for police officers (Baess 2005; Borum et al. 1998; Forchuk et al. 2010), strong partnerships between police and mental health program partners, and support from within organizations to guide implementation of program policies and activities (organizational buy-in) (Forchuk et al. 2010; Rosenbaum 2010) are program components that provide opportunities and resources to raise awareness of mental health issues among police officers. These aspects enhance police officers’ ability to recognize when a mental health issue is a factor and that engaging the coresponse program is an appropriate course of action rather than arrest. These mechanisms subsequently reduce the
123
need for officers to arrest individuals in crisis and facilitate diversion of these cases away from the criminal justice system (see Fig. 1). A second potential mechanism explains the key program outcome of diversion from hospitalization to health and social services. This hypothetical mechanism posits that programs with strong partnerships and collaboration across police and mental health system sectors facilitate the creation of strong partnerships with community agencies (Rosenbaum 2010; Steadman et al. 2000). Strong program partnerships with community agencies increase police and community awareness of the co-response program as a resource, as well as facilitating ease of linkage and referral of consumers to health and social services by co-response programs (Borum et al. 1998; Wood et al. 2011). These mechanisms increase utilization of the program and the treating of people in crisis in the community, which shifts practice norms to referral of consumers to community services rather than hospitalization (see Fig. 2).
Discussion This review firstly sought to synthesize the literature on seven desired outcomes of the co-responding police-mental health programs. Accordingly, evaluations of the coresponding police-mental health programs found that there is some evidence that this program model has the potential
Adm Policy Ment Health
Fig. 2 Mechanisms involved in increasing diversion from hospitalization to health and social services
to forge linkages with community services and mitigate the burden on the justice system (i.e. through reducing arrest rates and on-site handling time). Reducing ‘down time’ of officers is not only important for efficiency but also for police morale, as many officers find waiting in emergency departments to be an ineffective use of their time (Borum et al. 1998). The perception of the programs, particularly among consumers and their family members, was also found to be generally positive. Indeed, a recent study asked mental health consumers in Melbourne (Australia) about four different program models and found the co-responding police-mental health program (the so-called ‘ride along model’) to be the desired formal response (Boscarato et al. 2014). However, paradoxically, despite evidence that the co-responding police-mental health program has a low arrest rate and lower on-site handling time, many surveyed officers did not believe that their co-responding policemental health program reduces the time officers spend on calls, lowers the arrest rate, or improves the effectiveness of EDP calls (Borum et al. 1998, Ratansi 2004). Collecting program process and outcome data on a regular basis and improved sharing of data with program team members is important for gaining officers’ support for, and appreciation towards, co-responding police-mental health programs (Allen Consulting Group 2012; Hollander et al. 2012). In contrast, there have been limited evaluations (and therefore limited evidence) regarding whether coresponding police-mental health programs avert crisis
escalations and injuries, improve officers’ perception of individuals who have a mental illness, or are cost-effective. There is also mixed evidence regarding whether the coresponding police-mental health programs reduce the number of hospital admissions. The expectation that the coresponding police-mental health program reduces hospitalizations should therefore be low. However, it is not necessarily a shortcoming if individuals in crises are directed to hospital services when hospital admission is an appropriate response. It has also been argued that even if the co-responding police-mental health programs do not reduce the number of hospital admissions at all sites (as some studies have shown), emergency care staff will still spend less time evaluating, treating, and transferring inappropriate referrals because EDPs have been prescreened and assessed in the community by co-responding police-mental health programs’ healthcare workers (Allen Consulting Group 2012; Finn and Sullivan 1989). Given the increasing attention towards the management of mental health crises in the community, it is important to better understand whether the co-responding police-mental health programs are achieving their desired outcomes, and what might be the key ingredients of program success. As co-responding programs differ in composition and activities, and operate in diverse contexts, examining the key elements and mechanisms involved in program success can inform program design and resource allocation. Two potential mechanisms, thought to underlie intended
123
Adm Policy Ment Health
outcomes, were presented, and particular program activities thought to underlie the achievement of specific successful outcomes were hypothesized. Namely, it was hypothesized that co-responding police-mental health programs are likely to achieve outcomes such as reduced arrests and increased diversion away from hospital to community-based health and social services when they have strong organizational buy-in from all partners, good collaboration between police and health partners, provide adequate mental health training for police officers, engage community partners, and face fewer barriers when linking individuals to referral services. An enhanced understanding of mental health issues among police officers and increased awareness of the co-responding model may lead to improved ability to recognize when to utilize the program when addressing EDPs in the community, emerged as mechanisms in diverting these individuals from the criminal justice system by reducing the number of arrests. A review of literature on the police-based (CIT) response model has also highlighted similar key program elements to facilitate program success including: collaborative planning and implementation by partnering organizations, organizational support, specialized training for law enforcement personnel, and effective referral of consumers to community supports and services (Reuland, 2010). Another promising program element in police-based (CIT) response models that may minimize officer down time and mitigate arrests involves the use of psychiatric drop-off centers (particularly ones with a no refusal policy for police cases and streamlined intake) (Borum et al. 1998, p. 403; Steadman et al. 2000, 2001).
(Iacobucci 2014). Increasingly, the co-responding policemental health program is seen as potentially one of several components of an adequate crisis response system for people facing a mental health crisis, including warm and hot lines, highly trained dispatchers, Crisis Intervention Teams (CIT), and mental health crisis centers with streamlined intake processes. Regarding program implementation, there are several challenges highlighted in the literature, including establishing a partnership (rather than ‘ownership’) between two very different organizational cultures and perspectives, the scarcity of police drop-off centers, working collaboratively while upholding confidentiality, and forming appropriate organizational structures to support joint police mental health program operations (Bar-On 1995; Kirst et al. 2014; Patterson 2010; Steadman et al. 2000). Finn and Sullivan (1989) reviewed eight arrangements between law enforcement agencies and the social service systems and found that there have been problems initiating and sustaining these networks in some settings. Furthermore, police officers begin with relatively little formal knowledge and training about mental health problems and mental health professionals begin with relatively little awareness of law enforcement procedures and policies (Lamb et al. 1995), necessitating extensive and ongoing training for both groups. Despite the challenges and limitations of the program model, this literature synthesis points to several strengths that could potentially be built upon in efforts to design a system of care that safeguards both individual and community safety as well as timely and appropriate connection to mental health services and supports.
Challenges and Limitations of the Program Model Recommendations for Practitioners Glasgow et al. (1999) note that public health interventions should be evaluated not only in terms of their effectiveness, but also their reach, acceptability, implementation, and maintenance. Although the co-responding police-mental health program enjoys acceptability and has been implemented in different contexts and jurisdictions, there is a dearth of information on program sustainability and drift over time, and valid concerns about program reach. Indeed, as a result of the lack of clarity about the programs’ role among community partners, lengthy response times (Borum et al. 1998) and lack of capacity (Ratansi 2004), this program model responds to a relatively small proportion of EDP calls in some settings and this raises concerns about the programs’ added value. For example, the co-responding police-mental health program in Knoxville (Tennessee) only responded to 40 % of mental disturbance calls (Steadman et al. 2000) while the co-responding policemental health program in Toronto (Ontario) only reportedly responded to 11 % of mental disturbance calls
123
To support optimal care, avoid criminalization, and improve mental health consumer experience, programs should consider a number of factors impacting program delivery. First, it is important to consider how best to balance standardization of service delivery in a geographical area with optimal fit to the local context. Second, teams should consider adopting common outcome measures and shared definitions to facilitate routine outcome reporting and quality improvement efforts. Third, attention should be given to the ongoing training needs of all police officers, and mental health providers, and evidence supported standards for training and supervision should be established. Last, but most important, for the model to leverage its full potential, joint stewardship may be needed to bridge the two disparate cultures and perspectives and address each system’s limitation in managing crises and safety in the most acceptable way. In this regard, engaging stakeholders in discussions about the divergent tasks and
Adm Policy Ment Health
roles of the co-responding police mental health program workers is imperative during the development and evolution of the program (Patterson 2010). Limitations and Future Research The literature reviewed has a number of methodological limitations, including few published studies, non-experimental designs, and small sample sizes (Wood et al. 2011). For example, examining officer’s perception (or stigma) towards individuals who have a mental illness is challenging. Nevertheless, it is an important measure of the quality of policing, and future studies should evaluate attitude changes before and after the implementation of a co-responding police mental health program (rather than retrospectively surveying how introducing the program has impacted officers’ perceptions). Furthermore, there are few longitudinal studies that investigate the long-term impact that co-responding police-mental health programs have on EDPs such as continuity of care, arrests, or hospitalizations. For example, though the research indicates that the co-responding police-mental health program reduces arrest rates in the short-term, it is unclear whether the program reduces arrests in the long-term (Lamb et al. 1995). Future longitudinal evaluations should examine the long-term impact of the program on arrest rates, hospitalization, and service engagement. Another difficulty with the available literature is that a number of studies included in this review are dated or not generalizable, as they draw from a specific local context (e.g. Lamb et al. 1995). Given dynamic shifts in service contexts and program model drift over time, it is necessary to pursue high quality research to address fundamental questions of program effectiveness and consumer experiences with the program model. In particular, it is important to identify the key ingredients of the program as well as how the program works, for whom, and in what circumstances. Program drift over time creates an even greater need to understand key elements to enable local adaptations to optimize fit in different service contexts. However, there remains a dearth of studies that investigate contextual differences (Steadman et al. 1999). It would be valuable for controlled research to test the mechanisms that were presented in this review. As Pawson (2006) explains, to ascertain the evidence-base of policy it is important for research to understand causation (i.e. how a program achieve their effect), ontology (i.e. how a program works), and generalization (i.e. how we can inform future policy and practice). It would also be beneficial for future research to expand their scope of analysis. It is increasingly important to expand on ‘convenience data’ to also include variables that answer pertinent research questions on outcomes and
mechanisms. In order to do so, it would be necessary to not only examine questions of efficacy, but it is also important to assess other outcomes including program reach, adoption (i.e. the representativeness of the setting), implementation (i.e. the extent that a program is delivered as intended), and maintenance (i.e. of program-level measures and policies over time) (Glasgow et al. 1999). In order to address these outcomes it would be helpful to link hospital and justice data, while still protecting individuals’ anonymity. The extant literature does not consistently report on the presence or absence of model elements in coresponding police-mental health programs. There is a substantial need for future research to understand coresponding police-mental health programs’ mechanisms and outcomes in order to better inform practice and policy. In so doing, research that has examined the key model elements and common obstacles of the Crisis Intervention Team (CIT) model may serve as an informative template (see Compton et al. 2010; Cross et al. 2014; Dupont et al. 2007; McGuire & Bond 2011; Reuland 2004). It is becoming increasingly clear that improving the evidence base of the field, and program effectiveness, requires collaborations between police, health agencies, and academics (Dean et al. 2000; Matheson et al. 2005; Wood et al. 2011).
Conclusion This review synthesized literature on co-responding policemental health programs and identified potential mechanisms that may account for their success. Few quality studies exist, pointing to the need for future research to establish program outcomes and underlying mechanisms in diverse contexts. References Abbott, S.E. (2011). Evaluating the impact of a jail diversion program on police officer’s attitudes toward the mentally ill. (Doctoral dissertation). Retrieved from ProQuest Dissertations & Theses Full Text database (3450400). Adelman, J. (2003). Study in blue and grey. Police interventions with people with mental illness: A review of challenges. Vancouver: Canadian Mental Health Association, BC Division. Allen Consulting Group. (2012). Police, Ambulance and Clinical Early Response (PACER) evaluation: Final report. Melbourne: Department of Health, Victoria. Anderson, B., & Taylor, D. (2013). Surrey RCMP: Business case Car 67 program expansion. Personal communication with Bruce Anderson, Staff Sergeant, Surrey. Received 7/7/2013. Baess, E.P. (2005). Integrated Mobile Crisis Response Team (IMCRT): Review of pairing police with mental health outreach services. Vancouver Island Health Authority. Retrieved from http://www.pmhl.ca/webpages/reports/Pairing-report.pdf. Accessed 31 July 2014.
123
Adm Policy Ment Health Bar-On, A. (1995). They have their job, we have ours: Reassessing the feasibility of police-social work cooperation. Policing and Society: An International Journal of Research and Policy, 5, 37–51. Borum, R. (2000). Improving high risk encounters between people with mental illness and the police. Journal of the American Academy of Psychiatry and the Law, 28, 332–337. Borum, R., Deane, M. W., Steadman, H. J., & Morrissey, J. (1998). Police perspectives on responding to mentally ill people in crisis: Perceptions of program effectiveness. Behavioral Sciences & the Law, 16(4), 393–405. Boscarato, K., Lee, S., Kroschel, J., Hollander, Y., Brennan, A., & Warran, N. (2014). Consumer experience of formal crisisresponse services and preferred methods of crisis. International Journal of Mental Health Nursing, 23, 287–295. Brown, N.E., Hagen, C., Meyers, J., & Sawin, J. (2009). Report on comprehensive study of mental health delivery systems in Iowa. Retrieved from: http://www.nami.org/Content/Microsites230/ NAMI_Linn_County/Home218/Reports_Mental_Health_Care_ in_Iowa/League_of_Women_Voters_Study/lwvstudy.pdf. Accessed 31 July 2014. Canadian Police and Mental Health Liaison Committee. (2008). National list of Canadian crisis response services. Retrieved from: http://www.pmhl.ca/webpages/reports/CANADIAN%20 CRISIS%20RESPONSE.pdf. Accessed 31 July 2014. City of Toronto Mobile Crisis Intervention Team Coordination Steering Committee. (2013). MCIT program coordination in the city of Toronto. Submitted to the Toronto Central LHIN. Personal communication. Received 4/20/2013. Coleman, T. G., & Cotton, D. (2010). Police interactions with persons with a mental illness: Police learning in the environment of contemporary policing. Calgary: Prepared for the Mental Health and the Law Advisory Committee, Mental Health Commission of Canada. Compton, M. T., Broussard, B., Hankerson-Dyson, D., Krishan, S., & Stewart, T. (2010). System- and policy-level challenges to full implementation of the Crisis Intervention Team (CIT) model. Journal of Police Crisis Negotiations, 10, 72–85. Cotton, D. & Coleman, T.G. (2006). Contemporary policing guidelines for working with the mental health system. Prepared by the Police/Mental Health Subcommittee of the Canadian Association of Chiefs of Police (CACP) Human Resources Committee. Retrieved from: http://www.pmhl.ca/webpages/reports/Guide lines%20for%20Police.pdf. Accessed 31 July 2014. Cross, A. B., Mulvey, E. P., Schubert, C. A., Griffin, P. A., Filone, S., Winckworth-Prejsnar, K., et al. (2014). An agenda for advancing research on Crisis Intervention Teams for mental health emergencies. Psychiatric Services, 65(4), 530–536. Dean, C.W., Lamb, R., Proctor, K., Klopovic, J., Hyatt, A., & Hamby, R. (2000). Social work and police partnership: A summons to the village strategies and effective practices. North Carolina Governor’s Crime Commission. Criminal Justice Analysis Centre, United States of America. Retrieved from: https://www.ncjrs.gov/App/ Publications/abstract.aspx?ID=203853. Accessed 31 July 2014. Deane, M. W., Steadman, H. J., Borum, R., Veysey, B. M., & Morrissey, J. P. (1999). Emerging partnerships between mental health and law enforcement. Psychiatric services, 50(1), 99–101. Dupont, R., Cochrane, S., & Pillsbury, S. (2007). Crisis Intervention Team core elements. The University of Memphis School of Urban Affairs and Public Policy, Dept. of Criminology and Criminal Justice, CIT Center. Crisis Intervention Team website. Retrieved from: http://www.cit.memphis.edu/information_files/ CoreElements.pdf. Accessed 31 July 2014. Finn, P., & Sullivan, M. (1989). Police handling of the mentally ill: Sharing responsibility with the mental health system. Journal of Criminal Justice, 17(1), 1–14.
123
Fisher, W. H., Silver, E., & Wolff, N. (2006). Beyond criminalization: Toward a criminologically informed framework for mental health policy and services research. Administration and Policy in Mental Health and Mental Health Services Research, 33, 544–557. Forchuk, C., Jensen, E., Martin, M. L., Csiernik, R., & Atyeo, H. (2010). Psychiatric crisis services in three communities. Canadian Journal of Community Mental Health, 29, 73–86. Fry, A. J., O’ Riordan, D. P., & Geanellos, R. (2002). Social control agents or front-line carers for people with mental health problems: Police and mental health services in Sydney Australia. Health and Social Cray in the Community, 10(4), 277–286. Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American Journal of Public Health, 89, 1322–1327. Hails, J., & Borum, R. (2003). Police training and specialized approaches to respond to people with mental illnesses. Crime & Delinquency, 49(1), 52–61. Hollander, Y., Lee, S. J., Tahtalian, S., Young, D., & Kulkarni, J. (2012). Challenges relating to the interface between crisis mental health clinicians and police when engaging with people with a mental illness. Psychiatry, Psychology and Law, 19(3), 402–411. Hubbeling, D., & Bertram, R. (2012). Crisis resolution teams in the UK and elsewhere. Journal of Mental Health, 21(3), 285–295. Iacobucci, F. (2014). Police encounters with people in crisis: An independent review conducted by the Honourable Frank Iacobucci for Chief of Police William Blair, Toronto Police Service. Retrieved from: http://www.tpsreview.ca/docs/Police-Encoun ters-With-People-In-Crisis.pdf. Accessed 31 July 2014. Kean, R., Bornstein, S., & Mackey, S. (2012). Mobile mental health crisis intervention in the western health region of Newfoundland and Labrador. Rapid Evidence Reports. Newfoundland & Labrador Centre for Applied Health Research. Retrieved from: http://www.nlcahr.mun.ca/CHRSP/RER_MH_Crisis_Intervention. pdf. Accessed 31 July 2014. Kirst, M., Narrandes, R., Francombe Pridham, K., Yogalingam, J., Matheson, F., & Stergiopoulos, V. (2014). Toronto Mobile Crisis Intervention Team (MCIT) program implementation evaluation final report. Centre for Research on Inner City Health, St. Michael’s Hospital. Retrieved from: http://www.stmichaelshos pital.com/crich/wp-content/uploads/MCIT-evaluation-report-finalApril30-2014.pdf. Accessed 31 July 2014. Kirst, M., & O’Campo, P. (2012). Realist review methods for complex health problems. In P. O’Campo & J. Dunn (Eds.), Rethinking social epidemiology: Towards a science of change. New York: Springer. Kisely, S., Campbell, L. A., Peddle, S., Hare, S., Pyche, M., Spicer, D., et al. (2010). A controlled before-and-after evaluation of a mobile crisis partnership between mental health and police services in Nova Scotia. Canadian Journal of Psychiatry, 55(10), 662–668. Laing, R., Halsey, R., Donohue, D., Newman, C., & Cashin, A. (2009). Application of a model for the development of a mental health service delivery collaboration between police and the health service. Issues in Mental Health Nursing, 30(5), 337–341. Lamb, H. R., Shaner, R., Elliot, D. M., DeCuir, W. J, Jr, & Foltz, J. T. (1995). Outcome for psychiatric emergency patients seen by an outreach police-mental health team. Psychiatric Services, 46(12), 1267–1271. Lamb, H. R., Weinberger, Linda E., DeCuir, W. J, Jr, & Walter, J. (2002). The police and mental health. Psychiatric Services, 53(10), 1266–1271. Landeen, J., Pawlick, J., Rolfe, S., Cottee, I., & Holmes, M. (2004). Delineating the population served by a mobile crisis team: Organizing diversity. Canadian Journal of Psychiatry, 49(1), 45–50.
Adm Policy Ment Health Ligon, J. (1997). Crisis psychiatric and substance abuse services: Evaluation of a community program in an urban setting. (Doctoral dissertation). Retrieved from ProQuest Dissertations & Theses Full Text database (9726961). Ligon, J., & Thyer, B. A. (2000). Client and family satisfaction with brief community mental health, substance abuse, and mobile crisis services in an urban setting. Crisis Intervention and TimeLimited Treatment, 6(2), 93–99. Matheson, F. I., Creatore, M. I., Gozdyra, P., Moineddin, R., Rourke, S. B., & Glazier, R. H. (2005). Assessment of police calls for suicidal behavior in a concentrated urban setting. Psychiatric Services, 56(12), 1606–1609. McGuire, A. B., & Bond, G. R. (2011). Critical elements of the Crisis Intervention Team model of jail diversion: An expert survey. Behavioral Sciences and the Law, 29, 81–94. Mental Health Commission. (2009). Report of joint working group on mental health services and the police 2009. Retrieved from: http://www.garda.ie/Documents/User/report%20of%20joint% 20working%20group%20on%20mental%20health%20services% 20and%20the%20police%20%202009%20(mental%20health% 20commisson%20an%20garda%20siochana).pdf. Accessed 31 July 2014. Patterson, G. T. (2010). Police-social work crisis teams: Practice and research implications. Stress, Trauma, and Crisis: An International Journal, 7, 93–104. Pawson, R. (2006). Evidence-based policy: A realist perspective. London: Sage Publications. Peterson, J. K., Skeem, J., Kennealy, P., Bray, B., & Zvonkovic, A. (2014). How often and how consistently do symptoms directly precede criminal behavior among offenders with mental illness?. Law and Human Behavior: Advance online publication. Ratansi, S. (2004). Specialized response programs: Police handling of encounters involving persons with mental disorders. Retrieved from ProQuest Dissertations & Theses Full Text database (3168762). Reuland, M. (2004). A guide to implementing police-based diversion programs for people with mental illness. Delmar, NY: Technical Assistance and Policy Analysis Center for Jail Diversion. Reuland, M. (2010). Tailoring the police response to people with mental illness to community characteristics in the USA. Police Practice and Research, 11(4), 315–329. Rosenbaum, N. (2010). Street-level psychiatry-a psychiatrist’s role with the Albuquerque police department’s crisis outreach and support team. Journal of Police Crisis Negotiations, 10(1), 175–181.
Saunders, J. A., & Marchik, B. M. A. (2007). Building community capacity to help persons with mental illness: A program evaluation. Journal of Community Practice, 15(4), 73–96. Scott, R. L. (2000). Evaluation of a mobile crisis program: Effectiveness, efficiency, and consumer satisfaction. Psychiatric Services, 51(9), 1153–1156. Steadman, H. J., Deane, M. W., Borum, R., & Morrissey, J. P. (2000). Comparing outcomes of major models of police responses to mental health emergencies. Psychiatric Services, 51(5), 645–649. Steadman, H. J., Deane, M. W., Morrissey, J. P., Westcott, M. L., Salasin, S., & Shapiro, S. (1999). A SAMHSA research initiative assessing the effectiveness of jail diversion programs for mentally ill persons. Psychiatric Services, 50(12), 1620–1623. Steadman, H. J., Stainbrook, K. A., Griffin, P., Draine, J., Dupont, R., & Horey, C. (2001). A specialized crisis response site as a core element of police-based diversion programs. Psychiatric Services, 52, 219–222. Teplin, L. A., & Pruett, N. S. (1992). Police as streetcorner psychiatrists: Managing the mentally ill. International Journal of Law and Psychiatry, 15, 139–156. Thompson, S. (2010). Policing Vancouver’s mentally ill: The disturbing truth beyond lost in transition. Vancouver Police Department. Retrieved from: http://vancouver.ca/police/assets/ pdf/reports-policies/vpd-lost-in-transition-part-2-draft.pdf. Accessed 31 July 2014. Watson, A. C., & Fulambarker, A. J. (2013). The Crisis Intervention Team model of police response to mental health crises: A primer for mental health practitioners. Best Pract Ment Health, 8(2), 71. Wilson-Bates, F. (2008). Lost in transition: How a lack of capacity in the mental health system is failing Vancouver’s mentally ill and draining policing resources. Vancouver. Retrieved from: http:// vancouver.ca/police/assets/pdf/reports-policies/vpd-lost-in-transi tion.pdf. Accessed 31 July 2014. Wood, J., Swanson, J., Burris, S., & Gilbert, A. (2011). Police interventions with persons affected by mental illnesses: A critical review of global thinking and practice. Center for Behavioral Health Services Criminal Justice Research, Rutgers, The State University of New Jersey. Retrieved from: https:// www.temple.edu/cj/people/documents/Police_Interventions_ Monograph_March_2011.pdf. Accessed 31 July 2014. Zealberg, J. J., Christie, S. D., Puckett, J. A., McAlhany, D., & Durban, M. (1992). A mobile crisis program: Collaboration between emergency psychiatric services and police. Hospital & Community Psychiatry, 43(6), 612–615.
123
