Evidence for joint police and mental health responses for people in mental health crisis
Evidence for joint police and mental health responses for people in mental health crisis
Andrew Lancaster Community mental health nurse
NHS Tayside Community Mental Health Team, Perth, Scotland
Mental Health Practice. 19, 10, 20-26. doi: 10.7748/mhp.2016.e1067
Received on 03 March 2015
Accepted on 08 December 2015
Published in print 07 July 2016
This article has been subject to double-blind review and has been checked using automated software
In the UK, police officers who respond to people in mental health crisis must judge whether or not the people they are dealing with require assessment by mental health professionals. However, many police staff lack the skills and knowledge to make such decisions, which can lead to inappropriate use of mental health legislation or arrests. This article reviews the evidence on mobile crisis teams, which deliver a joint response from the police and mental health services to people in mental health crisis in the community. It describes how these teams can improve outcomes for people in mental health crisis, reduce inappropriate use of mental health legislation and hospital admissions, increase user engagement, strengthen relationships between the police and health services, and potentially reduce costs to public services.
IN THE UK, the police are increasingly the first responders to people in the community experiencing a mental health crisis (Bather et al 2008). A report completed by the Sainsbury Centre for Mental Health (Bather et al 2008) suggests up to 15% of incidents that the UK police respond to have a mental health aspect. Crises in the community have increased in recent years, partly because of the UK’s mental health laws, which state that people should be cared for in the least restrictive environment (Scottish Government 2003, Department of Health (DH) 2007), and partly because of the reduction in hospital beds because of constraints on health boards and trusts (Green and Griffiths 2014) and the closure of the former institutions (Green and Griffiths 2014). However, many police officers believe they do not have the skills or resources to respond effectively to such crises (McLean and Marshall 2010, Bradley 2009).
Police officers frequently refer the people they encounter to crisis resolution home treatment teams (Onyett et al 2006), which offer crisis assessments in the community (DH 2001). However, the structure of these teams, and the services they deliver, vary from area to area (Onyett et al 2006), and in most regions it is not standard practice for the police and crisis resolution home treatment teams to undertake joint assessments (Onyett et al 2006). Instead, the police are expected to transport people to an appropriate place of safety, usually a hospital but often a police station, for assessment (Bather et al 2008, Bradley 2009, Care Quality Commission (CQC) 2013).
Various approaches to joint assessments by the police and mental health professionals have been developed in the UK and elsewhere, to overcome some of these issues (Wood et al 2011). This article examines the evidence on mobile crisis teams (MCTs), which provide joint responses by the police and mental health services.
Overview of evidence
A search of Medline, PsychInfo and Cinahl highlighted a dearth of information about the development and effectiveness of MCTs in the UK, mainly because few of these services exist, and there is little high quality literature on those that do. The majority of UK research on the topic is qualitative, and the remainder of the evidence consists of practice literature, policies, government reports and anecdotes.
Most well-established MCTs are in the United States, and several models have been developed since the early 1990s (Steadman et al 2000, Erstling 2006, Morrissey et al 2009, but MCTs are also in operation in Canada (Kisley et al 2010) and Australia (Allen Consulting Group 2012). US models include specially trained police officers who respond to people in crisis, civilian police employees with specialist training, and mobile crisis response by the police and mental health services (Steadman et al 2000).
Although there are differences between the UK’s and other countries’ MCTs in terms of operating times and staffing structure (Table 1), they have similar procedures for responding to people in crisis (Scott 2000, Steadman et al 2000, Kisley 2010, Dean 2013).
When an emergency services call is received about a person in the community who might be experiencing mental distress or is displaying disturbing behaviour, some MCTs allocate a mental health professional to work directly with the police (Wood et al 2011), while in others the police contact a local crisis team and, if required, request a joint assessment accompanied by police officers (Wood et al 2011). MCTs provide a way of assessing people before there has been an arrest, and their broad aims are to reduce arrests and the inappropriate use of mental health legislation (Wood et al 2011).
Evaluation of the evidence supporting MCTs highlighted several themes, which are discussed below.
Availability and accessibility
Access to, and availability of, support for people with mental health needs at the pre-arrest phase requires development in the UK and abroad (The Council of State Governments 2002, Bradley 2009). Options for police support for people experiencing mental health crisis in the community are limited (Bradley 2009), and MCTs are one of several interventions that can provide enhanced support at the pre-arrest stage.
Kisley et al (2010) compared two areas in Nova Scotia, with and without access to a new MCT called the Crisis Outreach and Support Team (COAST). The police, emergency health and mental health services collaborated to form a team, which operates 24 hours a day, staffed by mental health professionals and the police. The mental health professionals triage incoming calls from a 24-hour hotline, and if a community assessment is required, plainclothes police officers accompany clinical staff.
The researchers measured the effect of the COAST on the availability of and access to services using referral rates and the number of crisis responses over a two-year period (Kisley et al 2010). Results show a dramatic increase in the number of contacts with the MCT, from 2,783 before its development to 7,558 in the second year. However, 92% of these contacts were by phone and did not result in a crisis response. The number of mobile crisis responses more than trebled over the same time period, from 162 to 613 in the second year.
An MCT in Australia, called the Police, Ambulance and Clinical Response (PACER) service, provided on-site assessments in a much higher percentage of cases (78%) (Allen Consulting Group 2012) than in the COAST study. However, it also received far fewer contacts (n=783) in a shorter evaluation period of 18 months. The PACER model is also different from the COAST model, in that it only operates for eight hours a day (Table 1).
The availability of the COAST improved access to services, highlighted by the number of referrals and subsequent interventions (Kisley et al 2010). However, as the study was undertaken in one geographical area, and examined only one service model and referral process, results might not be able to be generalised.
Some of the benefits of MCTs, in terms of collaborative working, include improved communication, sharing of information, and availability of mental health professionals to advise and assist police officers (Dean 2013). Lack of collaboration can lead to arrests resulting in criminal records, and inappropriate use of mental health legislation (Bather et al 2008).
One study identified barriers to effective collaboration when police officers attempt to contact mental health services for support and assessment of people in crisis (McLean and Marshall 2010). For example, some of the police officers interviewed in the McLean and Marshall study said they felt they were ‘filling a gap in services’, were viewed as ‘a problem… bringing a problem’. Overall participants were critical of current models of care.
McLean and Marshall’s (2010) findings correlate with concerns about current models of collaboration between the police and mental health services highlighted internationally and in the UK (The Council of State Governments 2002, Bradley 2009). More recently, Martin and Thomas (2013) reported that officers in five police stations in Victoria, Australia, felt frustrated and angry at the perceived lack of support from mental health services when referring people for assessment, and said they had to ‘shop around’ to find the care people required.
One of the goals of an MCT is to overcome some of the issues that arise when the police and mental health services attempt to work together (Scott 2000, Kisely et al 2010, Dean 2013). Scott (2000) noted that because MCTs provide joint responses by different agencies, success relies on those involved feeling satisfied with the working relationship. Scott (2000) looked at whether a new MCT in Georgia, US, called the Dekalb MCT, contributed to a more effective working relationship between the police and healthcare professionals, and reported that about 75% of police officers said they were ‘very satisfied’ with the performance and working relationship. Kisely et al (2010) also reported that police officers and MCT staff spoke positively about the collaboration, however some officers were critical about the fact that the MCT was sometimes unavailable to take their calls.
Comments from police officers in the Allen Consulting Group (2012) report were positive and many were keen to expand the trial. Police officers were particularly confident that the MCT could contribute to strengthening relationships between themselves and mental health services.
Mental health services are responsible for providing a comprehensive bio-psycho-social assessment of people’s needs (DH 2001, 2004), and mental health teams must also assess and manage risk (DH 2007, Scottish Government 2013).
Violent incidents between the police and people with mental health needs have been well documented over the years, and charities such as Rethink, Mind and Amnesty International, have condemned the use of methods such as using a Taser to de-escalate situations (Corrigan 2008, Mercer 2012).
Police officers who respond to emergency calls involving people with mental health needs assess the risk to self or others and decide on an appropriate outcome (Bradley 2009). However, McLean and Marshall (2010) found that some police officers believed the risk is heightened simply by their presence. To address this issue, most MCTs pair psychiatric nurses with police officers, and respond in unmarked police cars while wearing plain clothes instead of uniforms (Dean 2013), however there is a lack of evidence comparing standard police responses to MCT approaches.
Scott (2000) recorded the number of violent crises that one MCT responded to and the number of arrests, but did not clarify whether these numbers reduced in the intervention group because of the MCT’s response.
Allen Consulting Group (2012) was also unsure whether risk managed by the MCT was directly related to fewer instances of violence, therefore further research is needed. The effect of MCT intervention on suicide rates is not detailed in any of the literature reviewed, and also requires investigation.
When police officers encounter people experiencing mental distress in the community when no crime has been committed, it is their responsibility to direct the person to, or liaise with, local providers for assessment or treatment (Bradley 2009). However, McLean and Marshall (2010) suggest that police in the UK frequently feel unsupported by mental health services, and lack the knowledge and confidence to support people in crisis, which can result in inappropriate use of the Mental Health Act.
Martin and Thomas (2013) found that police officers in Australia misunderstood mental health legislation and falsely believed that people with certain conditions, specifically borderline personality disorder, were excluded from being treated under the current legislation. This confusion has been addressed, mainly by updating the language used to describe mental illness as laid out in the Australian state of Victoria Mental Health Act (Victoria State Government 2014)
In the UK and abroad, police can use mental health legislation to remove people from the community and take them to a place of safety for assessment (Wood et al 2011). In some areas this is a designated room in a hospital, but people are frequently taken to police cells when alternatives are not available (Docking et al 2008).
Police officers in England and Wales detained 9,378 people under section 136 of the Mental Health Act (1983) between April 2011 and March 2012 (CQC 2013).
Under English and Welsh legislation people can be held in a police cell while awaiting psychiatric assessment for up to 72 hours, despite not having committed a criminal offence (DH 2007), and for up to 24 hours in Scotland (Scottish Government 2003). People who have committed a criminal offence can be held for questioning for up to 24 hours.
The associated stigma and trauma of being held in a police cell can stay with people for the rest of their lives and they can feel criminalised by the system that is designed to protect them (CQC 2013).
Information about the effect of MCTs on the use of mental health legislation in the UK is limited, however some data are beginning to emerge. For example, an MCT in Leicester reported a 29% decrease in the use of section 136 in the first month of its implementation (Dean 2013), and when the team decided that someone did require detention under the act, the time spent in detention on section 136 was shorter than it was before the MCT started its work (Dean 2013). By contrast, Allen Consulting Group (2012) in its study found that a high percentage (37%) of incidents responded to by the MCT (n=610) required enactment of local mental health legislation by the police.
Mental health interventions such as MCTs, aimed at managing the pre-custody stage, could divert people away from the criminal justice system and reduce the unnecessary use of mental health legislation (CQC 2013, NHS England 2013). In the 2013 CQC report on the use of section 136 in England and Wales, police officers said that being able to engage with crisis teams before deciding to detain people made a significant difference to the outcome.
It could be argued, therefore, that mental health professionals who can provide advice and assistance on scene could have a similar effect on outcomes. Again, outcomes will differ depending on geographical area, model of MCT and legislation.
Admission to hospital
Admission to psychiatric facilities for assessment or treatment can have long and short-term negative effects on patients’ wellbeing (Moses 2011, Loch 2014, Rusch et al 2014). Stigma and discrimination are just some of the issues faced by people admitted to, and then discharged from, inpatient care (Loch 2014).
UK mental health legislation states that people should be cared for in the least restrictive environment possible, to reduce the number of unnecessary admissions and to avoid the associated issues described above (DH 2007, Scottish Government 2003).
MCTs, like crisis resolution home treatment teams, provide a way of delivering care to people in the community without admitting them to hospital. People are assessed at the scene and either discharged home or taken to hospital for further assessment. Some MCT assessments might result in admission, but assessments undertaken closer to the place of crisis may result in improved patient outcomes (Allen Consulting Group 2012).
Scott (2000) evaluated the effect of MCTs on rates of admission to psychiatric inpatient units by comparing assessments undertaken by the DeKalb MCT (n=73) to regular police contacts resulting from emergency 911 calls (n=58). In total, 33 people in the intervention group were admitted, 12 of whom required detention under mental health legislation, compared with 42 admissions in the control group, 28 of whom required detention under local legislation.
Although the sample size was relatively small (n=131), people who were assessed by the MCT were less likely to require admission, despite the fact that there were more people in the intervention than the control group. This suggests that assessments carried out by MCTs, compared with standard police responses, can result in fewer hospital admissions.
The amount of time taken to provide support for people experiencing a mental health crisis in the community varies widely. Police officers in England report average waiting times of between 6 and 8 hours for assessments under section 136, while one officer reported a wait of 52 hours (CQC 2013), although this is still within the allowed 72 hours (DH 2007). In Scotland, the Choose Life initiative (Scottish Government 2002) recommended that people in mental health crisis should have quick and easy access to help and support. However, this has not been implemented in some areas and local policies dictate that people must attend hospitals to be assessed, which increases delays in treatment (Tayside Police 2010, Tayside Multi Agency 2011).
Timely access to mental health services is one of the goals of MCTs. Kisley et al (2010) assessed the efficiency of the COAST scheme by measuring police time spent on visits and call-to-door-times. The researchers found that although the number of incidents the COAST team responded to increased significantly, as described earlier, police time spent on each visit decreased over the 2-year study period from 165 to 136 minutes. Furthermore, the length of time it took the COAST MCT to arrive at the scene halved from 73 to 36 minutes.
In Australia, the Allen Consulting Group (2012) found that the average time it took for the PACER MCT to respond was about 30 minutes, and that the time it spent on each call was reduced significantly compared with standard police responses.
These findings suggest that MCTs, when they arrive promptly at the scene, can help reduce police time taken up with responding to people in crisis. Call-to-door time data for the PACER study control group was not available so a comparison cannot be made. Further, comparison of MCT response times in rural and city areas needs investigation and more information is required on the time spent on each case, depending on whether or not the person was detained under mental health legislation.
Calculating the cost of healthcare is highly complex (McCrone et al 2008, Curtis 2011). Scott (2000) found that the average cost for each case for the DeKalb MCT, compared with standard police interventions, was about 23% lower. Information on the cost effectiveness of the few MCTs in the UK is limited, and Bather et al (2008) highlighted the need for further research into the cost of mental health problems on UK policing.
As discussed previously, established MCTs in other countries report a reduction in police time, arrests, and admissions to hospital. There is limited evidence of MCTs delivering cost savings in the UK. However, citing figures contained in a report completed by Sainsbury Centre for Mental Health (2009), Dean (2013) highlights that the average cost to arrest a person is £1,780. This figure does not include the cost of the mental health professionals required to assess and detain a person under section 136 of the Mental Health Act.
When people require admission to hospital following assessment, expenditure can increase dramatically. For example, the average cost of an inpatient bed in England in 2011 was £321 a day, excluding specialist care services (Curtis 2011). Allen Consulting Group (2012) provided a cost analysis of an MCT intervention compared with routine care, and included multiple scenarios involving different costs, such as with or without transport. Averaging expenditure related to each scenario gave an estimated cost per case of the equivalent of £566 for routine care and £437 for MCT intervention. Again, the price of routine care would increase if admission was required. It could be argued that MCTs which divert people from inpatient beds and reduce police time spent on each case, could make significant cost savings.
Service user engagement
Some people with severe and enduring mental health problems can be difficult to engage (Sainsbury Centre for Mental Health 1998). Therefore, people who come into contact with mental health services for the first time need to trust the teams they encounter and receive adequate services, or they could become ‘hard to reach’ (Sainsbury Centre for Mental Health 1998).
Mental health professionals teamed with police officers can provide the knowledge and skills required to respond effectively to people experiencing mental distress (Scott 2000, Kisely et al 2010, Allen Consulting Group 2012, Dean 2013), and a more empathetic and considered approach may help people engage positively with services in the future. One participant in Kisely et al’s (2010) study said they felt completely isolated apart from their ability to speak to the COAST staff, while a support service for people with mental health needs in the PACER study noted that people who are mentally ill are often afraid of the police, but that the PACER programme helped ‘build bridges’ (Allen Consulting Group 2012).
To quantify the effect of MCTs on service user engagement, Kisley et al (2010) recorded the study participants’ outpatient attendances at 12-months’ follow up. The researchers found that those who had been in contact with the COAST service (n=295) had a greater number of outpatient appointments at 12 months than the control group. These results suggest that people who receive care from MCTs are more likely to engage with mental health services in the long term.
Service user feedback
In Australia, Boscarato et al (2014) undertook one-to-one semi-structured interviews with 11 people who had received care from several teams that used different models of joint police and mental health service responses. Only 1 participant had received care in the form of an MCT, named in Boscarato’s 2014 study as the Ride Along Model. However, the person cared for by the MCT made positive comments about the immediacy of the service, the service’s specialist knowledge of mental health issues, its ability to engage the service user and offer alternatives to hospital admission (Boscarato et al 2014).
Participants in Scott’s (2000) study were asked to complete a consumer satisfaction questionnaire about their experience of working in, or receiving care from, the DeKalb MCT. Less than half the clients responded, but those who did (n=22) gave the MCT a mean rating of 27.4 ± 4.9, where the maximum score was 32, and family members (n=10) gave a mean rating of 27.7 ± 5.8. These scores indicate that consumers were generally satisfied with the care they received. However, the low number of responses could suggest that some were dissatisfied.
Car 87, an MCT in Vancouver, Canada (Vancouver Coastal Health 2014), is similar in structure to other MCTs discussed here. There is little empirical evidence of satisfaction with the service, however anecdotal evidence in comments on service-user internet forums and in the news about this MCT were negative (Downtown Eastside Enquirer 2008). Although anecdotal evidence should be used with caution (Enkin and Jadad 1998), these comments highlight potential problems that might not be reported in research.
This review of the literature on MCTs suggests that the UK is behind other countries in the development and implementation of these specialist services. Apart from the difference in mental health legislation, policies and resources between countries, one reason for the faster expansion of services elsewhere could be the potential lethal outcome of police responses to people in mental health crisis. For example, the fatal shooting of a person with mental health needs in Memphis in 1988 led to the rapid development of crisis intervention teams in the US (Cochran 2000). These are specialist voluntary education programmes for police officers to help them develop effective response skills to people with mental health needs. The programme has been highly successful and is expanding to other countries (Herrington et al 2009, Forchuk et al 2010).
MCTs can enable mental health professionals and police officers to overcome some of the issues described earlier and strengthen relationships between the services. Standard police responses can have a negative effect on people’s perception of situations, but the link between MCT intervention and the reduction of violent incidents and suicide is not clear and requires further study. However, MCTs can reduce the length of time people are detained by the police under mental health legislation, and could help reduce the amount of police time spent supporting people in mental health crisis.
The strongest evidence in support of MCTs is their ability to reduce inappropriate use of mental health legislation and hospital admissions, which were the original goals of the service. Furthermore, there is some evidence that MCTs can help reduce operating costs. Finally, service users who have received support from MCTs give positive reviews of the services, and are generally satisfied with the care they receive.
There are numerous differences between the services provided in the UK and abroad, therefore caution must be taken when generalising results of the studies reviewed here. There are only a few MCTs in the UK, but their underpinning values can be transferred to routine practice, for example improving inter-agency working between the police and mental health services.
MCTs are well established in some countries, but research supporting their use is lacking. Plans are in place to expand MCTs in the UK (DH 2013), so it is hoped that there will be high quality research of their effectiveness in the near future.
Implications for practice
» Mobile crisis teams can lead to better outcomes for people with mental health problems who come into contact with the police.
» There is evidence that they can reduce the use of the provisions of the Mental Health Act and lead to fewer hospital admissions.
» Follow-up studies suggest they can help to increase engagement with hard to reach people.
» The principles underpinning mobile crisis teams can be transferred to routine practice.
» Further research is needed to firmly establish their effectiveness in improving the overall care of people in crisis and reducing the involvement of the police in crisis care.
Thanks to mental health nursing lecturer Scott Kane and staff at the Abertay University School of Health and Social Sciences for their support during the completion of the Crisis Intervention and Resolution module
Council of State Governments (2002) Report of the Consensus Project. csgjusticecenter.org/cp/publications/the-consensus-project-report/ (Last accessed: 20 June 2016.)
Department of Health (2001) Mental Health Policy Implementation Guide. DH, London. Pubmed
Department of Health (2013) Extending the Street Triage Scheme: New patrols with Nurses and the Police. www.gov.uk/government/news/extending-the-street-triage-scheme-new-patrols-with-nurses-and-the-police (Last accessed: 20 June 2016.)
Victoria State Government (2014) Victoria Mental Health Act 2014. Victoria, Australia. Pubmed
Downtown Eastside Enquirer (2008) ‘Car 87’ Abuse Raised by Mayoral Candidate for the First Time in a Vancouver Election. downtowneastsideenquirer.blogspot.co.uk/2008/11/car-87-political-psychiatry-raised-by.html (Last accessed: August 15 2014.)
Kisley S, Campbell L, Peddle S et al (2010) A controlled before and after evaluation of a mobile crisis partnership between the police and mental health services in Nova Scotia. The Canadian Journal of Psychiatry. 55, 10, 662-668. Pubmed
Landeen J, Pawlick J, Rolfe S et al (2004) Delineating the populations served by a mobile crisis team: organizing diversity. Canadian Journal of Psychiatry. 49, 1, 45-50. Pubmed
Loch A (2014) Discharge from a mental health admission ward: is it safe to go home? A review of the negative outcomes of psychiatric hospitalization. Psychology Research and Behavior Management. 7, 137-145. Pubmed
Mercer D (2012) Police taser use on mentally ill slammed. Independent. www.independent.co.uk/news/uk/crime/police-taser-use-on-mentally-ill-slammed-8431150.html (Last accessed: August 15 2014.)
Rusch N, Muller M, Lay B et al (2014) Emotional reactions to involuntary psychiatric hospitalization and stigma-related stress among people with mental illness. European Archives of Psychiatry and Clinical Neuroscience. 264, 1, 35-43. Pubmed
Sainsbury Centre for Mental Health (1998) Keys to Engagement. Sainsbury Centre for Mental Health, London. Pubmed
Vancouver Coastal Health (2014) Vancouver Mental Health Emergency Services - Car 87. www.vch.ca/locations-and-services/find-health-services/?program_id=2184 (Last accessed: June 20 2014.)