Editorial Credits/ Crédits éditorials

Subscription Rates /Prix d'abonnements

Advertising Rates / Tarifs publicitaires (PDF)

Guest Editorial
Psychiatry, Technology, and the Corn Fields of Iowa
Harry Karlinsky
(PDF)

In Review
The Internet’s Impact on the Practice of Psychiatry
Rima Styra
(PDF)

Clinical and Educational Telepsychiatry Applications: A Review
Donald M Hilty, Shayna L Marks, Doug Urness, Peter M Yellowlees, Thomas S Nesbitt
(PDF)

Portable Computing in Psychiatry
John Luo
(PDF)

Original Research
Assessing and Monitoring Antipsychotic-Induced Movement Disorders in Hospitalized Patients: A Cautionary Study
Leonardo Cortese, Mandar Jog, T Jeffrey McAuley, V Kotteda, Giuseppe Costa
(PDF)

Sociodemographic Factors Associated With Comorbid Major Depressive Episodes and Alcohol Dependence in the General Population
JianLi Wang, Nady El-Guebaly
(PDF)

Delineating the Population Served by a Mobile Crisis Team: Organizing Diversity
Janet Landeen, Julie Pawlick, Steven Rolfe, Ian Cottee, Melanie Holmes
(PDF)

Detecting Women at Risk for Postnatal Depression Using the Edinburgh Postnatal Depression Scale at 2 to 3 Days Postpartum
Frédérique Teissèdre, Henri Chabrol
(PDF)

Review Paper
Ethics in Psychiatric Research: Study Design Issues
Gordon DuVal
(PDF)

Brief Communication
The Prevalence of Psychological Morbidity in West Bank Palestinian Children
Tanya L Zakrison, Amira Shahen, Shaban Mortaja, Paul A Hamel
(PDF)

Book Reviews
(PDF)

Beyond Technique in Solution Focused Therapy.
Reviewed by
Llewellyn W Joseph, MD, FRCPC

The Epidemiology of Schizophrenia.
Reviewed by
Mary V Seeman, MD

The Private Self.
Reviewed by
Paul Ian Steinberg, MD, FRCPC

Treatment Planning in Psychotherapy: Taking the Guesswork Out of Clinical Care.
Reviewed by
Gilbert Pinard, MD, FRCPC

Pharmacogenetics of Psychotropic Drugs.
Reviewed by
Gustavo Turecki MD PhD

Psychotherapy and Counselling in Practice. A Narrative Framework.
Reviewed by
Paul KB Dagg

Oedipus and Beyond: A Clinical Theory.
Reviewed by
Paul Ian Steinberg, MD, FRCPC

Letters to the Editor
(PDF)

Modafinil Treatment of Excessive Sedation Associated With Divalproex Sodium

Ziprasidone in Parkinson’s Disease Psychosis

Combined Oral Venlafaxine and Intravenous Clomipramine-A: Successful Temporary Response in a Patient With Extremely Refractory Depression

Doxepin Increases Serum Cholesterol Levels

Re: Suicide: The Persisting Challenge

Reply: Suicide: The Persisting Challenge

Depression and a History of Alcoholism

Gains in Speeded Information Processing Following Clozapine Treatment of Schizophrenia

Problems With Crystallizing Phenomenology and Nosology in Adolescent Psychiatry

Serotonin Syndrome With Prolonged Dysphagia

Original Research

Delineating the Population Served by a Mobile Crisis Team: Organizing Diversity

Janet Landeen, RN, PhD¹, Julie Pawlick, BScN², Steven Rolfe, BScN³, Ian Cottee, BScN⁴Melanie Holmes, BSc, MSc⁵

The past several decades have witnessed many changes in the care of those with serious mental illness. The waves of deinstitutionalization prompted the development of community-based programs (1). Service planners and researchers have identified the need for integrated, comprehensive programs that serve all individuals (2,3). However, gaps in service continue to exist, especially in the area of crisis intervention, for individuals who may or may not be connected to existing services (4). Increasing numbers of individuals with mental illness are homeless or incarcerated (4). The proliferation in the US of mobile crisis intervention teams attests to the growing awareness of the need for alternative approaches (5). In Canada, the Crisis Outreach and Support Team (COAST) was developed in Hamilton, Ontario (population 400 000), to address the need for comprehensive services.

Originally modelled after the Vancouver Car 87 Program (6), COAST has been responding to mental health crises since October 1997. This mobile program combines a specially trained police officer with a psychiatric nurse or social worker to respond to psychiatric crises in the community (7). COAST responds to the requests from individuals with mental illness, from families, and from social and mental health agencies, as well as from the police, for crisis intervention and support in the individual’s natural environment.

Despite a growing number of articles published about mobile crisis teams, systematic research studies into the evaluation of such programs are lacking (8,9). While mobile crisis teams have existed for decades, most studies have been of a descriptive nature, with little empirical data being produced (5). Perceived benefits of mobile crisis programs include accessibility of services, accuracy of assessments, cost effectiveness, improved public relations and education (10), and improved hospitalization and readmission rates (9,11). However, as Geller and colleagues conclude from their study of all state mental health agency directors, the proliferation of mobile crisis units is based on the belief and perception that such services are beneficial, rather than on any systematic evaluation of them (8). Only a single study was found that evaluated a program that paired police enforcement with mental health workers (12). Individuals served by this team were deemed to have a severe mental illness; were at high risk for violence, as indicated by prior arrest records; were homeless; or had substance abuse problems. The authors concluded that the team was effective in diverting individuals from the criminal system and into the mental health system. This study concludes that such a team could effectively deal with potential violence, substance abuse, and clients with criminal and mental health histories in a manner that decriminalized persons with mental illness.

Description of COAST

The COAST program is funded by the Ontario Ministry of Health and is administered through a community hospital, although it functions autonomously. Individuals, families, concerned neighbours, other service providers, and the community at large can call the designated crisis line 24 hours daily for service. The triage worker, who is a psychiatric nurse or social worker, answers the calls and determines the intervention. Often the situation can be handled over the phone through supportive telephone intervention and problem solving. If the impending crisis cannot be settled over the phone, the outreach team will respond to the crisis location. The outreach team consists of a plain-clothes police officer and a mental health worker (who is a nurse or social worker). The police officer has been specially trained through the program and works in conjunction with the mental health worker to conduct mental status assessments and provide crisis intervention in the client’s environment of choice. The police presence on the team relates to the unpredictable nature of a crisis and provides a degree of safety for clients and workers in unstable situations. The police officer also has the ability and responsibility to escort to hospital, for further psychiatric assessment, a person who meets the criteria for dangerousness under the Ontario Mental Health Act. This sharing of duties is similar to the model described by Lamb and colleagues (12).

Because such programs are scarce, evaluating the effectiveness of COAST is paramount. Developing general outcome measures is difficult because there are no age or diagnostic criteria for service, and the program responds to diverse situations in numerous environments. Initial attempts at program evaluation were hampered by the diversity of the population served. Various strategies aimed at measuring client satisfaction (that is, telephone survey, questionnaire, and focus group) did not yield useable data. Issues with data collection included difficulty in tracking clients, low response rates to questionnaires, and symptom severity impeding completion of interviews or questionnaires. Identifying appropriate outcomes that are valid for the entire population was impossible, given the diversity. Traditional approaches to identifying common characteristics of the population, such as by psychiatric diagnosis, are impractical owing to the nature of the services provided. The current research identifies specific cohorts of individuals served by COAST so that appropriate program evaluation can be designed and conducted for each unique grouping.

Methodology

We identified distinct cohorts of service users using a qualitative, text-based review adapted from grounded theory and schema analysis methodology (13). This retrospective study analyzed specific data from selected client records. Four sets of client records were reviewed in this process (n = 401). For the first set of records, every record that was opened in April 2000 was selected (n = 69). The second set of records consisted of every fourth record opened in September 2000 (n = 40). For these first 2 reviews, only data related to the referral source, sex, reason for referral, and the mental status assessment were known to the researchers. The third set of records consisted of every second chart opened in January 2001 (n = 16). This set of records included follow-up and outcome data, in addition to the material listed above, to determine whether additional information would influence the researchers’ assessments. In each instance, 2 researchers independently read and categorized each chart. After the independent reviews, 2 researchers met with a third to reach consensus on both the categories and the assignment of individuals to categories or clusters. The first set of records was read and reread until a pattern began to emerge. The second set of records, read by the same 2 researchers, was used to confirm the initial clusters. Consensus was reached on all records, and no new clusters emerged at that time. The third set of records was reviewed by 2 different researchers to confirm the usefulness of the cluster categorization tool. For the final set of records, the tool was used to categorize all new requests for service from April 1 to April 31, 2002 (n = 276). Two researchers independently assigned the categorization with an interrater reliability kappa of 0.906.

Results

The initial 2 chart reviews produced 5 preliminary clusters: Symptoms Disturbing Others, Symptoms Disturbing Self, Information Seeking, For Your Information, and Other. After the third review, the cluster Symptoms Disturbing Others was subdivided into Symptoms Disturbing Others–General and Symptoms Disturbing Others–Suicide. The volume of records reviewed, rather than the outcome data, prompted the division of this cluster. Further, inclusion criteria were developed to assist others in using the categorization tool (Table 1).

Symptoms Disturbing Others–General
Symptoms Disturbing Others is divided into 2 subcategories. In both categories, the caller is concerned about symptoms of mental illness displayed by the client. These symptoms may be threatening to the safety of the client or those around the client. Referrals come from various sources, including police, family, neighbours, and agencies, but this category excludes self- referrals from the client. The criterion that distinguishes the 2 subgroups is the nature of the symptoms. The General subcluster represents symptomatic actions and behaviours, expressed by an individual with serious mental illness, that cause others to worry about the safety of the client, as well as themselves, and possibly of others in the community. This cluster is defined as a crisis situation where a threat of harm to others is significant and the concerns are general in nature. The following scenario illustrates such a situation:

Police call and ask COAST to respond to a local homeless shelter to assist with a female client who is behaving in a bizarre manner. The client refuses to identify herself other than to say that her name is Cupid and she is married to James Bond. She is inappropriately dressed for the cold weather, and her speech is tangential, pressured, and somewhat aggressive. There is no evidence of substance abuse; others in the shelter, including staff, are afraid of her aggressive behaviour.

Symptoms Disturbing Others–Suicide
This cluster includes symptoms and behaviours that cause the referral source to worry about the individual; however, the cluster groups clients with threats, plans, or actions specific to self-harm or suicidal behaviours. As with the previous cluster, the referral source may include police, family members, neighbours, or agencies. This cluster represents crisis situations in which referrals come from anyone other than the individuals themselves.

A typical situation might be as follows:

Police were called to attend “Joe’s” home after his wife reported that Joe had taken too many pills and wandered off into the woods. When police arrived, Joe had returned home and stated that he was depressed and suicidal after losing his job but had not taken any pills. Joe agreed to stay safe in the home with his wife until COAST arrived.

Symptoms Disturbing Self
Symptoms Disturbing Self are exactly that—symptoms disturbing to the individual caller. In this cluster, the person who calls the program is self-referring and is experiencing some form of distress, either physical or emotional. Individuals in this cluster are strictly those who are in crisis and call for assistance themselves because their abilities to cope and solve problems are diminished by the crisis.

“Antonio” calls with concerns about his anxiety level. He reports that he spends up to 8 hours daily in a bathtub filled with hot water, and he believes he has sustained second or third degree burns to portions of the skin on his back. He cannot articulate why he feels compelled to do this, but he reports that his wife has threatened to leave him unless he obtains help. He refuses to see his doctor or go to the hospital, but he agrees to a mobile visit.

“Mary” calls COAST because she has suffered from depression since her husband’s death 4 months ago. She has few friends or family members in Canada and feels very isolated. She denies suicidal thoughts and just wants someone to talk to about her feelings. Mary has no history of mental health concerns but is experiencing disturbances in her sleeping and eating habits. She agrees to see her general practitioner and to call as required for support, or for a visit, if her symptoms become more intense.

Information Seeking
Information Seeking includes individuals who are calling the agency in search of information or direction related to connecting relatives to the mental health system. This can include information related to substance abuse issues, shelters, medication, and community resources. Referrals are often, but not exclusively, from persons who call for information to help advocate for a family member who is experiencing mental health concerns. This cluster represents a noncrisis situation.

“Kathlyne” calls to ask about services in town for peer support. She would not disclose what her issues were except to say that she was not interested in traditional psychiatric services. She was provided with phone numbers for 2 peer-support agencies in the vicinity.

“Pam” calls requesting information about local support groups for family and friends of individuals who have been diagnosed with bipolar disorder. Pam’s sister has bipolar disorder and Pam feels a support group would help the family become more educated about the illness itself and learn how best to provide support.

For Your Information
This cluster often consists of alerts where agencies forewarn the program that they might hear from, or about, a client who has recently been having a difficult time. Sometimes these individuals have left the helping agency without follow-up or have been released from jail refusing discharge planning. The alert serves to provide background in case a crisis call from the client or community is received. This cluster often concerns individuals who refuse services despite exhibiting psychiatric symptoms. Agencies making these referrals assess for themselves the necessity of breaching client confidentiality based on the perceived risks for violence to self or others.

A For Your Information Alert from the Hamilton Wentworth Detention Centre relays information regarding the targeted release date of a medium-risk offender. The alert describes “Joe” as an individual who has schizophrenia and has refused discharge planning. He has a habit of starting arguments with neighbours when he is noncompliant with his medications. He reluctantly agreed to take the COAST Crisis Line phone number for self-referral.

“Fred’s” case manager calls to say that Fred is decompensating and is refusing all interventions. He has been noncompliant with medications for 2 weeks and stormed out of the office today. Fred is to be evicted at the end of the month but is not actively seeking alternative residence. He does not wish to see COAST at this time, but he may come to the attention of the police or another community partner.

Other
Other calls are miscellaneous and do not fit the first 4 clusters. They may include substance abuse or housing concerns, and they have various referral sources, including police, agencies, families, and individuals themselves.

A public health nurse calls about a 55-year-old man who has Huntington’s disease and a baseball-sized cancerous growth on his forehead, for which he is not receiving treatment. He resides in a third-floor room in a rundown motel in the city core. He was recently in jail for assaulting the nurse (he struck her with his cane). Public health and all other outreach nursing services do not feel they can assist him at this time. They refuse to attend the residence owing to its hazardous location. COAST is to attend and assess the situation.

The public health inspector calls about “Vladimir,” who apparently resides alone in his own home. A furnace- repair person called public health to report that the home is uninhabitable and deplorable, with rotting food and human and animal excrement littering the residence. Vladimir appeared clear mentally, although the inspector suspects alcohol abuse. Public health is requesting that COAST attend and assess the situation.

In the final review, the records were categorized as follows: 30% Symptoms Disturbing Others–General, 19% Symptoms Disturbing Others–Suicide, 32% Self-Referrals, 2% Information Seeking, 10% For Your Information, and 7% Other. With the program receiving approximately 3000 requests for service yearly, close to 1000 calls would be self-referrals and 1500 would be from the community for Symptoms Disturbing Others–General and Suicide.

Discussion

Given the unique nature of this program, identifying groupings of clients could assist others conducting needs assessments for similar programs in their own communities.

The categorization of the client base at the COAST program has implications for further evaluation and outcome research. An understanding of the people who call the program and why, of the services offered and provided, and of the ultimate outcome of the contact may be used to identify service gaps, develop best practices, plan service delivery, determine staffing needs, and develop community education projects. Each cluster can be further examined for assessment and inter- vention practices to ensure best practices for the individual client population. Symptoms Disturbing Others–General (approximately 900 intakes yearly) represents the client population that the program was initially intended to service. These individuals often have a serious and persistent mental illness, are often vulnerable, and lack insight into their illness and the impact they have on others. Friends, family, neighbours, or police are most likely to contact COAST because of their behaviour. They often hover just under the Ontario Mental Health Act criteria for apprehension and are seen repeatedly by COAST until they meet the criteria and undergo further assessment in hospital, or until the crisis resolves. In approximately 75% of COAST apprehensions under the Mental Health Act, the individuals are admitted to a hospital bed, indicating that COAST is highly accurate in its assessment. These clients do not always pose a serious threat to the community, but they do stretch community tolerance when empathy evaporates as a result of bizarre or frightening behaviours. Some of the clients in this cluster might be well suited to a community treatment order or assertive outreach team after a period of hospital stabilization.

Those clients whose suicidal threats, gestures, or plans are distressing to others can be viewed as a distinct group. Yearly, the program can expect about 600 requests related to these clients, and the literature on suicide prevention can be useful in planning approaches to meet their unique needs.

Clients who ask for service (Symptoms Disturbing Self) are more likely to be willing to assist in evaluating the service that they receive. These clients (approximately 1000 yearly) are reaching out and, when provided with an array of inter- ventions, including assessment, support, and referral, can be discharged to other services for the long term. A satisfaction survey is currently being developed for use with this cohort.

Even though Information Seeking is not strictly a client-centred crisis, it is highly likely that calls that fit the criteria for this cluster will continue to represent a small (2%) but important portion of calls to COAST. Since the program’s mandate is to provide access to mental health services, providing information is one way of assisting individuals to avail themselves of appropriate services. With myriad services available in the community, this component of providing information about appropriate services, agencies, and places to start can be invaluable to a client, family, or friend in crisis. Indeed, crisis theory suggests that providing linkages to supports is one of the hallmark features of crisis intervention (9).

Most referrals that are For Your Information originate from other service providers. These intakes serve to reduce the anxiety other professionals feel when they have exhausted all other options with clients who are either unable or unwilling to accept services. Further research on this cohort may lead to the identification of service gaps and how to best serve these populations. For example, a retrospective study of the intakes from the jail system (that is, those clients uninterested in discharge planning and follow-up from the jail system) and their outcome after discharge may lead to an understanding of how these clients are best served within the criminal system.

The final category, Other, may be broken down with further investigation of outcomes. There will always be clients who have multiple, complex needs that do not fit within any single classification system and do not fit the mandate of other community services, making linkage difficult. These clients, while representing only about 7% of all annual intakes, tend to be complicated and consume a significant amount of time and energy because they are not well served in the community. These clients tend to require repeated outreach interventions and significant problem solving, as well as advocacy. Without aggressive outreach and creative advocacy, they “fall through the cracks.”

In the 5 years since the inception of the COAST program, referral source patterns have changed as the community, including service providers, police, clients, and families have become aware of the services provided. Initially, self-referrals represented only a small portion (about 15%) of the total requests for service. This figure has risen to about 30%, indicating that the program has become consolidated into the array of community services available to individuals with mental illness. Through examining the Other cluster, the program has gained an understanding of gaps existing in the community and has become a leader in advocating for the complex, underserved individuals who do not seem to fit any agency mandate.

Through an understanding of the diverse nature of the clients served by a program that melds the unique expertise and skills of both police and mental health service providers, service gaps can be filled and best practices developed. This research may help lead to an understanding of the priority populations the program should serve and how best to do this.

Acknowledgement

The authors thank the Father Sean O’Sullivan Research Centre, St Joseph’s Healthcare, Hamilton, Ontario, for support given to this research.

References

1. Anthony WA. Recovery from mental illness: the guiding vision of the mental health system in the 1990’s. Innovations and Research 1993;2:17–24.

2. Anthony WA, Cohen M, Farkas M. Psychiatric rehabilitation. Boston: Center for Psychiatric Rehabilitation. 1990.

3. Bachrach LL. Continuity of care and approaches to case management for long-term mentally ill patients. Hosp Community Psychiatry 1993;44:465–8.

4. Hylton J. Criminal justice and mental health. Network 1997; Spring: p 4–8.

5. Fisher WH, Geller JL, Wirth-Cauchon J. Empirically assessing the impact of mobile crisis capacity on state hospital admissions. Community Mental Health 1990;26:245–53.

6. Sladen-Dew N, Bigelow DA, Buckley R, Bornemann S. The greater Vancouver mental health service society: 20 years experience in urban community mental health. Can J Psychiatry 1993;38:308–14.

7. McGurk T, Pawlick J. COAST program. Emerg Psychiatry 1998;4:31–2.

8. Geller JL, Fisher WH, McDermeit M. A national survey of mobile crisis services and their evaluation. Psychiatr Serv 1995;46:893–7.

9. Scott RL. Evaluation of a mobile crisis program:effectiveness, efficiency, and consumer satisfaction. Psychiatr Serv 2000;51:1153–6.

10. Zealberg JJ, Santos AB, Fisher RK. Benefits of mobile crisis programs. Psychiatr Serv 1993;44:16–7.

11. Guo S, Biegel DE, Johnsen JA, Dyches H. Assessing the impact of community-based mobile crisis services on preventing hospitalization. Psychiatr Serv 2001;52:223–8.

12. Lamb HR, Shaner R, Elliott DM, DeCuir WJ Jr, Foltz JT. Outcome for psychiatric emergency patients seen by an outreach police-mental health team. Psychiatr Serv 1995;46:1267–71.

13. Denzon NK, Lincoln YS, editors. Handbook of qualitative research. 2nd ed. Thousand Oaks (CA): Sage; 2000.

Author(s)

Manuscript received August 2002, revised, and accepted June 2003.

1. Associate Professor, School of Nursing, McMaster University, Hamilton, Ontario; Investigator, Canadian Institute of Health Research, St Joseph’s Healthcare, Hamilton, Ontario.

2. Senior Mental Health Worker, Crisis Outreach and Support Team, St Joseph’s Healthcare, Hamilton, Ontario.

3. Community Nurse, St Joseph’s Healthcare, Mountain Site, Hamilton, Ontario.

4. Formerly, 4th year BScN student, McMaster University, Hamilton, Ontario.

5. Volunteer Reasearch Assistant.

Address for correspondence: Ms J Pawlick, Crisis Outreach and Support Team c/o St Joseph’s Healthcare, 50 Charlton Avenue East, Hamilton, ON L8N 1J6

e-mail: coast@coasthamilton.ca

1 | 2