Section B
Results of consultations and options for future data collection

View the most recent version.

Introduction
Police
Criminal Courts and Review Boards
Correctional System
Conclusion

Introduction

The issue of individuals with mental health issues in the criminal justice system is attracting increased interest and concern in Canada. While there have been pockets of studies and data collection activities trying to quantify the issue at the police, courts and corrections levels, there is presently a lack of comparative and comprehensive data to understand the extent of the problem, to inform decision-making regarding policy and action, and to measure outcomes of current initiatives and processes.

As a result, the Canadian Centre for Justice Statistics (CCJS), at the request of the National Justice Statistics Initiative, examined the feasibility of collecting data on the involvement of adults and youth with mental health issues in the criminal justice system.

Section B addresses two of the three main goals of the feasibility study:

1. to consult criminal justice stakeholders on their information priorities, data collection, barriers to data collection, and the feasibility of collecting data on the contact of individuals with mental health issues in the criminal justice system; and
2. to propose viable options for data collection involving police, courts, and corrections.

Method

The information contained in this document provides a summary of the consultations that were held between September 2007 and June 2008. The scope of the study encompasses all sectors of the criminal justice system, including police, courts, Review Boards and corrections. Consequently, a number of stakeholders were contacted to participate in the consultation process, including the following:

• Policing services;
• Courts – both traditional criminal courts and specialized mental health courts;
• Review Boards;
• Correctional services;
• Mental health organizations;
• Academics and researchers; and,
• Non-governmental organizations.

In total, over 100 individuals participated in the consultations by phone, in-person, or in writing (Appendix 4). Representatives from each sector were able to participate in the consultation. Some who were contacted for the consultations were unable to participate. While a large number of individuals from across Canada were involved in the consultations, the information presented in this report should not be considered representative of all viewpoints on the collection of mental health data in the criminal justice system.

Section B of the report is divided according to criminal justice sectors, namely police, courts/Review Boards, and corrections. Each section presents the stakeholders' views on general objectives of data collection, definition of mental health and mental illness, information needs and priorities, and data availability. Options that could be considered for future data collection are described for each sector of the criminal justice system.

Police

Objectives for data collection

Among police services, the most commonly identified objectives for data collection on persons with mental health issues in the criminal justice were 1) to facilitate information-based policy and programming decisions regarding responses to the issue and 2) to measure workload, performance and outcomes.

While only one police service ranked the need for consistent data recording practices as the most important objective among the five presented, most recognized this objective as worthwhile. Additional objectives were also mentioned, including the need to determine gaps in community resources and training and education needs of police and other criminal justice workers.

Definitional issues

In defining and responding to the issue of mental illness, there was a consensus that police must rely on observable behaviours and the environment to determine the presence or possibility of mental health issues. Diagnosed mental health conditions based on medical assessments are often not readily available or are only accessible as a result of police prior contact with the individual or information obtained from family or friends.

When asked which specific types of behaviours or conditions should be included in a definition of mental illness, the Canadian National Police Mental Health Liaison Committee,¹ a subcommittee of the Canadian Association of Chiefs of Police, argued that the clinical diagnostic criteria within the "Diagnostic and Statistical Manual for Mental Health Disorders" (DSM-IV-TR) can be useful. However, they argued that personality disorders that are criminogenic by nature, such as anti-social personality disorders, should be excluded in a definition of mental illness. In addition, many police services indicated that along with severe mental illness, such as schizophrenia, other conditions such as permanent brain injury/damage (e.g., Fetal Alcohol Spectrum Disorder) and substance abuse (due to evidence of co-morbidity with mental illness) should be included within the definition of mental illness.

At the operational level, a definition has been developed by some urban police services. Specifically, police services, such as the Ottawa Police Service and the Toronto Police Service, reported using the definition of "emotionally disturbed persons" when dealing with individuals they suspect as suffering from mental illness. This type of definition overcomes the difficulties in assessment and distinguishing between different types of mental disorders. It relies on visual cues and does not compel the officer to make a diagnostic assessment of the individual.

An observational definition, however, can pose some problems when applied to the issue of persons with mental health issues, including the following:

1. not all people who are emotionally disturbed are mentally ill;
2. the behaviour of an individual can be situational or fleeting as a result of a stressful situation;
3. it may be difficult to distinguish mental illness from substance use; and,
4. observations may be considered subjective.

Priority issues and information needs

A range of stakeholders, including police, provincial/territorial departments responsible for justice matters, academics, and non-government organizations, provided feedback on the issues relating to police contact with youth and adults with mental health issues. These issues, along with the corresponding data needs, are outlined below.

Overall, those consulted indicated that it would be ideal to have data at all geographic levels, including municipal, provincial, regional and national. Various arguments were presented in support of collecting and analysing data at different geographic levels, including legislative jurisdiction and budget allocation.

1. Workload of police

By far, police workload in responding to persons with mental health issues was identified as a leading priority issue. Police contact with persons with mental illness can include criminal and non-criminal contacts and can be the result of pro-active policing or call for service.

Consultation participants, including police and academics who have conducted research in the area, reported that there has been an increase in the number of police contacts with individuals with mental illness. It was said that in many cases, these interactions are associated with relatively minor offences, such as disturbances of the public peace, or non-criminal activities. For example, two major municipal police services, Regina and Vancouver police, reported that about half of their service calls involve non-criminal activities by individuals with a mental illness.

In addition, many participants reported that repeat contact involving the same individuals accounts for a substantial proportion of cases. This revolving door of individuals with mental illness was often attributed by participants to a lack of mental health monitoring and an absence of community resources.

It was also reported that the time taken to deal with a situation involving a youth or adult with mental illness is generally longer than a situation with no indication of mental illness. One academic consulted advised that results of an urban police study showed the average contact time was three times longer for mental health cases.

Consultation participants pointed to a number of reasons for the greater time spent on mental health calls. In non-criminal cases, the process to apprehend an individual under the provincial/territorial mental health act can involve a number of prolonged steps, depending on the proximity to hospitals.² The police officer is required to accompany and provide security at the local hospital while the individual waits for evaluation by a physician. In remote and northern communities, this process can be even longer, as a 2-person police escort must transport the individual to the nearest hospital setting, which can sometimes be located in the southern region of provinces.

Even when an apprehension under the mental health act is not undertaken and the incident is minor, participants indicated that locating appropriate resources in the community for diversion can be an onerous and lengthy task. In some cases, police may advise the individual with mental illness to voluntarily attend the hospital emergency room. This avoids an official apprehension under the mental health act and therefore eliminates wait times at the hospital and reduces the amount of paperwork.

In addition to police workload involved in apprehension, police are also responsible for transporting individuals who were charged and found not criminally responsible on account of a mental disorder. Depending on the location of the closest available forensic institution, which can vary based on the forensic bed capacity, police may be required to travel outside the jurisdiction of their local police service.

Measuring the workload of police with respect to mental health issues was seen as important to assessing the extent of the problem and to determining the policing costs and allocation of human resources. Through the consultations, the following workload indicators were identified as important:

• Number of contacts between police and individuals with mental health issues;
• Nature of contact: non-criminal or criminal. If criminal, type of crime (minor or violent crimes);
• Characteristics of individuals with mental illness, such as age, sex, occupation, medication use;
• Number of repeat contacts between police and individuals with mental health issues;
• Average length of time spent during interactions with mentally ill individuals;
• Identification of areas with higher incidence of mental illness within policing boundaries;
• Criteria used by police in making decisions on appropriate actions;
• Action taken (e.g., no action, charge, divert to social services, apprehend under mental health act);
• Number of apprehensions under the provincial/territorial mental health acts;
• Number of individuals with mental illness brought to hospital and average wait time at the hospital; and,
• Outcome from hospital visit (e.g., admission, discharge).

In most cases, stakeholders suggested the use of comparison groups (i.e., mentally ill contacts versus non-mentally ill contacts) in the collection and analysis of data.

2. Absence of community resources

Closely related to the issue of police workload was the availability of resources in the community. Consultations indicated that the suspected increase in mental health calls may be linked to an absence of adequate mental health services in the community and the subsequent reliance on police as the access point for services. In addition, the lack of resources was said to limit the non-charging options for police.

Along with a stated scarcity of readily accessible mental health treatment programs, it was reported that the absence of social supports, such as appropriate housing and drug treatment programs, is seen to exacerbate mental health co-morbidity issues. A lack of adequate resources to address compromised social well-being, such as homelessness, and to address addictions was viewed as contributors to police contact.

When police involvement is initiated with individuals suffering from mental illnesses, consultation results indicated that police participants believed they have limited options in providing appropriate responses. Diversion for minor offences or referrals was seen as difficult to manage in an environment with a perceived absence of or an actual lack of community resources. A need for greater education and awareness on alternatives to charging was seen as critical to limiting individuals' involvement in courts and corrections. The absence of community resources was also linked to the issue of police re-contact, as the likelihood of re-involvement was argued to increase with a lack of adequate health care plan for those with mental illness, combined with limited access to services.

For youth, consultation participants commented that while the Youth Criminal Justice Act (YCJA) contains provisions to assist police with mental illness cases, it was felt that there was a lack of services or protocols to put these provisions into practice. Participants also stated a difficulty in addressing service needs of older adolescents aged 16 to 17 years.

To inform the community resource issues, stakeholders have indicated the following data needs:

• Identify community resources that can and would be willing to provide services to individuals with mental illness that come into contact with police;
• Number and percentage of cases where the police would have diverted/referred the individual with mental illness to community service, if services were available;
• Number of individuals with mental illness who come into contact with the police who previously accessed mental health services and/or who were on prescribed medications;
• Type of treatment needs among mental health contacts; and,
• Measurement or assessment of the effectiveness of mental health services in decreasing future involvement in the criminal justice system.

3. Assessment of mental illness and training issues

While police participants viewed mental health issues as a priority, many participants expressed the view that police are not always equipped to assess mental illness, particularly the specific type of mental illness or diagnosis. As visual cues are the primary tools used by police during an incident, it was thought that basic education on symptoms and even myth-busting would be useful.

Some stakeholders argued that the lack of recognition of mental illness could be due to minimal police training on mental illness. As a result, there was an identified need for properly trained police to deal with mental health cases. At the same time, it was also noted that in some larger police services, crisis intervention teams, comprised of a police officer and health worker, are specially trained to deal with individuals presenting symptoms of mental illness. These teams are dispatched to suspected mental health calls and attempt to keep individuals with mental health issues out of the court system, either by referral, pre-charge or post-charge diversion.

To address the information priority surrounding assessment and training/qualifications, consultation results reveal the following data needs:

• Information on available responses based on specific mental health condition;
• Number of calls for specialized mental health crisis intervention teams; and,
• Action taken by police services with mental health crisis intervention teams.

4. Information sharing

Consultation participants said that one of the difficulties in easily identifying and responding to persons with mental illness is the challenge of information sharing between police and health authorities. While recognizing and respecting that privacy legislation is necessary to protect individual's privacy rights, the question of privacy and confidentiality was identified as a barrier to sharing client information for the purpose of coordinating and improving service.

In addition, in some jurisdictions, mental health files are legislated for deletion after a set period of time. This introduces issues for long-term tracking and monitoring of individuals.

To address this information priority, stakeholders have proposed:

• Data that would integrate police and health information for research and operational purposes.

5. Nature of interactions between police and persons with mental illness

Due to the complex nature and dynamics of the interactions between police and persons with mental illness, police participants commented that it was important to track both positive and negative encounters with individuals presenting mental illness. An example of a positive police encounter includes outreach with the homeless population who are mentally ill, while the use of force against individuals with mental illness represents an example of a negative encounter. The element of officer safety in incidents involving individuals with mental illness was also raised as an important issue.

The data needs for documenting these interactions include the following:

• Recording and reporting on positive interactions between police and persons with mental illnesses;
• Number of complaints against the police regarding persons with mental illness, including use of force and deaths in police custody (suicide and deaths by officers); and,
• Data on officer safety involving individuals with mental illness.

Data availability

Consultations revealed that information on police contact with individuals with mental health issues is being collected by some police services. The purpose of capturing these data, however, is varied and includes legal requirements relating to the mental health acts, operational needs (e.g., safety of attending officer) and data analysis. Given the differing goals underlying data capture, there is no standardization in the methodology or in the type of information captured. In all, eight types of data collection approaches related to mental health were identified as current methods of data collection. These included mental health act calls, calls for service, involuntary admission  of an individual  to a mental hospital, activity reporting systems, use of force reports, specialized data collection tools for emotionally disturbed persons, cross-sectional studies and the homicide survey. The majority of these data are stored electronically.

1. Calls for service – Apprehensions under the Mental Health Act

In general, provincial and territorial mental health acts provide powers to police to apprehend a person when the officer has reasonable and probable cause to believe that the person is a threat to him/herself, a threat to others, or shows a lack of competence to care for him/herself. The action of compelling a physician's assessment and possible involuntary admission to hospital legally requires police to complete a mental health act apprehension form. These forms are submitted by officers to their respective police detachment.

Mental health act calls can also be recorded on the police service's Computer Aided Dispatch (CAD) system and the Records Management System (RMS). The Royal Canadian Mounted Police (RCMP) specifically uses the Police Reporting and Occurrence System (PROS) to score mental health act calls for service as "Mental Health Act Non-Criminal" and "Mental Health Act - other activities". This reporting, which is recorded electronically, includes all mental health act calls for both adults and youth. This information is then used to track the specific time requirements and workload issues. Known as the Police Resourcing Methodology, the RCMP and Ontario Provincial Police (OPP) use the system to indicate the prevalence of mental health act events per detachment and zone, as well as time spent on the call.

2. Flagging calls for services as mental health

For some police services, when calls are received through emergency dispatchers, the dispatcher may enter a mental health flag in the Computer Aided Dispatch (CAD) system and for some police services, this information is automatically transferred into the Records Management System (RMS). The primary objective is to alert attending officers of the nature of the current incident, as well as any future incidents involving the same individual. This operational type of information, however, has its limitations for data analysis in terms of the type of data capturing system and the way in which the mental health flag is maintained.

Within the CAD system, a number of calls can be entered for a single incident. For example, separate individuals can place calls into the police station for the same incident, all of which can be recorded by the dispatcher. This could lead to an exaggeration of the volume of mental health flags. The CAD system, however, has the advantage of capturing all calls for services, including non-criminal activities. This may help to inform police workload issues.

In terms of reporting mental health flags, there is an absence of consistent reporting across jurisdictions and by individual dispatchers and officers. This prevents a reliable assessment on the actual prevalence of mental illness among all calls for service. Furthermore, the flag often does not undergo updates based on the outcome of the call. It may not be removed if it is determined that the incident was unrelated to mental health issues and alternatively, a mental health flag may not be added when a mental health issue is determined after police appearance and investigation on the scene. Depending on the police service, this absence of updates may be due to the authorization level required for modifications. Specifically, for some police services, individual officers cannot make changes to the records and can only be modified by the records manager.

3. Involuntary admission to a mental hospital

Information on involuntary admissions to mental hospitals is forwarded to the Canadian Centre for Justice Statistics (CCJS) as part of the Incident-based Uniform Crime Reporting Survey (UCR2). Police services are mandated by the Statistics Act to report if an accused person is not available for prosecution because: a) they are in a mental institution without the hope of early release or b) as per conditions set by the court or Review Board under C.C. 672.54(b).

4. Activity reporting system

Information on mental health may also be recorded in the activity reporting system, which describes the activities of an officer during the course of his or her shift. A duty code related to mental health, however, it is not consistently reported. Also, if an offence occurred, the offence code takes precedence over the mental health code. A secondary code of mental disorder will not be captured.

5. Use of force reports

All police services are required to complete a "use of force" form when force is used by officers against a suspect. This form includes information on persons designated as emotionally disturbed. This data source for individuals with mental illness is limited to a small number of cases and would not serve as an indicator of prevalence.

6. Specialized data capture for mental health calls

Among police services consulted, some indicated that data on mental health cases are maintained on a regular basis. An "emotionally disturbed persons" (EDP) form was the most common data collection tool. Although there are variations in the form depending on the police service, the form typically contains observational information on the incident and the individual, the individual's thinking (e.g., displaying disorganized thinking, abnormal speech, and odd beliefs), the type of behaviour exhibited, dwelling information, and substance use. In addition to differences in the type of information collected, police services also vary in when EDP forms are completed. For some, the forms are completed for all interactions involving persons with a suspected mental illness. Others only complete the form when the police officer is apprehending the individual under the mental health act or when the officer is referring the individual to specialized crisis intervention teams, which are teams comprised of police officers and health workers.

Besides EDP forms, a few police services reported tools for capturing information on the mental health of contacts. Again, the type of information is based on observations, as opposed to mental health diagnosis. One division of the RCMP currently uses a standardized Police Template that provides physicians/nurses with data from mental health calls for service.

7. Cross-sectional studies on police contact involving individuals with mental illness

In addition to ongoing data collection, some police services have undertaken special one-time studies to determine the extent of police contact with individuals suffering from mental illnesses. For example, the Vancouver police service conducted a 16-day prevalence study with a sample of police officers. Using a paper-based system, officers indicated whether the incident, both criminal and non-criminal, was related to an individual's mental health condition. This type of study captured the prevalence but did not examine characteristics of the incident or individual.

8. The Homicide Survey administered by the Canadian Centre for Justice Statistics

The Canadian Centre for Justice Statistics manages the annual Homicide Survey, which collects information on all homicides in Canada. All police services are mandated to provide incident, victim and accused information on homicides. Within the accused questionnaire, police services must indicate if the suspect was suffering from a mental or developmental disorder at the time of the homicide. The variable is based on the investigating officer's assessment and does not need to be based on an assessment or diagnosis of a medical professional. In addition, the information is not subject to release under the Access to Information Act.

This type of status information on the mental health of the offender is not currently captured with the Incident-based Uniform Crime Reporting Survey. This police-based survey, which is also managed by the Canadian Centre for Justice Statistics, records a variety of detailed accused, victim and incident information for all crimes reported to police and substantiated through police investigation.

Feasibility of data collection involving the police

Those consulted were asked to provide feedback on a possible approach to gathering information on police contacts involving individuals with mental health issues. Specifically, all participants, including police and experts who felt they were in a position to comment on the policing sector, were asked about the feasibility of conducting a survey with a sample of police services. It was proposed that the study would be conducted for a period of time during the year, whereby the sample of police services would be asked to complete a one-page paper survey on incidents involving persons with mental health issues.

While stakeholders expressed a number of challenges with the proposed data collection, nearly all participants indicated that some type of data collection involving the police would be feasible. Their views differed on the actual approach. The challenges and suggested modification to the approach are presented below.

Challenges in proposed approach

The most frequently expressed concerns of any form of data collection were related to respondent burden and defining mental illness. Participants indicated that front-line officers are already over-burdened with calls for services and work within limited resources. A few stakeholders further indicated that police are over-studied. Some police services were apprehensive that an additional survey would duplicate existing data collection efforts.

It was also expressed that completing a survey on mental health contacts would be time consuming and would heighten officers' workload. As a result, individual officers may be unwilling to participate. Integrating the survey within existing records management systems was identified as a possible way to reduce respondent burden and to monitor changes in mental health contacts over time. Specifically, it was suggested that a drop-down mental health template screen be added to Record Management System (RMS). This would ensure standardization across all police services.

Determining a practical and reliable definition of mental illness was seen as critical for successful data collection. For some stakeholders this also meant providing training for officers, such as training in diagnostic tools. Among academic stakeholders, it was felt that researchers should accompany police officers, while for police stakeholders, partnering between police and health workers was seen as most beneficial to collecting data on individuals with mental health issues.

Any new data component would have to be simple and reflect the expertise of police personnel, namely questions based on observational/visual cues. A few of those consulted noted that mandating data collection would increase reliability of the data; that is, requiring completion of the survey under the authority of the Statistics Act.

To further ensure quality data capturing, many stakeholders recommended proper monitoring of data collection and follow-up, particularly at the initial stages of implementation. This data quality step could be done by researchers or the records management team. The importance of accurate and appropriate data analysis was also stressed during consultations.

It was also proposed that a police mental health survey should be representative of urban and rural police services, as well as police services with and without specialized mental health crisis teams. The challenges for these latter police services, such as access to community resources and action taken would be different. Others mentioned the importance of data capture at different times of day and during the year to account for variations.

Other concerns for data collection included consistency of reporting in larger detachments, as well as between individual officers with differences in experience and training.

Modified data collection approach by the police

Based on the consultations, it was determined that a data collection strategy relying on a paper-based survey would result in a high respondent burden, especially if implementation involved all police services in Canada. Such an approach would also require site management of paper forms prior to processing by the Canadian Centre for Justice Statistics. This would be particularly challenging for police services without specialized crisis intervention teams.
 
Consultation participants' feedback suggests that future data collection should look toward a long-term solution by integrating questions within a standardized and existing crime survey. Realizing this vision would essentially require working within the records management systems and modifying the existing Incident-based Uniform Crime Reporting Survey (UCR2). In doing so, data on prevalence of incidents involving emotionally disturbed persons in Canada would be available on an annual basis, along with data on the detailed characteristics of the incident, victim and accused.

The disadvantage of using the UCR2 survey would be the inevitable exclusion of non-criminal incidents, since the UCR2 program is not mandated to record other non-criminal calls for service. In other words, a crime must have occurred for any information to be recorded.

While the integrated approach would seemingly have less respondent burden for the police, previous UCR2 data elements that relied on police observations and interpretation have been plagued by reporting problems and questions of data quality. For example, earlier versions of the UCR2 survey contained a data element on alcohol/drug consumption to understand the prevalence of drug and/or alcohol consumption by accused persons. The criterion for police completing this information for the UCR2 Survey was police observations of apparent alcohol or drug consumption prior to the time of the incident. Despite an identified need for this information, this variable was rarely completed by officers for two reasons. First, the police were not always able to determine through observation if the accused had consumed drugs, alcohol or both. Second, there was concern among officers that the information based on observation could subsequently be used or challenged in a subsequent court case. The problems led to the eventual removal of this data element from the UCR2 survey.

Considering the observation-based similarity between alcohol/drug consumption and emotionally disturbed persons, it is likely that introducing a variable to measure emotional disturbance and indicators of possible mental health issues could suffer the same data quality issues. In addition, to accurately document the indicators of emotionally disturbed persons, a new observation-based data element on the UCR2 would require numerous data fields relating to appearance, thinking, mood, behaviour, dwelling/housing and personal hygiene. This is because a range of indicators, interpreted as a whole, are needed to ensure the accuracy of the accused status as emotionally disturbed. The number of required data fields would be substantially higher than any current data element on the UCR2 survey. This would further heighten the possibility of incomplete data.

Moreover, in some police services, the information that is entered on Records Management Systems is actually completed by records management personnel who review narratives written by the attending officers and then complete the required UCR2 data fields. As such, for these police services, the scoring of a UCR2 variable on emotional disturbance would require the interpretation of the records management personnel and, depending on the depth of the narrative taken by police, may not include the details necessary to accurately score such a variable. Again, this runs the risk of incomplete data or compromised data quality.

Therefore, notwithstanding the desire for integration within the existing RMS, the feasibility of successful implementation within the UCR2 is questionable. In other words, the UCR2 does not appear to be the appropriate tool for future data collection on individuals with emotional disturbance or possible compromised mental health and their contact with police.

Possible options for future data collection

1. Option for data collection by police

To successfully implement a future data collection tool by the police, it is key to consider current approaches that have been developed, tested and implemented. To date, police services that have been systematically collecting mental health data rely on "Emotionally Disturbed Persons" (EDP) forms. This tool can include both criminal and non-criminal contacts with police. While EDP forms are similar in design across police services, their development has been specific to the needs of particular police services and the community. There are variations in the types of information collected and the way in which the information is collected and stored.

Facilitating questionnaire standardization across police services that currently use EDP forms could be a short-term goal, which could be accomplished through collaboration between the participating services. Such collaboration could examine possibilities for consistent methods of data capture and storage, which in turn could lead to comparable data. In the end, standardization in questionnaire design and data storage may permit comparison of police contact with emotionally disturbed persons across police services and over time.

Collaborating police services may also consider options for sharing EDP data publicly to provide stakeholders, including other members of the justice community, governments, the health sector and the general public with information on the prevalence and nature of police contacts with emotionally disturbed persons. In the long-term, extending expertise and knowledge on data collection with other police services, namely those with an interest in mental health data collection could be considered.

One option is for police services currently using "Emotionally Disturbed Persons" forms to collaborate to standardize the data collection tool, as well as data capture and storage systems.

Once standardized, police services using the EDP forms could make this form available to other police services to assist with increasing the collection of standardized data.

2. Option for data collection using a household survey

A household survey of the general Canadian population would facilitate an examination of the mental health status of Canadians, juxtaposed with their previous contact with the criminal justice system, including the police. It could also capture both criminal and non-criminal contacts with police. Ideally, the household survey would be developed based on an existing survey tool that has already been tested and implemented in the field. One common limitation of household surveys is the exclusion of the homeless population and the population in institutions, such as prisons and hospitals.

In 2002, Statistics Canada as part of its Canadian Community Health Survey (CCHS) conducted a cycle devoted to aspects linked to the mental health of Canadians. Cycle 1.2 named the "Canadian Community Health Survey-Mental Health and Well-being" interviewed Canadians aged 15 years and older living in private occupied dwellings in the ten provinces, representing approximately 98% of the population aged 15 and older in the provinces. The homeless and institutionalized population were excluded from the survey. One of the primary objectives of the CCHS Mental Health and Well-being survey was to provide cross-sectional estimates of mental health determinants, mental health status and mental health system utilization across Canada. For the first time, this survey was able to provide provincial level estimates of past 12-month and lifetime prevalence of mental disorders in the population, as well as information on the utilization of mental health services.

Given this survey's demonstrated ability to capture the extent and nature of selected mental disorders in Canada, developing and integrating questions on previous contact with the criminal justice system within such a survey would enable an understanding of both criminal and non-criminal contacts of mentally ill individuals with the criminal justice system. Indeed, such an approach would require examining the feasibility of adding new questions to the survey without unduly increasing the respondent burden, as well as assessing the necessity of existing survey questions on non-mental health measures. That said, it is expected that adding a module on the criminal justice system to such a survey would only require the addition of key questions on previous contact with different components of the system.

Specifically, new questions on criminal justice contact could be modelled on questions contained in another household survey, namely the General Social Survey (GSS) on Victimization. The GSS asks a series of questions on criminal justice system contact in relation to individuals' views of the system. For a future mental health survey, these contact questions could be asked and then analyzed alongside the mental health status of respondents. The following police contact questions from the GSS could be considered.³

During the past 12 months, did you come into contact with police:
…for a public information session

During the past 12 months, did you come into contact with police:
…for a traffic violation

During the past 12 months, did you come into contact with police:
…by being arrested

During the past 12 months, did you come into contact with police:
…as a victim of crime

During the past 12 months, did you come into contact with police:
…as a witness of crime

During the past 12 months, did you come into contact with police:
…for any other reason?

Other, please specify.

These questions could be posed regardless of whether the person had a profile consistent with mental illness. By doing so, comparisons of police contact could be made between individuals with mental illness in the last 12 months and those without any mental illness over the same time period.

For a mental health survey similar to cycle 1.2 of the CCHS to provide provincially representative data on the prevalence of police contact among Canadians with and without profiles consistent with mental illness, the size of the sample to be surveyed would need to quite large. If the sample size is too small, the size of the sampling error would be substantial and would, therefore, make the estimates too unreliable to publish. (The sampling error is the difference between an estimate derived from the sample and the one that would have been obtained from a census that used the same procedures to collect data from every person in the population.)

Consideration of the sample size is critical given that findings from the 2004 General Social Survey on Victimization with a sample size of approximately 24,000 showed that a small proportion of all respondents were arrested by police (1%). The size would decrease when the data are analysed by mental health status and when examining individuals who proceed to court.

Therefore, in order to produce national level estimates for mental illness and contact with the criminal justice system, it would be necessary to conduct a mental health survey with a large sample size, greater than the previous sample of 37,000 for cycle 1.2 (mental health and wellbeing) of the CCHS. This would improve the possibility of obtaining reliable estimates with low levels of sampling errors.

Another limitation to conducting such a survey to improve our understanding of the rate of contact with the criminal justice system among people with a profile consistent with mental illness is that the homeless and populations in institutions would be excluded due to operational and financial constraints.

It is recommended that, if a household survey on the mental health and wellbeing of Canadians is conducted in the future, questions on contact with police be considered for inclusion.

Criminal Courts and Review Boards

Objectives for data collection

According to participants from criminal court areas, including traditional courts and specialized mental health courts, as well as Review Boards, three objectives for mental health data collection were considered most important: 1) the need for greater public awareness on mental illness and the criminal justice system; 2) information-based policy and programming decisions; and 3) measuring workload and performance outcomes.

While similar to the goal of making information-based decisions, the chair of one provincial Review Board expressed the importance of evaluating legislation, notably the Criminal Code as it relates to mentally disordered offenders. Another Review Board chair indicated that data collection should aim to improve an understanding of mental health and the role of the Review Board among those working in the criminal justice system. Finally, some stakeholders from courts and Review Boards indicated that the issue of assessing resources in the community should be a separate objective of data collection.

Definitional issues

With respect to data collection at the court level, most stakeholders representing the criminal courts and Review Boards indicated that a definition of mental illness should be based on section 2 of the Criminal Code which defines "mental disorder" as a disease of the mind. This broad understanding of mental disorder is further refined in the court's consideration of mental disorder within the context of the Criminal Code provisions on fitness to stand trial and criminal responsibility.

It was argued that the legal test for fitness to stand trial and the defence of mental disorder differs from determining if the accused person is suffering from a mental illness. That is, not all persons who are mentally ill will are found unfit to stand trial or not criminally responsible on account of mental disorder. The legal determination of fitness and criminal responsibility is based on the results of an expert health assessment and diagnosis, rather than observational or experiential knowledge of the court. Furthermore, beyond a finding of mental illness, fitness assessments must indicate if the accused is not fully capable of instructing counsel or is not capable of understanding the nature and consequences of the trial. For criminal responsibility, the assessment must indicate if the person was incapable of appreciating the nature and quality of the act or omission or of knowing it was wrong.

Other stakeholders specifically listed behaviours or mental health conditions, such as cognitive disorders, addictions, and depression that should be used to determine mental illness. This broader application of mental disorder is often used following a court decision to inform sentencing and conditions. It may be used when considering pre-sentencing reports, the use of stays and withdrawals, court-sponsored diversion programs, and Review Board orders (e.g., addiction treatment). The mental health conditions tend to be less severe than the thresholds used to determine fitness to stand trial and the finding of not criminally responsible on account of mental disorder.

Priority issues and information needs

A range of stakeholders, including courts, provincial/territorial Review Boards, provincial/territorial and federal departments responsible for justice matters, academics, and non-government associations provided feedback on the issues affecting courts with respect to youth and adults with mental health issues. In most cases, these stakeholders indicated that the data would be most useful at the provincial level, followed by intra-provincial regions and the national level.

1. Prevalence of accused with mental illness in the court system

Some court stakeholders indicated that based on their own observations, they had witnessed an increase in the number of accused with mental health problems appearing in court. It was suggested that along with a growth in the number of fitness hearings and applications for the defence of not criminally responsible, there had been an increase in the volume of accused persons who were mentally ill but did not reach the legal threshold for unfit to stand trial or not criminally responsible.

To quantify this assumption of an increase, stakeholders identified the following data needs:

• Prevalence of fitness to stand trial and not criminally responsible on account of mental disorder, including calculating rates per 1,000 population;
• Number of accused persons with mental illness appearing in court over time;
• Number of accused persons with mental illness found guilty and their sentence type;
• Types of mental illnesses among accused persons with mental health problems; and,
• Number of accused persons on mood-altering drugs.

2. Forensic assessments

The reliance on court-ordered assessments, as opposed to visual or experiential knowledge was the key reason for the identification of forensic assessments as a priority issue for courts and Review Boards. Consultation participants argued that there are limited resources for forensic assessments. Specifically, a lack of forensic specialists and psychiatrists to adequately deal with the current caseload for assessments was linked to longer wait times for assessments and subsequent longer lengths of remand for accused waiting to be assessed. In addition, it was argued that there are unacceptable delays in getting accused into hospital for court-ordered assessments and treatment. This was attributed to under-resourcing of 'designated hospitals', as defined in Part XX.1 of the Criminal Code.

The consequences of assessment delays can be far-reaching. Participants indicated that delays can lengthen the elapsed time of case processing, reducing the overall efficiency of the criminal court system and impacting the right of the accused to have their case processed through the court system in a timely manner. It was suggested that data could help inform decision-making on this matter.

Some stakeholders expressed the view that there had been an increase in the volume of referrals for forensic assessments, which has had an effect on the costs of administering the court system. The perceived increasing volume of forensic assessments is also seen as an issue for non-forensic community-based assessment. Court-ordered referrals for assessments were felt to pre-empt assessments for individuals in the community who are not in contact with the criminal justice system.

In terms of informing the issue of court-ordered forensic assessments, stakeholders indicated the following data needs:

• Number and type of court-ordered forensic assessments for adults and youth over time;
• Percentage of forensic assessments that are false positive (e.g., an accused person is deemed unfit to stand trial through an assessment but in fact, the assessment is incorrect and they are fit);
• Number of times an accused person is sent for forensic assessments; and,
• Number of forensic assessments ordered before and after the 1992 Criminal Code amendments.⁴

3. Community resources and social supports

Among stakeholders, there was the suggestion that a gap in community services and social supports can be linked to a greater number of adults and youth with mental illnesses appearing in court. In other words, it was argued that individuals suffering with mental illnesses may not receive proper treatment and support in the community, which brings them in conflict with the law. Stakeholders expressed concern that once in contact with police, individuals who could have been diverted away from the criminal justice system are often charged and appear before court, especially in cases where police do not feel comfortable releasing mentally ill individuals back into the community and are unable to locate suitable community alternatives.

The perceived absence of community resources, including mental health treatment, housing and social supports, also extends to supports following court appearances. Some of those consulted expressed the need for better follow-up and community support after court release, such as in the case of absolute discharge, to limit re-contact with the criminal justice system. There was a sense among participants that services dwindle following the persons' involvement in the criminal justice system.

At the Review Board level, a leading priority issue involved access and availability of resources to support the Review Board mandate. Appropriate treatment and care for mental health conditions were identified as challenges given a perceived under-staffing of psychologists and case managers to work with forensic patients. Consultation participants further suggested that the scarcity of forensic resources also requires Review Board clients and their families to travel long distances. Moreover, the remote location of some psychiatric facilities was argued to contribute to difficulties in accessibility.

As with the court system, social and housing supports is a concern for the Review Boards. According to stakeholders, there are often difficulties in community placements for forensic patients. This was explained by the fact that special care homes are owned privately and in turn, it can be difficult for patients to pay accommodation costs.

To address these priority issues, stakeholders identified the following data needs:

• Proportion of accused before courts and Review Boards who had previous contact with community resources;
• Social determinants of mental health, including homelessness, education, substance abuse, lack of employment opportunities;
• Socio-demographic and offence differences between individuals under Review Board supervision who lack mental health treatment and those who refuse mental health treatment; and,
• Outcomes after Review Board discharge, such as progress and repeat contact/recidivism.

4. Court decisions on bail for persons with mental illness

As with all accused persons, criminal courts must consider whether an accused person with mental illness should be held in custody while awaiting further court appearances or be released in the community on bail or their own recognizance. Those consulted indicated that mentally ill individuals may be more likely than other accused to be remanded into custody. The perceived higher likelihood of remand was attributed to waiting times for assessments and forensic beds. That is, accused persons with mental health problems may be held in remand while awaiting psychiatric resources. Some stakeholders indicated that this is particularly an issue for youth accused. Depending on the waiting times for services, the time spent in remand can be lengthy.

In addition, participants felt that accused with mental illness often do not have sufficient community, social and/or family supports. With an absence of supervised bail programs, stakeholders indicated that accused persons with mental health illnesses are denied bail and are remanded to custody. They are deemed flight risks.

To inform the issue of the perceived over-use of remand for mentally ill accused, stakeholders indicated the following data needs:

• Number of accused persons with mental health issues remanded to custody, compared to accused persons with no mental health issues.

5. Structure of traditional criminal courts and mental health courts

Stakeholders identified differing priority issues for traditional criminal courts and specialized mental health courts. Many times, the priority concerns related to traditional criminal courts were non-issues for mental health courts, since these same issues were specifically addressed in the development of the specialized courts. In other words, mental health courts are designed to address the unique needs of accused with mental illnesses and therefore, do not always face the same challenges as other criminal courts.

Non-criminal court stakeholders expressed the view that traditional criminal courts are not well informed about the mental health services available in the accused persons' community. This appears to contrast the operations of mental health courts. For example, mental health court workers stated that Crown prosecutors with the Ontario mental health courts are responsible for finding a physician in the community to treat the accused with mental illness.

Another example of a priority issue affecting traditional courts that did not seem to affect the specialized courts is legal representation. Some stakeholders indicated that the refusal of legal aid for accused with mental illnesses clogs the traditional court system, making it inefficient. Mental health courts, on the other hand, are often tasked with finding a lawyer to represent the accused with mental health issues; thereby making it a non-issue. However, some consultation participants argued that the ability of the mental health courts to provide these additional services can be difficult with insufficient funding and inadequate staffing.

Both types of courts are also limited to the parameters of the Criminal Code and the Youth Criminal Justice Act (YCJA). For example, one stakeholder argued that the YCJA places limits on sentencing for youth with mental health issues, as sentences must be offence-based not risk-based.
 
To better understand the benefits and limitations associated with traditional criminal courts and specialized mental health courts, the following data needs were suggested:

• Need for evaluation of mental health courts;
• Identification of components of mental health courts that are successful (e.g., greater access to forensic beds, more resources, operation structure); and,
• Outcomes of traditional criminal courts and specialized mental health courts for accused with mental health problems.

6. Information sharing between sectors

To properly coordinate services and resources for accused persons with mental health problems, stakeholders consistently stated the need for a collaborative approach involving all relevant sectors. This is not strictly limited to justice sectors but also encompasses the education and social systems. The primary barrier to information sharing was related to privacy legislation and policies regarding confidentiality.

No data needs were identified for this priority issue.

7. Forensic patients under the authority of the Review Boards

Specific to the provincial/territorial Review Boards, priority issues centered on forensic patients' legal representation, their time spent under the authority of the Review Board and issues related to the severity of their offence.

Some Review Board chairs were concerned that the courts were referring too many individuals involved in minor criminal offences to the Review Board. It was felt that these accused persons might be better dealt with diversion and/or other programs. It was also argued that it was important to know the characteristics of individuals under the authority of the Review Board, such as the accused person's Aboriginal and immigrant status to fully understand the clientele and their needs.

The following data needs were identified as priority issues related to forensic patients under the authority of the Review Boards:

• Number of admissions to provincial/territorial Review Boards;
• Average time forensic patients spend under Review Board supervision; and,
• Types of offences and demographic information on forensic patients (e.g., sex, Aboriginal status, immigrant status, country of origin).

Data availability

An examination of data availability for the courts and Review Boards must consider the data collection activities of traditional criminal courts, specialized mental health courts and Review Boards. For traditional criminal courts, the Integrated Criminal Court Survey (ICCS) provides standardization in the type of information captured.⁵ No such national data collection tool is specifically available for mental health courts and Review Boards.

1. Traditional criminal courts – Integrated Criminal Court Survey

Among the traditional criminal courts, those consulted indicated that very little information is currently collected on the mental health condition of accused persons, including both youth and adults. Most often, data on fitness hearings and the outcome of "not criminally responsible on account of mental disorder" are the two data elements that are recorded by the criminal courts. This information is then forwarded to the Canadian Centre for Justice Statistics as part of the Integrated Criminal Court Survey (ICCS). However, not all jurisdictions report these variables in the same way, and some jurisdictions do not submit complete data.

The ICCS, managed by the Canadian Centre for Justice Statistics, is a micro-data survey that collects detailed information pertaining to adults and youth who are processed through the criminal courts in Canada. At this time, coverage of the ICCS includes most jurisdictions, with the exception of Quebec adult courts and Saskatchewan courts.

1.a. Fitness to stand trial

The ICCS contains two data fields dealing with the accused person's fitness to stand trial: 1) the type of appearance and 2) the appearance result. The first variable, appearance type, indicates the reason for the court appearance (i.e., fitness hearing). This data element has the benefit of indicating how often the issue of fitness is raised during court proceedings. Reporting on this variable to the ICCS is high, as only Quebec youth court does not currently report when the code value is a fitness hearing.

However, there are issues of consistency in reporting on the "appearance type" variable. First, there is a possibility that courts may not submit data on fitness hearings if during the hearing, the accused person is found fit to stand trial. Second, not all jurisdictions will report fitness hearings if there was more than one reason for the appearance. Some jurisdictions do not have the capacity to report multiple reasons for an appearance. Finally, some jurisdictions report the scheduled reasons for an appearance, while others report the actual reason. These reasons can differ.

The ICCS appearance result variable indicates the result of the court appearance. In the case of fitness to stand trial, there are two code values: decision to find the accused fit and the decision to find the accused unfit. The results may not be considered final since an accused may be initially found unfit to stand trial but with medication to treat the mental illness, the same accused person can be later determined fit. The decision of "unfit to stand trial" is more consistently reported than "fit to stand trial". In particular, nine jurisdictions report the code value of unfit to stand trial, while only four jurisdictions report data on the code value for fit to stand trial. The remaining jurisdictions either do not report any data on these code values or report information as a residual value code under another data field (i.e., "other" code value under Decision variable), making it analytically unusable.

The main explanations for the inconsistency in reporting across jurisdictions are due to variations in data capturing systems and the identified data collection priorities. The data capturing systems were originally developed by jurisdictions for operational purposes prior to their participation in the ICCS and as such, the systems are not always setup to capture all ICCS variables. Furthermore, despite a variable being listed as an ICCS national data requirement, the jurisdictions must first identify this type of information as a priority before the courts will take steps to collect the data. In some cases, mental health variables are not considered priorities for the day-to-day operations of the court. For example, for the appearance decision of "fit to stand trial", there is often a lack of willingness to collect the information, since operationally, the courts are only concerned if the case can proceed.

1.b. Defence of not criminally responsible on account of mental disorder

Another set of data elements on the ICCS deals with the defence of not criminally responsible. As with fitness hearings, there is an indicator when the defence of not criminally responsible is raised during court proceedings. Under the data field "appearance type", the courts can indicate if an application for the mentally disorder defence was submitted to the court. However, all applications, including those for mentally disordered offender, dangerous offender status and publication bans, are not yet released as individual appearance types. The consistency in reporting application values within the appearance type data field has to be evaluated and better understood.

Another variable, "type of decision", has better quality information on the defence of mental disorder. Nearly all jurisdictions, with the exception of Quebec youth court and Yukon courts, report whether the decision was an acquittal on account of mental disorder. With this variable, it is possible to determine the volume of decisions of not criminally responsible along with the number of not criminally responsible offenders sent to provincial/territorial Review Boards. Case characteristics, such as the offence type, age and sex, can be analyzed alongside the decision of not criminally responsible.

1.c. Other ICCS variables on mental disorder

In addition to fitness hearings and the mental disorder decision, the ICCS contains a variable relating to medical/psychological/psychiatric reports ordered by the courts. This variable can be considered a strong indicator of the volume of court-ordered assessments. Most jurisdictions comply with reporting requirements on this variable, with the exception of Manitoba. Beyond the data collected by the ICCS, some jurisdictions indicated that they collect and internally retain data on pre-sentencing reports, which may contain information on programming recommendations for offenders who are suffering from mental health conditions.

Following a conviction, an offender with mental health issues may receive a sentence other than custody, probation, fine, restitution/compensation, conditional type sentences, discharge, and suspected sentence. Under the ICCS data field "other sentence type", courts can indicate if the offender was issued a hospital order. However, the majority of jurisdictions do not report any data on this code value.

A final ICCS mental health variable strictly deals with youth. Under the Youth Criminal Justice Act (YCJA), courts can sentence serious violent young offenders suffering from a mental or psychological disorder to a specialized custodial sanction. Known as the Intensive Rehabilitative Custody and Supervision order (IRCS), this custodial sanction is currently captured under the ICCS. Only one jurisdiction, Saskatchewan, does not report any data on this decision, since Saskatchewan youth courts do not currently report to ICCS.⁶

2. Specialized mental health courts

When data beyond fitness hearings and criminal responsibility outcomes are collected by mental health courts, most stakeholders indicated that they record the number of cases heard and the number of court appearances in the specialized court. Some mental health courts collect additional data on the outcome of court proceedings, diagnosis, diversion (if any), and residence before and after court appearance.

The main data collection challenge for mental health courts is accurately distinguishing mental health courts from other criminal courts. This is because mental health courts are not always dedicated facilities. Similar to other specialized courts, mental health courts take place in courtrooms with multiple uses and are therefore difficult to identify. Multi-use courtrooms also prevent analysis of differences in appearances and outcomes within the Integrated Criminal Court Survey (ICCS). While the ICCS currently allows jurisdictions to indicate individual courtroom numbers that can then be mapped to the type of specialized court, no jurisdictions currently provide this information.
 
During consultations, it was indicated that at least one jurisdiction was conducting its own evaluation of mental health courts and trying to overcome the identification challenge.

3. Review Boards

Provincial and territorial Review Boards do not have a standardized approach to data collection for individuals with mental health issues. Variations exist in the type of data collected and the method of data collection and storage. Review Boards with small caseloads tend to keep paper-based administrative files or transcripts of proceedings for each individual case. This information is retained by Review Board members until the forensic patients are discharged. More comprehensive data on forensic patients are sometimes retained by forensic hospital facilities. For example, the forensic facility in Nova Scotia has a database on all unfit to stand trial cases, as well as cases of not criminally responsible on account of mental disorder.

Among Review Boards with larger caseloads, such as British Columbia and Ontario, electronic stand-alone case management databases are maintained. Some of the data fields contained within the British Columbia database include: number of active cases, number of cases closed, caseload by verdict (not criminally responsible on account of mental disorder and unfit to stand trial), reasons for case closures (absolute discharge, sent out of province, appeal verdict, death), number of hearings, and demographic information.

Privacy and confidentiality were raised as possible issues for sharing information with the Canadian Centre for Justice Statistics.

A new standardized survey of Review Boards could address the different reporting practices across Review Boards. This administrative survey could be completed for each Review Board hearing and could include hearings for all persons admitted prior to and during the start of data collection. This approach would allow for the production of both admission and count data. Administrative forms would be completed by Review Board members.

Despite the benefit of this approach in providing ongoing data for the examination of trends in prevalence and case characteristics, it does not seem feasible in the immediate future. When consulted regarding the feasibility of this data collection approach, Review Board members perceived the research design as too onerous, since it was felt that Review Board members, as opposed to other staff members, would need to complete the survey to ensure accuracy. Review Board members indicated that they are unable to add any additional tasks to their regular workload.

In order to make the approach possible, there would be a need for ongoing funding to employ a trained researcher for each review board. These researchers would be responsible for data collection and submission to the Canadian Centre for Justice Statistics (CCJS). Ongoing funding would also be needed for CCJS to process, analyse and disseminate the data. There are currently no funding sources for a data collection endeavour of this magnitude.

The National Trajectory Project (NTP) is a project currently underway and funded through the Mental Health Commission of Canada and aims to examine the operation of current criminal justice provisions for individuals declared not criminally responsible on account of mental disorders (NCRMD) who are under the authority of a provincial or territorial review board. It will examine the antecedents and course of accused who end up in the Review Board System. The study will include the three largest provinces (Ontario, Quebec and British Columbia), which comprise the majority of NCRMD cases. The project is based on an ongoing study in Québec funded by the Fonds de recherche en santé du Québec. 

The project is comprised of three arms: a quantitative arm, a qualitative arm and a legislative review. The objectives of the quantitative arm of the National Trajectory Project are to: 1) Explore the demographic, psychosocial, and criminological profile of people declared NCRMD in Canada, as a function of geographic region and type of institution of detention; 2) Evaluate the importance and systematization of assessments of risk of violence presented to the Review Boards; 3) Report the rationale for decisions made by the Review Boards as a function of the disposition rendered (absolute discharge, conditional discharge, or custody); 4) Establish rates of criminal recidivism and psychiatric rehospitalization of discharged offenders as well as track positive outcomes; 5) Examine the migration trajectories or sedentary patterns of people declared NCRMD; 6) Identify the individual and organizational factors associated with these trajectories; 7) Determine the use and predictors of mental health services by this clientele prior to the NCRMD verdict, during the jurisdiction of the Review Board, and following discharge. An overarching objective will be to examine each of these findings with respect to culture and gender of NCRMD individuals. To do so, a file-based trajectory study of individuals declared NCRMD (examining mental health services received and criminality prior to being declared NCRMD, during NCRMD tenure and following absolute discharge from the provincial review board) is ongoing in Quebec. The study is currently being put in place in British Columbia and is being developed in Ontario. Based on the results, qualitative interviews with key stakeholders in the NCRMD and mental health services provision (e.g. persons declared NCR, caregivers, victims, family, review board members, psychiatrists and lawyers who have NCR clients, hospital administrators) will be carried out in order to contextualize the quantitative arm as well as report on current practices in various jurisdictions. Finally, a legislative review committee on the Part XX.1 provisions of the Criminal Code will be established. Legislative review of the history, development, implementation and consequences of the 1992 reforms to the mental disorder provisions of the Criminal Code will be done by legal experts familiar with these developments. The objective is to ensure that the National Trajectory Project maintains a practical focus on the operation of the current law. 

Possible options for future data collection

Based on the consultations, it would be advantageous to improve the consistency and coverage of the data being reported to the existing national micro-data survey - the Integrated Criminal Court Survey (ICCS). For Review Boards, one data collection option could fulfill some of the key data gaps. Finally, a household survey on mental illness and contact with the courts could provide information on accused persons with mental health issues, regardless of whether or not they were found unfit to stand trial or not criminally responsible on account of mental disorder.

1. Opportunities for improvements to Integrated Criminal Court Survey data

In considering adult and youth traditional criminal courts, the stated data need to determine the prevalence of individuals with mental health issues in the court system can be partly addressed with improvements to data that are reported to the Integrated Criminal Court Survey (ICCS). It is recognized that the full extent of the issue of mental illness in the courts cannot be solely based on fitness hearings, findings of not criminally responsible, hospital sentencing orders, and youth Intensive Rehabilitative Custody and Supervision orders. However, these court actions relating to mental illness can be useful indicators of court and Review Board workload issues.

While the ICCS has the capacity to measure the above indicators of mental illness, there is significant jurisdictional disparity in the way ICCS data are captured and stored. In addition, there are coverage limitations. Therefore, there is a need to promote consistency in ICCS reporting across jurisdictions and to increase overall survey coverage and coverage for ICCS variables relating to mental illness.

1.a. Promote consistency

Work undertaken to develop standards for capturing and storing mental health data, particularly as it relates to appearance types (fitness hearings) and appearance results (fit or not fit to stand trial), would be an important addition to the ICCS.

1.b. Survey coverage and coverage for mental health variables

Working towards full implementation of the ICCS, including participation from Quebec adult court and Saskatchewan courts, would increase survey coverage and would generate national level data.

Working with courts in Quebec (youth) and the Yukon to report the acquittal on account of mental disorder code value for the "type of decision" (DECISION) variable would add to available information on the defence of not criminally responsible on account of mental disorder.

Working with Manitoba courts to report data on court-ordered medical/psychological/ psychiatric reports (MPREPORT) would increase information on clinical assessments of accused persons.

Consultation results also revealed that data specific to mental health courts are not collected in a standardized manner. This is not surprising given the relatively recent emergence of specialized courts. To overcome the lack of standardization and allow for the possibility of comparison of court proceedings and outcomes between mental health courts and traditional criminal courts, it is important to consider the capabilities of the Integrated Criminal Court Survey (ICCS). As with any specialized courts, such as family violence courts and drug courts, ICCS allows jurisdictions to indicate a courtroom number associated with the specialized court.

Working with jurisdictions that currently have mental health courts to identify courtroom numbers would be useful in distinguishing mental health courts from traditional criminal courts.

2. Options for data collection involving Review Boards

The data collection gaps regarding individuals subject to the jurisdiction of a provincial or territorial review board suggest a need to consider data collection tools that could address the information needs expressed by consultation participants. Some of these key data needs include the number of admissions to Review Boards and the characteristics of offences and accused persons.

Based on the consultations, the following is a viable option for future data collection involving the Review Boards. The advantages and disadvantages of this option are described below.

Justice Canada to repeat their 2006 study and expand the number of provincial/territorial Review Boards included in the study

In the previous Justice Canada study, data were compiled on a one-time basis through manual extraction of administrative Review Board files from seven provinces and territories.⁷ A data collection form was used to collect information on a number of different fields, such as current criminal offence, criminal history, mental disorder and treatment information. Additional provinces and territories could be added to the study to ensure representation across Canada. Funding for the project would need to be secured.

The advantages of such an approach include:

• The possibility of comparing results with previous findings;
• The tools for data collection have been developed; and,
• The Review Boards' endorsement of a trained researcher studying case files,⁸ rather than Review Board members collecting information.

The disadvantages of the approach relate to the fact that the one-time study would not fulfil the need for ongoing data on the prevalence and nature of individuals under the authority of the Review Boards. In addition, the previous Justice Canada study did not include all provinces and territories. The lack of involvement from Review Boards in the Prairie Provinces precluded a complete understanding of Aboriginal issues. This could be addressed with an expansion in the number of participating Review Boards.

3. Option for data collection on courts using a household survey

With an adequate sample size, a general household survey on mental illness with questions on respondents' contact with the criminal court system might address limitations of the proposed data approaches involving the courts and Review Boards. These limitations relate to the inability of the Integrated Criminal Courts Survey (ICCS) and a questionnaire for the Review Boards to capture data on accused persons who were never sent for a forensic assessment, as well as those accused whose fitness or criminal responsibility was never questioned. The inclusion of questions to measure contact with criminal courts to a household survey on mental illness could make it possible to examine the prevalence of mental illness among all persons who came into contact with the courts. An example of such a survey, Statistics Canada's Cycle 1.2 of the Canadian Community Health Survey (CCHS), is provided in the section on police. While the inclusion of court-related questions in a household survey on mental illness may fill a data need not currently filled by the ICCS, a possible limitation is that populations who are in institutions or who are homeless are usually excluded due to operational and financial constraints.

As with the questions on police contact, new questions on criminal court contact could be modeled on questions contained in Statistics Canada's General Social Survey (GSS) on Victimization. Responses to these questions could be analyzed according to the mental health status of respondents. The following screening question on criminal court contact modeled on a GSS question could be included in a future mental health household survey.⁹

Have you had contact with the Canadian Criminal courts within the last 12 months?

If respondents answer 'yes' to this question, a follow-up question, which does not currently appear on the GSS, would be asked to understand the nature of the court contact. The question would ask:

Was this contact because: (check all that apply)
…you were on jury duty
…you were charged with a crime
…you were a victim of a crime
…you were a witness to a crime
…your friends or family members were charged with a crime, were witnesses to a crime, or were victims of a crime
…other reason (specify)

These questions could be posed regardless of whether the person had a profile consistent with mental illness. By doing so, comparisons of court contact could be made between individuals with mental illness in the last 12 months and those without any mental illness over the same time period.

With an adequate sample size, results of a mental health survey similar to cycle 1.2 of the CCHS, with questions on criminal justice contacts, might provide provincially representative data on the prevalence of court contact among Canadians with a profile consistent with mental illness and those without.

A significant limitation to this approach, however, is that it would be necessary to conduct such a mental health survey with a very large sample size to produce reliable estimates of contact with criminal courts among Canadians with a profile consistent with mental illness. This is because a small proportion of individuals overall are arrested and proceed to court (see section on police for full discussion). In addition, this data collection tool would exclude the institutionalized and homeless populations due to operational and financial constraints.

It is recommended that, if a general household survey on the mental health and wellbeing of Canadians is conducted in the future, questions on contact with the criminal courts be considered.

Correctional System

Objectives for data collection

As with other sectors of the criminal justice system, correctional service respondents were asked to provide comments on objectives for data collection relating to mental illness. Nearly all provincial, territorial, and federal correctional respondents indicated that data collection on mental health should primarily aim to assist those in the field and in policy to make information-based decisions regarding responses to the mental health issue, including program development. Specifically, those consulted indicated that data can inform treatment, rehabilitation and correctional intervention programs.

Also of interest were data that can help plan and allocate resources for correctional institution medical and psychiatric services. Less commonly expressed objectives were to improve public awareness regarding the issue of mental illness and to measure workload, performance and outcomes.

Definitional issues

Provincial/territorial and federal correctional system participants tend to prefer a broad definition of mental illness. For this group, a definition of compromised mental health or mental illness should include depression, suicide ideation, substance abuse, emotional disturbance, behavioural disorders, cognitive disorders, personality disorders, and permanent brain damage (e.g., Fetal Alcohol Spectrum Disorder).

However, stakeholders consulted disagreed on whether the definition should be strictly based on an actual health assessment/diagnosis or one based on observable or reportable behaviours. Proponents of a diagnostic definition of mental illness view the need for complete accuracy in assessing and treating mental illness. Relying on medical and psychiatric reports was argued to reduce the likelihood of over-predicting the incidence of mental illness in the correctional population.

On the other hand, most argued that the observation-based definition of mental illness overcomes the need for psychiatrists and forensic experts, which are often costly and not readily accessible to all offenders. Several participants commented that mental health treatment is based on symptom reduction irrespective of the actual diagnosis. For instance, not all people with schizophrenia experience the same symptoms and therefore, the treatment is not identical and will not necessarily be based on the diagnosis of schizophrenia. It was further argued that this observation-based definition better ensures that offender's mental health needs and problems are addressed in the correctional setting. Under this definition, self-injurious behaviour and isolation can also be captured.

Priority issues and information needs

Provincial/territorial and federal correctional systems, provincial/territorial and federal departments responsible for justice matters, researchers, academics, and non-government association participants suggested that in consideration of the structure of corrections in Canada, data would be most useful at the provincial/territorial and federal levels. The specific data needs and priority issues are described below.

1. Prevalence of mental illness among offenders

One of the leading priority issues for corrections pertained to an increasing number of adults and youth with mental health issues entering the correctional system. Respondents indicated that the growing prevalence, combined with the changing types of mental illnesses has required correctional services  to take measures  to meet the needs of this population.

In particular, those who participated in the consultation expressed concern over the number of individuals in correctional services with cognitive or brain disorders, notably Fetal Alcohol Spectrum Disorder, and challenges with respect to their rehabilitation. Individuals with dual diagnoses or co-morbidity issues were also identified as a challenge in terms of treatment, since, as it was argued, multiple mental health issues must be treated at the same time. Failing to address all mental health conditions was connected to a higher risk of re-contact with corrections. The most commonly identified co-existing mental health condition was substance abuse. Participants explained this association by a tendency for offenders to self-medicate the initial mental illness, especially when resources are unavailable in the community.

The following data were suggested as good indicators of the prevalence of mental illness in the correctional system, particularly at the provincial/territorial level:

• Number of youth and adults with mental illness in correctional systems and by type (s) of mental illness;
• Number of suicide incidents, attempted and completed;
• Number of offenders with substance abuse issues; and,
• Number of repeat contacts of offenders with mental illnesses.

2. Appropriate type of custody for individuals with mental illness

The appropriateness of different types of correctional placement for individuals with mental illness was raised as a priority issue during consultations. This issue was identified for both adult and youth offenders with mental health issues.

Aside from remanding individuals to await future court appearances, the additional reasons for remanding individuals with mental illnesses raised concerns among stakeholders. In particular, the practice of remanding offenders while waiting for diagnostic assessments or the availability of forensic beds was seen as problematic. It was conveyed that the special needs of the individuals with mental illnesses cannot be met while remanded because of the actual absence of treatment or the inability to provide treatment due to the accused person's indeterminate period of custody. A few stakeholders further commented that there has been an increase in the number of individuals with mental illness who have been admitted to remand custody.

For non-remand custody, it was reported that the lack of psychiatric units results in segregation of offenders with mental illness away from the general population. A few participants communicated concern that this isolation may further exacerbate existing mental health conditions. Others expressed concerns about the use of secure custody for young offenders with mental health issues in the absence of appropriate specialized forensic youth facilities. A greater co-ordination with children and youth services was seen as a possible solution.

To inform the custodial placement issue, consultation participants stressed the importance of the following data needs:

• Number of individuals with mental health issues in corrections, including remand and sentenced custody;
• Length of remand for individuals with mental illness; and,
• Location of remanded accused persons with mental illness – hospital versus custodial setting.

3. Mental health services in corrections

Closely related to the appropriate placement of individuals with mental illnesses was the overwhelming concern of participants with the level of mental health services in corrections, principally in provincial/territorial community and custodial corrections. Since each province and territory develops its own approach to address the needs of offenders with mental health problems, there are provincial/territorial differences in the delivery of programming and treatment. One academic argued that the lack of consistency in mental health services is especially pronounced between provincial/territorial and federal offenders. He attributed the lack of program standardization to provincial/territorial mental health acts, which do not set medical parameters for correctional institutions, as they do for hospitals.

Several consultation participants reported that offenders with mental health problems have a willingness to participate in treatment programs but unlike the hospital setting, the appropriate therapy is not always available in provincial/territorial corrections. The lack of mental health resources in custody was seen as creating problems with case management and planning, as well as an ability to prevent re-offending and return to corrections. A few stakeholders suggested that another consequence of the absence of sufficient services is the use of medication to control rather than treat offenders in custody. For young offenders in smaller provinces, it was mentioned that the absence of in-patient treatment programs results in referrals to services outside the youth's home province.

At the community corrections level, some consultation participants suggested better management of offenders with compromised mental health. The concern was primarily rooted in long wait times for treatment in the community and the refusal of non-voluntary referrals by mental health and addiction services in the community. It was felt that offenders on probation and parole were unable to receive timely and proper treatment, including diagnostic assessments. This was said to impact case management and planning.

Consultation participants wanted the following data to inform the issue of mental health services in corrections:

• Type of mental health and nursing services in correctional facilities compared to the community;
• Number of users of mental health services within corrections;
• Programming differences in provincial/territorial custody and federal custody;
• Identification of characteristics of offenders that increase probability of success in programming; and,
• Outcomes for offenders who received mental health programming.

4. Knowledge of mental illness by correctional services

Effective treatment of offenders with mental illnesses requires proper recognition of symptoms and problems. During consultations, it was argued that mental health disorders can be unidentified or misdiagnosed. This challenge in assessing and treating offenders was associated with an insufficient number of mental health experts within corrections, a lack of access to assessments, and lack of diagnostic procedures. There was a stated need to train corrections staff on mental illness and to determine indicators of mental health issues for both youth and adults, which would in turn, ensure more appropriate delivery of services.

To inform the issue, the following data need was expressed:

• Differences in persons receiving diagnosis in prison versus those diagnosed upon admission.

5. Continuity of mental health care before correctional admission and after discharge

The flow into and out of corrections and its impact on offenders' mental health treatment was raised. In cases where offenders were receiving support in the community, there was some concern that this same level of care did not always continue upon admission into the correctional system. Some academic researchers stressed that the mental health condition of offenders should not worsen while in corrections because of an inability to continue treatment while in custody.

Monitoring and providing treatment also extends to offenders' release from the correctional system. Many consultation participants argued that there is a lack of continuity of mental health care from the correctional system to the community, since the access to community services can be limited for individuals with criminal records. In addition, a lack of transportation to and from services, as well as proper housing was argued to impede the ability and willingness of offenders to participate in programming outside of corrections.

Many discussed the need for quality discharge planning, which would connect youth and adults to resources in the community. While a need for correctional services to work outside of institutions was identified, it was acknowledged that correctional services are not obligated or permitted to provide follow-up when offenders are released from correctional supervision.

To address the issue of continuity of mental health care, stakeholders identified the following data needs:

• Information on the state of the offenders' mental health before and after admission to corrections;
• Comparison of treatment before and after incarceration for both youth and adults; and,
• Presence of discharge planning and quality of support.

Data availability

The majority of data collection undertaken by correctional services occurs upon intake or soon afterwards. While the point of data collection is similar across all jurisdictions, wide variations exist in the type of information collected and the method of collection and storage. Some standardization exists with the Integrated Correctional Services Survey, which is managed by the Canadian Centre for Justice Statistics. This survey, however, is limited in its application to the issue of mental illness, since there are no strong indicators of the overall prevalence and characteristics of those with mental illness in corrections. A description of the variables within this survey, along with data collection conducted by the provinces/territories and the Correctional Service of Canada is described below.

1. Integrated Correctional Services Survey, Canadian Centre for Justice Statistics

The Integrated Correctional Services Survey (ICSS), managed by the Canadian Centre for Justice Statistics (CCJS), is a micro-data survey that collects detailed information pertaining to those who fall under the supervision of the youth and adult correctional systems. At this time, coverage of the ICSS includes six jurisdictions: Newfoundland and Labrador, Nova Scotia, New Brunswick, Ontario, Saskatchewan (adults only), and Alberta (community corrections only). Therefore, the results are not nationally representative of offenders with mental illness.

Within the data element "security concern", which identifies security or supervision concerns with respect to a specific inmate, correctional systems can indicate whether the offender was suffering from a mental illness and could not function in the general prison population. This data element is limited to inmates with mental illnesses who are displaying coping problems within the prison environment and not those inmates with mental illness who are not experiencing any outward adjustment problems. Also within the data element of "security concern", corrections can indicate symptoms not necessarily directly related to mental illness, namely suicidal tendencies and substance abuse. Substance abuse responses also appear in numerous other ICSS data elements, including needs assessment, conditions of sentence, and event type (attending treatment).¹⁰

2. Data within provincial/territorial information systems

Aside from the ICSS, provincial/territorial correctional systems collect their own information on mental health of offenders. Generally speaking, information on community corrections is collected by probation officers or case managers. This information is primarily used to assess individual offender's programming requirements and is stored in case files. Aggregate data are not consistently compiled and analysed.

In most provincial and territorial facilities, information is collected on suicidal risk and risk of re-offending and escape from custody. Depending on the province or territory, this information is either stored on paper files or electronically. Additional information, which may or may not be directly related to mental health, is collected by some provinces/territories and can include previous hospital admissions, level of previous mental health services, type of medications, substance use and abuse, number of assessments ordered, and history of family violence.

Medical units within custody maintain patient mental health records; however, these records are typically not stored electronically. Correctional clinical services also collect information on mental health services but data on the characteristics of offenders with mental health issues are not systematically captured.

3. Data on federally sentenced inmates – Correctional Service of Canada

Consultations revealed that nursing staff at Correctional Service of Canada (CSC) conduct an assessment of mental health indicators during the intake assessment.  This information is paper-based.  In addition, CSC is developing a computer-based screening tool that will provide indicators of possible areas of mental health concern.  

The nursing intake process involves assessment of mental health indicators, which are retained in the inmates' medical files. This information is more comprehensive than the Offender Intake Assessment and includes a set of detailed questions on suicide ideation, mental health history, current mental status, and mental health impression (e.g., physical appearance, mood and behavioural and emotional state). These records are paper-based. In addition to the information collected at the nursing intake process, other indicators of mental health were collected via questions asked by correctional and parole officers at intake.  These data were included in the CSC's offender management system.  However, in February 2008, a privacy audit concluded the collection of these data were in contravention of privacy legislation. 

In addition, the CSC is developing an individualized assessment tool, known as the Computerized Mental Health Intake Screening System (ComHiss).  Implementation began in September 2008 with the intention of implementing the system at all 16 CSC intake assessment sites. The psychological test assesses nine different indicators of psychological problems, such as psychosis and obsessive compulsive disorder, and three global indicators of psychological distress using the Brief Symptom Inventory (BSI). In addition, the Depression, Hopelessness and Suicide (DHS) Scale and the Paulhus Deception Scale are used to screen for dissimulation and life-threatening behaviours. While these indicators will not provide psychiatric diagnoses, they will provide indicators to psychology staff who can then follow up with arriving inmates. The BSI and Paulhus are available to psychologists from test suppliers, and the DHS is available from its authors.

Possible option for future data collection

Consultations with correctional systems revealed that despite numerous data needs, provincial and territorial correctional systems have few reliable, automated indicators of mental illness. The status of data may be due to the provisions of privacy legislation regarding the collection and storage of such information. Generally, the collection of such information must be completed by a qualified professional and access to the data must be restricted. This gap in the collection of mental health information provides an opportunity to underscore the need for the electronic collection of standardized indicators of mental illness which are collected according to the provisions of privacy legislation. A starting point to work toward this goal could be for the provinces and territories and the CSC to share information on the types of data they collect and identify how these data could be stored in an automated fashion that would abide by privacy legislation.

To successfully work toward standard questions to measure mental health issues, the Canadian Centre for Justice Statistics, CSC and the provincial/territorial correctional systems would benefit from collaboration to achieve the following tasks:

1. Share information across the federal, provincial and territorial systems regarding standard questions on mental health issues and discuss their advantages and disadvantages as basic indicators;
2. Identify how data could be collected and stored electronically without contravening privacy legislation;
3. Secure appropriate funding to move forward on automated, standardized data collection;
4. Examine the technical requirements needed to automate the collection of standard information;
5. Determine training requirements for intake assessment personnel;
6. Identify one or more provinces/territories to initiate implementation;
7. Aid the identified provinces/territories and their system vendors with the changes required; and,
8. Determine ways in which final data could be shared with the CCJS or ways in which analysis and findings from the data could be made available to the public and stakeholders in order to fill gaps in information.

Conclusion

Quantifying the issue of the involvement of persons with mental health illness in the criminal justice system presents a significant challenge. A combination of factors ranging from the lack of standard definitions, assessment and resourcing issues, operational requirements of the justice sectors, issues of privacy and confidentiality and challenges with information sharing, and current design of databases, all place challenges on efforts to quantify the issue. Nevertheless, sound statistics provide quality information to governments, the justice community, the health sector and the general public to understand and set priorities, and make policy decisions regarding the involvement of persons with mental illness in the criminal justice system.

Based on the results of this feasibility study, several options for data collection are presented for the criminal justice sectors that have the potential to generate useful and reliable statistics on the issue of mental illness and the criminal justice system. The options varied widely, from emphasizing data co-ordination and collection by policing services to improving data capture for an existing micro-data survey to creating new survey tools to address current data gaps. Implementing proposed options means that funding would have to be secured to move forward and partnerships would have to be developed and strengthened.

---

Notes

1. As of November 2008, the Canadian National Committee for Police/Mental Health Liaison has been disbanded.
2. All provinces and territories have versions of mental health acts. These acts provide direction on the treatment of individuals with mental health issues, including provisions relating to apprehension and detention/admission for psychiatric examination.
3. The addition of questions on contact with other criminal justice sectors is addressed in their respective sections.
4. The provisions of the Criminal Code that relate to Mentally Disordered Accused changed significantly in 1992 with the proclamation of Bill C-30. Before that time, law and policy regarding persons found not guilty by reason of insanity was not codified and detention was at the discretion of the Lieutenant Governor (Steller, 2003).
5. Within the ICCS, mental health courts are not differentiated from other criminal courts.
6. A small number of youth are sentenced each year to this type of sentence. For example, in 2006/2007, only 5 youth received an IRCS order (excludes Saskatchewan).
7. These seven jurisdictions were Prince Edward Island, Quebec, Ontario, Alberta, British Columbia, Nunavut and the Yukon.
8. Based on consultations with Review Boards .
9. The addition of questions on contact with other criminal justice sectors is addressed in their respective sections.
10. The Adult Correctional Services Survey, which is also managed by the Canadian Centre for Justice Statistics, contains aggregate data on correctional systems in Canada. All provinces/territories report to this survey. The survey includes a data element on suicides within and outside of custody.