FORUM on Corrections Research

Psychological intake assessment: Contributing to contemporary offender classification

Systematic case-based risk/needs assessment² has become a cornerstone for identifying treatment needs of offenders, assisting decision making by providing risk assessments and management strategies and, more recently, informing policy makers in the Correctional Service of Canada. Surprisingly, while the expertise for the development of such assessments has principally been provided by psychologists, their clinical practice within the Service appears to be determined by individual preferences. Psychologists' varying familiarity with the literature, their background and professional training may account for this apparently unsystematic approach to assessment. This is not to imply that specific institutions have ignored the need for such development, but across institutions there has been little agreement as to how to make the specific guidelines reflected in Commissioner's Directive 840 and other related documents, such as the National Parole Board policy guidelines and the Mental Health Task Force, operational. This article describes the development of a Psychological Intake Assessment (PIA) protocol intended to address these concerns.

Over the last two years, specific changes have occurred to set the stage for the development of a contemporary assessment protocol for psychologists. First, Correctional Service of Canada psychologists authored Forensic Psychology: Policy and Practice in Corrections,³ a clinically oriented text which described best practices for psychologists working in correctional settings. Second, these psychologists facilitated changes to the referral criteria for psychological assessments, reflecting offender case needs and mental health concerns. These changes are now part of the revised Commissioner's Directive (CD) 840, which governs the delivery of psychological services in the Correctional Service of Canada. Third, these initiatives provided the backdrop for the development of standards of practice, which inform psychologists about the important questions and content areas involved in the various types of assessments needed by the consumers of psychological services. Finally, the success of the Offender Intake Assessment (OIA) implementation (see Motiuk this issue) indicated that there was a need to incorporate computerized standardized psychological assessments into the intake process. As a result, the Research Division decided to develop the PIA protocol as a pilot project in the intake units.

Before developing the specific content of the PIA, it was important to:

• consider how it would integrate with OIA, (avoid marked duplication of services and information gathering);
• identify overall content areas consistent with OIA and employ consistent terminology;
• contribute value-added information to the intake process, with specific emphasis on offender needs and risk; and
• provide a process for mental health screening.

Furthermore, the information in PIA needed to be organized into a database to facilitate individual and collaborative research initiatives by the intake unit psychologists, in addition to Research Division requirements. It was also important that the PIA not be prescriptive regarding how the information should be gathered, while still ensuring that the final report met standards of practice regarding content and completeness. Last, the PIA protocol was intended to be considered a minimum standard. Psychologists are encouraged to consider additional assessment areas according to their training and experience.

Content areas

Several prominent reviews of the literature on psychological risk assessment and treatment planning⁴ suggest content areas representative of contemporary practice. Current risk assessment strategies⁵ also assist in further delineating information that should be included in a detailed assessment protocol.

The PIA protocol was developed to reflect this literature, and efforts were made to make it sensitive to both culture and gender. The protocol was organized into four specific core content areas consistent with the OIA approach. Each area also represents specific domains (see Figure 1). The core content areas, with domains in parentheses, are criminal risk (criminal history, use of violence, sex offence history), mental health (institutional adjustment concerns, mental health screening), case needs (intelligence and neuropsychological impairment, developmental history, lifestyle stability) and supplementary information (substance abuse, personality and clinical presentation, treatability, risk management issues).

Figure 1

PIA: Domains within Content Areas
Criminal Risk	Criminal history Use of violence Sex offence history
Mental health	Institutional adjustment Mental health screening
Case needs	Intelligence/neuropsychological Developmental Lifestyle stability
Supplementary	Substance abuse Personality/clinical presentation Treatability Risk management

Within several of the domains there are "flags" or exemplars of critical information necessitating more detailed assessment when endorsed. These "hits" route clinicians through a series of more exhaustive items for a particular domain. When not endorsed, the clinician can proceed to the next stage in the process. Given that the assessment is in a Windows environment, the clinician can navigate through the protocol in any order, selecting core content areas and domains easily.

Also, efforts have been made to ensure that the PIA reflects multimethod assessment techniques, including offender self-reports, clinical ratings and behavioural observations. Self-report information results from psychological testing and interviews; clinical ratings are structured judgments following agreed-on guidelines; and behavioural observations include historical information and staff comments. These strategies should provide cumulative information regarding needs and risk assessments. Optimally, there would be convergence among these strategies and case-specific analysis reflected in the final PIA report. Specific anchors for ratings and scoring should assist in reliability and limit any rater drift, although a user's manual will eventually also be required.

The consultative process

Before meetings were held with psychologists from all the intake units, steps were taken to ensure that the protocol was valid. First, the protocol was shared with several Correctional Service of Canada psychologists with many years of correctional experience. Collectively, these colleagues provided positive reviews and helpful recommendations for item content, clarity for enhanced item reliability and expansion of some items. The psychologists were able to provide detailed feedback by actually using the protocol as part of assessment duties in their respective positions. Additional comments regarding branching of items and content flow were considered for subsequent revision. Finally, the psychologists concluded that, despite a learning curve, the assessment protocol reflected the content areas already considered in clinical interviews, but did suggest an accompanying interview schedule.

Concurrently, the draft protocol was shared with two external pre-eminent academics/ clinicians in forensic psychology and psychiatry. They were asked to review the material for completeness, and their comments were encouraging. They particularly noted the merits of such a detailed assessment at intake and for various decisions throughout an offender's sentence.

The first consultation phase yielded positive comments and the observation that the development of an empirically derived and clinically relevant protocol was a distinct advantage. The content was considered defensible and consistent with standards of practice. They believed such a strategy, if adopted, could well deflect undue criticism of assessments, without being overly prescriptive regarding clinical skills.

The second phase involved psychologists from all the intake units who met for three days in March 1996 to review the most recent draft of the PIA protocol. Advance copies were provided to psychologists in both official languages to ensure informed discussion. By the end of the meeting, following considerable discussion on all items, there was consensus regarding the content and scoring of the protocol. Final revisions were completed by late July following further suggestions by members of this working group for changes to specific domains.

The assessment process

The development of the PIA protocol, and its automation, permits the systematic assessment and reassessment of offender needs and risk for treatment planning and risk management. For some content areas, all items will be considered (for example, criminal risk and criminal history); for others, computerization permits branching so that "hits" lead to more detailed investigation. The priority for PIA is to establish a base line psychological measure of risk and need which is then incorporated into a correctional treatment plan used to make placement decisions. With respect to treatment, psychological assessments should contribute to the correctional strategy by commenting on the level of intervention required for a particular offender. PIA is intended to be value added and to complement existing case management assessment procedures. Some duplication of information (criminal and social history) is inevitable, but this has advantages. For example, criminal and social history is often used to establish a rapport with the offender during the clinical interview. As well, the comparison between psychology and case management has the potential to address concerns in this setting about malingering. Furthermore, rationales for discrepancies in opinions should be more easily discernible if all parties use standardized assessment protocols.

The process, as presented in Figure 2, would be to have offenders referred for PIA according to the criteria in CD 840. Psychologists would complete file reviews and clinical interviews before completing the PIA protocol. Completion of the protocol would yield a menu of report options from which the psychologist would select for expansion into a final report. This PIA report would be entered into the Offender Management System (OMS) and shared with Case Management.

A remaining issue is the development of a parallel battery of psychological tests to measure constructs reflected in the core content areas. It is important that there be consensus regarding these standardized tests across all sites. As well, the tests could be hierarchical, so a climb up the hierarchy would be indicative of a problem area requiring more specialized assessment. Finally, it is recommended that psychologists support the use of a standardized psychological test battery without limiting those who wish to augment the testing.

Figure 2

PIA Flowchart

Case and file review

| | |

Structured interview

| | |

Psychological testing

| | |

Case consultation

| | |

Completion of protocol

| | |

Report generation

| | |

Debrief offender and put in the Offender Management System (OMS)

Information sharing and data management

The PIA protocol is a road map for psychologists to follow in the systematic assessment of offenders. The critical information reflected in the PIA protocol is to be incorporated into the final report and distributed according to guidelines for psychological reports. There is also an accountability framework for considering quality assurance issues in psychological reports. The protocol database, however, should have restricted access: its purpose is to assist researchers within an intake unit or across sites to investigate research questions. The software has been developed to facilitate input of data into statistical analysis programs or merging data from several sites.

Status of the PIA initiative

The PIA software was developed over a two-year period with considerable consultation regarding content, format and applications. The present version is still considered a draft, and pilot testing began at intake units in January 1997. Future development and implementation will be reviewed by the working group following the initial pilot tests. This initiative highlights the important contribution of psychological assessments to offender classification by informing case management about offender treatment needs and management of offender risk throughout an offender's sentence.

---

1. Research Branch, Correctional Service of Canada, 340 Laurier Avenue West, Ottawa, Ontario, K1A 0P9.

2. L.L. Motiuk, Community Risk/Needs Management Scale, CRNMS (Ottawa: Correctional Service of Canada, 1996). See also L.L. Motiuk and S.L. Brown, The Validity of Offender Needs Identification and Analysis in Community Corrections, Research Report R-34 (Ottawa: Correctional Service of Canada, 1993).

3. T. Leis, L.L. Motiuk and J. Ogloff, Forensic Psychology: Policy and Practice in Corrections (Ottawa: Correctional Service of Canada, 1996).

4. D.A. Andrews and J. Bonta, The Psychology of Criminal Conduct (Cincinnati, Ohio: Anderson Publishing Company, 1994). See also R. Blackburn, The Psychology of Criminal Conduct (Chichester, England: Wiley & Sons Ltd., 1993). And see Leis, Motiuk and Ogloff, Forensic Psychology. And see P. Gendreau, "The Principles of Effective Intervention with Offenders," in Choosing Correctional Options that Work: Defining the Demand and Evaluating the Supply, A.J. Harland (ed.) (Thousand Oaks, California: Sage [in press]). See also M.E. Rice, G.T. Harris and V.L. Quinsey, "Treatment of Forensic Patients," in Mental Health and Law: Research, Policy, and Services, B. Sales and S.A. Shah (eds.) (Durham, North Carolina: Academic Press, 1996) p. 141-189.

5. R.D. Hare, The Hare Psychopathy Checklist (Revised) (Toronto: Multi-Health Systems, 1991). See also C.D. Webster, D. Eaves, K. Douglas and A. Wintrup, The HCR-20: Assessment of Dangerousness and Risk (Burnaby, British Columbia: Simon Fraser University and the Forensic Services Commission of B.C., 1995). And see D. Andrews and J. Bonta, The Level of Service Inventory - Revised (LSI-R) (Toronto: Multi-Health Systems, Inc., 1995). And see MacArthur Research Network on Mental Health and Law. See also J. Monahan and H. H. Steadman, Violence and Mental Disorder: Developments in Risk Assessment (Chicago, Illinois: University of Chicago Press, 1994). And see H.J. Steadman, J. Monahan, P.C. Robins, P. Applebaum, T. Grisso, D. Klassen, E.P. Mulvey and L. Roth, "From Dangerousness to Risk Assessment: Implications for Appropriate Research Strategies," in Mental Disorder and Crime, S. Hodgins (ed.) (Newbury Park, California: Sage Publications, 1993) p. 39-62.