Clinical judgment in the assessment of offender risk contrasts with mechanical assessment. When clinical judgment is used, the assessment is based on clinical observations or experience, formal theory or informal hypotheses about criminal behavior, grasp of the current status of research on risk assessment, or similar intuitive understandings of risk. Clinical judgment can also be used following the completion of a mechanical or otherwise structured assessment of previously selected risk factors, commonly referred to in the criminal risk assessment field as structured professional judgment. This article briefly compares unaided clinical risk assessment with mechanical risk assessment tools and considers the nature and role of structured clinical judgment in current risk assessment approaches.
Clinical Judgment Compared With Mechanical Assessment
Clinical judgment in the absence of a formal procedure or guide to conducting an assessment is known as unstructured or unaided clinical judgment. It maximizes the use of intuition, which itself may be influenced by the individual assessor’s experience with similar cases as well as the assessor’s attention to, memory for, and ability to integrate feedback from the outcome of previously assessed cases. In this way, the assessor inevitably draws upon a broad class of offenders to generalize to the individual case.
In contrast, mechanical risk assessment uses a preselected collection of risk factors and derives a total score based on the item scores. Alternatively, mechanical risk assessment might classify risk based on a cutoff score or the presence of certain items. The numeric or categorical result of a mechanical risk assessment tool is used, in the case of actuarial tools in particular, to refer to a chart of absolute (i.e., probability) and relative (i.e., percentile rank) risk of reoffending. These statistics are derived from follow-up studies of large samples of offenders and are used to inform decisions about risk management at the individual or policy level.
The Current Status of Unaided Clinical Judgment
Unaided clinical judgment is no longer considered an appropriate method of risk assessment because of its poor reliability and validity. Research indicates that assessors who use unaided clinical judgment do so inexpertly: They show little agreement with each other, they do not predict criminal outcomes well (especially when predicting that violence will occur), and even experienced and highly trained clinicians use the same kinds of information about an offender that educated laypersons use to assess offenders’ risk of reoffending. In addition, unaided clinical judgment shows poor predictive validity. In 2001, John Monahan and colleagues decried the prospect of further research into the validity of unstructured clinical assessment, declaring “That horse was already dead” (p. 7).
Clinical Judgment in Structured Risk Assessment
Some offender risk assessment procedures combine a mechanical approach with a summary rating of risk based on overall clinical judgment. In these structured instruments that incorporate clinical assessment, the risk factors might be scored to derive a total that informs the clinical judgment, or the clinical judgment might be a more intuitive deduction of risk based on all the information previously considered. The final judgment is expressed in terms of a category such as high risk, moderate risk, or low risk. Depending on the type of assessment, the assessor might judge the risk of a specific outcome such as sexual violence or violence against an intimate partner. Clinical assessment might also be used to judge the severity or imminence of a criminal outcome. In any case, use of structured clinical judgment in criminal risk assessment results in the use of categorical, nonnumerical methods of risk communication.
Some risk factors on a mechanical risk instrument might themselves require clinical assessment skills. For example, it is common for instruments to include items pertaining to psychopathy or mental illness in offender risk assessment. These measures might require psychological testing and diagnostic skills. When clinical judgment is used to administer a mechanical risk assessment tool, it only contributes to the evaluation of specific items and is not used to adjust the total score. Thus, assessment tools that incorporate clinical judgment differ from strictly mechanical risk assessment tools most substantially in the extent to which clinical judgment is used in deriving the final appraisal of an offender’s risk.
Another key characteristic of tools that use structured clinical assessment is that they might incorporate risk factors on the basis that they are theoretically informed, rather than only items selected through their empirical relationship with criminal outcomes. Such tools may also encourage assessors to consider additional or unusual factors that are thought to contribute to offending behavior in the individual being assessed.
Using Clinical Judgment in Forensic Decision-Making
Following risk assessment, some professional judgment is necessary to determine how to intervene with criminogenic needs or take other steps toward risk management because no existing offender risk assessment tool indicates the decisions that must be taken to prevent offending. Treating an offender as higher risk than indicated by a structured risk assessment tool is acceptable in exceptional circumstances, such as a credible threat made by an offender to physically harm another person. Furthermore, when the predictive validity of a mechanical risk assessment is less than optimal, adding an overall clinical judgment has the potential to improve risk assessment. However, routine adjustment of structured risk assessment using clinical judgment is not supported because clinical overrides have been shown to reduce the accuracy of validated risk assessments in several studies.
References:
- Cohen, T. H., Pendergast, B., & VanBenschoten, S. W. (2016). Examining overrides of risk classifications for offenders on federal supervision. Federal Probation, 80, 12–21.
- Monahan, J., Steadman, H. J., Silver, E., Appelbaum, S., Robbins, P. C., Mulvey, E. P., . . . Banks, S. (2001). Rethinking risk assessment: The MacArthur study of mental disorder and violence. Oxford, UK: Oxford University Press.
- Quinsey, V. L., & Ambtman, R. (1979). Variables affecting psychiatrists’ and teachers’ assessments of the dangerousness of mentally ill offenders. Journal of Consulting and Clinical Psychology, 47, 353–362.