The Structured Assessment of Protective Factors (SAPROF) for violence has been under development since the early 2000s. Arguably, it represents one of the most systematized attempts at creating an adult-focused protective factor assessment tool that mirrors the structure commonly found in other risk assessment tools. It joins a relatively small but growing group of forensic tools incorporating protective or strength-based factors (i.e., any characteristics of an individual person, his or her situation, or environment that may reduce the risk of future violence) into violence risk assessment, such as the Structured Assessment of Violence Risk in Youth, the Inventory of Offender Risk, Need, and Strength, and the Short Term Assessment of Risk and Treatability. This article addresses the basic structure of the SAPROF and the relationship between protective factors and risk factors, evaluates the role played by the SAPROF in predicting future violent offending, and explores other possible uses of the SAPROF.
Basic Structure
The SAPROF is comprised of 17 hypothesized protective factors, which are subdivided into five internal characteristics, seven motivational factors, and five external factors. Two internal characteristics (i.e., intelligence and secure attachment in childhood) are viewed as static variables (i.e., generally unchanging with time or intervention), whereas the remaining 15 items are hypothesized to be dynamic variables (i.e., potentially changeable).
Each item is rated as 2, 1, or 0 to represent whether the factor is present, partially present, or absent (as per the operational definitions and rationale in the administration manual). A total protection score can then be calculated. Final protection ratings are assigned using structured professional judgment rather than relying on actuarial derived protection bins. Additionally, the authors, Vivienne de Vogel, Corine de Ruiter, Yvonne Bouman, and Michiel Vries Robbé, argue that the SAPROF is best used in combination with a risk-focused assessment tool, such as the Historical, Clinical, and Risk Management-20 (HCR-20). Should clinicians choose to pair the HCR-20 with the SAPROF, the SAPROF protocol allows clinicians to make integrated final risk ratings after reviewing the HCR-20 and SAPROF together. Finally, clinicians can also identify which factors are key to the offender’s evaluation, which factors should be treatment goals for the offender, and whether other considerations are worth noting.
Due to its recent publication, the current literature on the SAPROF’s psychometric properties, predictive accuracy, and usefulness is relatively small but rapidly growing. Studies examining the SAPROF have demonstrated support for its use with a range of populations. Reception of the SAPROF by the academic community has been largely positive; however, it has not been without criticism.
Protective Factors Versus Risk Factors
Protective factors have long been included as an under-addressed and under-researched component of effective correctional treatment. The SAPROF conceptualizes protective factors as any characteristics of an individual person, his or her situation, or environment that may reduce the risk of future violence. A major criticism of the literature on protective factors is a lack of a mutually agreed upon conceptual and operational definition regarding what constitutes a protective factor, and by what mechanism they impact offending. Some authors view protective factors and risk factors as the same construct in that the presence of one categorically means the absence of the other (e.g., the presence of the risk factor dysfunctional parenting would translate to the absence of the protective factor secure childhood attachment). Many researchers use the terms protective factor and risk factor interchangeably. If such a relationship is true, then protective factor tools would not, in fact, assess protection but rather provide reverse assessments of risk. Thus, any relationship between protection and offending would actually just be a proxy of the relationship between risk and offending.
Such a criticism could be made about some components of the SAPROF. In numerous published studies, many of the SAPROF items as well as its total score have been found to correlate highly with the risk factors and total score of the HCR-20. For example, the SAPROF’s self-control protective factor is defined as the absence of impulsivity; impulsivity is a risk factor on the HCR-20 and is largely defined as the lack of self-control and inhibition. Correlations between these 2 items generally range between −.60 and −.76. Additionally, many of the SAPROF items share marked similarities with the Big 4 and Central 8 risk factors. The authors of the SAPROF acknowledge this criticism in the administration manual and note that most of the SAPROF’s protective factors could also be considered risk factors. The authors acknowledge that the concept of protective factors is ambiguous, and they openly discuss the conceptual difficulties in defining protective factors.
Other possible relationships between protective factors and risk factors warrant discussion:
- Protective factors and risk factors could represent opposite poles of a continuum; that is, the factor could be protective at one end of the continuum, risky at the other end, and neither risky nor protective toward the middle of the continuum. For example, the absence of the risk factor antisocial peers does not necessarily mean the presence of the protective factor prosocial peers.
- Protective factors and risk factors could represent completely unique and independent factors. Therefore, they would not correlate with each other and an offender could be high in both protection and risk. Currently, the research literature does not support this possibility.
- Protective factors are actually resilience factors. Resilience can be defined as the capacity to adapt successfully in the presence of risk and adversity. Research on resilience factors in offending similarly struggles with mixed findings and its own conceptual issues (i.e., Are they traits? Are they teachable? Are they universal?).
Subsequent to resolving these conceptual and operational definition issues, mechanisms of action still need to be clarified. Possible mechanisms include direct effects, buffering or mediating effects (either fully or partially), and moderating effects. It is also possible that specific protective factors may only have protective effects in the presence of a specific risk factor. That is, protective factors may not have a global protective effect on offending but rather have smaller moderating effects only when specific risk factors are present. Further research elucidating these processes is clearly required.
Role of the SAPROF in Predicting Future Offending
Although conceptual issues remain regarding the exact nature of the SAPROF’s protective factors, the research literature has been supportive of its ability to predict future nonviolent, violent, and sexually violent offending. This relationship has been observed in a number of populations including male and female offenders; correctional, not criminally responsible, and mentally disordered offenders; psychiatric and incarcerated populations; and over many different follow-up periods (ranging from 1 to 10+ years post-assessment). Predominantly, this research has been conducted in European settings; however, recent research in Canada seems to be equally favorable. The magnitude of the SAPROF’s predictive accuracy appears similar to that of other risk assessment tools. As such, it is unlikely that the SAPROF is superior or inferior to the other mainstream risk assessment tools in its ability to predict offending.
Since the SAPROF’s predictive accuracy has largely been established, current research seems to focus on testing its internal assumptions as well as whether it adds uniquely to the prediction of offending that is not otherwise accounted for by risk. A major assumption of the SAPROF is that 15 of its protective factors are dynamic. In order to establish dynamism, one must measure the factor at a minimum of two time points, that an intervention occurs between these two time points, that the factors change in response to the intervention, and accordingly an increase or decrease in an outcome variable is observed (in this case, increases or decreases in offending). Research in both European and Canadian samples has demonstrated that scores on the SAPROF’s dynamic factors do significantly change with time and intervention; however, demonstrating that these changes reflect actual changes in future offending rates has been less positive. For example, in a study of high-risk/ low-protection Canadian offenders who underwent an intensive RNR-based violence treatment program, level of protection on the SAPROF was significantly higher posttreatment; yet, these increases in protection only minimally to moderately corresponded with reductions in future reoffending. Clearly, more investigation is necessary.
A second major area of research on the SAPROF focuses on whether the SAPROF uniquely adds to the prediction of offending beyond that captured by other risk assessment tools. Studies examining whether the SAPROF adds incrementally to the prediction of the offending over the HCR-20 has been mixed. Many studies have demonstrated incremental contributions for a subset of offending types, but which type of offending the SAPROF incrementally predicts appears to vary from study to study. Recently, in a Canadian sample of incarcerated high-risk violent offenders, the SAPROF did not add incrementally to the prediction of any offending types over the HCR-20 despite showing strong independent predictive accuracy. These results do raise the question that even if the SAPROF does add incrementally to the prediction of reoffending, such contributions may be small and may not markedly change a clinician’s final risk ratings. As such, more research is needed to establish whether the SAPROF consistently adds both incremental and clinical meaning to the prediction of all offending or only to specific populations and to specific offense types.
Other Uses and Future Directions of the SAPROF
Although the SAPROF has been demonstrated to predict offending, its authors argue that its greatest value relates to guiding treatment planning. This line of research is in its fledging stages. Integration of a protective factor focus into RNR-based programming has largely been absent from the research literature. Further, it is unclear whether treatment planning using the SAPROF would be superior to treatment plans generated from risk-based tools such as the Violence Risk Scale. Currently, the authors of the SAPROF have not produced guidelines detailing which treatments to implement in the presence or absence of certain protective factors. Importantly, the authors acknowledge in the administration manual that they plan to make continuous improvements and revisions to the SAPROF.
Additionally, some researchers argue that the inclusion of protective factor tools in risk assessment may have other benefits beyond predictive accuracy. Such researchers have speculated that risk assessments based solely on risk factors could hypothetically lead to a negative connotation or bias in reports, thereby generating pessimism among clinicians and unbalanced final evaluations by parole boards. Potentially, this could lead to lengthier periods of detention. Although scant research exists examining these potential impacts of the risk-focused assessment, these hypothetical impacts would be costly to all parties involved.
Further, current research tends to have a narrow focus toward the overprediction of offending rather than evaluating other potentially positive community outcomes. Indeed, it makes logical sense that protective factors may be uniquely situated to provide accurate assessment of positive community outcomes (regardless of whether the offender reoffends) including returning to school, reduced hospitalizations, holding consistent employment, or maintaining safe and stable housing. One could argue that an offender who reoffended and had maintained stable employment and housing is more successful than an offender who reoffended and was unable to attain employment or safe housing. Again, scant research exists in this area, but this could represent a unique research avenue in the burgeoning forensic positive psychology literature.
Final Thoughts
The SAPROF represents an important addition to the risk assessment literature and is arguably one of the best developed protective factor tools currently published. Although many conceptual and research questions remain, this would be expected for any new risk assessment tool. Given the authors’ stated commitment to continually improve and revise the tool as new research is published, the SAPROF is a useful tool that may become valuable in forensic assessment.
References:
- de Vogel, V., de Ruiter, C., Bouman, Y., & de Vries Robbé, M. (2009). SAPROF Structured Assessment of Protective Factors for Violence Risk (Administration Manual-English version). Utrecht, the Netherlands: Forum Educatief.
- de Vries Robbé, M. (2014). Protective factors. Validation of the structured assessment of protective factors for violence risk in forensic psychiatry. Utrecht, the Netherlands: Van der Hoeven Kliniek. Retrieved from www.saprof.com/content/2014/03/PhD-thesis-SAPROF