Predicting Violence

Given the complexity of the human mind, how are we, as practitioners, to know if the patient in our office is likely to commit a violent crime? What if we are wrong? Once deemed violent, can someone ever be considered fully safe? How long must someone demonstrate mental health “stability” to be considered safe and how often should they be assessed? What of individuals who do not meet criteria for a particular disorder, but nevertheless cause a practitioner profound concern about the safety of others? Should everyone be assessed, whether or not there is an obvious cause for concern?

Exploring the answers to these questions is a large undertaking that must be met through many smaller steps. Through a series of questions, this post merely seeks to understand and handful of existing tools which aim to predict violence in individuals and reviews their reliability and validity where possible. This post does not attempt to be an exhaustive list of available predictive measures; rather, it serves as an entry point to continue the exploration of these topics in greater depth over time. From any perspective, implications for nursing and other professionals (health care, teachers, law enforcement, among others) are varied and profound. As arbiters of mental health diagnoses, having reliable and valid tools with which to practice and protect both the public and our selves is essential.

As a history of violence is the greatest predictor of future violence (Scott & Resnick, 2006), the Violence Risk 10 item scale (V-RISK-10) was studied in populations of both voluntary and involuntarily admitted patients with acute psychiatric disorders. Though they found the V-RISK-10 had good predictive validity in patients with a known history of violence (Roaldset, Hartvig, & Bjorkly, Mar 2011), a drawback to this study was the acknowledgement “violent persons with a personality disorder” are more likely to encounter law enforcement and incarceration. Therefore, there are a significant portion of violent individuals may be missed, should the focus of assessment dwell solely within mental health and psychiatric facilities.

In contrast, 1.794 Polish prisoners received the Psychopathic Personality Traits Scale (PPTS) (Boduszek, Debowska, Dhingra, & DeLisi, 2016). This 20-item, self-report scale attempts to identify violent tendencies despite age, cultural background, gender or criminal history. The scale seeks to quantify the prevalence of 4 factors: affective responsiveness, Cognitive responsiveness, interpersonal manipulation and egocentricity. Individual questions are then further categorized as belonging to one of two groups, either knowledge/skills or attitudes/belief. The study indicated that questions categorized as attitude/belief were much more indicative of violence in the respondent. A drawback to this study is the violence measured was historical, thus, to be utilized as a measure of predictive violence, the study must be replicated in persons who are not already detained for violent acts.

So where do these attitudes/beliefs come from? A study of juvenile offenders in Florida sought to evaluate the association between childhood trauma and future violence, using the Adverse Childhood Experiences (ACE) measure (Fox, Perez, Cass, Baglivio, & Epps, 2015). The study indicates the ACE measure is useful in identifying those who are at most risk to engage in violent activity (Fox et al., 2015). Early identification may enable providers to implement behavioral and mental health treatments aimed at violence prevention. Determining which interventions are effective and proving that intervention provided a reduction in violence is an obvious additional area of exploration.

What if we tackled these issues earlier? Can violent tendencies be identified in children? A 2015 study (Hong, Tillman, & Luby, 2015) attempted to distinguish between transient and normative conduct problems in children using the Kiddie Disruptive Behavior Disorder (KDBD) Scale. 2,232 children were followed from birth to 10 years old and evaluated through puppet interviews and questions for parents (mostly mothers), regarding their child’s behaviors (Hong et al., 2015). The results found that what constitutes “normal behavior” is a matter of degree. For instance, low-intensity defiance is not predictive of future violence, but high-intensity defiance is. Aggression to people or animals, however, were both associated with future violence regardless of intensity (Hong et al., 2015). The conclusion of this study suggests that use of the KDBD may be a useful tool in referring children to treatment and services.

Unfortunately, the costs of identifying children, adolescents and adults at risk for violence can be very steep, whether or not treatment is ever implemented. A 2013 study (Zagar et al., 2013) explored the question of whether computer testing would be equivalent to “paper and pencil” testing, as the use of computer tests is far less expensive and more readily available to those who might wish to administer such tests. Using five different measures, the tests were administered by paper/pencil or internet (Zagar et al., 2013). Findings concluded that the cost of internet testing was 70-80% less than paper tests, with the added benefit of immediately available results (Zagar et al., 2013). Immediate access to results may assist providers to make timely decisions about patient needs, referrals and safety.

It appears that there are numerous available scales with which to predict the likelihood of violence at many different ages and it is important to know that internet or computer-based testing may be cost-effective, immediate and accurate for many of these measures. These tools may help providers in areas with limited resources or access to services identify and refer individuals for psychiatric treatment. Of course, the incidence of violence with which we currently contend in this country, means there is no shortage of need for reliable, valid assessment measures and appropriate treatment. Nor will there be a shortage of opportunity to research the effectiveness of those scales and interventions. Nurses, ever on the front lines of patient care and intervention, be they in the schools, home, office or institutions, will undoubtedly be on the front lines of this effort, too.


Boduszek, D., Debowska, A., Dhingra, K., & DeLisi, M. (2016). Introduction and validation of psychopathic personality traits scale (PPTS) in a large prison sample. Journal of Criminal Justice, 46, 9-17. doi:10.1016/j.jcrimjus.2016.02.004

Dodge, K. A., Bierman, K. L., Coie, J. D., Greenberg, M. T., Lochman, J. E., McMahon, R. J., & Pinderhughes, E. E. (2015). Impact of early intervention on psychopathology, crime, and well-being at age 25. American Journal of Psychiatry, 172, 59-70. doi:10.1176/appi.ajp.2014.13060786

Fox, B. H., Perez, N., Cass, E., Baglivio, M. T., & Epps, N. (2015). Trauma changes everything: Examining the relationship between adverse childhood experiences and serious, violent and chronic juvenile offenders. Child Abuse and Neglect, 46, 163-173. doi:10.1016/j.chiabu.2015.01.011

Hong, J. S., Tillman, R., & Luby, J. L. (2015). Disruptive behavior in preschool children: Distinguishing normal misbehavior from markers of current and later childhood conduct disorder. Journal of Pediatrics, 166, 723-730. doi:10.1016/j.jpeds.2014.11.041

Im, D. S. (2016). Template to perpetrate: An update on violence in autism spectrum disorder. Harvard Review of Psychiatry, 24, 14-35. doi:10.1097/HRP.0000000000000087

Roaldset, J. O., Hartvig, P., & Bjorkly, S. (Mar 2011). V-RISK-10: Validation of a screen for risk of violence after discharge from acute psychiatry. European Psychiatry, 26, 85-91. doi:

Scott, C. L., & Resnick, P. J. (2006). Violence risk assessment in persons with mental illness. Aggression and Violent Behavior, 11, 598-611. doi:10.1016/j.avb.2005.12.003

Zagar, R. J., Kovach, J. W., Basile, B. B., Hughes, J. R., Grove, W. M., Busch, K. G., . . . Zagar, A. K. (2013). Finding workers, offenders, or students most at-risk for violence: Actuarial tests save lives and resources. Psychological Reports, 113, 685-716. doi:10.2466/16.03.PR0.113x29z3