Practitioners of forensic psychiatry often assess individuals who are involved with the legal system. As such, they may be required to determine those individuals’ past mental states or predict their future behavior.

Risk Assessment of Stalkers Still in Development

Debra A. Pinals, MD, University of Massachusetts Medical School, Worcester, Massachusetts, USA, discussed clinical risk management in stalking, noting that women are at a higher lifetime risk of being stalked (8% to 33%) than men (2% to 7%); also at increased risk are mental health professionals, college students, and Native Americans. Typical stalker characteristics include being white, male, and unemployed; lacking relationships or having disturbed relationships; and having a history of criminal behavior, substance abuse, and psychiatric diagnoses.

Those stalking an ex-intimate are most likely to engage in violence. The majority of those who assaulted victims had made prior threats, and morbid jealousy may be a risk factor. Various risk assessment tools are available or in development (Table 1).

The dilemma of intervention is that anything that is done has the potential to escalate or defuse the stalker’s activity. Victims should convey a single, explicitly clear message that no further contact is desired and then should make no further contact. Victims might consider seeking guidance from law enforcement before acting, although, unfortunately, between 50% and 80% of restraining orders are violated. The victim should have input into actions taken, and the response should be targeted to the individual situation because risk may change over time. If the stalker is the client, therapists should understand their legally required obligations to notify victims at risk.

Authentic and Malingered Hallucinations Can Be Differentiated

Phillip J. Resnick, MD, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA, discussed how to detect malingered hallucinations. The presence of delusions, not hallucinations, is pathognomonic of psychosis; individuals who are not psychotic can hallucinate.

Most genuine hallucinations are associated with delusions and gender-based insults. The voices ask questions, are seen as omniscient, but do not see the hallucinator as all-knowing. A voice reported as asking “Why are you smoking?” is more likely to be genuine than one reported as asking “What time is it?”

Those experiencing auditory hallucinations usually develop strategies to reduce them, such as seeking company, taking extra antipsychotics, humming, telling voices to go away, sleeping, or praying [Farhall J et al. Clin Psychol Rev. 2007]. Being alone, watching television, or listening to the radio, especially news programs, worsens hallucinations, said Dr Resnick.

Most visual hallucinations are in color and of normally sized, humanoid forms; a minority are animals or objects. Initial response is often fear, and visual hallucinations are rarely small. Individuals usually remember their first visual hallucinations.

Those with suspect auditory hallucinations have no strategies to reduce malevolent voices, saying that they follow all the voices’ commands (commands are obeyed in only one-third of true hallucinations). Voices in malingered auditory hallucinations are reported as always yelling, as vague or mumbling, or only as female or children’s voices; true hallucinations are usually in a normal speaking voice and could be only men’s voices [McCarthy-Jones S, Resnick PJ. Int J Law Psychiatry. 2014]. Malingered visual hallucinations may be reported as being in black and white or otherwise atypical, and the individual will be unable to recall the first one.

In any medicolegal context, clinicians should have a suspicion of malingering. In the case of hallucinations, some evidence-based data can help distinguish the suspect from the genuine.

Psychiatrists Continue to Play a Role in Death Penalty Issues

Charles L. Scott, MD, University of California, Davis, Davis, California, USA, discussed how psychiatry contributed to the US Supreme Court decisions stipulating that executing those who were aged < 18 years or who were intellectually disabled was cruel and unusual punishment and so violated the Eighth Amendment to the US Constitution. There is no constitutional ban on executing individuals with mental disorders.

There is no US professional guideline that prohibits psychiatrists from testifying in competence cases of capital defendants. Testimony regarding dangerousness at sentencing is also allowed.

The US Supreme Court has yet to address the constitutionality of the situation where an individual is psychotic when not medicated but, when treated (even involuntarily), could become competent to be executed. Is it in the best interest of the prisoner to either receive medication, even involuntarily, and cure the psychosis even if it leads to execution or be left untreated and actively psychotic? Dr Scott said that there is no consensus among board-certified psychiatrists on how to handle this situation.

Participation of psychiatrists in competency, sanity, and sentencing phases of trials involves examining personal ethics related to these forensic duties.

Dialectical Principlism as a Method of Analyzing and Resolving Complex Ethical Dilemmas

Robert Weinstock, MD, University of California at Los Angeles, Los Angeles, California, USA, presented a system that he is developing with William Connor Darby, MD, at his institution that analyzes the ethics of situations with conflicting obligations, which they call “dialectical principlism.” The concept deals with the various principles that may arise in a typical situation faced by a psychiatrist and the synthesis of these considerations into a whole.

He discussed 3 hypothetical situations that a general or forensic psychiatrist might experience: (1) thinking that a patient might reveal committing child abuse, having forgotten his warning that it would have to be reported; (2) having to decide if a patient will be disabled for a year and therefore be a candidate for disability payments; and (3) being asked by an insurance company to agree with its evaluation of a case rather than do an assessment, thereby keeping a source of lucrative referrals.

Dr Weinstock presented a balancing process of pros and cons for 2 possible actions for each dilemma, and he explained how using his system could result in different, sometimes opposite conclusions. Therefore, although consideration of the ethics of a situation is frequently called for, it is unclear what dialectical principlism contributes to decision making.

Improved Observation Practices Reduce Inpatient Suicide Risk

Predictors of inpatient suicide have not been identified. Jeffrey S. Janofsky, MD, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA, discussed reducing risk of inpatient suicide in the absence of defined predictors.

Inpatient suicide represents 5% to 6% of US suicides a year, mostly by hanging and jumping. Mitigation strategies include initiating structural safeguards, such as installing breakaway shower/closet rods and safety glass, measures that reduced inpatient suicide rates from an average of 2 every 9 months to 1 every 14 months [Watts BV et al. Arch Gen Psychiatry. 2012].

Another risk mitigation strategy uses procedural safeguards, such as doing a complete mental status evaluation and establishing a good therapeutic relationship. Dr Janofsky noted that a significant number of inpatient suicides occur while patients are on some type of observation status. There are no agreed-on best practices for the observation of psychiatric patients, and observation practices may vary within the same facility across shifts and units.

Failure modes and effects analysis (FMEA) is proactively used to identify and prevent human process errors before they occur by observing the steps in each process that are vulnerable to error. The Institute for Healthcare Improvement tool to aid in FMEA development allows institutions to share their analyses.

In Dr Janofsky’s hospital, 42 process steps were identified with the help of all inpatient staff, including nursing assistants. Observation of patients was identified as the crucial element in keeping those who were at high risk for suicide safe. Solutions to failure mode causes included development of a patient support sheet that the nurse discusses with observers, as well as an observer feedback form that the observer discusses with the nurse.

Other institutions can identify potential critical process failures. Rather than using an after-the-fact cause analysis, FMEA allows system solutions to prevent errors before harm occurs.

Progress Is Being Made in Predicting Juvenile Violence

Dr Scott also discussed how to assess the risk of juvenile violence. Future dangerousness is determined by a combination of risk factors and social context. Early age of onset of involvement with the police predicts a likelihood of continued criminal activity. Conduct disorder is the DSM-V disorder that is the best predictor of juvenile violence; other risk factors include having weapons, gang involvement, substance use, and family conflict and aggression.

An actuarial risk assessment model—the Youth Level of Service/Case Management Inventory 2.0—is best suited for use in the juvenile justice system. Other models include structured professional judgment tools that guide clinicians on risk and need for intervention. The structured assessment of violence risk in youth is based on the structured professional judgment model and is one of only a few instruments that includes protective factors along with historical, social/contextual, and individual/clinical factors, and it can be administered by a range of professionals.

Whether a child should be assessed as a psychopath is controversial. Some argue that youths should not be given this label, whereas others believe that some children have a high risk for recidivism that should not be ignored. A meta-analysis of 53 studies with > 10 000 participants came down on the side of early screening for prediction of recidivism [Asscher JJ et al. J Child Psychol Psychiatry. 2011].

It is difficult to predict which students will become homicidal. Dr Scott emphasized that communication of intent prior to attack is important; youths should be educated to tell authorities. Virginia has developed a student threat assessment model; although it needs to be validated via a randomized controlled study comparing alternative approaches, it appears promising, said Dr Scott. Universal school-based programs where everyone receives antibullying and antiviolence training also have a positive effect.

Juvenile and school violence have decreased over the past decade. The use of structured instruments can improve risk assessment and management as they have for adults. Early intervention and matching the intervention to the level of threat are important.