This Month’s Highlights

This month’s lead article describes a study in which researchers asked inpatients to predict whether they would engage in violent behavior in the coming weeks. As Jennifer L. Skeem, Ph.D., and her coauthors note, “The principle behind this approach is simple: ‘If you want information from someone, the best way to get it is to ask them.’ ” During each interview, the researcher also rated the patient’s violence potential with two brief risk assessment tools. All of the patients had co-occurring disorders, and all were considered to pose a high risk of violence. When these individuals were reinterviewed in the community two months later, the researchers found that the patients’ predictions were more accurate than those made with the assessment tools (page 410). A research team in Ireland wanted to learn whether psychiatric inpatients who were about to be discharged would be more candid about their inpatient experiences—especially about self-perceived coercion—if they were interviewed by someone they knew to be a fellow service user who had previous inpatient experience. Brian O’Donoghue, M.Sc., M.R.C.Psych., and colleagues randomly assigned 161 voluntarily and involuntarily admitted patients to discharge interviews conducted by service users or by clinicians. Most inpatients gave positive accounts of their hospital stays, and no differences were found by interviewer or by admission status (voluntary or involuntary) (page 416). In a Taking Issue commentary, Laura Van Tosh notes that studies such as these are helping mental health providers better understand the perspectives of service users (page 401).

Assertive community treatment (ACT) is associated with an array of positive outcomes for people with serious mental illness, but reducing arrest and incarceration is not one of them. To gain a better understanding of the failure of traditional ACT models to improve criminal justice outcomes, Craig Beach, M.D., M.Sc., and colleagues analyzed five years of data for more than 4,700 ACT clients with and without forensic histories in New York State. Clients with recent forensic histories experienced an array of adverse outcomes, particularly during their first year of ACT. Findings highlight the need for ACT teams to implement additional strategies for these high-risk clients, the authors note (page 437). In the Veterans Affairs (VA) health system, ACT is called mental health intensive case management. A team of VA researchers led by Marcia Valenstein, M.D., M.S., examined 2001–2004 data for VA patients with schizophrenia who were new ACT enrollees and a matched sample of those not enrolled in ACT. ACT enrollment significantly improved medication adherence. The effect persisted for at least 24 months and was higher among patients with greater ACT participation (page 445).

Inpatient psychiatric wards seek to create a safe, calm atmosphere that is conducive to therapeutic care. Most wards that attain this goal have low rates of containment (coerced medication, seclusion, and restraint) and low rates of conflict (aggression, absconding, and substance use). A London-based research group investigated psychiatric wards with counterintuitive combinations: wards with low containment and high conflict and those with high containment and low conflict. Len Bowers, R.M.N., Ph.D., and colleagues sorted 136 acute wards across England into a quadrant that reflected varying levels of conflict and containment and examined variables significantly associated with each of the four types. They found, for example, that high-conflict, low-containment wards had higher proportions of male staff and lower-quality environments (page 423).

Individuals with severe mental disorders who do not adhere to community-based treatments are often subject to pressures to leverage participation, including pressures related to the criminal justice system, housing, and money. Research on the association between leverage and participation has yielded mixed results. Dale E. McNiel, Ph.D., and colleagues examined whether features of the treatment relationship, such as the working alliance and perceived coercion, act to mediate the effects of leverage. Among nearly 200 patients interviewed in San Francisco, 40% reported having experienced some form of leverage in the past six months. Those who reported greater coercion to participate in treatment were more likely to be medication adherent. But higher perceived coercion came at a cost in terms of satisfaction with care (page 431).