The epidemic of prescription opioid abuse requires early recognition and engagement of patients through motivational interviewing. Family reports or clinical suspicion based on a vague history or lack of rapport, accompanied by signs of acute opioid intoxication (constricted pupils, slurred speech, itching, euphoria or agitation, dry mouth, drowsiness, and impaired judgment) or withdrawal (dysphoric mood, nausea or vomiting, muscle aches, runny nose, watery eyes, dilated pupils, goosebumps, sweating, diarrhea, yawning, fever, and insomnia) should prompt urine toxicology. Many abused prescription opioids also contain acetaminophen, whose toxicity can be detected as elevated liver transaminase levels. Dodrill et al. (CME, p. 466) recommend buprenorphine treatment to address both the addiction and chronic pain. Methadone may be required for severe dependence. Maintenance of clinical gains requires pain management and behavioral counseling in addition to medication.

Initiation of antidepressant treatment was associated with increased risk of stroke in a Taiwanese national sample examined by Wu et al. (CME, p. 511). About 7% of adults hospitalized for a first stroke had received an antidepressant in the preceding year. The odds ratio of stroke during initiation of antidepressant use, compared to previous recent history of antidepressant use, was 1.48, adjusted for other concurrent medications and mood disorder. Antidepressants with high inhibition of the serotonin transporter, such as SSRIs, were associated with greater risk. Smoller (p. 457) points out that results are similar to those in the Women's Health Initiative, but neither study can fully differentiate the effects of depression versus medication.

Nonsuicidal self-injury predicts subsequent suicide attempts in depressed adolescents nearly as strongly as previous suicide attempts themselves. Wilkinson et al. (p. 495) found that poor family function is another major predictor of suicidality. Nonsuicidal self-injury is predicted by hopelessness, anxiety disorder, and being younger and female, but not by poor family function. Improvement in depression during treatment reduces nonsuicidal self-injury more than suicide attempts, as pointed out by Brent (p. 452). Because nonsuicidal self-injury occurs more frequently in younger patients than suicidal behavior itself, it should be seen as a sentinel, but potentially treatable event for later suicide risk.