The duration of untreated psychosis is one of the few risk factors for poor outcomes that has the potential to be modified. Because of this, there is growing interest in the primary care provider's role in the care of patients with psychosis. By identifying those at clinical high risk for psychosis, these providers can help to expedite referral to a specialist and early treatment. Yet, many of the symptoms of psychosis are nonspecific and frequently missed. Recently, investigators studied whether certain clinical high-risk symptoms are associated with a subsequent psychosis diagnosis and have potential clinical utility.

In this case-control study, the researchers analyzed the data of 93,483 people that was collected from 530 primary care practices. The cases included 11,690 people with psychosis and 81,793 matched controls who had never received this diagnosis. The researchers identified a set of clinical high-risk symptoms: bizarre behavior, blunted affect, problems associated with cannabis use, depressive symptoms, role-functioning problems, social isolation, symptoms of mania, symptoms similar to those of obsessive-compulsive disorder or attention deficit-hyperactivity disorder, disordered personal hygiene, sleep disturbance, problems associated with cigarette smoking, and suicidal behavior (including self-harm).

They found an association between a psychosis diagnosis and all of those symptoms except for disordered personal hygiene. The strongest association was with suicidal behavior. Positive predictive values were highest for suicidal behavior in men ages 24 and younger and women ages 25 to 34. Symptom pairs increased the positive predictive value of single symptoms, with the researchers noting this was particularly marked with suicidal behavior. They found that reports of symptoms occurred more frequently in the five years before a psychosis diagnosis; for some symptoms, there was a sharp increase in the number of consultations in the three months before a diagnosis.

The authors point out that the primary care data utilized in this study may reflect a selection bias, because not all patients consult a primary care provider before a diagnosis of psychosis.

REFERENCE