Social contact problems warn of future psychosis

Difficulties with social contact in adolescents from the general population are associated with later development of psychosis, say researchers.

This is consistent with studies in people at high risk for psychosis, in whom social withdrawal is considered to be part of the prodromal psychosis state, say study author Pirjo Mäki (University of Oulu, Finland) and co-workers.

The team used the PROD screen, focusing on four items thought to be of particular relevance to prodromal psychosis. Two items related to negative social symptoms: difficulty or uncertainty in making contact with other people and social withdrawal; and two to positive symptoms: feeling that events or other people’s behavior specifically concern oneself and a feeling of being followed or influenced.

These symptoms were relatively common in a cohort of 6274 adolescents, aged between 15 and 16 years, from the general population, Mäki et al report in European Psychiatry. Among the 6162 study participants who were not hospitalized during follow-up, between about 5% and 20% endorsed one of these symptoms.

Endorsement rates were about 5% for the feeling of being followed or influenced, about 10% for difficulty making contact and social withdrawal, and almost 20% for the feeling that events or people’s behaviors specifically relate to oneself.

During the 6 years of follow-up, 89 participants were hospitalized with a nonpsychiatric disorder, and endorsement rates among these people were similar to those among individuals who were not hospitalized. But they were significantly higher for three items among 23 participants who were hospitalized for a psychotic disorder, with each item except being followed or influenced endorsed by 30–40% of these participants.

Participants who developed psychosis were 3.6-fold more likely than controls to have endorsed at least two PROD items. The sensitivity of at least two PROD items for psychosis was low, however, at 34.8%, but the specificity was high, at 89.7%. The positive and negative predictive values were 1.2% and 99.7%, respectively.

The researchers say that it could “be useful to assess the presence of these symptoms in clinical work when detecting patients possibly at high risk of psychosis.”

But they add that the low positive predictive values of these symptoms in young people in the general population make them an unsuitable screening tool.