NEW YORK – A further decline in social functioning of adolescents who already display significant social deficits might be a risk marker for psychosis in adulthood, according to a researcher.

"The inability to function socially appropriately seems to have tremendous potential to be a very early and clear predictor of early mental illness," said Barbara A. Cornblatt, Ph.D., of Zucker Hillside Hospital in New York City.

Dr. Cornblatt distinguishes between functional impairment in the social domain – difficulties making friends, increasing social isolation, lack of social support networks – and role of impairment, which is the "inability to maintain age-appropriate roles in the community."

She and her team follow adolescents and young adults aged 12-22 who are at clinical high risk for psychosis as part of their center’s RAP (Recognition and Prevention) program. They also participate in the NAPLS (North American Prodrome Longitudinal Study) consortium.

They have observed that among clinical high-risk subjects, moderate social problems often are visible early in development, as children join in social activities and try to make friends. Most of these patients appear to have deficits that are stable across development, putting them at risk for future moderate clinical problems and potential disability.

Additionally, there appears to be a subgroup of those with functional social deficits who display a large decline in social interactions typically beginning in their mid-teens, and these progressive deficits might predict later emerging psychosis, Dr. Cornblatt said at the annual meeting of the American Psychiatric Association.

Data from the RAP cohort show on the 10-point social domain of the GAF (Global Assessment of Functioning) Scale, where 1 is extreme social isolation and 10 is superior functioning in a wide range of social and interpersonal relationships, subjects at clinical high risk scored approximately 6, defined as moderate impairment in social functioning, with moderate impairments and moderate difficulty in age-appropriate intimate relationships. In contrast, healthy controls tend to score around 9, deemed to have good functioning in all social areas.

During a mean of 3 years of follow-up, the social scores for clinical high-risk patients remained stable, hovering around 6. The investigators also found that of the measures at baseline, including neurocognition, negative symptoms, role functioning and social functioning, only social functions was predictive of social outcome at follow-up.

"What this says to me is that intervention here is very important. At baseline, when kids are younger, we can intervene and probably affect social functioning using noninvasive, nonpharmacological techniques, for example, neurocognitive remediation or trying to deal with negative symptoms," Dr. Cornblatt said.

As the individual ages, untreated social impairment might become entrenched and much harder to modify, she added. However, she also pointed out that these findings apply to general functioning, because the majority of subjects with social deficits do not go on to develop schizophrenia or other forms of psychosis.

Sharp social declines

There might be a subset of patients who exhibit declines in social skills and interactions from the mid to late teens (16-19) who appear to be at especially high risk for later psychosis, she said.

Dr. Cornblatt noted that in the NAPLS 1 study (Arch. Gen. Psychiatry 2008;65:28-37), decline in social functioning was one of five predictors of psychosis. Similarly, in the RAP phase I cohort, a Cox proportional hazard model controlling for age showed that a decline in social functioning over follow-up also was predictive of psychosis, along with disorganized thoughts, suspiciousness, and verbal memory deficits.

Among 169 patients in the NAPLS 2 cohort who have completed 2 years of follow-up, those who had a social decline of 2 or more points on the GAFS were significantly more likely to convert to psychosis than were patients with a decline of less than 2 (P = .001) (Schizophr. Res. 2012;142:77-82).

"Going down 2 points, from a 7 to a 5, or a 6 to a 4, is a really big change. We’re not talking about trivial social fluctuations; we’re talking about kids (who) have substantial deterioration in their functioning," Dr. Cornblatt said.

She noted that early evidence for a biological underpinning of the observation comes from unpublished data looking at clinical high-risk subjects in NAPLS2 who converted to psychosis. The data showed that growth of ventricles and decrease in gray matter correlated with decline in social functioning. In other words, the larger the ventricles became over time, the worse the decline in social functioning.

The findings "suggest that social problems are one of the areas where it doesn’t matter if the risk is specific for schizophrenia or other forms of psychosis, or just for bad outcome. We should be working to improve the social skills of kids considered at risk from very early ages on," she concluded.

The RAP program is supported by the National Institute of Mental Health and private foundations. Dr. Cornblatt reported having no financial disclosures.

Decline in social functioning in teens predicts psychosis