Schizophrenia Research Forum - A Catalyst for Creative Thinking

Predicting Functional Outcome in High-Risk Teens: Looking Beyond Psychosis

September 18, 2013. In a new, prospective study of adolescents and young adults showing early warning signs of psychosis, nearly 20 percent of whom later "converted" to a full-blown psychotic disorder, researchers identified several factors that predict social and occupational functioning years down the road. The findings, published online September 4, 2013, in JAMA Psychiatry, suggest that the early treatment of these difficulties could ward off long-term functional disability. Led by Ricardo Carrión of the Zucker Hillside Hospital in Glen Oaks, New York, the researchers also found that poor functional outcomes were largely independent of the development of full-fledged psychosis and pointed to a critical need for early interventions even in those who do not convert.

Some adolescents and young adults show mild signs of psychosis (such as suspicious beliefs or unusual perceptual experiences) that are troubling, but aren’t strong enough to meet the criteria for a full-blown psychotic disorder. Sometimes referred to as the illness prodrome or attenuated psychosis syndrome, these symptoms can substantially affect quality of life, and studies have shown that this clinical high-risk group is more likely to eventually "convert" to a major psychotic disorder such as schizophrenia than the general population (see SRF related news story). Most research in this population has focused on predicting the minority of patients that will eventually receive a formal diagnosis, but increasing evidence suggests that forecasting the far more common outcome of functional impairment may be a more effective way to prevent long-term disability.

In an effort to identify predictors of functional outcome, Carrión and colleagues examined 92 high-risk adolescents and young adults and 68 healthy controls, recruited as part of the Recognition and Prevention Program, an ongoing longitudinal investigation. Subjects in the current study received a baseline assessment of symptoms, neurocognitive performance, and functional outcome that was repeated every six to nine months for an average of three years until treatment ended (n = 77) or a conversion to psychosis occurred (n = 15).

The investigators used a cognitive battery of neuropsychological tests to measure performance in eight domains (including processing speed, working memory, and attention). To assess functional outcome, they focused on social and role functioning measured by two Global Functioning subscales—Social and Role—developed specifically for use in this population and not subject to the same limitations and confounds of more global measures (Cornblatt et al., 2007). The Social scale looked at the quantity and quality of relationships with peers, romantic partners, and family members, while the Role scale assessed performance in an appropriate academic or occupational position.

Follow-up function and the search for predictors
Baseline cognitive impairments and impaired social and role functioning in these subjects compared to healthy controls have been previously reported (Carrión et al., 2011), and the new study found that for many subjects these deficits persisted. At the end of the follow-up period, 48 percent of the at-risk group was classified as having poor social outcome, while 49 percent had a poor role outcome. One-third of subjects had a poor outcome in both domains, while 36 percent had a good outcome in both.

Statistical analyses demonstrated several differences in baseline cognitive performance between groups. At-risk subjects in the poor social outcome group performed worse on baseline tests of verbal memory, processing speed, executive function, motor speed, and language than those who later achieved a good social outcome. Similarly, those in the poor role outcome group performed worse on baseline tests of verbal memory, executive function, sustained attention, and language. Healthy controls also performed better than the at-risk subjects with poor social outcome in nearly all domains.

With an eye toward early intervention, the investigators used logistic regression models to look for baseline variables that predicted poor social and role outcomes. Three significant predictors emerged for each: Processing speed, social functioning, and disorganization symptoms predicted social outcome, while verbal memory, role functioning, and motor symptoms predicted role outcome. Thus, those who initially performed worse on cognitive tests were already having functional difficulties or displayed attenuated non-psychotic symptoms at baseline.

Psychosis a minor player
Notably, positive symptoms at baseline were not predictive of future social or role functioning. Those who later convert to psychosis tend to have worse cognitive performance and functioning at baseline than those who do not—a potential confounder of the association between predictors and functional outcome. To explore the independence of functional outcome from conversion, the researchers added conversion status to the prediction models. Although receiving a formal diagnosis of psychosis was significantly associated with poor social outcome, it did not solely account for later social difficulties because two of the three predictor variables (baseline social functioning and disorganized symptoms) remained significant contributors even after conversion status was accounted for. There was no effect of conversion status on later role functioning, as all three of the previously identified variables continued to be predictors.

"Functional disability is not solely dependent on the progression of full-blown psychosis, because many individuals who did not convert continued to present with impairments in social and role functioning," note the authors. In fact, a substantial 40 percent of non-converters had a poor social outcome, and the number was even higher (46 percent) for role outcome, meaning that both converters and non-converters could benefit from early interventions.

The finding that poor functional outcomes were largely independent of both positive symptoms and the development of psychosis suggests that prevention and treatment strategies should be aimed not only at emerging psychotic symptoms, but also difficulties in relationships and at school/work. Early identification and treatment of these troubles could help to reduce or prevent the progression to long-term functional disability.—Allison A. Curley.

Carrión RE, McLaughlin D, Goldberg TE, Auther AM, Olsen RH, Olvet DM, Correll CU, Cornblatt BA. Prediction of Functional Outcome in Individuals at Clinical High Risk for Psychosis. JAMA Psychiatry. 2013 Sep 4. Abstract

Comments on Related News

Related News: Attenuated Psychotic Symptoms: Risky, But Not Predictive of Psychosis

Comment by:  William Carpenter, SRF Advisor (Disclosure)
Submitted 8 March 2012
Posted 8 March 2012

Werbeloff et al. make a valuable contribution with an epidemiological cohort that permits estimates of self-reported attenuated psychosis-like experiences as a risk factor for a future psychotic disorder in the non-ill population. The possibility that attenuated psychosis syndrome would be included in DSM-5 has created an intense and interesting debate. Jim van Os and I, both members of the responsible DSM-5 Work Group, provided an update on the controversy last year (Carpenter and van Os, 2011), and I have offered a rebuttal to some of the objections elsewhere (Carpenter, 2011). As this is a work in progress, I will briefly state where we are, at the moment, in the DSM-5 process.


Carpenter WT and van Os J. Should Attenuated Psychosis Syndrome Be a DSM-5 Diagnosis? Am J Psychiatry 2011 168: 460-463. Abstract

Carpenter Jr. WT. Criticism of the DSM-V risk syndrome: A rebuttal. Cognitive Neuropsychiatry, 16(2):101-106, 2011. Abstract

Fusar-Poli P, Bonoldi I, Yung AR, Borgwardt S, Kempton MJ, Valmaggia L, Barale F, Caverzasi E, McGuire P. Predicting Psychosis: Meta-analysis of Transition Outcomes in Individuals at High Clinical Risk. Arch Gen Psychiatry. 2012; 69(3);220-229. Abstract

View all comments by William Carpenter

Related News: Attenuated Psychotic Symptoms: Risky, But Not Predictive of Psychosis

Comment by:  William Carpenter, SRF Advisor (Disclosure)
Submitted 9 March 2012
Posted 9 March 2012

Editor's note: This is an addendum to Will Carpenter's previous comment above.

Jim van Os and colleagues now report on transition to psychosis from a non-clinical population sample. Transition rates are elevated for those with psychotic-like experiences, and persistence of these experiences increases risk for transition. Rates are substantially higher than the representative population without psychotic experiences, but much lower than clinical referral populations reported in studies to date.


Kaymaz N, Drukker M, Lieb R, Wittchen HU, Werbeloff N, Weiser M, Lataster T, van Os J. Do subthreshold psychotic experiences predict clinical outcomes in unselected non-help-seeking population-based samples? A systematic review and meta-analysis, enriched with new results. Psychol Med. 2012 Jan 20;1-15. Abstract

View all comments by William Carpenter