Neurocognitive Assessments Are More Important Among Adolescents Than Adults for Predicting Psychosis in Clinical High Risk

Background: Few studies have examined the effects of age on neurocognition to predict conversion to psychosis in individuals at clinical high risk (CHR). This study aimed to compare the extent and predictive performance of cognitive deficits between adolescents and adults at CHR.

Methods: A comprehensive neuropsychological battery was performed on 325 CHR individuals and 365 healthy control (HC) subjects. The subjects were first divided into 189 CHR adolescents (age 12-17 years), 136 CHR adults (age 18-45 years), 88 HC adolescents, and 277 HC adults. CHR subjects were then divided into converters (CHR-Cs) (adolescents, n = 43; adults, n = 34) and nonconverters (CHR-NCs) (adolescents, n = 146; adults, n = 102) based on their 2-year follow-up clinical status.

Results: The adolescents and adults at CHR performed significantly worse than their control groups on all neurocognitive tests, except for performance on the continuous performance test in adolescents. In the comparison between adolescents and adults, patterns of neurocognitive deficits seemed to vary in HC subjects rather than in CHR subjects. In the comparison between CHR and HC subjects, the rank order of effect sizes across the neurocognitive tests was similar for the top two tests of symbol coding and verbal learning. Comparison between CHR-Cs and CHR-NCs revealed that adolescent CHR-Cs performed significantly worse than CHR-NCs on seven of eight neurocognitive tests; however, adult CHR-Cs performed significantly worse than CHR-NCs only in the visuospatial memory test.

Conclusions: The role of neurocognitive dysfunction may have different patterns and weights during the onset of psychosis in adolescents and adults at CHR, implicating the development of specific strategies that could monitor and improve cognitive function in CHR adolescents.