Individuals at clinical high risk (CHR) often present with a mixture of dif culties in addition to subthreshold psychotic symptoms, such as neurocognitive decline, premorbid dysfunction, and anxious/mood disorders.1–4 Heterogeneity impedes research by obscuring potentially discrete subtypes, which hinders clinical research, evalua- tion, and treatment.

Latent subgroup models are a novel approach in expli- cating risk in CHR and are within a group of statisti- cal methods known as latent variable mixture modeling (LVMM5). LVMM, such as latent pro le analysis (LPA), aims to identify homogenous subgroups within hetero- geneous cohorts, each with independent symptom con- stellations and differential associations with conversion and functional ability.5,6 LVMM may improve accuracy in identifying who among the CHR group will subsequently convert to psychosis. Imaging studies have provided sup- port for latent CHR subgroups, nding signi cant neuro- biological heterogeneity in gray matter volume.7

LVMM has been applied in 2 CHR studies with mixed results.8,9 Velthorst et al8 used a modi ed latent class fac- tor analysis to investigate symptom pro les of 288 CHR and unaffected control (UC) individuals. “At risk” and “healthy” classes emerged, but classi cation did not enhance prediction of conversion. Possible reasons for this include incorporation of UCs with limited variability and lack of diversity in predictive indices.

View or download the PDF.