Premorbid functioning trajectories and the one-year course of cognitive performance in first-episode psychosis: a cluster analysis in PSYSCAN

Le résumé fourni par la source

Background: We examined the course of cognitive performance in first-episode psychosis (FEP) compared to healthy controls (HC), and whether this varied across subgroups of patients defined by premorbid functioning (PMF) trajectories, using a clustering approach. Methods: Data were collected in 302 FEP and 136 HC subjects participating in PSYSCAN (HEALTH.2013.2.2.1-2-FEP). K-means clustering (Euclidean distance) was used to cluster longitudinal trajectories of different PMF domains simultaneously. Since PMF was assessed retrospectively using the Premorbid Adjustment Scale (PAS), findings should be interpreted with caution, although PAS ratings have shown reasonable validity against prospective data. Results: As expected, FEP showed impaired performance across all cognitive domains compared to HC. We identified four trajectories of PMF: a normal premorbid developmental trajectory (globally-normal, 21 %), stable intermediate PMF across domains (stable-intermediate, 29 %), stable poor or deteriorating PMF in the academic domain (normal-social/poor-academic, 29 %), and a globally impaired group with poor/deteriorating PMF across domains (globally-poor, 21 %). These clusters showed distinct levels of post-onset impairments in sustained visual attention, visual working memory and emotion recognition. Conclusions: This study confirms a positive association between PMF and cognitive performance in the early years following psychosis onset. It aligns with findings that individuals later diagnosed with schizophrenia already show developmental deficits/lags from childhood to early adolescence compared to normally developing children. As PMF can be considered a proxy for cognitive reserve, our results suggest that higher reserve acts as a buffer against cognitive decline and supports better performance on sustained visual attention, complex visual working memory, and aspects of emotion recognition.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.