Hiroki Okada, Kazuki Yoshida, Ryota Hayashi, Seiichi Watanabe
A three-class solution was identified. Because one class was small, interpretation focused on the two larger classes. In one class, personal recovery was associated with social functioning, depressive symptoms, positive symptoms, and metacognition, reflecting a multidimensional association pattern consistent with previous findings. In contrast, the other class showed a different association pattern despite similar levels of symptoms and functioning. In this class, no single dominant predictor emerged, and recovery appeared to involve a distributed association pattern. Negative symptoms were associated with personal recovery in both larger classes.
OBJECTIVE: Previous studies have identified multiple factors associated with personal recovery in schizophrenia, including clinical symptoms, functioning, cognitive abilities, and psychosocial resources. However, findings regarding the strength and direction of these associations have been inconsistent, suggesting that different association structures may exist within the population. This study examined whether such inconsistencies reflect latent classes with different association structures underlying personal recovery.
METHODS: Participants were 101 individuals with schizophrenia or schizoaffective disorder. Personal recovery was treated as the dependent variable. Psychiatric symptoms, depressive symptoms, cognitive functioning, social functioning, and metacognition were explanatory variables. Finite mixture regression identified latent classes with different association patterns. Interaction and simple slope analyses examined class-specific associations.
RESULTS: A three-class solution was identified. Because one class was small, interpretation focused on the two larger classes. In one class, personal recovery was associated with social functioning, depressive symptoms, positive symptoms, and metacognition, reflecting a multidimensional association pattern consistent with previous findings. In contrast, the other class showed a different association pattern despite similar levels of symptoms and functioning. In this class, no single dominant predictor emerged, and recovery appeared to involve a distributed association pattern. Negative symptoms were associated with personal recovery in both larger classes.
CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: These findings suggest that personal recovery in schizophrenia cannot be adequately explained by a single average-based model. Instead, different association structures may exist within the population. Psychiatric rehabilitation may benefit from approaches that consider heterogeneity in recovery processes and tailor support to factors associated with each individual's recovery. (PsycInfo Database Record (c) 2026 APA, all rights reserved).