Cognitive impairment and associated factors in elderly patients with schizophrenia: a retrospective observational study with phenotype analysis

Objective To examine the prevalence of cognitive impairment and factors associated with cognitive impairment among elderly patients with schizophrenia, and to explore potential clinical phenotypes using a clustering approach. Methods A retrospective cross-sectional study was conducted among 149 elderly inpatients with schizophrenia. Cognitive impairment was defined as Montreal Cognitive Assessment (MoCA) score < 26. Clinical characteristics, metabolic indicators, activities of daily living (ADL), and perceived social support (MSPSS) were collected. Multivariate logistic regression was performed to identify factors associated with cognitive impairment. Model classification performance was evaluated using a 2 × 2 table. K-means clustering was applied to explore cognitive and social functioning phenotypes. Results The prevalence of cognitive impairment among elderly patients with schizophrenia was 47.7%. Multivariate logistic regression analysis showed that higher MSPSS total score (OR = 0.609, 95% CI: 0.469-0.791, p < 0.001) and higher ADL score (OR = 0.553, 95% CI: 0.379-0.806, p = 0.002) were independently associated with lower odds of cognitive impairment. Diabetes mellitus was associated with increased odds (OR = 7.735, 95% CI: 1.129-52.979, p = 0.037), although the wide confidence interval indicates limited precision. The model demonstrated good classification performance (accuracy = 96.6%). K-means clustering identified three phenotypes (optimal functioning, mild-to-moderate impairment, and severe impairment), which were significantly associated with cognitive impairment status (χ² = 137.58, p < 0.001). Conclusion Cognitive impairment is prevalent among elderly patients with schizophrenia and is associated with reduced social support, lower functional independence, and diabetes. Cluster analysis suggests heterogeneity in cognitive and psychosocial profiles, supporting the potential value of stratified nursing management. Prospective studies are warranted to validate these findings.