253. NEUROPSYCHOLOGICALLY-BASED CLUSTERING OF PSYCHOSOMATIC PATIENTS

Abstract Background The purpose of this study is to explore whether patients in psychosomatic clinics can be grouped based on their performance in neuropsychological tasks and to compare the demographic characteristics, psychopathological features, personality traits, and quality of life (QOL) among the groups. Aims & Objectives Since neuropsychological features are more physiologically grounded, we aim to identify a grouping model that transcends phenomenology and could guide diagnosis and treatment. Method A total of 182 adult patients attending a psychosomatic clinic in Taiwan (primarily presenting with somatic symptoms, depression, and anxiety) were included in this study. All participants underwent eight computerized tests from the Cambridge Neuropsychological Test Automated Battery (CANTAB), covering dimensions such as attention, executive function, memory, and social cognition. They also completed a range of self-report questionnaires, including the Patient Health Questionnaire-15 (measuring somatic distress), the Health Anxiety Questionnaire (measuring general aspects of illness-related anxiety), the Cognitions about Body and Health Questionnaire (CABAH, measuring cognitive aspects of illness-related anxiety), the Beck Depression Inventory-II (measuring depression), the Beck Anxiety Inventory (measuring the somatic aspect of anxiety), the Tridimensional Personality Questionnaire (measuring personality), and the WHOQOL-BREF (measuring QOL). Based on 27 indicators from the CANTAB, a two-step cluster analysis was conducted. The two clusters identified were compared in terms of demographics, psychopathologies, and QOL using independent t-test and Chi-square test. Analyses adjusted for sex and age were also considered. Results The cluster analysis identified two groups, with 106 individuals in cluster 1 and 76 in cluster 2. Among the CANTAB indicators, spatial working memory, intra-dimensional/extra-dimensional set shifting, and emotion recognition task had the greatest influence on the clustering; overall performance was lower in cluster 1. Intergroup comparisons showed that cluster 1 had older participants, a higher proportion of females, fewer years of education, higher CABAH scores, and lower harm avoidance levels. No significant intergroup differences were found in other demographic, psychopathological, personality, or QOL measures. After adjusting for sex and age, the significant relationship between harm avoidance and clustering disappeared, but the association between CABAH and clustering remained significant. Discussion & Conclusions Patients in psychosomatic clinics can be grouped into two clusters based on overall neuropsychological test performance. The most significant psychopathology distinguishing the two clusters is the cognitive aspect of illness-related anxiety. This finding may have implications for assessment and treatment strategies.