427. VALIDITY OF A SMARTPHONE APPLICATION FOR SELF-MONITORING PSYCHIATRIC SYMPTOMS IN PATIENTS WITH SCHIZOPHRENIA

Abstract Background Despite increasing research on digital technologies for psychiatric disorders, studies specifically examining self-monitoring of symptoms with smartphone applications by patients with schizophrenia remain limited. Aims & Objectives This study aims to evaluate the validity and reliability of self-monitoring psychiatric symptoms using a smartphone application among patients with schizophrenia at Mindlink, a community-based early intervention center. Method Fifty-three young patients with schizophrenia spectrum disorders participated. They rated their psychiatric symptoms across five domains-delusions, hallucinations, anxiety, depression, and perceived stress-using an 11-point Likert scale at baseline, 1 week, 8 weeks, and 16 weeks. Test-retest reliability was assessed using intraclass correlation coefficients (ICCs) between baseline and 1-week ratings. Concurrent validity was determined by correlating app-based ratings with established self-report and clinician-administered scales, including the Eppendorf Schizophrenia Inventory, Hamilton Program for Schizophrenia Voices Questionnaire, Beck Depression Inventory, Generalized Anxiety Disorder-7, and Perceived Stress Scale. The accuracy of the app's depression rating was assessed using receiver operating characteristic (ROC) analysis. Results ICCs for test-retest reliability were high across all symptom domains, ranging from 0.741 to 0.876 (p < 0.001). Significant correlations were observed between app-based ratings and formal assessments at all time points. ROC analysis for single-item self-ratings using the app yielded an area under the curve of 0.829 (p = 0.002), indicating good accuracy. Discussion & Conclusions This study demonstrates that self-monitoring of key symptoms and stress using a smartphone application is valid and reliable for patients with schizophrenia. These findings support the app's potential to enhance symptom management and enable early detection of relapse in this population.