Discordance between clinician-rated and patient-reported outcomes during 48-week community care for schizophrenia: longitudinal patterns and associations with insight

Background Recovery in schizophrenia is multidimensional, but whether clinician-rated and patient-reported outcomes (PROs) improve concordantly during community care is unclear. We examined their longitudinal discordance and its association with insight and treatment attitudes. Methods In this secondary analysis of a 48-week, multicity, naturalistic cohort, 306 community-dwelling adults with schizophrenia in Guangdong, China, were assessed at baseline and weeks 12, 24, 36, and 48 using the Positive and Negative Syndrome Scale (PANSS), Personal and Social Performance Scale (PSP), Schizophrenia Quality of Life Scale (SQLS), and Insight and Treatment Attitudes Questionnaire (ITAQ). Changes in PANSS, PSP, and SQLS were standardized using baseline standard deviation (SD) units, with positive values indicating improvement. Study-defined symptom-PRO and function-PRO recovery gaps were estimated using mixed-effects models adjusted for age, sex, city, and baseline treatment pathway. A joint week-48 bootstrap test was conducted, and ITAQ was decomposed into within-person and between-person components. Results At week 48, adjusted improvements were 0.30 SD for PANSS (95% CI, 0.23 to 0.37), 0.20 SD for PSP (95% CI, 0.12 to 0.28), and 0.46 SD for SQLS (95% CI, 0.37 to 0.54). The symptom-PRO gap was -0.16 SD (95% CI, -0.27 to -0.05), and the function-PRO gap was -0.26 SD (95% CI, -0.38 to -0.14), indicating greater patient-reported improvement than clinician-rated improvement (joint Q = 24.58, df = 2, p < 0.001). The gaps were evident from week 12; the symptom-PRO result was sensitive to city composition. At the 0.5-SD threshold, patient-reported-only improvement was more frequent than clinician-rated-only improvement for symptom-PRO (20.7% vs. 11.3%) and function-PRO (22.4% vs. 12.3%) comparisons. For each 1-SD occasion-specific increase in ITAQ, the gaps shifted 0.08 and 0.06 SD toward the clinician-rated direction, but the joint test was inconclusive (p = 0.064). Conclusion During routine community care, patient-reported quality of life improved more than clinician-rated psychopathology and social functioning, with the discrepancy larger and more robust for functioning. Parallel monitoring may better characterize symptom, function, and quality-of-life trajectories, while the role of time-varying insight and treatment attitudes warrants further study.