OBJECTIVE: The aim of this study was to estimate the potential bias by personality traits for ratings on the Positive and Negative Syndrome Scale (PANSS).METHODS: Personality dimensions (five factor model), personality traits (SCID-II) and PANSS scores were assessed prospectively in 45 patients with schizophrenia spectrum disorders (SSD).RESULTS: Borderline (r=0.34; p=0.021), avoidant (r=0.66; p<0.001) and depressive (r=0.51; p<0.001) personality traits were significantly correlated with the PANSS total score. There were significant correlations for all PANSS subscores with the exemption of PANSS positive. In multivariate analyses, the final models for PANSS total score and PANSS depressive explained a total of 45.3% and 54.3% of the variance. Avoidant traits could lead to a difference of 13.1 (95% CI: 5.6-20.7) points regarding PANSS total score, depressive traits could cause differences of 4.8 points (95% CI: 2.2-7.3) for PANSS depressive subscore.CONCLUSION: Although PANSS positive subscore and PANSS excited component are relatively robust against bias by personality traits, PANSS total score and the remaining subscores are affected to a clinically relevant degree. Outcome studies in SSD patients should control for personality traits.