OBJECTIVE Recent studies have suggested a beneficial role of insulin on brain function and psychological well-being. This study was undertaken to examine whether fasting serum insulin levels are associated with the psychopathology profile in a cross-sectional sample of acutely ill non-diabetic inpatients with schizophrenia. METHODS Subjects were recruited from a county hospital. Each subject underwent a psychopathology assessment with the Positive and Negative Syndrome Scale (PANSS). A fasting blood sample was taken to measure serum insulin, plasma glucose and lipids. RESULTS Twenty-six subjects (7 females, 19 males) were included in the study. Pearson correlation analysis showed significant inverse relationships between serum insulin level and PANSS-Total, Positive Symptom subscale, and General Psychopathology subscale scores (r=-0.41, p=0.037; r=-0.49, p=0.010; r=-0.45, p=0.023, respectively). However, there was no significant relationship between serum insulin level and PANSS-Negative Symptom subscale score (r=-0.13, p=0.53). Partial correlation analysis showed that the inverse relationships between serum insulin levels and PANSS-Total, Positive Symptom subscale, and General Psychopathology subscale scores became even stronger after controlling for potential confounding variables including age, gender, race, family history of mental illness, age of illness onset and body-mass index (BMI). CONCLUSIONS Higher fasting serum insulin levels are associated with a better psychopathology profile in acutely ill non-diabetic inpatients with schizophrenia. It is speculated that insulin might improve clinical symptoms of schizophrenia by interacting with dopamine and other neurotransmitter systems.