OBJECTIVE: A recent consensus conference has proposed guidelines for the monitoring for diabetes in patients with schizophrenia and also identifies the need of long-term prospective studies. METHOD: A large scale prospective study on metabolic risks of antipsychotic medication is currently ongoing. At baseline, patients get a full laboratory screening, ECG and an oral glucose tolerance test (OGTT). Baseline data on 100 non-diabetic patients at study inclusion and stable on medication for at least 6 months are presented. RESULTS: Glucose abnormalities are found in 22% of patients at baseline. A monitoring protocol based only on fasting glucose would not have detected 63.6% of these patients with classifiable glucose abnormalities in our sample. Fasting insulin and measures for insulin resistance have a high predictive value for abnormalities late in the OGTT. CONCLUSION: Already at baseline, metabolic problems are frequently present in patients with schizophrenia treated with antipsychotics. Adding assessment of fasting insulin in a monitoring protocol improves detection of glucose abnormalities late in an OGTT.