Using epidemiologic data, in an earlier study we formulated the hypothesis that estrogens can delay the onset of schizophrenia in females by raising the vulnerability threshold for this disease. In animal experiments, Häfner and colleagues found evidence that chronic estradiol treatment reduces the sensitivity of dopamine (D2) receptors in the brain. In the clinical study presented in this article, as a further step we examined the antipsychotic properties of estradiol in human females by testing whether schizophrenic symptomatology varies with estradiol serum levels throughout the menstrual cycle. We examined 32 acutely admitted female schizophrenia patients (Present State Examination/CATEGO diagnosis, ICD-9) with a history of regular menstrual cycles, ages 18 to 43 (mean = 30.5), during their hospital stays (3-8 weeks), analyzing hormonal parameters and applying various rating scales for psychopathology every 7 days. In all patients, estradiol serum levels were markedly reduced as compared with the normal population, and fluctuations throughout the cycle were dampened. Nevertheless, a significant association emerged between estradiol levels, on the one hand, and psychopathology scores, on the other--that is, the psychiatric symptomatology as assessed by the clinical psychiatrist (Brief Psychiatric Rating Scale, p < or = 0.01), behavior on the ward as assessed by the nursing staff (Nurses' Observation Scale for Inpatient Evaluation p < or = 0.01), paranoid tendencies and general well-being as assessed by the patients themselves (Paranoid-Depressivitäts-Skala paranoid score p < or = 0.05; Befindlichkeits-Skala p < or = 0.05). Psychopathology seems to improve when estradiol levels rise, and vice versa. These findings can be interpreted as further evidence for a protective effect of estrogens in schizophrenia, possibly due to the known anti-dopaminergic activities of these hormones.