186 tures, though at this stage in the study data do not demonstrate a significant difference in prognosis. Rqli'rence Haas G.L., Sweeney J.A. ( 1992} Premorbid and onset features of first episode schizophrenia. Schizophrenia Bulletin, 18, 373 86. ARE SYMPTOMS OBSERVED DIFFERENTLY IN MALE AND FEMALE FIRST-EPISODE PATIENTS.'? M. Hambrecht, H. H'afner Department q/"PsychiatJ3' and Pa3'chotherapv, Universi O' o/ Kgiln. D-50924 KgihT, German.r Ol?jective." Preventive strategies in first-episode schizophrenia rely on the early and valid detection of signs and symptoms of the disorder. This study investigated whether symptoms are observed differently in male and female patients. Methods. After remission of acute psychosis, information on the early course was assessed systematically from 171 patients and their families by means of the IRAOS (Interview for the Retrospective Assessment of the Onset of Schizophrenia). Then measures of congruity between patients and families were calculated separately for male and female patients. Results: Highly significant Pearson correlations for data on the time of the first sign of the disorder were found in both sexes, but correlations were higher for female patients (0.78 vs. 0.73 ). As for the presence of symptoms, mean kappa and mean specificity scores over all symptoms did not show significant gender differences, but for female patients the sensitivity for observing symptoms was higher (0.56 vs. 0.52 p<0.07). Regarding single symptoms, higher agreement rates on presence and higher temporal correlations were found for female patients for aggressive affects, delusional symptoms, hallucinations, and formal thought disorders. For male patients, the congruity of self reports and relatives" reports was higher for anxiety, persecutory delusions, abnormal behaviours, and social deficits. Conclusions: While basic information on the early course of schizophrenia is valid in both sexes, specific symptoms are observed differently in male and female patients mostly due to gender specific role expectations. SEXUAL DYSFUNCTION AND SERUM PROLACTIN LEVELS IN PATIENTS USING CLASSICAL ANTIPSYCHOTICS OR RISPERIDONE H. Knegtering, M. Boks, C. ten Brink I~#liversi(v Ho,spital Groningen. P.O. Bo.v 30.00l, 9700 RB Groningen. The Netherlands. Email rikus@iname, corn In 1997 all patients using classical antipsychotics or risperi- done for 4 to 6 weeks were asked to participate in a open study evaluating sexual dysfunction, We used a questionnaire assess- ing changes in libido, erection, orgasm, ejaculation, menstrua- tion and galactorrhea related to the recent use ofantipsychotics. Serum levels of antipsychotics and prolactin were determined the same day, 55 out of 57 patients agreed to participate in the study (27 classical antipsychotics, 28 risperidone). There were no statisti- cal significant differences in socio-demographical or clinical parameters between both patient groups. The mean dosage in haloperidol equivalents for patients on risperidone was 7.1 mg/day SD 3.3 and for patients on classical antipsychotics 6.0 rag/day SD 4.8 (difference not significant). Serum prolactin levels were in patients receiving risperidone significantly higher tmean 1235 ME/I SD 991) versus patients receiving classical antipsychotics (mean 683 ME/I SD 695) (t=2.4 p=0.02). Less than 10% of the patients complained spontaneously about sexual dysfunctions. In response to the questionnaire 65% of the patients reported sexual problems related to the use of antipsychotics (55% in classical group and 75% in risperidone group). Patients using risperidone experienced significantly more sexual problems (mean number of problems 2.21 SD 0.56 problems/patient) compared to patients using classical antipsy- chotics (mean number of problems 0.75 SD I. 11 ) T-test p =0.01. Com'h4sion. Sexual dysfunctions are seldom spontaneously reported, but very common side effects when patients use antipsychotics. Clinicians should be actively asked for these dysfunctions. Risperidone causes more prolactin elevation and sexual dysfunctions than classical antipsychotics. SEX AND NEUROENDOCRINE DIFFERENCES IN RESPONSE TO TREATMENT WITH OLANZAPINE: A PRELIMINARY ANALYSIS S. Andersen, L. Cohen, J. Goldstein, M. Tohen, G. Tollefson Eli Lilly and Company, Indianapolis, Lil(v Corporate Center huliana, 46285, USA Recent studies of gender differences in treatment response to atypical neuroleptics such as clozapine have failed to demon- strate patterns of enhanced response in women (contrary to published studies of typical neuroleptics). These atypical studies may have included more chronically ill women thus generating a sampling bias. The current investigation attempts to identify gender differences in response to treatment with olanzapine and haloperidol and to unconfound sex from factors which might affect treatment response including chronicity of illness, positive and negative symptom expression, and prolactin secretion. In a large clinical trial of olanzapine versus haloperidol, 1996 patients were treated in a double-blind fashion over a six week acute therapy phase followed by a double blind extension phase. Preliminary analyses failed to demonstrate significant overall sex differences in treatment response to olanzapine as compared to haloperidol. Current analyses will be presented which uncon- found the impact of sex on treatment response from selection bias resulting from factors such as chronicity of illness, duration