Numerous studies have reported a lower mean age at first hospitalization for schizophrenia in males than in females. For this finding not only a gender difference in age at first onset of schizophrenia, but also other factors can be responsible. With the aim of providing a comprehensive analysis of gender differences in onset, symptomatology and course of schizophrenia, we started by testing the hypothesis postulating a gender difference in mean age at first hospitalization. By using the Danish and the Mannheim psychiatric case registers we analysed all hospital admissions for schizophrenia and related diagnoses and all previous admissions for other diagnoses of the Danish population in 1976 and those of the inhabitants of the German city of Mannheim in the period of 1978-80. Artefacts were controlled for systematically. The impact of intervening variables such as selection factors as well as the influence of gender on the ascription of a diagnosis of schizophrenia for the first time were assessed. We found a mean difference of 5 to 6 years in age at first hospitalization between males and females in both countries when a broad definition of the diagnosis was used and of 4 to 5 years when a restrictive definition was applied. The higher mean age at first hospitalization among females is not attributable to artefacts, diagnostic procedures or to any essential extent to gender differences in help-seeking behaviour or occupational status. When a distinction was made between 'single' and 'married', the significant difference in age at first hospitalization between the sexes disappeared in singles. With case register data and without knowing the chronological order of marriage and onset of the disease, it remains an open question whether this finding can be explained by purely correlative associations between sex, marital status and age of onset or by causal effects.