BACKGROUND: Parental characteristics influence the risk of offspring suicide. In this study we wanted to separate the hereditary from the environmental influence of such factors by comparing their effects in the adopted versus non-adopted.MethodA register study was conducted in a national cohort of 2 471 496 individuals born between 1946 and 1968, including 27 600 national adoptees, followed-up for suicide during 1987-2001. Cox regression was used to calculate hazard ratios (HR) for suicide of socio-economic indicators of the childhood household and biological parents' suicide, alcohol abuse and psychiatric morbidity separately in the adopted and non-adopted. Differences in effects were tested in interaction analyses. RESULTS: Suicide and indicators of severe psychiatric disorder in the biological parents had similar effects on offspring suicide in the non-adopted and adopted (HR 1.5-2.3). Biological parents' alcohol abuse was a risk factor for suicide in the non-adopted group only (HR 1.8 v. 0.8, interaction effect: p=0.03). The effects of childhood household socio-economic factors on suicide were similar in adopted and non-adopted individuals, with growing up in a single parent household [HR 1.5 (95% confidence interval 1.4-1.5)] as the most important socio-economic risk factor for the non-adopted. CONCLUSIONS: The main familial effects of parental suicide and psychiatric morbidity on offspring suicide are not mediated by the post-natal environment or imitation, in contrast to effects of parental alcohol abuse that are primarily mediated by the post-natal environment. Social drift over generations because of psychiatric disorders does not seem likely to explain the association of socio-economic living conditions in childhood to suicide.