While early intervention strategies have been developed for psychotic disorders, affective psychoses and bipolar disorders have been neglected by this movement. However, when considering that outcome of bipolar disorders is often not as favorable as previously thought and that delay between illness onset and introduction of an adequate treatment is often very long, such developments seem clearly justified. In this paper we briefly review arguments supporting early intervention in bipolar disorders, the practical and theoretical obstacles that still need to be overcome, the strategies that may already now contribute to decrease treatment delay, and we describe current state of research regarding identification of the prodromal phase of bipolar disorders.