Bipolar disorders are frequently not diagnosed until long after their onset, leaving patients with no or correspondingly inadequate treatment. The course of the disorder is all the more severe and the negative repercussions for those affected all the greater. Concerted research effort is therefore going into learning how to recognize bipolar disorders at an early stage. Drawing on current research results, this paper presents considerations for an integrative Early Symptom Scale with which persons at risk can be identified and timely intervention initiated. This will require prospective studies to determine the predictive power of the risk factors integrated into the scale.