The evidence for the integrative cognitive model of bipolar disorder is yet to be reviewed. We evaluate the evidence for four hypotheses derived uniquely from the model: extreme positive and negative appraisals of internal states are associated with (1) risk of developing bipolar disorder; (2) bipolar disorder diagnosis; (3) relevant clinical and functional outcomes including hypomanic and depressive mood symptoms; and (4) outcomes over time. Research involving individuals with diagnosed mood disorders as well as non-clinical populations is reviewed. The hypotheses were broadly supported and several consistent findings were not accounted for by alternative psychological models of bipolar disorder. The evidence base is limited by a relative paucity of prospective studies. Implications for theory, research and clinical practice are discussed.