Levels of residual morbidity in mood disorder patients followed up long-term under community conditions of treatment are remarkably high [1] . Both unipolar major depressive disorder (MDD) and bipolar disorder (BD) patients were ill 40–50% of follow-up time; in BD patients, three-quarters of that residual morbidity was depressive [1] . These striking findings encourage further consideration of aspects of mood disorders that might contribute to such outcomes in an era of supposedly effective modern treatments. Selected topics considered here include basic conceptualizations and diagnoses of mood disorders, risk factors associated with their natural history, and characteristics of available treatments, with an emphasis on BD.