Prediction of Treatment Outcome With Lithium

The fundamental issue underlying the question of predicting treatment outcome with lithium is whether there are meaningful subgroups within the broad spectrum of patients with affective disorders. A pessimistic view of our progress in resolving this question would be that we have not progressed in our collective research efforts beyond the point reached by Cade¹ 30 years ago and by Schou² 20 years ago. Many clinicians have discussed the use of drugs to identify specific disorders "hidden" within broad diagnostic categories: the examples of penicillin in cerebral syphilis, and of cyanocobalamin in pernicious anemia are often cited to support this approach. For our purposes, the major limitation of these analogies is that cerebral syphilis and pernicious anemia rarely show spontaneous remissions, whereas the affective disorders usually do. So these analogies are not entirely apt: when dealing with the affective disorders, we are faced with much more difficult methodological