Lithium in treatment failures.

Thirty-five patients with chronic and incapacitating mental illness who had not responded to the usual pharmacological and interactional therapies were treated with lithium. None of these patients appeared to be suffering from manic-depressive illness. If a trial of lithium resulted in unexpected improvement, lithium's contribution was assessed by double blind substitution of a placebo followed by lithium in an A-B-A-B design in which the patient served as his own control. Five patients (14 per cent) improved dramatically; in retrospect, four of these five patients suffered from nonremitting forms of manic-depressive illness, and the fifth patient suffered from a severe obsessive compulsive neurosis. Six other chronically hospitalized patients improved to the point of unexpected discharge. A trial of lithium therapy is recommended for the "backward" or intractable patient.