This study evaluated specific influences of a cognitive marital treatment (CMT) for depression based on an integration of cognitive theory of depression and systems theory. The effects of CMT on variables representing cognitions, emotions, and behaviors were compared with the effects of traditional cognitive therapy (CT), pharmacotherapy (PT), and no treatment (NT). Subjects were Major Depression Disorder and Dysthymic outpatients and their spouses. The changes considered were for pre–post treatment and pre-treatment–follow-up. CMT affected a wider range of variables than other treatments. In most instances it was also superior to the NT condition. Most of the gains were manifested at termination and lasted through follow-up. The variables affected by CMT were patients' and spouses' cognitions and emotions. PT had best effects on patients' emotions. CT affected patients' cognitions, but did not achieve superiority on any of the compared variables. None of the treatments produced significant changes in behaviors.