To what extent are treatment choices made on the basis of individual patient traits (eg, acuity of symptoms, comorbidity of conditions), rather than features or advantages of the modality alone?
The work of Myrna Weissman and colleagues has shown that combined treatment involving psychotherapy and medication has the best outcome. It is to my mind important to remember that MDD is a heterogeneous disorder-- as Nemeroff and Heim showed, certain patients who have a history of child maltreatment with little family history of depression respond very differently to medication (less well in general) than those patients who have experienced relatively little adversity but have a clear biological loading for depression based on family history. And then there is a gamut in between of course. Another factor is clinical reality: Some patients will not want to engage or cannot engage well in psychotherapy and only want medication. Whereas others refuse medication. And then there is also varying degrees of comorbidity... Evaluating on a case by case basis with knowledge of the existing empirical literature is for me the best approach to treating depressed patients.
It is a question of adaptive indication. This means both are important. Psychopharmacology in some cases can be a conditio sine qua non because for some patients it is necessary to use SSRI to get a minimum of activation which is needed for collaboration in tasks of psychotherapy. Antidepressiv medication and psychotherapy can therefore interact positively in activating common factores (treatment expectancy, prior symptom change, insight, allegiance, relationship bonding and working alliance etc.)