I suppose consensus might not be the right word; maybe something closer to bias or even superstition might be appropriate. But basically, I'm synthesizing my own view of several trends here. That being the case, I don't have a link that neatly sums this up. Perhaps this is even my own little hallucination. I'm only an expert by comparison to the general public, in the sense that I periodically try to survey the literature with the most open mind I can manage. But the components I'm thinking of are mostly:
1) A sense that antidepressants tend to work better the more severe the depression is. This often comes up as a counterpoint to the anti-antidepressant talking point that SSRIs "don't outperform placebo" (often, in turn, in reference to studies that directly or indirectly exclude or otherwise under-represent people with severe depression).
2) A sense that the more severe a person's condition is, the less likely that psychotherapy is to be effective, at least as a first-line treatment. There's a lot of heat over whether or why this is the case, with a common refrain among advocates of psychotherapy being that some people need antidepressant drugs temporarily in order to be able to "do the work" of psychotherapy.
3) A longstanding belief among various people in the field that "melancholic" depression is both more severe/treatment-resistant/chronic/personality-based and more amenable to biologically-oriented interventions. There is a particular thread of HPA-axis (hypothalamic-pituitary-adrenal) [1] dysfunction that has been a subject of research for decades and continues to attract interest, with a more specific clinical focus being the dexamethasone suppression test (e.g. [2]).
1) A sense that antidepressants tend to work better the more severe the depression is. This often comes up as a counterpoint to the anti-antidepressant talking point that SSRIs "don't outperform placebo" (often, in turn, in reference to studies that directly or indirectly exclude or otherwise under-represent people with severe depression).
2) A sense that the more severe a person's condition is, the less likely that psychotherapy is to be effective, at least as a first-line treatment. There's a lot of heat over whether or why this is the case, with a common refrain among advocates of psychotherapy being that some people need antidepressant drugs temporarily in order to be able to "do the work" of psychotherapy.
3) A longstanding belief among various people in the field that "melancholic" depression is both more severe/treatment-resistant/chronic/personality-based and more amenable to biologically-oriented interventions. There is a particular thread of HPA-axis (hypothalamic-pituitary-adrenal) [1] dysfunction that has been a subject of research for decades and continues to attract interest, with a more specific clinical focus being the dexamethasone suppression test (e.g. [2]).
[1] https://en.wikipedia.org/wiki/Hypothalamic%E2%80%93pituitary...
[2] https://www.ncbi.nlm.nih.gov/pubmed/27736954