Biomarkers in psychiatry need to be mechanistic, not simply predictive of course and outcome? —Yes.

It is well acknowledged that a major problem facing psychiatric research is the greater complexity of the brain and psychiatric illness compared with organs and diseases that are the focus of other medical specialties. Nonetheless, psychiatry might achieve a needed paradigm shift by more closely aligning its research approach with those used in other branches of medicine. Such an alignment has generated progress for psychiatry in the past.¹ Considerable medical research targets biomarkers closely linked to mechanisms, illustrated by studies of infections, heart disease, and cancer. Psychiatry should also prioritize studies of biomarkers that not only predict course or treatment outcome but provide substantial insights about mechanisms.