BACKGROUND Treatment-resistant depression (TRD) is severely disabling with no treatments proven to provide sustainable benefit when multiple medications, psychotherapy, and electroconvulsive therapy fail. (Malone, 2008). OBJECTIVES The perform a systematic review of published literature to evaluate recent health economics and outcomes research related to TRD to inform future research initiatives. METHODS English language citations in PubMed and Cochrane Library from January 1, 2000–October 1, 2008 were systematically reviewed. Standard search terms defined disease states (including TRD, severe depression (SD), chronic depression (CD), major depressive disorders (MDD), and suicidality); interventions (deep brain stimulation, electroconvulsive therapy, transcranial magnetic stimulation and vagus nerve stimulation); and outcomes (treatment outcomes, presenteeism, symptom burden, resource utilization, productivity, cost, economics, and quality of life(QOL)). RESULTS Initial searches yielded 8811 citations, narrowed to 5626 using date and language criteria. Title review further narrowed literature (reasons: not related to depression ~70%, clinical trials of antidepressant medications ~20%, and review/case studies ~10%), resulting in 411 unique abstracts for review and forty publications for abstraction. Clinical studies were most frequent (n=16) and economic models least frequent (n=3); a majority (n=22) reported a U.S. patient population. Thirteen publications reported on TRD, thirteen on MDD, and six reported on a “general depression” population. Reported definitions of TRD varied greatly (one medication failure (n=2), two medication failures (n=6), HAM-D score (n=3), diagnosis/treatment algorithm (n=1), not defined (n=1)). HAM-D (n=12) and MADRS (n=9) were the most frequently reported clinical outcomes measures. Cost outcomes were discussed in thirteen reports (direct costs (n=11), indirect costs (n=2)). CONCLUSIONS Definitions of TRD or CD reported in the literature vary substantially, and studies examining productivity of patients were confined to generic depression or MDD populations and would not be directly comparable to a CD or TRD population. Additional research is necessary to better define the cost and patient burden of TRD.