OBJECTIVE To systematically review the research evidence on the effectiveness of homeopathy for the treatment of depression and depressive disorders. METHODS A comprehensive search of major biomedical databases including MEDLINE, EMBASE, CINAHL, PsycINFO and the Cochrane Library was conducted. Specialist complementary and alternative medicine (CAM) databases including AMED, CISCOM and Hom-Inform were also searched. Additionally, efforts were made to identify unpublished and ongoing research using relevant sources and experts in the field. Relevant research was categorised by study type and appraised according to study design. Clinical commentaries were obtained for studies reporting clinical outcomes. RESULTS Only two randomised controlled trials (RCTs) were identified. One of these, a feasibility study, demonstrated problems with recruitment of patients in primary care. Several uncontrolled and observational studies have reported positive results including high levels of patient satisfaction but because of the lack of a control group, it is difficult to assess the extent to which any response is due to specific effects of homeopathy. Single-case reports/studies were the most frequently encountered clinical study type. We also found surveys, but no relevant qualitative research studies were located.: Adverse effects reported appear limited to 'remedy reactions' ('aggravations') including temporary worsening of symptoms, symptom shifts and reappearance of old symptoms. These remedy reactions were generally transient but in one study, aggravation of symptoms caused withdrawal of the treatment in one patient. CONCLUSIONS A comprehensive search for published and unpublished studies has demonstrated that the evidence for the effectiveness of homeopathy in depression is limited due to lack of clinical trials of high quality. Further research is required, and should include well-designed controlled studies with sufficient numbers of participants. Qualitative studies aimed at overcoming recruitment and other problems should precede further RCTs. Methodological options include the incorporation of preference arms or uncontrolled observational studies. The highly individualised nature of much homeopathic treatment and the specificity of response may require innovative methods of analysis of individual treatment response.