A valid method to diagnose depression in palliative care has not been established. In this study, we aim to determine the prevalence of depression and the discriminant validity of the items of four sets of diagnostic criteria in palliative care. This is a cross-sectional study on 240 palliative care patients where the presence of depression was based on the Diagnostic and Statistical Manual of Mental Disorders, DSM–IV Criteria, Modified DSM–IV Criteria, Cavanaugh Criteria, and Endicott’s Criteria’s. Anxiety, depression, and distress were measured with Hospital Anxiety and Depression Scale and Distress Thermometer. The prevalence of depression among the palliative care patients was highest based on the Modified DSM–IV Criteria (23.3%), followed by the Endicott’s Criteria (13.8%), DSM–IV Criteria (9.2%), and Cavanaugh Criteria (5%). There were significant differences (p < 0.05) in the depressive symptoms showed by DSM–IV item 1 (dysphoric mood), item 2 (loss of interest or pleasure), and Endicott’s criteria item 8 (brooding, self-pity, or pessimism) among the palliative patients, even after adjustment for the anxiety symptoms and distress level. We found that dysphoric mood, loss of interest, and pessimism are the main features of depression in palliative patients. These symptoms should be given more attention in identifying depression in palliative care patients.