Objective: Underdetection and undertreatment of depression is a serious problem in palliative care. The aim of this study was to investigate the screenings ability for depression of a single question about depressed mood and secondly to find somatic or disease or treatment related variables that indicate a depression in terminally ill cancer patients.
Methods: A cross-sectional study was performed among 141 cancer patients with a life expectancyB/3 months, in a tertiary referral centre for cancer patients in the Netherlands. Depression was defined as a score higher than 20 on the Hospital Anxiety and Depression Scale. The course of the disease and sort of tumour, its treatment and demographic variables were recorded from medical files. The physician rated the patient's functional level on the Karnofsky Performance Rating Scale. Physical symptoms and the single question about depression were part of The Edmonton Symptom Assessment Scale. We performed logistic regression to test which combination of variables best predicted depression.
Results: Depression was present in 23%. All somatic symptoms were statistically significantly related to depression, most profoundly drowsiness. The question about depressed mood (ESAS) most strongly discriminated patients with a depression from those without a depression (the area under the ROC curve was 86%). None of the demographic or disease variables (including the Karnofsky Performance Scale) was a useful indicator of depression in this group.
Conclusion: In this study we found that asking patients for a depressed mood reasonably discriminated patients with a depression compared to an often used screenings instrument in palliative care. Somatic symptoms were found to be related to depression, even though depression was measured by an instrument that does not include physical symptoms to diagnose depression. Somatic symptoms and most notably drowsiness should therefore alert clinicians to think of a depression.