Validation of the Greek version of the distress thermometer compared to the clinical interview for depression

Funding text This study was partly supported by research grants funded by the Cyprus University of Technology (statistical analysis) and Papageorgiou General Hospital (copyright expenses to NCCN). No funding sources were used in the design of the study, interpretation of data and in writing of the manuscript. AcknowledgementsBackground: The Distress Thermometer (DT) is worldwide the most commonly used instrument for quick screening of emotional burden in patients with cancer. In order to validate the Greek version of the DT in the Greek population we aimed to explore the capacity of the DT to identify patients with comorbid depressive diagnosis. Methods: We analyzed the routinely collected clinical data from 152 patients with cancer who had been evaluated by the consultation-liaison psychiatric service and had received a diagnosis of either depressive disorder or no psychiatric diagnosis. The score of the DT accompanied by the list of problems in the Problem List, the depression status, and the clinical and demographic characteristics entered the data sheet. Results: The ROC analysis revealed that the DT achieved a significant discrimination with an area under the curve of 0.79. At a cut-off point of 4, the DT identified 85% of the patients with an ICD-10 depressive diagnosis (sensitivity) and 60% of the patients without a psychiatric diagnosis (specificity). The positive predictive value was 44%, the negative predictive value 92% and the diagnostic odd ratio 8.88. Fatigue and emotional difficulties were the most commonly reported problems by the patients. Conclusion: The Greek version of the DT has a sufficient overall accuracy in classifying patients regarding the existence of depressive disorders, in the oncology setting. Therefore, it can be considered as a valid initial screening tool for depression in patients with cancer; patients scoring ≥4 should be assessed by a more thorough mental evaluation