Screening tools for psychiatry disorders in cancer setting: Caution when using

•Sensitivity indicators: use with caution in cancer setting (false negative rates: 50%).•Studies to determine scales’s psychometric properties in cancer population are necessary.•More specific criteria for screening psychiatric disorders in cancer population is vital.•Scales validated for other clinical conditions should be used with caution for cancer samples.

This study evaluated sensitivity/specificity of self-report instruments for the screening of psychiatric disorders/symptoms in cancer outpatients like: current/past major depression, dysthymia, alcohol abuse and dependence, tobacco abuse and dependence, panic disorder, social anxiety disorder, generalized anxiety disorder, obsessive compulsive disorder, posttraumatic stress disorder, phobias, current mania, delusion and hallucination. First, 1384 patients responded to several self-assessment instruments. Then, 400 patients, were then interviewed by telephone to confirm the presence/absence of psychiatric diagnosis. The ROC analyses showed moderate/excellent specificity (Patient Health Questionnaire-4 (PHQ-4)=0.75–0.88, Generalized Anxiety Disorder (GAD-7)=0.77, Fast Alcohol Screening Test (FAST)=0.83–0.86, Fagerström Test for Nicotine Dependence (FTND)=0.72, Brief version of the Patient Health Questionnaire-Panic Disorder Module (Brief-PD)=0.75, and Self Reporting Questionnaire – psychosis items=(0.68–0.91) but low sensitivity (PHQ-4=0.53–0.54, GAD-7=0.52, FAST=0.48–0.58, FTND=0.97, and Brief-PD=0.66)). These results suggest that sensitivity indicators should be used with caution in the cancer clinical setting.