Assessment and screening of panic disorder in cancer patients: performance of the PHQ-PD.

OBJECTIVE This study's objective was to promote the transcultural adaptation of the Patient Health Questionnaire-Panic Disorder Module (PHQ-PD) for Brazilian Portuguese and to evaluate the discriminative validity of this scale in detecting PD among cancer patients. METHODS Adult cancer outpatients (n=400) from a specialized cancer hospital (61.50% female; 68.40% married; 56% incomplete elementary education or elementary school as the highest educational level) were assessed with the Structured Clinical Interview for DSM-IV and PHQ-PD. Using receiver operating characteristic (ROC) analyses, we determined the sensitivity and specificity values for the original PD algorithm and the PD screening. RESULTS The prevalence of PD in cancer patients (8.75%) was higher than the prevalence of PD for the general population. The original PD algorithm demonstrated an accuracy of 0.66, sensitivity of 0.31 and specificity of 0.94. The PD screening question in the PHQ-PD had a sensitivity of 0.66 and a specificity of 0.75 (accuracy=0.80). CONCLUSION PD screening questions in the PHQ-PD may be useful for identifying cancer patients with PD because of the high prevalence of PD in this population and because the questionnaire's sensitivity is greater than that of the original PD algorithm. Nevertheless, researchers and clinical practitioners should consider the original PD algorithm (five items) in the PHQ-PD when they investigate PD in patients because of the algorithm's high specificity. Individuals who are found to be positive for PD on screening should be referred for assessment and a thorough psychiatric interview that focuses on the differential diagnosis of an anxiety disorder relating to cancer.