Abstract
Objective: To investigate the diagnostic efficiency of the Hospital Anxiety and Depression Scale (HADS) in patients with breast cancer.
Methods: Women (303) recently diagnosed with stage I or II breast cancer completed the HADS and were psychiatrically assessed. Recommended cut-off scores for Depression (D) and Anxiety (A) Scales identified probable caseness and the accuracy was compared with DSM-IV diagnoses. Cut-off scores were varied to establish optimal accuracy.
Results: One hundred and eleven (36.6%) met criteria for depressive disorders and 25 (8.3%) met criteria for anxiety disorders. At the recommended cut-off score of 11 for the D scale, positive and negative predictive values were.75 and.64, respectively, while sensitivity and specificity were.05 and.99, respectively. For the A scale, positive predictive value (PPV) was.15; negative predictive value (NPV),.90; sensitivity,.08; and specificity,.87. Reducing the cut-off score to 5 produced PPV of.63, NPV of.74, sensitivity of.49 and specificity of.83 for the D scale;.14,.94,.75 and.45, respectively, for the A scale. Composite scores (cut-off of 22) had a PPV of.92, NPV of.57, sensitivity of.09 and specificity of.99.
Conclusions: Recommended cut-off scores for the HADS may result in under-reporting of psychiatric morbidity among women with early stage breast cancer. These results are consistent with other recent studies of the HADS. The type of constructs measured by the scale is the likely explanation for its limited utility when screening for psychiatric morbidity in early stage breast cancer.