A cross‐sectional study of agreement between the Hospital Anxiety and Depression Scale and patient‐ and radiation oncologist–reported single‐item assessment of depression and anxiety. Issue 7 (16th May 2018)

• Record Type: Journal Article
• Authors: Mackenzie, Lisa; Carey, Mariko; Suzuki, Eiji; Yoshimura, Michio; Toi, Masakazu; D'Este, Catherine; Sanson‐Fisher, Rob
• Is Part Of: Psycho-oncology. Volume 27:Issue 7(2018)
• Page Start: 1840
• Page End: 1846
• Publication Date: 2018-05-16
• Subjects: anxiety -- cancer -- depression -- oncology -- psychosocial support systems -- questionnaires -- radiotherapy; Cancer -- Psychological aspects -- Periodicals; Cancer -- Social aspects -- Periodicals; Neoplasms -- psychology -- Periodicals; 616.9940019
• DOI: 10.1002/pon.4736
• Languages: English
• ISSNs: 1057-9249

Abstract

Abstract: Objective: To describe among radiation oncology patients: (1) the proportion likely to be experiencing symptoms of depression and anxiety as identified by ( a ) the Hospital Anxiety and Depression Scale (HADS; standardised tool), ( b ) patient‐reported single items (ultrashort tool), and ( c ) radiation oncologist–reported single items (clinician judgement); (2) preferences for being offered psychological support; and (3) agreement between single‐item measures and the HADS. Methods: Adult cancer patients (n = 152; consent rate 58%) receiving radiotherapy completed a touchscreen tablet survey assessing symptoms of anxiety and depression (HADS and a single‐item tool) and support preferences. Each participant's treating radiation oncologist completed a survey assessing his or her perception of whether the patient was anxious or depressed. Results: Prevalence estimates for likely depression (6.9‐18%) and anxiety (17‐33%) overlapped across the 3 measures. Overall, only 9.9% of patients (95% CI, 5.6%‐16%) wanted to be offered psychological support. For depression, agreement between the HADS and ultrashort tool was fair (κ = 0.37, P < 0.0001); agreement between the HADS and clinician judgement was slight (κ = 0.14, P < 0.05). For anxiety, agreement between the HADS and clinician judgement was not significantly greater than chance alone (κ = 0.04, P = 0.33), and agreement between the HADS and ultrashort tool was moderate (κ = 0.49, P < 0.0001). Conclusions: These findings highlight the important role that oncology consultations play in interpreting assessment tool results and responding to individual patient's history and preferences for psychological support.