This post is based on a peer-reviewed journal article (may require subscription to access) that is available online and is due to be published in the Journal of Affective Disorders in March 2018 as: Evaluation of the Hospital Anxiety and Depression Scale (HADS) in screening stroke patients for symptoms: Item Response Theory (IRT) analysis. Authors: Salma A. Ayis, Luis Ayerbe, Mark Ashworth, Charles DA Wolfe. The authors are from King’s College London, Guy’s and St Thomas’ NHS Foundation Trust, King’s College Hospital NHS Foundation Trust, and Queen Mary University. The authors have not reviewed this post.
Many stroke survivors experience depression and anxiety, which impacts on function and recovery (as well as the mental health burden) (e.g. see post Long-term outcomes: survivors’ experiences up to 15 years after stroke). However, estimates for rates of depression and anxiety vary considerably. The variation could be due to several things: for example, the way the patient is assessed, and what the ‘cut-off’ points are for diagnosing someone with anxiety or depression.
The standard tool used in primary and secondary care for determining whether someone has depression and/or anxiety is the Hospital Anxiety and Depression scale (HADS). In an evaluation of HADS (paper referenced at top), researchers used data from the South London Stroke Register (SLSR) to understand what information the HADS items give about anxiety and depression in the SLSR population up to 5 years after stroke.
Statistical methods (factor analysis and Item Response Theory methods) were used with the SLSR data. The findings confirmed that HADS measures depression and anxiety as two distinct domains, but also suggest that some HADS items are more useful than others for clinicians who want to determine whether someone has anxiety and/or depression, or to measure the severity of symptoms. The study also showed that some items seemed to give similar information about symptoms. Potentially, clinicians/researchers could use items selected because of their properties, rather than all the items, enabling more precise patient screening. The researchers suggest that more needs to be known about how stroke survivors perceive the items, to determine which items might not be necessary.