Further research needed: A comment on Coyne and van Sonderen's call to abandon the Hospital Anxiety and Depression Scale

摘要

Coyne and van Sonderen [1] provide a commentary to Cosco et al.'s [2] systematic review of the Hospital Anxiety and Depression Scale (HADS) that calls for the abandonment of this widely used assessment of psychological distress. This judgement is based mainly on their perception of inconsistent findings with regard to its factor structure. We argue that the inconsistency is due to the hierarchical structure of anxiety and depression and that interpretation of literature from this perspective resolves the issue.There is actually considerable consistency bearing in mind the statistical method employed. Cosco et al. identify 50 studies examining the HADS latent structure: all four of the studies employing item response theory (IRT) support a unidimensional structure; 18 of the 22 (82%) studies employing solely exploratory factor analysis (EFA) support a two-factor anxiety-depression structure; and 16 out of 24 (67%) studies employing confirmatory factor analytic (CFA) support a three-factor structure, with a further 7 (29%) supporting a two-factor structure. Notably, the three-factor structure commonly relates to a tripartite model consisting of autonomic anxiety, anhedonic depression and negative affectivity. It is this implicitly hierarchical structure that hints to the reason for inconsistent findings across methods.