Psychometric properties of The Hospital Anxiety and Depression Scale and The General Health Questionnaire‐20 in COPD inpatients

Aim : To compare the psychometric properties between the H ospital A nxiety and D epression S cale ( HADS ) and G eneral H ealth Q uestionnaire–version 20 ( GHQ ‐20) in detecting psychological distress in COPD patients referred to pulmonary rehabilitation, and to examine the factor structure of GHQ ‐20. Methods: The study comprised 161 consecutive patients with mild to very severe COPD . For comparison of mean scores between the HADS and GHQ ‐20, one sample t ‐test was used. Potential differences in the detection of possible and normal cases were analysed using P earson C hi square test. We report P earson's correlations within and between the questionnaires, and internal consistency was assessed through C hronbach's alpha. The factor structure of the GHQ ‐20 was examined through principal axis factoring ( PAF ) with oblique rotation and eigenvalue >1. Results: There were no differences in mean scores of psychological distress between HADS and GHQ ‐20 (12.03 vs. 24.73, p = 0.000), as well as no differences in the prevalence of possible cases of psychological distress (34.6 vs. 36.9, p = 0.000) and normal cases (65.4 vs. 63.1, p = 0.000). The observed difference between HADS and GHQ ‐20 regarding internal consistency was marginal, with C hronbach's alpha coefficients of 0.91 and 0.94, respectively. The PAF analysis resulted in a three‐factor solution for GHQ ‐20, notably with only two items loading on the third factor, giving an internal consistency <0.70. A two‐factor solution, comprising anxiety/depression and coping, may therefore be more appropriate. Conclusions: This study demonstrates no significant differences between the HADS and GHQ ‐20 in their ability to detect possible cases of psychological distress in a rehabilitation setting for COPD patients. Although the HADS and GHQ ‐20 are measuring different concepts of psychological distress, both questionnaires can be recommended as screening tools for detection of psychological distress in COPD inpatients. The GHQ ‐20 appears to be two‐dimensional, comprising anxiety/depression as one dimension, and coping as the other dimension.