Psychometric Properties of the Hospital Anxiety and Depression Scale (HADS) in Previously Hospitalized COVID-19 Patients

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus is associated with psychological/emotional disturbances. This study aimed to assess internal consistency, reliability, and construct validity of the Hospital Anxiety and Depressive Scale (HADS), as a patient-reported outcome measure (PROM) for evaluating emotional consequences of SARS-CoV-2 in hospitalized COVID-19 survivors with long COVID. The LONG-COVID-EXP-CM is a multicenter cohort study including patients hospitalized by COVID-19 during the first wave of the pandemic in five hospitals in Madrid. A total of 1969 (age: 61 ± 16 years, 46.5% women) COVID-19 survivors experiencing post-COVID symptoms a mean of 8.4 ± 1.5 months after hospital discharge completed HADS. Internal consistency (Cronbach α), reliability (item-internal consistency, item-discriminant validity), construct validity (confirmatory factor analysis), and floor effect and ceiling effect were calculated. The mean time for fulfilling HADS was 65 ± 12 s. A ceiling effect ranging from 1.99% to 13.74% and a floor effect ranging from 43.05% to 77.77% was observed. Based on the item-scale correlation coefficients, the Cronbach’s alpha values reflecting the internal consistency reliability were 0.890 for the anxiety scale (HADS-A) and 0.856 for the depressive scale (HADS-D) The correlation coefficient between HADS-A and HADS-D scores was excellent (r: 0.878). The confirmatory factor analysis revealed that five out of the seven fitness indexes were excellent: CFI = 0.969, NNFI = 0.963; TLI = 0.963; AGFI = 0.951; GFI = 0.972), supporting good construct validity. In conclusion, this study indicates that both anxiety and depressive symptoms scales of HADS had overall good psychometric properties to be used for assessing psychological and emotional stress in COVID-19 survivors with long COVID.