Background: The Hospital Anxiety and Depression Scale (HADS) is frequently used in cancer studies, yet its utility for comparing people with cancer with people in the community is uncertain.
Methods: HADS scores were obtained from population-based samples of women with (n = 731) and without (n = 158) early-onset breast cancer. Psychometric properties were examined using differential item functioning (DIF) which is the presence of systematic group differences in certain response items independent of the trait being measured.
Results: Women with breast cancer scored lower than reference women on anxiety (mean (SD) 7.5 (4.3) vs. 8.2 (4.0); p = 0.06) and depression (3.3 (3.2) vs. 4.2 (3.0); p = 0.003). Group differences remained following adjustment for demographics. Time since diagnosis was not related to anxiety or depression scores. DIF was present in two anxiety and five depression items. Adjustment for DIF did not substantially change the anxiety or depression group differences.
Conclusion: Specific sampling or DIF effects do not explain the observation that women with breast cancer have lower levels of anxiety and depression than population controls. The psychometric properties of the HADS appear to be acceptable in these groups.