Validation of Dysfunctional Attitude Scale and Patient Health Questionnaire-9 Cognitive interviewing and Differential Item Functioning analysis

Abstract
Background and Objectives
Depression is a worldwide health problem that affects millions and causes personal suffering and economic strain. Existing assessment tools like PHQ-9 and DAS fail to fully capture the intricacies of individual experiences, especially when considering socioeconomic factors. By addressing these limitations, we can enhance the accuracy and significance of depression measures, resulting in more effective and personalised interventions. Through a mixed method approach, the study aimed to investigate both lived experiences and potential bias to enhance tool validity.
Method
The study employed both cognitive interview and Differential Item Functioning (DIF) analysis. Eight university students were interviewed to gain qualitative insights into their understanding and responses to DAS and PHQ-9 items. In addition, the data from 3500 patients who underwent psychological treatment at NHS AIPTs service was retrospectively analysed. The cognitive interviews examined participant interpretations of survey items, while DIF analysis looked at response biases related to socioeconomic status (SES) and specifically deprivation decile.
Results
Participants in cognitive interviews frequently relied on their personal experiences when answering questions about depression, highlighting the importance of relatable and clear items. The perception of ambiguity in multiple DAS items led to a decrease in their relatability and meaningfulness. The DIF analysis revealed significant variations on some items in how different socioeconomic groups respond to DAS and PHQ-9, suggesting potential biases. Therefore, this suggests that certain items in the scale may not capture the complete range of lived experiences of individuals from diverse backgrounds.
Conclusions
The study emphasises the importance of incorporating the patient's personal experience in creating depression assessment tools. The current measures, such as DAS and PHQ-9, may not effectively capture the full range of individual experiences, particularly regarding socioeconomic factors. By addressing these limitations, we can create assessment tools that are accurate and meaningful, leading to better interventions. Ongoing research should focus on exploring item bias and the role of socioeconomic and cultural factors in depression assessment, to enhance the development of assessment tools.