Validating the 15-item stress anxiety depression scale (SAD-15) using Rasch analysis

Stress, anxiety, and depression frequently co-occur in primary care, complicating accurate diagnosis and treatment. Existing screening tools typically evaluate each condition separately, which can be time consuming and inefficient. The 15-item Stress Anxiety Depression scale (SAD-15) was designed as a brief instrument to assess all three conditions simultaneously. This study examines its structural validity using Rasch measurement theory.

Data from 222 general practice patients were analyzed to assess the clinimetric properties of the SAD-15 and its subscales. Rasch analyses included tests for dimensionality, item and person fit, person separation reliability, local item dependency, differential item functioning (DIF), scale targeting, and identification of cut-off scores using the WHO-5 Well-Being Index as the criterion standard.

Initial analysis of the full SAD-15 revealed poor overall fit due to local item dependency. However, when analyzed as three composite scores, the scale and subscales demonstrated excellent Rasch model fit, unidimensionality, acceptable item and person fit, and high person separation reliability (PSI 0.77-0.93). No DIF was observed across gender or age groups. The scales were well-targeted, and ROC analysis yielded optimal cut-offs of 17 for the total scale, 7 for stress, and 5 for both anxiety and depression subscales.

The SAD-15 shows potential as an initial screening tool for simultaneously assessing stress, anxiety, and depression in primary care. Its composite structure supports its use as a first step before condition-specific diagnostics. Further validation in clinical populations and over time is recommended.