Development of an Item Bank for the Assessment of Depression in Persons With Mental Illnesses and Physical Diseases Using Rasch Analysis

Objective: The calibration of item banks provides the basis for computerized adaptive testing that ensures high diagnostic precision and minimizes participants' test burden. The present study aimed at developing a new item bank that allows for assessing depression in persons with mental and persons with somatic diseases. Method: The sample consisted of 161 participants treated for a depressive syndrome, and 206 participants with somatic illnesses (103 cardiologic, 103 otorhinolaryngologic; overall mean age = 44.1 years, SD = 14.0; 44.7% women) to allow for validation of the item bank in both groups. Persons answered a pool of 182 depression items on a 5-point Likert scale. Results: Evaluation of Rasch model fit (infit < 1.3), differential item functioning, dimensionality, local independence, item spread, item and person separation (> 2.0), and reliability (>.80) resulted in a bank of 79 items with good psychometric properties. Conclusions: The bank provides items with a wide range of content coverage and may serve as a sound basis for computerized adaptive testing applications. It might also be useful for researchers who wish to develop new fixed-length scales for the assessment of depression in specific rehabilitation settings.