Measuring palliative care self-efficacy of intellectual and developmental disability staff using Rasch models.

Abstract

The objectives of this study were to evaluate the psychometric properties of a palliative care self-efficacy instrument developed for intellectual and developmental disability (IDD) staff using Rasch analysis and assess the change in palliative care self-efficacy between 2 time points using Rasch analysis of stacked data. Staff from 4 nonprofit IDD services organizations in a US Midwestern state (n=98) answered 11 questions with Likert-style responses at baseline and 1-month follow-up post training. Rasch analysis was performed to examine rating scale structure, unidimensionality, local independence, overall model fit, person and item reliability and separation, targeting, individual item and personal fit, differential item functioning (DIF), and change in palliative care self-efficacy between 2 time points. The rating scale structure improved when 5 response categories were collapsed to 3. With the revised 3 response categories, the instrument demonstrated good psychometric properties. Principal components analysis of Rasch residuals supported the assumption of unidimensionality. Model fit statistics indicated an excellent fit of the data to the Rasch model. The instrument demonstrated good person and item reliability and separation. Gender-related DIF was found in 1 item, and work tenure-related DIF in 3 items. Overall palliative care self-efficacy improved between 2 time points. Rasch analysis allowed for a more thorough examination of this palliative care self-efficacy instrument than classical test theory and provided information on rating scale structure, targeting, DIF, and individual persons and items. These recommendations can improve this instrument for research and practical contexts.