A generalizability study of the instrumental activities of daily living profile.

Abstract

To establish generalizability estimates of the Instrumental Activities of Daily Living (IADL) Profile when administered to persons with a moderate or severe traumatic brain injury (TBI) within their home and community environment. Generalizability theory was used to estimate generalizability and dependability coefficients as well as the relative contribution of identified sources of measurement error to total measurement error. Decision studies were used to enable the investigators to determine the optimal measurement design. The IADL Profile was administered in subjects' homes and community environments. Convenience sample of adults with a moderate or severe TBI (N=30, aged 16-65y). Not applicable. IADL Profile scores include 6 factor scores (going to the grocery store/shopping for groceries, having a meal with guests/cleaning up, putting on outdoor clothing, obtaining information, making a budget, and preparing a hot meal) and 1 total score. The greatest sources of measurement error were the subject-item interactions (3-random-facet design) and the subject-rater interactions (2-random-facet design). One hundred percent of generalizability coefficients of factor scores indicated acceptable to excellent reliability. Indices of dependability confirmed that 1 evaluator could reliably score the tool on a single occasion after having received a 3-day training workshop. The IADL Profile administered to persons with a moderate or severe TBI provides occupational therapists with a reliable set of measures of IADL independence capable of both capturing and analyzing the complex interactions between personal and environmental factors.