""""""""Health and Literacy in Child and Adult Assessment,"""""""" a proposal to become a PROMIS Primary Research Site by The University of North Carolina at Chapel Hill and its partners, addresses 3 important problems in using computer adaptive testing (CAT) to measure patient-reported chronic diseases outcomes (PROs) in clinical research: (1) PRO assessment in children 8-17 years of age (Independent Project), (2) Development and testing items in low literacy populations in English and Spanish, and (3) Rigorous assessment of variables (e.g., many aspects of functional capacity or adherence to treatment) most appropriately conceptualized as dyadic (e.g., wife-husband, parent-child) rather than individual. Our Independent Project will (1) use IRT to evaluate item parameters for the PedsQLTM4.0 in 8,000 children in 4 core domains (physical, emotional, social, school/cognitive) and an asthma-specific domain, taking this widely used instrument as a starting point for a pediatric item bank, (2) develop new items and collect data in 1,060 children in North Carolina and Texas, using IRT to estimate item parameters, (3) develop and pilot test a CAT tool and use it to survey 375 children in NC and Texas. Our long-term objective is to develop CAT to assess PROs in clinical research across a wide range of chronic childhood diseases. However, this initial data collection will be in children with asthma and in healthy children. Our Network Project on literacy proposes developing a methodology for creating and testing PRO items in English and Spanish that function consistently across languages and literacy levels. We will provide leadership for (1) assessing patient literacy, (2) testing item comprehension in low literacy populations, (3) devising a strategy to develop new items that can be understood by people with low literacy, (4) testing all items in diverse samples of English- or Spanish-speaking people to evaluate differential item functioning by language or literacy in our large ethnically diverse patient population representing all literacy levels. Our Network Project on dyadic assessment proposes (1) specifying health variables that are largely interpersonal, (2) identifying existing measures of these and assessing their validity, (3) modifying measures that fail to assess interpersonal processes in a way that captures their interpersonal nature, (4) assisting the Network in collecting data on the modified measures and assessing their validity.
Showing the most recent 10 out of 124 publications
