: Background: There is an urgent need to develop a practical yet precise system to measure health status for children with chronic health conditions across a continuum of medical care settings and services. This planning grant will be used to resolve conceptual, sampling, and data collection issues in order to develop a generic, norm-based computer adapted testing (CAT) system for measuring health and functioning for children and youth.
The specific aims of this planning grant are to: (1) develop a consensus conceptual framework, construct definitions, and specific item content for a broad conceptualization of health that includes multiple domains of physical functioning, psychological well-being and social and role participation (represented here by """"""""health status""""""""); (2) demonstrate the ability to collect item pool data on the Internet from both caregivers and children; (3) build a prototype web-based Pediatric Health Status CAT; and (4) evaluate the feasibility, efficiency and accuracy of a prototype web-based Pediatric Health Status CAT. Research Design and Methods: The investigative team, which includes experts in CAT technology, clinical pediatrics, and instrument development, will guide a team of content specialists in developing consensus definitions of concepts through a series of parent and child focus groups, review of existing child health measures, secondary analyses of child health data from three sources - the Pediatric Evaluation of Disability Inventory (PEDI), Child Health Questionnaire (CHQ), and RAND Health Insurance Experiment (HIE) databases; and expert panel consensus meetings. In year 2, the PIs will explore strategies to collect item pool data from a representative sample of caregivers and children via a web based data collection portal and identify any effects of sampling bias on the structure of the item pool. Item response theory and other statistical methods will be used to evaluate the structure of item pools, establish item calibrations, and create preliminary algorithms for a prototype CAT system. The accuracy and precision of CAT-based score estimates and reductions in respondent burden will be tested using computer simulations. In year 3, 100 caregivers/children will be recruited for a field study to evaluate the feasibility, efficiency and accuracy of a prototype Pediatric CAT. Significance: In order to build a norm-based CAT for children, one needs to find an inexpensive means to link content and cross-calibrate scores across instruments and to norm items in large samples of children with and without chronic health conditions. By demonstrating the utility of the Internet to build and administer a web-based CAT system, one can take advantage of the efficiencies it offers to construct a fully developed norm-based CAT system in future work. A web based CAT used to assess pediatric health and functioning may eliminate the current fragmentation of widely-used generic instruments used across age and diagnostic groups and achieve a substantial improvement in measurement breadth and practicality as required for routine use of health status tools in medical research and practice.
