: A substantial number of persons with disabilities are physically inactive and are therefore at increased risk for developing secondary conditions that may result in further functional loss. Efforts to monitor trends in physical activity levels in this population as well as to assess the efficacy of health promotion interventions have been limited by the absence of a valid, reliable and sensitive measure of physical activity. The proposed planning grant has four specific aims (1) develop a measure of physical activity that is specific to individuals with physical disabilities and that will accurately reflect physical activity levels; (2) pretest the instrument to identify problems with item interpretation and recall; (3) scale and validate the instrument using a Rasch measurement model; and (4) develop a dynamic prototype of the instrument using a computerized adaptive testing methodology. Each of these four objectives represents one phase of the proposed project. In Phase 1, an initial pool of physical activity items will be developed based in large part on the Compendium of Physical Activities, which consists of 587 physical activities ranging from the lowest levels of energy expenditure (i.e., standing) to the highest levels of physical activity (i.e., running """"""""x"""""""" mph). Additional items will be added to the pool to reflect disability-specific modes of activity (e.g., use of a wheelchair and other assistive devices). In Phase 2, cognitive interviews will be conducted with 12-15 individuals with physical disabilities in order to identify problems in recall of past physical activity and item interpretation. The initial pool of items will be reviewed and, if necessary, revised on the basis of data collected from these interviews. In Phase 3, survey forms consisting of subsets of items from the item pool will be mailed to 2,000 persons with lower extremity impairments (i.e., spinal cord injury, stroke, poliomyelitis, and multiple sclerosis). Data will be analyzed using a Rasch partial credit measurement model. In Phase 4, a computer adaptive testing prototype of the instrument will be developed and tested. The proposed planning grant will lay the foundation for a phase II initiative involving the implementation of the dynamic physical activity measure in rehabilitation settings in order to evaluate long-term outcomes of rehabilitation programs in increasing physical activity behavior and functional performance in physically disabled populations.
