COMPUTER ADAPTIVE TESTING OF FUNCTIONAL STATUS - DESCRIPTION (provided by applicant): There is a pressing need expressed increasingly by health care providers, consumers, and policy makers for new and improved systems of functional outcome assessment that can be used to monitor health care quality across post-acute care settings and services. These monitoring systems must meet the difficult, dual requirement of precision and practicality. Function, in this context, refers to an individual's ability to carry out meaningful activity and participate in important daily routines. Traditional, fixed-form, functional outcome instruments designed for use in post-acute care settings are either too long, and therefore impractical for widespread use, or too brief and, therefore, lack adequate precision for outcome monitoring. The competing requirements of precision and practicality have long presented a dilemma to the post-acute care field and for outcomes researchers. The projects proposed during the period of this award will address these issues by developing and validating computer-adaptive testing (CAT) outcome instruments that are ideally suited for the challenge of precise and practical monitoring of functional status within post-acute health care settings. By releasing the PI from substantial administrative and management obligations as Director of the Center for Rehabilitation Effectiveness at Sargent College, this award will allow the PI to pursue a focused research program on this critical area of need. This award will allow the PI to also increase his research mentoring activities with pre- and postdoctoral trainees, further support the development of health services research careers for fellows and doctoral students, increase the number of peer-reviewed papers he will complete, and enhance his technical skills in the use of CATs. The three general long-term objectives of this award will be supported by a series of current and pending grant proposals: 1) CAT development for adult (R01HD43568-03) and pediatric post-acute care (R44 pending) settings, 2) studies to determine the effect of multiple respondents (clinician, patient, proxy) and modes (self-report vs. interview) of CAT scoring (R21 pending), and 3) projects to determine the responsiveness and minimal clinical important differences of CATs for long-term follow-up and quality monitoring. In collaborative studies, the PI will pursue new developments in design of CATs for health care applications, including improved algorithms for item selection and evaluation of various item response theory models for scoring and report functions.
