The use of patient-reported outcome (PRO) measures is growing in health care research, and demand for high-quality data from PROs has intensified with today’s focus on patient-centered care and outcomes.
A new paper by Assistant Professor Tam Nguyen and collaborators at Johns Hopkins University and the University of Texas at Austin introduces a research framework, item response theory (IRT), and its potential for improving the efficiency and accuracy of measuring PROs—questionnaires completed by patients about their perceived health status. The paper was published online in the journal Patient in January 2014.
An alternative to classical test theory, IRT “is a set of mathematical models that describe the relationship between an individual’s ‘ability’ or ‘trait’ and how they respond to items on a scale,” the authors write. Already used extensively in education, IRT is gaining traction in health care, they say; for example, it is being used to develop computerized adaptive testing, which bases items for test-takers on responses to previous questions (think MCATs).
Their article outlines IRT concepts and illustrates typical applications using existing data from 636 Korean- and Vietnamese-American adults who responded to two questionnaires with different formats, one assessing health literacy in the context of high blood pressure, and another measuring depression. Nguyen and her colleagues also provide a handy glossary of IRT-related terms, such as ability invariance, function curves, and monotonicity.
—Research summary by Debra Bradley Ruder