A framework for selecting outcome measures for ambulatory care research

Abstract

Selecting outcome measures must be considered in the context of the application the researcher has in mind, that is, describing the health of a population, predicting health status, or evaluating health outcomes. Most important, the outcome measure selected must be conceptually defensible--it needs to make sense. A general framework for selecting measures is discussed in the context of five critical characteristics: (1) sensibility, (2) reliability, (3) validity, (4) responsiveness, and (5) burden. A commonsense approach to outcome measurement selection weighs the relative importance of each of these characteristics with the framework and application clearly in mind.