This paper explores philosophical and methodological issues involved in determining 'What counts as making a meaningful difference?'--the fundamental question in health outcomes research and evidence-based practice. Eight diverse stakeholders are identified along with the competing agendas they bring to the debate. The power to define what counts as meaningful change in health status is typically rooted in disciplinary socialization, linguistic traditions and an orthodox consensus that circumscribes acceptable research foci and methods. The various implicit assumptions and sites of definitional authority have consequences both for health researchers and for the public being served. For researchers, their designs may be driven by available instrumentation, short political time lines, and statistical rather than clinical significance. Because methods to capture salient health outcomes are still being developed, failure to capture significant outcomes may be methodological in origin. The public may experience the consequences of subsequent programmatic funding decisions and the exclusionary potential of different categorization schemes. Despite this, the public seldom has a voice in deciding what constitutes a meaningful health outcome. Researchers are urged to involve their target population in designing studies to help rebalance relative stakeholder power, and to consider the consequences of their methodological decisions.