Accountability for health outcomes and the proper unit of analysis: What do the experts think?

Abstract

Background. An invitational conference was held in Dearborn, MI, in April of 1998 to discuss technical and conceptual issues related to the general topic of using outcomes data to compare plans, networks, and providers. Approximately 150 researchers, clinicians, purchasers, and representatives of accreditation bodies and government agencies attended. Survey of participants. At the opening session, attendees participated in an electronic survey exercise designed to identify areas of agreement or disagreement on controversial issues related to the main conference topic. Main findings. There was general agreement about the basic concept of health plan and provider accountability for health outcomes, and about the need for further development of data sources and Case-mix adjustment models. There was disagreement about other issues, including questions of who should bear the cost of collecting outcomes data and whether results should be analyzed at health plan, network, or individual clinician level. Conclusion. A group of experts agreed on the importance of reporting comparative outcomes data, but disagreed on many of the technical details of how that could best be done.