Abstract
Although enthusiasm for physician profiling dates back to the origins of scientific medicine, the track record of profiling as an intervention is not strong. Profiling appears to be most valuable for simple interventions and processes, but its acceptance, usefulness, and impact will be influenced by future trends. Health care organizations will continue to be complex, and multiple overlapping profiles will be the norm. Technology will advance, but concerns over privacy and confidentiality will influence the social, political, and regulatory approaches to profiling. Public sector purchasers and consumers will increasingly demand profiling, but data will continue to lack the ability to interpret complex profiling data. Medical practice will increasingly accept profiling as part of ongoing quality improvement efforts, but profiling may become focused on key processes, not outcomes. Statistical, cognitive, and epistemological challenges to profiling will remain, and profiling may simultaneously become more complex and more simplified with the rise of information brokers. Greater attention to the human side of profiling will enhance its effectiveness.