Performance Measurement in Search of a Path

In this issue of the Journal, Landon and colleagues report on a Herculean undertaking — a study of quality-improvement interventions conducted at 44 community health centers.1 This study showed a modest improvement in some process measures and no improvement in intermediate or end-stage outcomes — results that are similar to those of most previous large-scale quality-improvement initiatives. As the authors correctly note, improved processes may not be accompanied by discernible improvements in outcomes for several reasons. A particularly important problem is that the majority of Health Plan Employer Data and Information Set–style performance measures used to guide most large-scale quality-improvement . . .