Measuring preventable harm: helping science keep pace with policy.
This commentary discusses the shortcomings in policy measures to improve patient safety. To a large extent, policy measures are limited by an inability to distinguish between "inevitable" and "preventable" harm, and current policies (such as the Center for Medicare and Medicaid Services' [CMS] no pay for errors policy) assume that certain harms are entirely preventable. While this may be appropriate for certain adverse events, the authors argue that research should focus on the basic science of patient safety—determining which types of errors are truly preventable and developing error measurement methods that are reliable, scalable, and useful for comparing hospitals. Policy measures should be informed by this research, so that payment policies appropriately incentivize organizations to eliminate preventable harm. The merits of the CMS error payment policy were discussed in a prior editorial.