Addressing missing data in patient-reported outcome measures (PROMs): implications for comparing provider performance

摘要: Patient-reported outcome measures (PROMs) are now routinely collected in the English National Health Service (NHS) and used to compare reward hospital performance within a high-powered pay-for-performance scheme. However, PROMs prone missing data. For example, hospitals often fail administer pre-operative questionnaire at admission, or patients may refuse participate return their post-operative questionnaire. A key concern with is that individuals complete information tend be an unrepresentative sample of each provider, inferences based on cases will misleading. This study proposes strategy for addressing data survey using multiple imputation techniques, investigates its impact assessing provider performance. We find about relative sensitive assumptions made reasons