作者: Yue Li , Xueya Cai , Dana B. Mukamel , Peter Cram
DOI: 10.1097/MLR.0B013E318270BC13
关键词:
摘要: OBJECTIVE : To determine the effect of postoperative length stay (LOS) on 30-day readmission after coronary artery bypass surgery. DATA SOURCES/STUDY SETTING We analyzed a final database consisting Medicare claims cohort (N=157,070) all fee-for-service beneficiaries undergoing surgery during 2007-2008, American Hospital Association annual survey file, and rural urban commuting area file. STUDY DESIGN regressed probability LOS using (1) (naive) logit model that controlled for observed patient hospital covariates only; (2) residual inclusion instrumental variable (IV) further unobserved confounding. The IV was defined measure hospital's risk-adjusted patients admitted gastrointestinal hemorrhage. PRINCIPAL FINDINGS naive predicted 1-day reduction in median (ie, from 6-5 d) lowered rate by 2 percentage points. increased 3 CONCLUSIONS findings indicate is associated with an risk among surgery, both confounding effects are corrected.