Multistate utilization, processes, and outcomes of carotid endarterectomy.

OBJECTIVES: The purpose of this study was to describe variation in utilization, care processes, and outcomes for carotid endarterectomy (CEA) procedures in 10 states. METHODS: We reviewed the medical records of Medicare patients who underwent 10,561 CEA procedures between June 1, 1995, and May 31, 1996, in 10 different states to determine indications, care processes, and outcomes. This study also included medical record review of hospital readmissions within 30 days of the procedure and identification of out-of-hospital deaths from the Medicare beneficiary files. RESULTS: Utilization rates of CEA varied from 25.7 to 38.4 procedures per 10,000 Medicare beneficiaries among states. The overall combined event rate (30-day stroke or mortality) was 5.2% for primary CEA alone (n = 9945). The mortality rate was 1.5%, and the nonfatal stroke rate was 3.7%. Combined event rates (CEA alone) by surgical indication were 7.7% for stroke (n = 1037), 7.4% for transient ischemic attack (n = 1304), 5.3% for nonspecific symptoms (n = 3713), and 3.7% for asymptomatic patients (n = 3891). The combined event rates (CEA alone) among states ranged from 4.1% to 7.7% with the event rates in asymptomatic patients ranging from 2.3% to 6.7%. In a multivariate analysis (correcting for indication), the use of preoperative antiplatelet agents (odds ratio [OR], 0.70), intraoperative heparin (OR, 0.49), and patch angioplasty (OR, 0.73) was significantly associated with lower combined event rates. There were significant differences among states in the use of preoperative antiplatelet therapy (range, 56%-70%) and patch angioplasty (range, 11%-49%). Combined event rates for repeat procedures (n = 380) and CEA combined with coronary artery bypass grafting (n = 236) were 6.3% and 17.4%, respectively. CONCLUSIONS: The striking variation among states suggests that there is room for improvement in the utilization, care processes, and outcomes of CEA. All surgeons performing CEA should participate in outcome assessment and adopt protocols that include the routine administration of antiplatelet agents preoperatively, the use of heparin intraoperatively, and patch angioplasty of the endarterectomy site.

Improving the outcomes of carotid endarterectomy: results of a statewide quality improvement project.

OBJECTIVE: The purpose of this study was to establish the statewide outcomes for carotid endarterectomy (CEA) and to facilitate improvement in outcomes through feedback, peer discussion, and ongoing process and outcome measurement. METHODS: The Medicare Part A claims files were used to identify all Medicare patients undergoing CEA in Iowa during two 12-month time periods (January 1994-December 1994 and June 1995-May 1996). Medical record abstraction was used to obtain surgical indications, perioperative care process, and outcome information. Confidential reports were provided to each hospital (N = 30) where the procedure was performed. Surgeons performing the procedure (N = 79) were invited to meetings to discuss care process variation and outcomes. Voluntary participation was solicited in a standardized program of ongoing hospital-based data collection of CEA process and outcome data. RESULTS: The statewide combined stroke or mortality rate decreased from 7.8% in 1994 to 4.0% in the 1995 to 1996 time period (P <.001). Fourteen hospitals, accounting for 74% of the statewide cases, participated in ongoing data collection. The combined stroke or mortality rate in these hospitals decreased significantly (P <.05) over time from 6.5% (1994) to 3.7% (1995-1996) to 1.8% (June 1997-May 1998). The use of intraoperative assessment of the operative site (20% in 1994, 46% in 1997-1998) and patch angioplasty (14% in 1994, 30% in 1997-1998) increased significantly during this time in the participating hospitals. CONCLUSIONS: Confidential feedback of outcome and process data for CEA may lead to change in perioperative care processes and improved outcomes. Standardized community-based outcome analysis should become routine for CEA to ensure that optimum results are being achieved.