This JAMA Guide to Statistics and Methods details use of the Society for Vascular Surgery’s Vascular Quality Improvement Program, a robust database that provides detailed data of common vascular procedures.
The Vascular Quality Initiative (VQI) was developed by the Society for Vascular Surgery in 2011 to improve the safety and effectiveness of 12 common vascular procedures (Box 12). The VQI operates within the structure of a patient safety organization, which protects the quality improvement activities as patient safety work product and thus provides a degree of privilege and confidentiality for the data. Because the VQI is a member of the Society for Vascular Surgery patient safety organization, comparisons with regional and national institutions can be performed.
BOX 12 Best Practices for Using the Vascular Quality Improvement Program
Use a flow diagram to demonstrate how the target population was selected.
Clearly delineate sample sizes, statistical techniques to mitigate selection bias, and the efficacy of the predictive model.
Emphasize practical clinical findings instead of incidental statistically significant results.
Include a power calculation.
Ensure clear methods to permit reproduction of results.
The VQI registries include carotid artery stenting, carotid endarterectomy, endovascular abdominal aortic aneurysm repair hemodialysis access, infrainguinal bypass, inferior vena cava filter, lower extremity amputations, open abdominal aortic aneurysm repair, peripheral vascular interventions, suprainguinal bypass, thoracic and complex endovascular abdominal aortic aneurysm repair, and varicose vein treatment. As of July 2017, 390 270 procedures were captured.1 A total of 431 participating institutions represent more than 3200 physicians throughout the United States and Canada. There are 18 regional quality improvement groups that promote ownership of the quality improvement process and practical implementation of new clinical processes. Approximately 40% of the participating institutions are community hospitals, 29% are teaching hospitals, and 31% are academic hospitals.2 Multiple physician specialties are represented within the database.
Each VQI registry tracks demographic, physician, hospital, and patient-specific factors that are pertinent to the procedure being performed. Clinical care details are collected for the index procedure hospitalization and at 1 year, thus providing data on outcomes including mortality, reintervention, and postoperative complications. The VQI uses patient identifiers to match with other data sets such as the Social Security Death Index or Medicare claims. These data sources are used in conjunction with periodic billing data to ensure that 100% of the sample is being captured by each member institution.
The web-based system used for data entry is provided by M2S (M2S Inc), a subsidiary of Medstreaming, and is used to generate deidentified benchmark reports that allow participating centers and physicians to compare their outcomes with the regional and national benchmarks. Multiple users are able to navigate the data forms, and data capture can be integrated with some electronic health record systems.
The limitations of the data set include a selection bias manifest ...