Variation in hospital and surgeon level reoperation rates following lumbar decompression for herniated disc and the influence of practice volume

There is a significant need for new research using empirical performance data to describe the safety of lumbar surgery. Low back pain is the second most common reason for inpatient surgery in the United States. New technological innovation, increased imaging, and increased willingness of surgeons to perform surgery may all contribute to the increased rate of spine surgery over the last two decades. These increases are accompanied by early complications and late failures leading to reoperations, a well-accepted indicator of treatment failure, complication, or progression of disease. Back surgery has been identified as a quality improvement priority by the National Quality Forum (NQF) National Priorities Partners (NPP).;This dissertation presents three studies that examine variation in reoperations following lumbar decompression for herniated disc. The first is entitled " Assessing Quality of Decompression Surgery for Lumbar Herniation Using Hospital and Surgeon Level Variation in Repeat Surgery Rates." It uses multilevel modeling to describe the variation in reoperation rates from an inpatient discharge registry, and compares these rates to those reported in clinical trials. We report wide variation in reoperation rates even after controlling for patient characteristics, and found that a large proportion of hospitals' and surgeons' reoperation rates exceeded quality benchmarks based on randomized control trials.;The second study responds to recent reviews that have highlighted the need for a better understanding of the association between procedure volume and outcomes following spine surgery. Although patients seen by high-volume spine surgeons appeared to be healthier that those seen by low volume surgeons, they had a significantly higher risk of undergoing a reoperation.;The third study presents findings to show that hospital's historical risk-adjusted reoperation rates are strongly associated with future reoperation rates. We present a model to predict the probability of a hospital to have a 1-year reoperation rate greater than 10%.;Collectively, these studies lay the foundation to improve spine surgery outcomes by motivating providers to improve their practice, helping to inform pay-for-performance or tiered payment incentive programs, and informing patients about the risks of surgery. This research may translate to other spinal procedures where indications less well defined.