Expensive surgical procedures including spinal fusions have become targets for cost reduction programs. It is well-established that patient comorbidities, more complex surgeries, and complications increase costs. Less well-studied is the effect of individual surgeons on costs. In order to better explore this, Dr. Sielatycki and colleagues from Vanderbilt University compared the costs and patient reported outcomes associated with ACDF among 5 spine surgeons at their institution. They included 431 elective ACDF cases (1 to 3 levels) and determined the inpatient and 90-day costs based on billing data and average Medicare reimbursements for the inpatient, surgeon, and outpatient services. Patient reported outcomes were recorded at baseline and 3 months and included the neck disability index, EQ-5D, and numeric rating scale for arm and neck pain. Baseline patient characteristics were similar for all 5 surgeons, and predicted costs adjusted for patient characteristics were comparable across surgeons as well. However, the actual costs were significantly different among surgeons, with inpatient costs ranging from $10,522 for the least expensive surgeon to $15,366 for the most expensive. Differences were not quite as pronounced for 90 day costs but still varied by 25%. Logistic regression controlling for patient and surgery factors demonstrated that complications were the strongest predictor of higher costs, followed by number of levels fused. Individual surgeon was also a strong predictor of cost, with one surgeon having a 150% increase in the odds of having a patient in the highest cost quartile than the median cost surgeon. There were no significant differences in patient reported outcomes across surgeons.
The authors have done a nice job demonstrating significant variation in costs for routine ACDF among five different surgeons at the same institution. What the paper does not tell us are the reasons behind the variation. The methods for determining costs were somewhat indirect, based on average Medicare reimbursements at the DRG and CPT level. While this might approximate costs on a large scale across institutions, actual costs can vary substantially on a patient-by-patient basis. Given that one of the surgeons operated on only 22 patients, these cost estimates may not have been accurate. The authors also did not attempt to explain why inpatient costs varied so much. One would expect similar inpatient cost estimates based on DRG and surgeon fee, but the costs varied by 45%. The main drivers of inpatient cost differences among surgeons are probably implant and bone graft choices, and these were not evaluated. Length of stay, operative time, and post-operative care including imagining, office visits, and PT could also vary among surgeons and affect costs. Complications, reoperations, and readmissions are major cost drivers and can vary among surgeons, though these tended not to vary much across surgeons in this study. While it has significant limitations, this study shines a light on the role individual surgeons can play in driving costs. It comes as no surprise that increased costs were not associated with better outcomes. Hopefully this paper will encourage more detailed cost analyses that evaluate the role of specific surgeon decisions (such as implant selection) on costs.
Please read Dr. Sielatycki's paper in the August 15 issue. Does this change how you view the role of the individual surgeon in driving costs? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor