Medicare Payments for Spine Surgery Vary Across the Nation
There is variation in the size of Medicare payments to hospitals for surgical treatment of three spine conditions based on evaluation of the Medicare Provider Analysis and Review. The most expensive hospitals received episode payments that were more than twice as much as those received by the least expensive hospitals. After adjusting for regional differences in prices, indications for surgery, and case mix, the most expensive hospitals received Medicare payments that were 47% higher than what the least expensive hospitals received.
Adjusting for the choice of procedure lowered this to a 28% difference. Interestingly, 40% of the variation in Medicare payments was accounted for by index hospitalization, and about 40% were accounted for by postacute care.
Differences in patient characteristics were small compared to the differences in payments between the two groups of hospitals. Only Twelve percent (12%) of the payment difference between the most expensive and least expensive hospitals was attributable to differences in the cost of physician services. Differences in payments for index hospitalization were almost entirely eliminated once the performance of fusion was taken into account.
Researchers at the University of Michigan and Harvard Medical School analyzed Medicare data collected between 2005 and 2007. Their study utilized a query of an administrative database of International Classification of Diseases, 9th revision (ICD-9) codes to identify the primary diagnosis of patients undergoing spine surgery. Reforms in healthcare and health insurance may put pressure on both hospitals and spine surgeons to evaluate their practices both locally and nationally, the authors concluded.
The study, published in Spine Journal, examined the variation in episode payments from Medicare for surgery for three conditions, spinal stenosis, spondylolisthesis, and lumbar disc herniation.
- Episode payments are also called bundled payments or episode-based payment and represent a potential vehicle to lower the cost of patient care.
- The surgical procedures performed were categorized as decompressions (such as discectomy or laminectomy without fusion) or fusions.
- The data was collected on nearly 186,000 episodes of spine surgery.
- Demographic data was also collected, as well as information on comorbidities.
- Hospitals were ranked and then divided into quintiles by cost.
Hospitals that were expensive tended to be expensive for all three spinal conditions, the study found suggesting that other causes may exist not evaluated in this study.
More than $287 billion was spent on fusion-based spine surgeries in the United States in the first 10 years of this century. In patients covered by Medicare, the rate of complex spinal surgery has increased 15-fold since 2002. Despite these large expenses, there is no consensus on the accepted indications for which spinal fusions are performed. The study did not evaluate indications aside from analysis of ICD-9 codes and did not attempt to evaluate surgical outcomes.
"If payments to hospitals in the most expensive quintile in this study were reduced to the national average per episode of care; a saving of over $162 million could be realized," they concluded.
The performance of fusion surgeries made a difference in costs of postacute care. The difference in payments for index hospitalizations was almost eliminated if fusion surgery was factored in, but the residual difference in payment after this adjustment was largely attributable to postacute care.
"Hospitals and surgeons with particularly high rates of fusion-based procedures may consider external benchmarking and recalibration of practice if appropriate," the authors stated. The indications for postacute care after spine surgery are also not well-established, they added. The analysis found that the average total payment per episode of spine surgery and care was $24,100. This reduced to $23,877 when the price variation and case mix of the hospitals and indication was taken into account. However, the actual episode payment to hospitals in the highest quintile was $34,171, twice as much as the actual episode payment of $15,997 to hospitals in the lowest quintile. Even after adjusting for case mix and indication, the total episode payments to the hospitals in the highest quintile were $9,210 higher than those in the lowest quintile.
The authors noted that this was only the second study to look at differences in Medicare payments to hospitals for spinal surgery.
The authors performed an evaluation of an administrative database that utilizes ICD-9 codes to establish diagnosis. The three categories of diagnosis were disc herniation, spinal stenosis, and spondylolisthesis. Episode-based payments are being explored as a method to potentially contain costs within the Medicare patient population. The study does represent an interesting insight into the cost of care associated with spine surgery, including the relatively small variation in the cost of physician services and relatively high contribution of postacute care.
Significant caution is needed with respect to conclusions made based on the use of administrative databases and indications for surgery. These databases are often created by hospital-based coders, and a recent study by Gologorsky et al. found that the data fidelity of these codes when reviewed by the operative surgeon is less than 50%.1 Conclusions on the utilization of fusion procedures cannot be made if the initial indication data entered is inaccurate more than half the time. The study does not look at outcomes, and the additional cost associated with a fusion procedure is anticipated given the additional cost of implants not utilized in a simple decompression of microdiscectomy.
Episode-based payments are one model being explored to potentially lower the cost of care. It remains to be seen whether it will reduce care and maintain or improve quality. More detailed databases have been established and will hopefully provide valuable insight into surgical indications and outcomes.
1. Gologorsky Y, Knightly JJ, Chi JH, Groff MW. The Nationwide Inpatient Sample database does not accurately reflect surgical indications for fusion. Journal of Neurosurgery: Spine. 2014;21(6):984-993.