Editor Choice: Performing a Cost Analysis in Spine Outcomes Research
Background: Medical cost analysis is increasingly important, but the methodology is complex and varied.
Objective: To illustrate how different cost analysis methodologies influence conclusions generated from data from a prospective nonrandomized trial for treatment of cervical spondylotic myelopathy.
Methods: Patients 40 to 85 years of age with degenerative cervical spondylotic myelopathy were enrolled from 7 sites over 2 years (2007-2009). Patients were treated with ventral or dorsal fusion surgery, and outcomes were measured to 1 year postoperatively. A hospital-based cost analysis was performed using Medicare cost-to-charge ratios (CCRs) multiplied by hospital charges from the index hospitalization (CCR method). A society-based cost analysis was performed by estimating costs from the index hospitalization using Medicare coding reimbursement (the Medicare reimbursement method). A separate outpatient cost analysis was performed on a subset of 20 patients.
Results: Of the 85 patients analyzed, 72 had 1-year follow-up. The CCR method showed a difference in upfront direct costs between the dorsal and ventral approaches ($27 942 ± 14 220 vs $21 563 ± 8721; P = .02). Overall upfront direct costs with the Medicare reimbursement method were not different. With the CCR method, the ventral approach dominates an incremental cost-effectiveness ratio analysis. With the Medicare reimbursement method, the incremental cost-effectiveness ratio for ventral surgery is $34 533, the cost of 1 additional quality-adjusted life-year gained by using ventral instead of dorsal surgery. In the subanalysis, outpatient costs were less after ventral surgery than dorsal surgery ($1997 ± 1211 vs $4734 ± $2874; P = .006).
Conclusion: The choice of cost methodology may substantially influence the final results of an economic study.
From: Performing a Cost Analysis in Spine Outcomes Research: Comparing Ventral and Dorsal Approaches for Cervical Spondylotic Myelopathy by Whitmore et al.