Analysis of Treatment Cost Variation Among Multiple Neurosurgical Procedures Using the Value-Driven Outcomes Database.

Abstract

Health care costs comprise a substantial portion of total national expenditure. Although interest in cost-effectiveness analysis in neurosurgery has increased, there has been little cross-comparison of neurosurgical procedures. The aim of this study was to compare costs across elective neurosurgical procedures to understand whether drivers of cost differ. The Value Driven Outcomes database was used to evaluate treatment costs for resection of vestibular schwannoma, intracranial meningioma, gliomas, and pituitary adenoma; anterior cervical discectomy and fusion and lumbar spinal fusion; and aneurysm treatment. A total of 1997 patients (mean age 54.6 ± 14.5 years; 45.2% male) were evaluated. The mean length of stay (LOS) was 4.0 ± 4.4 days. For cases involving hardware implantation, including spine fusion or aneurysm treatment, supplies and implants (49.1%) accounted for the largest fraction of costs followed by facility costs (37.9%). For cases that did not involve hardware, including tumor cases, facility costs (63.9%) were the largest fraction, followed by supplies and implants (16.2%). Aneurysm treatment and lumbar fusion were 1.5-3 times more costly than cranial tumor resection and anterior cervical discectomy and fusion per patient. Multivariate linear regression demonstrated that LOS (β = 0.7, P = 0.0001) and patient treatment type (β = 0.2, P = 0.0001) had the greatest effect on costs. LOS correlated with cost differently depending on case type; its effect was largest for patients with meningioma and smallest for patients with vestibular schwannoma. Costs across time increased similarly for all case types. Costs for neurosurgical procedures vary widely depending on treatment type and correlated directly with LOS. Strategies to reduce cost may require different approaches depending on procedure type.