Cost and Utilization Trends of Lumbar Fusion
By Admin | March 04, 2026
Question What changes in lumbar fusion surgery utilization and costs occurred between 2002 and 2023?
Findings In this cross-sectional study of more than 5 million lumbar fusion admissions, adjusted inpatient hospital and mean inpatient per-procedure costs increased. There were increased trends toward fusions involving 2 or more disc levels and in anterior-posterior column fusion in more recent years as well as a trend toward procedures in outpatient facilities.
Meaning In this study, lumbar fusion trends were marked by greater utilization of multilevel and anterior-posterior approaches and greater use in the outpatient setting.
Importance The increasing cost of lumbar fusion has invited payment reforms, such as mandatory price limits by Medicare in 2026.
Objective To examine the cost, utilization, and procedural case-mix trends for different types of lumbar fusion from 2002 to 2023 in the United States.
Design, Setting, and Participants This cross-sectional analysis used survey-weighted data from the 2002 to 2023 National Inpatient Sample (NIS) and the 2016 to 2022 Nationwide Ambulatory Surgical Sample (NASS). From this nationally representative sample of inpatient and hospital-owned outpatient discharges, information on US adults aged 20 years and older undergoing lumbar fusion for any indication from January 2002 to December 2023 were included.
Exposures Lumbar fusion of any type (1-disc level or multilevel as well as single vertebral column or both anterior-posterior columns) with nonfusion surgery as a comparison.
Main Outcomes and Measures The main outcomes were the survey-weighted annual total of procedures, the mean age of patients undergoing lumbar fusion, the inflation-adjusted hospital costs, and the annual procedure rates per 100 000 population.
Results A total of 5 033 772 lumbar fusion admissions between 2002 and 2023 were included. In 2023, the cohort of patients undergoing 274 750 procedures had a mean (SD) age of 63.2 (12.9), with 142 815 (52.0%) female patients. Excluding 54 620 complex fusions, which were mostly multilevel anterior-posterior column fusions, there were 164 105 (50.1%) multilevel fusions, and 109 130 (51.3%) combined anterior-posterior column fusions. The age-adjusted population rate of inpatient fusion procedures increased from 60.1 (95% CI, 58.8-90.3) per 100 000 in 2002 (148 823 admissions) to a peak of 89.9 (95% CI, 89.6-90.3) in 2016 (284 180 admissions), before declining to 80.0 (95% CI, 79.7-80.4) by 2023 (273 235 admissions). Lumbar fusion performed in hospital-owned outpatient facilities was minimal in 2016 (6132 procedures, or 2.1% of total lumbar fusions) and 6.9 per 100 000 (27 331 procedures, or 9.8% of total lumbar fusions) in 2022. Adjusted inpatient hospital costs increased 265.3% from $3.86 (95% CI, $3.81-$3.92) billion in 2002 to $14.1 (95% CI, $13.9-$14.2) billion in 2023, and mean inpatient per-procedure cost increased from $25 849 (95% CI, $25 684-$26 015) in 2002 to $45 458 (95% CI, $45 207-$45 709) in 2023. Lumbar fusion primarily shifted from single column at 1 or 2 disc levels in 2002 (mean cost, $24 515; 95% CI, $24 361-$24 669) to multilevel anterior-posterior column fusion in 2023 (mean cost, $55 034; 95% CI, $54 420-$55 650).
Conclusions and Relevance In this cross-sectional study, lumbar fusion trends were marked by greater utilization of procedures overall, and especially involving multilevel and combined anterior-posterior column approaches and by greater use in the outpatient setting. Costs also increased at both the national and per-procedure levels.
Spinal fusion surgery has invited increasing scrutiny because of widespread use, high costs, unexplained differences in surgical rates across geographic regions, and significant risk of complications. Furthermore, its effectiveness for certain common conditions has been called into question. The trend toward more fusions involving multiple intervertebral disc levels or both the anterior and posterior vertebral columns (AP fusion) has made spinal fusion a target for payment reforms, such as Medicare’s Transforming Episode Accountability Model (TEAM).1
While multilevel and AP fusion may improve spinal alignment and bone healing for specific pathology, small randomized clinical trials suggest they have higher early complication rates and similar patient-reported outcomes compared with single-level, single-column techniques for some common spine-related disorders.2,3 Nevertheless, a recent study showed greater reliance on fusion procedures irrespective of surgical indication (eg, disc herniation or stenosis without spondylolisthesis).4
We sought to describe recent trends in costs, utilization, and procedural case-mix for lumbar fusion in the United States. We hypothesized that the accelerated costs of lumbar fusion would be mainly attributed to the greater adoption of multilevel and AP fusions.
Following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies, we performed an annual cross-sectional analysis of inpatient and hospital outpatient discharges available from the Agency for Healthcare Research and Quality’s Health Care Utilization Project (HCUP). Our study was exempted from institutional review board review and the requirement for informed consent by the University of Utah, which designated HCUP as public data.
To assess population rates for each type of lumbar fusion performed in an inpatient setting, we analyzed the National Inpatient Sample (NIS)5 from January 2002 to December 2023, using age-specific population data from the US Census as the denominator.6 A 20% sample of all hospitals in the United States are included in NIS every year. All discharges from the sampled hospitals are included. The sampling strategy ensures that discharges are representative of hospitals designated as community hospitals in the American Hospital Association Annual Survey.7 Survey weighting and design variables requiring complex sampling statistics are included with NIS to ensure an unbiased national estimate.
To assess lumbar fusions performed in outpatient settings, we analyzed the Nationwide Ambulatory Surgical Sample (NASS)8 from 2016 to 2022. Outpatient data started including spinal fusion in 2016 when International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) procedure codes were adopted. NASS only includes discharges from hospital-owned ambulatory surgery centers (less than 10% of total). As with NIS, NASS uses complex sampling methods to generate national estimates.
Adults aged 20 years or older undergoing inpatient lumbar, lumbosacral, and thoracolumbar fusion were selected using Medicare diagnosis related groups (DRGs) for spinal fusion (eTable 1 in Supplement 1). DRGs combine ICD diagnosis and procedure codes into groups defined by Medicare to set hospital payments. Each admission is coded by a single DRG. DRG codes for spinal fusion were revised in 2025 to separate 1-level from multilevel fusions. To characterize changes in multilevel and AP fusions, we applied the 2025 DRG definitions to data since 2016, when ICD-10 procedure codes were adopted. Fusion operations combined with discectomy, laminectomy, interspinous spacer, or dynamic stabilizing device procedures are included. We excluded 2.4% of fusion operations because they were associated with DRGs not specific for fusion, such as “Soft tissue procedure with major complication or comorbidity” (DRG 500) and “Back and neck procedures except spinal fusion without (major) complication or comorbidity” (DRG 520).
Hospital outpatient department discharges in NASS are defined by up to 25 Current Procedural Terminology fields. We searched all these fields to identify spinal fusion and nonfusion (eg, decompressive laminectomy without fusion) procedures.
Using Medicare DRG nomenclature inpatient lumbar fusions were categorized as noncervical spinal fusion, complex noncervical spinal fusion, and combined anterior-posterior spinal fusion. Because the latter group included both cervical and lumbar fusions, we used ICD-10 codes to restrict the cohort to only patients with lumbar diagnoses. We applied the 2025 revised DRG definitions to data from 2016 to 2023 to distinguish single-level from multilevel (2-7–level) fusions within single-column (either anterior or posterior) and within AP fusion techniques. Complex fusions are those involving 8 or more levels or fusions for fracture, infection, deformity, or cancer. For comparison, we also report trends in inpatient lumbar surgery not involving fusion.
Age group (5-year increments), sex, race, primary insurance payer, discharge disposition, and income quartile for the zip code of patient residence were included in the discharge registries. (However, NASS lacked race data until 2019, and NIS did not provide race in 2023.) To simplify reporting, we recoded race as Black, White, or additional groups (combining American Indian or Alaska Native, Asian, Hispanic ethnicity, Pacific Islander, and any other race and ethnicity not listed). Race and ethnicity are provided by HCUP partner organizations, and their reporting can vary by hospital. They are included to further characterize the sample. Primary payer was grouped as Medicare, Medicaid, private insurance, and other. The latter category (8.7% in NIS and 8.0% in NASS) included self-pay and charity or was unavailable. We estimated comorbidity using the enhanced version of the Charlson Comorbidity Index (CCI) from Quan et al9 based on all ICD-10 diagnosis codes included on the claim. In addition, we identified claims with diagnosis codes for osteoporosis, osteoarthritis, or dementia because they are relevant to spinal fusion but not included in CCI.
Each fusion case was classified by surgical indication using a validated algorithm that demonstrated high sensitivity and specificity for grouping back pain–related ICD-10 codes into clinically distinct conditions (eTable 2 in the Supplement).10 The algorithm uses a hierarchy to group lumbar degenerative diagnoses into 5 categories: deformity (scoliosis or kyphosis), then spondylolisthesis, then spinal stenosis, then disc herniation, then disc degeneration. A patient with a diagnosis code for scoliosis is grouped into deformity even if he or she also has diagnosis codes for spondylolisthesis or spinal stenosis or disc herniation or disc degeneration associated with...(More)
For more info please read, Cost and Utilization Trends of Lumbar Fusion, by JamaNetwork
