The purpose of the study is to determine the cost-effectiveness of different surgical strategies to treat cervical spondylotic myelopathy. The study will use data generated from the CSM-S Trial (NCT02076113).

1. To determine if laminoplasty is more cost-effective compared to dorsal fusion or ventral fusion surgery.
2. To determine the relative cost-effectiveness between anterior cervical discectomy and fusion (ACDF), posterior instrumented cervical fusion (PCF), and cervical laminoplasty.

In this study, the investigators aim to cover a significant gap in the literature, namely, to compare anterior and posterior surgical approaches for the management of CSM on the cost of care. The investigators will conduct a post-hoc analysis of data from the randomized trial "Cervical Spondylotic Myelopathy Surgical Trial" (www.ClinicalTrials.gov; identifier: NCT02076113). This trial was conducted in collaboration between multiple academic institutions in North America. The results of the CSM-S RCT in terms of clinical and patient-reported outcomes suggested similar effectiveness with anterior and posterior surgical approaches. Notably, the study showed that cervical laminoplasty, a relatively low instrumentation-related cost procedure, was effective in managing CSM. The findings of the analysis proposed herein will add a new dimension to clinical decision-making by informing patient selection and improving the quality of care.

Post-hoc analysis will be conducted on the following data that were collected during the trial.

I. Functional outcomes determined by well-known quantitative scales (SF-36, Oswestry Neck, mJOA, and EuroQol-5D) at pre-op, 3 months, 6 months, 1 year, 2 years, 3 years, 4 years, and 5 years.

II. Health resource utilization information was collected using patient diaries along with copies of all medical bills and receipts at 1 month, 3 months, 6 months, and 12 months post-operatively for all patients. Participants were asked to monitor days missed from work and days unable to perform usual activities, in addition to days missed from work for medical treatments or evaluations.

The overall cost will be estimated for each patient from a societal perspective and will include the following components:

I. Index Hospitalization Cost - performed on all patients:

• Hospital Costs: Total hospital charges (TotalHC) will be used to estimate costs using hospital-specific and year-specific Medicare cost-to-charge ratios (CCR). The costs will be adjusted for inflation to the latest year of the study (2018) without a 3% discount. - Assumptions: cost-to-charge ratios for all hospitals and all relevant years will be provided. Data will be complete for all patients from all sites.

  o As a sub-analysis, hospital charges for specific revenue centers (i.e., radiology, pharmacy, physiotherapy, and occupational therapy) will be converted to costs using hospital-specific and year-specific CCRs. This will allow the identification of the major drivers of the difference in cost between the surgical strategies. A crosswalk between the revenue center and the cost center will have to be created. - Assumptions: Total hospital charges will be segmented out into the relevant revenue centers. All patients will have charges segmented in this manner. No major differences in how total hospital charges will be segmented by each site will exist.

• For professional services (identified with either HCPCS code or CPT-4 codes) provided to patients during the index hospitalization, costs will be calculated using national reimbursement amounts from the appropriate Medicare fee schedules for the respective year. - Assumptions: Detailed line-item services of all HCPCS codes and CPT-4 codes (including any necessary billing modifiers) will be provided for all professional services. Data will be complete for all patients from all sites.

II. One-year All-Cause Cost - performed on patients from sites capturing 1 year utilization:

• One-year all-cause costs will be calculated on hospital and professional services in the manner described above. - Assumptions: The same underlying assumptions described above.

III. Time to Return to Work - performed on all patients reporting employment:

• Investigators will assume that the work halted for patients on the procedure date).
• For patients to whom this outcome applies, the investigators will compare different cohorts using return to work as a time-to-event outcome.
• Lost wages will be estimated using the median weekly earnings in the US as reported by the Bureau of Labor Statistics.
• Patients providing no employment information will be assumed to have zero lost wages.
• Job-specific wages will not be estimated.

Index hospitalization costs and one-year all-cause costs will be combined with costs due to loss of productivity (indirect costs) to estimate total costs for each patient from a societal perspective. This will be performed on the subpopulation for which one-year costs were captured.

Inclusion Criteria:

-Patients with Cervical Myelopathy (> or equal to 2 levels of spinal cord compression from C3 to C7) with 2 or more of the following symptoms/ signs: clumsy hands, gait disturbance, hyperreflexia, upgoing toes, bladder dysfunction.

Exclusion Criteria:

• C2-C7 kyphosis>5 degrees (measured in standing extension radiograph)
• segmental kyphotic deformity (defined by greater than 3 osteophytes extending dorsal to a C2-C7 dorsal-caudal line measured on cervical spine CT)
• structurally significant ossification of the posterior longitudinal ligament (OPLL - measured on cervical spine CT)
• previous cervical spine surgery, or significant active health-related co-morbidity (Anesthesia Class IV or higher)

Study population consists of patients with multi-level cervical spondylotic myelopathy without kyphosis and who satisfied clinical equipoise criteria (ventral vs. dorsal approach) as judged by a panel of at least 15 spine experts.

De-identified data from the CSM-S trial including cost data will be available to other researchers who provide a valid study protocol.

Within 6 months of primary publication of the data and for 5 years thereafter

Researchers with a clear and valid research protocol will be able to gain access to the data.