What Are the Costs of Cervical Radiculopathy Prior to Surgical Treatment?

Cameron Barton, MD; Piyush Kalakoti, MD; Nicholas A. Bedard, MD; Nathan R. Hendrickson, MD; Comron Saifi, MD; Andrew J. Pugely, MD


Spine. 2019;44(13):937-942. 

In This Article


Data Source

The Humana Inc. claims dataset, a longitudinal commercial cohort, for the years 2007 to 2015 was utilized as the data source for the current study. The dataset was accessed through the PearlDiver Research Program (PearlDiver Technologies, Colorado Springs, CO) and contains over 20 million HIPAA compliant medical billing records. In comparison to the commonly utilized administrative databases and registries exploring spine outcomes, Humana claims dataset is unique and enables longitudinal tracking of patients over study years using a combination of International Classification of Diseases, ninth Clinical Modification (ICD-9-CM) and Concurrent Procedural Terminology (CPT) codes. Our group[15,16] and other contemporary researchers[17–19] have previously utilized this database for assessing spine surgical outcomes.

Study Design and Cohort Definition

In a retrospective, observational cohort study, patients that underwent ACDF surgery (CPT 22551) for cervical radiculopathy (ICD-9 diagnoses code 721.0, 722.0, 722.4, 722.91, 723.0, 723.1, 723.4) were identified. The cohort was subjected to further filtering to identify patients that were active for the entire year prior to ACDF with the insurance provider.

Outcome Measure

The primary outcome measure was assessment of overall and per-capital costs within 1-year prior to ACDF for cervical radiculopathy. Subgroup assessments were performed by examining costs for diagnostic modalities and non-surgical treatment. Diagnosis specific linking allowed abstraction of specific costs related to diagnoses of cervical radiculopathy. Total costs were extracted from the database and were later separated to identify inpatient costs (those linked to ACDF CPT-code hospital encounter) and those incurred during non-inpatient (nonoperative) visits.

The nonoperative costs included both diagnostic and therapeutic costs as categorized in the North American Spine Society (NASS) cervical radiculopathy current Clinical Practice Guidelines (CPG).[1] Diagnostic costs were segregated for radiographs, CT, MRI, and electromyogram (EMG) or nerve conduction studies (NCS) based upon CPT coding definitions (Online supplement: Table S1, http://links.lww.com/BRS/B410). Likewise, therapeutic costs were examined for nonoperative interventions such as physical training (PT), chiropractic, bracing, injections, and medications for alleviating cervical radiculopathy symptoms. Except for medications, all nonoperative treatments were defined using the CPT definitions (Online supplement: Table S1, http://links.lww.com/BRS/B410). The database was mined for pharmacological data associated with billing records for prescription narcotics (Online supplement: Table S2, http://links.lww.com/BRS/B410), NSAIDs (Online supplement: Table S3, http://links.lww.com/BRS/B410), tramadol, gabapentin and muscle relaxants (Online supplement: Table S4, http://links.lww.com/BRS/B410), and costs associated with each subgroups of medication type was computed.

Overall cumulative costs for nonoperative management was computed as summation of the total diagnostic and treatment costs for procedures and interventions described above. Per-capita costs for each category of diagnostic or treatment modalities was computed by averaging the total costs for the respective category with the number of patients undergoing the procedure or intervention. The Humana dataset tracks reimbursement payments only and does not track total charges or other aspects of hospital billing including patient payments (deductibles). Therefore, this study defined the cost of care as reimbursement fees paid by the insurance provider without the facility fees. Costs of reimbursement were funds paid for services provided that do not necessarily account for the actual costs incurred by the treating entities.

Secondary outcome measure was to determine the cost for ACDF including those incurred for hospital stay during surgery and professional payments by the insurer.


Descriptive statistical comparisons were performed considering the objectives of the study of the data. Categorical data are reported as count and/or proportions. Quantitative variables (costs) are depicted as means. As all data derived from the data source were de-identified and Health Insurance Portability and Accountability Act of 1996 compliant, our institution deemed this study exempt from review by the Institutional Review Board (IRB).