Background: Cervical spine surgeries are important for relieving burdensome pain in the cervical spine. Hospital billing data can provide safety and effectiveness insights on medical devices used in cervical spine surgery. However, hospital data follow patients longitudinally in the same hospital only. Comparison of outcome rates (e.g., readmission and revision) in hospital versus claims data is important to understand any potential loss of data capture when evaluating outcomes for devices using hospital data.
Objectives: The objective of this study was to evaluate the difference in readmission and revision risk after cervical spine surgery between hospital billing and claims databases.
Methods: Patients that underwent a cervical fusion surgery from 2010-2021 were identified in PINC AI™ Premier Healthcare database (hospital) and Medicare Standard Analytic file (claims). Hospital patients were required to have Medicare fee-for-service insurance. Within both databases, the same inclusion criteria were used (age≥65, specific spinal diagnoses). Outcomes were all-cause readmission at 90-days and revision surgery at 90- 365-, and 1,825-days post-surgery. Adjusted risk, risk difference (Medicare-Premier) and 95% Confidence Intervals (CI) were estimated.
Results: There were 91,205 hospital patients and 150,459 claims patients included. Risks of readmission at 90- and revision at 365-days post-surgery were 8.8% and 2.5% in the hospital database and 13.9% and 2.9% in the Medicare claims database. The difference in readmission risk (95% CI) was 5.1% (4.7%, 5.5%), while for revision, the difference in revision risk was < 1% (0.4% (0.2%, 0.6%)) between the hospital and claims data for the cumulative 365 days post-index. As time increased to ≥2 years, the difference in risk increased, with the cumulative 1,825-day difference in revision risk at 1.9% (1.5%, 2.2%).
Conclusions: Among cervical fusion patients, the risk difference for 90-day readmission was large, aligning with the expectation that patients may not use the same hospital for all their healthcare needs. By contrast, the risk differences for 90- and 365-day revision were small ( < 1%) indicating the vast majority of patients return to the same hospital when undergoing additional cervical spine care. With longer follow-up times, the risk differences for revision became larger as patients dispersed to other hospitals for cervical spine care, suggesting that hospital billing data were most effectively used to assess revision risk for time periods up to two years following index spine surgery. This study helps inform outcome selection when working with hospital data.
