S.A. Rao1, A. Lundberg1, T. Lagu1, A.W. Heinemann1, A. Stey1 1Feinberg School Of Medicine – Northwestern University, Chicago, IL, USA
Introduction: Spinal cord injury (SCI) is a debilitating condition affecting an estimated 294,000 Americans. SCI has detrimental impacts on employment, social reintegration, and healthcare use. Importantly, neurologic losses are mitigated by intensive physical therapy provided at inpatient rehabilitation facilities (IRFs). We hypothesized that IRF care reduces healthcare costs and utilization among SCI patients in the long-term. We aimed to assess 1) the baseline incidence of SCI, 2) one-year costs and healthcare utilization following SCI, and 3) the association of insurance status and receipt of inpatient rehabilitation care on these metrics.
Methods: Adults newly diagnosed with SCI between 1/1/2016 and 12/31/2017 were identified in the California Department of Healthcare Access and Information (HCAI) database. Patients 65 years and older were excluded to reduce confounding from SCI-unrelated healthcare use. Costs were obtained from charges using year- and facility-specific cost-to-charge-ratios. One-year metrics were analyzed by insurance type and whether patients received up-front rehabilitation after their first hospitalization. Costs were adjusted to June 2024 dollars using month-specific Consumer Price Indices. Significance was assessed with non-parametric tests.
Results: A total of 2,377 patients were hospitalized with a new diagnosis of SCI during the study period (1,366 publicly insured, 849 privately insured, and 162 self or other pay). A collective 3,244 admissions were identified within the first year after injury. Costs of initial admission after SCI were significantly higher than those for readmission (median $73,728.55 versus $39,771.40, p < 0.001). Only 619 patients (25%) were directed to IRFs immediately following initial hospitalization. Patients referred to IRFs had significantly higher one-year inpatient costs over the course of the year (median $116,731.90 versus $63,878.41, p < 0.001) despite comparable admission rates (median 1.316 versus 1.333, p = 0.973). Receipt of public insurance was associated with 0.791 odds of receiving IRF care compared to privately insured individuals (p = 0.004, 95% CI [0.674, 0.929]). Despite having lower rates of IRF care, publicly insured patients had higher one-year inpatient costs than the privately insured (median $81,965.90 versus $73,710.37, p = 0.001) even with comparable admission rates (median 1.314 versus 1.330, p = 0.699). Length of stay after initial injury was greater for the publicly insured (median 6 vs 9 days, p < 0.001) and for those directed to IRFs (median 6 versus 12 days, p < 0.001).
Conclusion: SCI poses a severe cost burden to public payors. Inpatient cost differences at one year between subgroups defined by insurance type or receipt of IRF care are not explained by differences in readmission rates but are associated with length of stay. Further work examining the effect of individual patient characteristics on these measures is planned to assess this discrepancy.