Article: Insurance-Related Disparities in Withdrawal of Life Support and Mortality After Spinal Cord Injury

The existing literature indicates a consistent association between health insurance status and health outcomes across various disease contexts. Generally, individuals who are uninsured or covered by public insurance tend to experience poorer health outcomes compared to those with private insurance. This trend is particularly evident in the case of cancer, where uninsured patients or those with public insurance are more likely to be diagnosed at a later stage and exhibit higher mortality rates than their privately insured counterparts. Furthermore, disparities in the types of coverage for screening and preventive care contribute to variations in the utilization of these essential services among patients with different insurance statuses. Studies demonstrate that uninsured adults are less likely to receive timely preventive care compared to those with any form of insurance coverage. Additional factors influencing health outcomes related to insurance status include access to emergency services, trauma care, and chronic disease management.

Similarly, in the realm of spinal cord injury (SCI), research reveals notable disparities in treatment management and outcomes based on insurance status. For instance, a recent study indicated that SCI patients with Medicaid experienced delays in spinal fusion compared to those with private insurance. Other investigations have shown that Medicaid recipients with SCI had poorer mobility and greater psychological distress than those with private insurance during follow-up.

This month, we will examine an article published in JAMA Surgery in August, titled “Insurance-Related Disparities in Withdrawal of Life Support and Mortality After Spinal Cord Injury” by Shakil et al. This study adds to the growing body of literature documenting disparities in outcomes and treatment based on insurance status among individuals with SCI.

This study employed a retrospective observational cohort design, utilizing data from the American College of Surgeons (ACS) Trauma Quality Improvement Program (TQIP) spanning from 2013 to 2020. The study included patients aged 16 and older with cervical SCI that were classified as American Spinal Injury Association (ASIA) Grade A (indicating complete loss of sensory and motor function), totaling 8,421 patients. Among those with private insurance, 14.3% underwent withdrawal of life-sustaining treatment (WLST), whereas 22.9% of patients with public insurance and 16.6% of uninsured patients experienced WLST. In terms of mortality, 21.5% of privately insured patients died in the hospital, compared to 30.6% for those with public insurance and 33.8% for the uninsured. Multivariate analysis revealed that both patients with public insurance and those who were uninsured had higher adjusted odds of WLST compared to privately insured patients; however, only the uninsured showed significantly greater adjusted odds of mortality. A propensity score matching analysis further corroborated these findings, indicating that patients with public insurance exhibited increased odds of WLST, higher mortality, greater incidence of in-hospital complications, and longer hospital stays relative to those who were privately insured.

Overall, these results suggest that key aspects of clinical management, such as WLST, and outcomes in SCI are significantly influenced by insurance status. Identifying modifiable factors associated with healthcare disparities, including insurance status, is essential for uncovering areas of inequity within the healthcare system that may be addressed through policy reforms. With traumatic SCI affecting approximately 1.3 million individuals in the United States and Medicaid covering around 80.6 million people, this issue warrants considerable attention. The decision regarding WLST is complex, and it is plausible that variations in preferences and values among patients and healthcare providers are influenced by insurance status. Nonetheless, this study highlights the potential for policy changes to enhance coverage and benefits for public insurance in trauma care.

The study’s strengths include its large sample size across multiple institutions, which enhances the generalizability of the findings. However, large quality improvement databases, like the one used in this study, limit studies to the variables collected, potentially overlooking relevant confounding factors in multivariate analysis. While investigations such as this one elucidate the associations between insurance status and clinical management and outcomes, future research should delve deeper into the underlying reasons for these associations. Are the observed treatment and outcome disparities primarily driven by differences in coverage plans? Or do individuals with public insurance, who often come from lower socioeconomic backgrounds, encounter specific challenges that influence their treatment choices? The current study, along with future ones, will provide valuable insights to better address these healthcare disparities, whether through insurance policy reform or by implementing support through other means for those with public insurance or the uninsured.

Attending Response

Anthony DiGiorgio, DO, MHA

Austin Lui provides a thoughtful analysis of the article on insurance-related disparities in withdrawal of life support and mortality after spinal cord injury (SCI).  He captured the essential findings well, particularly regarding the higher rates of WLST and mortality among uninsured and publicly insured patients. There are a few other areas that deserve attention.

One area that requires more detail is the definition of insurance categories used in the study. The classification of “public insurance” doesn’t differentiate between Medicare alone, Medicare with private supplemental plans, Medicare Advantage, Medicaid fee for service, Medicaid managed care, or Dual Eligible Medicare (Medicare and Medicaid). These distinctions matter because each of these subgroups likely has different levels of coverage, out-of-pocket costs, and access to care, which could affect both treatment decisions and outcomes. Medicare, being more common in older populations, introduces additional variables, such as age-related health issues, which admittedly the propensity matching should have accounted for.  Even within Medicaid, the program varies tremendously between states and if they utilize Managed Care or Fee for Service, a variable which is absent in the analysis.

Additionally, the covariates used in the study, particularly the Charleston Comorbidity Index (CCI), don’t sufficiently capture the socioeconomic factors that disproportionately affect Medicaid and uninsured patients. Factors such as social support networks, housing stability, nutritional status, substance use, and trust in healthcare institutions play a crucial role in shaping health outcomes. Moreover, the study did not include the Area Deprivation Index (ADI), which could have provided a more comprehensive view of the patients’ socioeconomic environment. These unmeasured social determinants of health likely contribute to the poorer outcomes observed in Medicaid and uninsured patients, beyond what is captured by the CCI.

That said, the study’s findings on complications and mortality are valuable, especially when considering the systemic disadvantages faced by Medicaid and uninsured patients. These patients often enter the healthcare system with worse baseline health, and hospitals serving them tend to be under-resourced, which can lead to delays in care, fewer specialists, and substandard facilities. This context is essential for understanding the outcomes observed, as they reflect not only individual health status but also the structural challenges faced by both patients and the institutions that treat them.

In summary, while the study highlights significant disparities, a more nuanced understanding of insurance classifications and a broader set of covariates would provide a deeper insight into the social and structural factors influencing these disparities. I encourage readers to consider these aspects in future analyses to enhance critical evaluation of studies on healthcare disparities.