Socioeconomic Determinants of Guideline-Concordant Therapy for Early-Stage Non-Small Cell Lung Cancer: A Population-Based Analysis from Appalachian and Non-Appalachian Ohio, 2004-2015

Technical Summary: Socioeconomic Determinants of Guideline-Concordant Therapy for Early-Stage NSCLC in Ohio

Problem Statement
Despite established guidelines recommending surgical resection or definitive radiation for early-stage non-small cell lung cancer (NSCLC), a significant proportion of patients fail to receive guideline-concordant therapy. Previous literature has documented disparities linked to race, age, and rural residence, with specific concerns regarding the Appalachian region, characterized by persistent poverty and limited healthcare access. However, it remains unclear whether observed disparities in treatment receipt are driven by geographic designation (Appalachian vs. non-Appalachian) or by the concentration of socioeconomic disadvantage (e.g., Medicaid insurance, county-level poverty) within those regions. National datasets often lack the granularity to disentangle these factors or exclude patients treated outside Commission on Cancer-accredited facilities.

Methodology
This study employed a retrospective, population-based cohort design using the Ohio Cancer Incidence Surveillance System (OCISS), a statewide registry capturing all cancer diagnoses regardless of treatment location.

• Cohort: The analysis included 26,756 adults diagnosed with early-stage NSCLC (in situ, localized, or regional with direct extension) between 2004 and 2015.
• Primary Outcome: Receipt of guideline-concordant local therapy, defined as surgery or definitive radiation therapy.
• Exposure Variables: The primary exposure was Appalachian residence (based on Appalachian Regional Commission county designations). Key covariates included insurance status, county-level poverty index, rural-urban classification (USDA RUCC), race/ethnicity, marital status, age, and tumor characteristics.
• Statistical Analysis:
  • Multivariable logistic regression was used to identify predictors of therapy receipt, adjusting for predisposing, enabling, and need factors based on the Andersen behavioral model.
  • Cox proportional hazards models assessed all-cause and cancer-specific mortality, with survival capped at 60 months.
  • Sensitivity analyses included E-values to quantify unmeasured confounding, race-stratified models, inverse probability of treatment weighting (IPTW), and restricted analyses to potentially operable candidates (excluding those with documented medical contraindications).
  • Missing data for smoking history and tumor grade were handled via "Unknown" categories and multiple imputation.

Key Results

• Treatment Rates: Overall, 83.6% of patients received guideline-concordant local therapy (59.6% surgery, 24.0% radiation). Unadjusted rates were slightly higher in Appalachian counties (84.6%) compared to non-Appalachian counties (83.4%).
• Socioeconomic Determinants: In adjusted analyses, Medicaid insurance (Adjusted OR 0.53; 95% CI 0.44–0.63) and county-level poverty >20% (Adjusted OR 0.77; 95% CI 0.68–0.87) were the strongest independent predictors of lower receipt of guideline-concordant therapy. Unmarried status and older age were also associated with lower receipt.
• Geographic Findings: After adjusting for insurance, poverty, and other socioeconomic factors, Appalachian residence was not a barrier to treatment; in fact, it was associated with modestly higher receipt (Adjusted OR 1.17; 95% CI 1.06–1.29).
• Temporal Trends: A significant interaction between year and Appalachian status (p<0.001) indicated that treatment rates converged between regions over the study period, with no significant difference remaining by 2012–2015.
• Survival Outcomes: Lack of local therapy was the strongest predictor of mortality (Adjusted HR 4.33). Medicaid insurance (HR 1.25) and high county poverty (HR 1.13) remained independently associated with worse survival. While Appalachian residence showed a modest association with all-cause mortality (HR 1.07), this association attenuated to null in cancer-specific competing-risks analyses.
• Racial Heterogeneity: Race-stratified models revealed that the poverty gradient was significant only among non-Hispanic White patients; no significant association between poverty and therapy receipt was found for non-Hispanic Black patients in this cohort.

Significance and Claims
The authors conclude that socioeconomic factors, specifically insurance status and county-level poverty, are the primary drivers of disparities in guideline-concordant lung cancer treatment in Ohio, rather than Appalachian geography itself. The study challenges the framing of treatment disparities as purely geographic, suggesting that interventions targeting "Appalachian" designation alone may be less effective than those addressing underlying socioeconomic barriers.

Key claims include:

1. Insurance as a Modifiable Barrier: Medicaid status is the most significant modifiable predictor of treatment failure, conferring roughly 47% lower odds of receiving guideline-concordant therapy compared to private insurance.
2. Decoupling Geography from Disparity: In the Ohio context, the concentration of poverty and underinsurance within Appalachian counties explains treatment gaps, not the geographic designation itself.
3. Policy Implications: The findings support targeted interventions that address insurance-related and poverty-related barriers in high-poverty communities regardless of their geographic label. The authors note that while Ohio's 2014 Medicaid expansion occurred late in the study window, the data suggests that insurance coverage alone may be insufficient without parallel investments in care navigation and infrastructure to overcome structural barriers in high-poverty areas.

The paper emphasizes that its findings are specific to the Ohio context, where a Commission on Cancer-accredited regional center serves the Appalachian counties, potentially limiting the generalizability of the "no geographic deficit" finding to other states with different healthcare infrastructures.