1013 - High Deductible Health Plans and Survival among Cancer Survivors and the Role of Cost-Driven Delayed or Forgone Medical Care

J. M. Barnes^1,2, P. Santos³, S. Wallingford², A. Gupta⁴, M. Ragavan⁵, and F. Chino^6,7; ¹Washington University School of Medicine in St. Louis, Department of Radiation Oncology, Saint Louis, MO, ²Costs of Care, Boston, MA, ³Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, ⁴University of Minnesota, Minneapolis, MN, ⁵Kaiser Permanente, San Francisco, CA, ⁶Affordability Working Group, Memorial Sloan Kettering Cancer Center, New York, NY, ⁷Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY

Purpose/Objective(s): High-deductible health plans (HDHP) offer low premiums at the expense of potentially high out-of-pocket costs. HDHPs are associated with later stage diagnoses and Delayed or Forgone Care due to cost (DFC) among cancer survivors. Our objectives were to 1) quantify associations of HDHPs and DFC with survival among cancer survivors and 2) determine whether DFC mediates the association between HDHPs and survival. Materials/

Methods: Cancer survivors ages 18-74 years with non-Medicaid insurance were identified from the 2011-2018 National Health Interview Survey (NHIS) with linked mortality files from the National Death Index. HDHP status was defined by the NHIS as a yearly deductible =$1200-1350 (individual) or =2400-$2700 (family) with values dependent on survey year with increases over time. DFC was defined as delaying or forgoing any medical care due to costs within the prior 12 months. Logistic regression quantified associations of HDHPs with DFC, and Cox proportional hazards models quantified associations of HDHPs and DFC with overall survival (OS), using age as the time scale. Models accounted for the NHIS survey design and were adjusted for insurance, marital status, sex, comorbidities, education, income, region, and cancer site and time since diagnosis. Mediation analyses tested whether DFC mediates the effect of HDHP status on survival. Sensitivity analyses evaluated 1) overall mortality using time since diagnosis as the time scale, enabling adjustment for age; 2) mortality due to cancer; and 3) mediation of cost-related medication underuse rather than DFC.
Results: A total of 9,799 cancer survivor NHIS respondents were identified, of which 25.6% had a HDHP and 6.8% experienced DFC at the time of survey participation. Those with HDHPs were more likely to experience DFC (OR 2.08, 95% CI = 1.67 – 2.58). Both having a HDHP (HR: 1.46, 95% CI = 1.19 - 1.79) and experiencing DFC (1.84, 1.48 – 2.28) were associated with worse OS. In mediation analyses, the effect of HDHP on survival mediated through DFC (ie, natural indirect effect) had an HR of 1.75 (95% CI = 1.31 – 2.34), while the remainder of the effect of HDHP on survival (ie, not mediated by DFC, or natural direct effect) had an HR of 1.41 (1.14 –1.74). Experiencing DFC mediated 9.5% (P<.001) of the total association between HDHP and survival (up to 62.0% in exploratory analyses with alternate estimation strategies). Results were similar in sensitivity analyses.
Conclusion: In this analysis of nationwide data, high deductible health plans and forgone/delayed care due to cost were associated with worse survival among cancer survivors. Delayed/forgone care due to cost is likely an important mechanism by which high deductible health plans decrease survival. Insurance coverage that financially discourages medically necessary care may decrease utilization and ultimately adversely impact outcomes in cancer survivors.