Receipt of treatment by insurance and race/ethnicity.

Adjusted odds of undergoing surgical resection or receiving adjuvant chemotherapy.

Objective: A one third reduction of premature deaths from non-communicable diseases by 2030 is a target of the United Nations Sustainable Development Goal for Health. Unlike in other developed nations, premature mortality in the United States (US) is increasing. The state of Oklahoma suffers some of the greatest rates in the US of both all-cause mortality and overdose deaths. Medicaid opioids are associated with overdose death at the patient level, but the impact of this exposure on population all-cause mortality is unknown. The objective of this study was to look for an association between Medicaid spending, as proxy measure for Medicaid opioid exposure, and all-cause mortality rates in the 45–54-year-old American Indian/Alaska Native (AI/AN45-54) and non-Hispanic white (NHW45-54) populations.Methods: All-cause mortality rates were collected from the US Centers for Disease Control & Prevention Wonder Detailed Mortality database. Annual per capita (APC) Medicaid spending, and APC Medicare opioid claims, smoking, obesity, and poverty data were also collected from existing databases. County-level multiple linear regression (MLR) analyses were performed. American Indian mortality misclassification at death is known to be common, and sparse populations are present in certain counties; therefore, the two populations were examined as a combined population (AI/NHW45-54), with results being compared to NHW45-54 alone.Results: State-level simple linear regressions of AI/NHW45-54 mortality and APC Medicaid spending show strong, linear correlations: females, coefficient 0.168, (R2 0.956; P < 0.0001; CI95 0.15, 0.19); and males, coefficient 0.139 (R2 0.746; P < 0.0001; CI95 0.10, 0.18). County-level regression models reveal that AI/NHW45-54 mortality is strongly associated with APC Medicaid spending, adjusting for Medicare opioid claims, smoking, obesity, and poverty. In females: [R2 0.545; (F)P < 0.0001; Medicaid spending coefficient 0.137; P < 0.004; 95% CI 0.05, 0.23]. In males: [R2 0.719; (F)P < 0.0001; Medicaid spending coefficient 0.330; P < 0.001; 95% CI 0.21, 0.45].Conclusions: In Oklahoma, per capita Medicaid spending is a very strong risk factor for all-cause mortality in the combined AI/NHW45-54 population, after controlling for Medicare opioid claims, smoking, obesity, and poverty.

IntroductionUse of oral anticoagulants (OACs) for stroke reduction in atrial fibrillation (AF) varies by race and geography within the United States. We seek to better understand the relationship between OAC underutilization, race, and US geography.MethodsPatients with AF were selected from the US Centers for Medicare & Medicaid Services claims database from January 1, 2013, to December 31, 2016. The final population consisted of patients with 12 months of health plan enrollment before and after their index AF diagnosis, with a baseline CHAD2S2-VASc ≥2 and of either Black or White race (other races are underrepresented in the data). Among those with AF that met the inclusion criteria, patients who were prescribed warfarin or DOACs within 12 months after the index date were extracted. Each patient was assigned to a US county based on their 5-digit zip code and OAC use was stratified by race. Statistically significant differences were determined by student’s t-test and chi-square.ResultsOf the 2,390,830 final patients, 94.1% were White and 5.9% were Black patients. Mean (SD) age and HASBLED scores were 78 (9) and 3.9 (1.2) respectively, for Black patients and 80 (9) and 3.3 (1.2), respectively, for White patients (p