(A-002) Patterns and predictors of adherence to direct acting antivirals treatment among Medicare beneficiaries with chronic hepatitis C.

Background: Past studies on adherence to direct acting antivirals (DAAs) in treating hepatitis C virus (HCV) were in clinical settings, lacked real-world evidence, especially in Medicare beneficiaries and did not include the latest pangenotypic DAAs (sofosbuvir/velpatasvir, sofosbuvir/velpatasvir/voxilaprevir, glecaprevir/pibrentasvir).

Objectives: To determine patterns and predictors of DAA treatment adherence in Medicare beneficiaries with chronic HCV.

Methods: We conducted a retrospective cohort study using Medicare claims data to examine patterns and predictors of DAA treatment adherence. Patients were included if they were ≥18 years old, newly diagnosed with chronic HCV, and initiated DAAs between 2014 and 2020 after a one-year period with no DAA claims. Patients continuously enrolled in Medicare parts A, B, and D were followed up to 365 days before and 108 days after the DAA initiation following the documented chronic HCV diagnosis. We used a validated completion of treatment metric as a measure of DAA adherence. Descriptive statistics were used to describe patterns of DAA adherence and multivariable relative risk regression was used to assess predictors of DAA adherence. All analyses were done with SAS 9.4 (SAS Institute, Cary, NC).

Results: Among 15,543 DAA initiators, 13,138 (84.5%) were adherent and 2,405 (15.5%) were non-adherent. Most of the patients were non-Hispanic White (60.6%), male (56.8%), < 65 years old (61.9%), and had Medicaid (54.9%). Patients receiving non-pangenotypic combination DAAs (sofosbuvir-daclatasvir, sofosbuvir-simeprevir, sofosbuvir/ledipasvir, and ombitasvir/paritaprevir/ritonavir & dasabuvir, or elbasvir/grazoprevir) had the highest proportion of adherent patients (88.8%) followed by pangenotypic combinations (78.8%) and single DAA users (71.8%). Multivariable regression analysis found DAA regimen type (P < 0.001) and income status (P < 0.001) as the most influential predictors of adherence. Compared to single DAA users, patients receiving non-pangenotypic combination and pangenotypic combination had 17% (Adjusted Relative Risk (aRR): 1.17, 95% CI: 1.12–1.23) and 11% (aRR: 1.11, 95% CI: 1.06–1.16) higher rates of adherence, respectively. Compared to patients without Medicare subsidy enrollment or Medicaid, patients with Qualified Medicare Beneficiary (QMB), Specified Low-income Medicare Beneficiary (SLMB)/Qualified Individual (QI), and Qualified Disabled and Working Individual (QDWI) coverage had 7% (aRR: 1.07, 95% CI: 1.05–1.09), 9% (aRR: 1.09, 95% CI: 1.07–1.12), 5% (aRR: 1.05, 95% CI: 1.02–1.09), and 8% (aRR: 1.08, 95% CI: 1.03–1.12) higher rates of adherence, respectively.

Conclusions: Rates and predictors of DAA adherence were found in the study population. Opportunities to improve adherence remains particularly among patients with single DAA use and patients not enrolled in Medicare’s income subsidy programs. Patients with non-adherence risk may need strict monitoring to achieve adherence and subsequent treatment success.