OBJECTIVES: To develop quality of care (QC) metrics using claims data in hepatitis C (HCV) Medicare patients with disability, a vulnerable population facing increased access barriers and representing the majority of HCV patients in Medicare, and quantify metrics' correlation with treatment receipt. METHODS: We adapted 14 Veterans Affairs-developed quality metrics (QM) for measurement in a cohort of 1,936 disabled HCV patients (2006-2009) with 6 months continuous Medicare parts A, B, D enrollment before diagnosis and no previous treatment. Based on the machine-learning principle of recursive partitioning, the proposed algorithm implements a random forest model of conditional inference trees, identifies the forest’s representative tree, and aggregates its terminal nodes into QC patient groups. Using linked county-level data from the Area Health Resource Files, we compared contextual characteristics across QC groups. RESULTS: On average, 10.4% received peg-interferon. The five strongest predictors of treatment included “having received liver biopsy”, “HCV genotype testing”, “visit to specialist”, “confirmation of HCV viremia”, and “iron overload testing”. High QC (n=360; treated=33.3%) was defined for patients who had at least 2 from the abovementioned metrics. Good QC patients (n=302; treated=12.3%) had either "HCV genotype testing” or “visit to specialist”, while fair QC patients (n=282; treated=7.1%) only had “confirmation of viremia”. Patients with low QC (n=992; treated=2.5%) had none of the selected QMs. The algorithm accuracy of predicting treatment was 70% sensitivity and 78% specificity. Compared to those with fair or low QC, more high and good QC patients lived in rural or small town areas with lower access to specialized hospital and physician services and lower rates of insurance and education. CONCLUSIONS: Higher quality of care correlated with higher treatment rates. Limited healthcare access among Medicare disabled patients with HCV was not associated with lower quality. Future research is needed to assess pre-treatment QM with newer HCV therapies.