Cost-Effectiveness of Hepatitis C Screening Strategies in the Presence of New Direct Acting Anti-Viral Therapies

Background: Decreasing costs for effective direct-acting antiviral therapies (DAAs) used to treat individuals chronically infected with hepatitis C virus (HCV) are on the horizon. Historically, screening for HCV was limited to sub-populations of individuals considered at high risk (i.e. injectable drug users). More recently, screening expanded to include baby boomers (born 1945-1965) among whom prevalence was shown to be high. Given the recent epidemic of opioid use affecting all age groups, it is important to examine if screening the whole population may become a cost-effective option.

Objectives: 1) Develop deterministic cost-effectiveness models accounting for patient and provider behaviors known to influence screening participation such as patients not disclosing high-risk behaviors and providers inconsistently assessing for high-risk. Model to include deterministic estimates of costs and effectiveness reflective of the complex disease progression of HCV. 2) Determine the levels of sub-population prevalence indicating a cost-effective move from risk-based to birth cohort or universal screening in the presence of highly efficacious DAAs.

Perspective: Healthcare payer or system

Setting: Clinic-based screening.

Base case: Screening of the general United States population considered high-risk (risk-based screening).

Comparators: Risk-based plus birth cohort and universal screening.

Methods: Using population, diagnosis, disease, and transition probability and cost estimates from peer-reviewed literature or estimates derived from the National Health and Nutrition Examination Survey and the United States Census bureau, our deterministic decision analytic model included paths leading to 1) treatment, 2) patient declining treatment, 3) patient HCV seronegative, 4) patient HCV seropositive without viremia or 5) patient HCV seropositive undiagnosed. Path probabilities accounted for 1) population and sub-population prevalence, 2) provider and patient behaviors, 3) test sensitivity and specificity, and 4) the ability of DAAs to attain sustained virologic response (SVR). Accounting for sub-population state transition differences, state transition probabilities estimated mean duration in each HCV fibrosis stage then combined with population proportions of HCV fibrosis stage at diagnosis to create weighted 20-year costs and benefits. Number of cases cured and quality-adjusted life years (QALYs) were effectiveness measures, and costs (2016 USD) reflected the accumulation of medical care received. Assumptions included liver disease did not progress in patients achieving SVR, patients were screened once, and HCV positive patients not screened remained undiagnosed. Two-way sensitivity analysis tested prevalence ranging from 0.011 to 0.75, and comprehensive cost of treatment evaluated from $500 to $50,000.

Results: Cost per QALY gained were $665, $573, and $680 for risk-based, birth cohort, and universal screening, respectively. Birth cohort screening 1)dominated other strategies regardless of population and sub-population prevalence, translated to a gain of 98.6 quality-adjusted days, and a net monetary benefit of $837,108 (willingness-to-pay threshold $50,000). With cases cured, risk-based screening was extendedly dominated by universal screening, but net monetary benefits for all strategies were negative.

Conclusions: Sensitivity analysis of prevalence suggests birth cohort screening is the optimal strategy. Limitations include the fluctuation in the distribution of fibrosis state at diagnosis over twenty years, the inclusion of more than one high-risk population, and the translatability of identified state transition probabilities to the general and sub-populations.