Am J Manag Care. 2016;22(5 Spec Issue No. 6):SP183-SP184

Although the medical prospects facing patients with hepatitis C virus (HCV) have never been better, the prospect of gaining access to a cure is another matter for many patients. About 2.7 million Americans suffer from chronic HCV, with 170 million worldwide.

 After decades of research and billions of dollars in investment, we now have a definitive treatment—direct-acting antivirals (DAAs)&mdash;that can cure this serious illness.

 Better still, the pills are relatively cheap to produce and future improvements in production technology are likely to make them even cheaper in the future.

Despite this obviously great news for patients with HCV, there is good reason to worry about the high costs of DAA treatment: the average cost for a complete 12-week course of one such treatment is $84,000.

 At this price, many insurers are reluctant to provide full access to DAAs for all patients with HCV; Medicaid programs provide only limited access.

 presents important new peer-reviewed research about the consequences of this dismal arithmetic of HCV treatment. The papers cover ground ranging from the out-of-pocket costs of a DAA treatment course to the real-world consequences—both economic and clinical&mdash;of failing to treat.

 documents substantial out-of-pocket costs for patients with HCV on Medicare. They show that such patients face up to almost $11,000 in out-of-pocket costs for a complete course of treatment with DAAs. Even low-income patients with HCV who qualify for low-income subsidies from the federal government can face out-of-pocket costs of up to $1200 for a course of treatment. Another paper, by Karmarkar et al,

 shows that insurance coverage greatly increases adherence to therapy; conversely, high out-of-pocket costs greatly increase the likelihood of prematurely abandoning therapy.

Two papers document the clinical effects of DAAs on disparate populations in the real world. Younossi et al

 find that the high rate of cures observed in clinical trials of patients with HCV are realized in real-world practices for treatment-naïve patients. In contrast, Lakshmi et al

 find that cure rates with DAAs are lower in a population coinfected with HIV and HCV than what would be expected based on the clinical trial data. They argue that improved adherence to therapy, including follow-up clinic visits, greatly increases the probability of a cure in this population.

Multiple papers also document the considerable direct and indirect benefits that would come from universal coverage for DAA treatment. Moreno et al

 calculate savings over a longer period, and find that curing patients with HCV would save billions of healthcare dollars that would otherwise be spent on traditional therapies in the coming decades in the absence of curative treatment with DAAs. Jena et al

 identify an unanticipated boon from HCV cures: treatment with DAAs benefits non-HCV patients with liver failure who are awaiting transplant by reducing wait times and mortality rates.  This happens because curing patients with HCV reduces the number of those patients who will need a transplant. Finally, Linthicum et al

 calculate that a strategy of widespread screening of patients for HCV, combined with treatment to cure with DAAs at an early stage of the disease, generates nearly $1 trillion in social benefit.

Together, this collection of research makes a strong case that providing patients with HCV with access to curative treatment at an affordable price would create enormous benefits for both the patients and for society at large. Two commentaries

 also published in this special issue argue that given these facts, insurance should cover these drugs and that patients with HCV should face little or no cost sharing to access treatment. Although this notion makes sense given the benefits of DAA treatment for all patients with HCV, there is still a risk of higher insurance premiums if insurers follow this advice.

The central problem is that the price for a course of DAA treatment is so high.  The question is, why do we tolerate such a price? The reasoning is that it may be needed to provide an incentive for research. Prior work shows that investment in research and development in the biopharmaceutical industry are responsive to profitability of new drugs.

 Therefore, if we want cures for other diseases in the future, we need to make sure investing in finding cures is profitable.

However, there is a way out of this policy dilemma, and a back-of-the-envelope calculation demonstrates the possibility. At current prices, let us say the firms that produce DAAs stand to earn $10 billion per year, to pick a round number. The drugs would generate these revenues for a few years until competition from other drugs and patent expiration eroded revenues and profits. If revenues erode at about 10% a year, and we pick an 8% discount rate to reflect the cost of capital in the pharmaceutical industry, the present value of this future revenue stream for the drug manufacturer would be about $40 billion to $50 billion. This number represents—roughly&mdash;the risk-free value of the intellectual property rights to the new drugs DAA.

Thus, my proposal is the following: a benevolent buyer could purchase the intellectual property rights to these drugs and make them public. Without patent protection, the cost of the pill would likely fall precipitously toward the relatively low marginal costs of production, subsequently making universal access to treatment achievable not only in the United States, but in many other countries as well. A $50-billion investment by an altruistic philanthropist would unleash more than $1 trillion in value for society, and it would do so without jeopardizing incentives for finding future cures.

One practical problem with the solution, of course, is finding someone with the capacity to buy the intellectual property rights for the new hepatitis drugs. A possible option is for private charities or philanthropists to play a role; charitable giving in the United States surpassed $350 billion in 2014 alone, so the capacity to make major investments is there. Another option is for governments or international aid organizations that already invest billions of dollars in healthcare around the world to take up the challenge.

What about the potential pitfalls? If no new resources are brought to the table, this investment in universal access for HCV might mean a diversion of resources from other worthy causes. If government borrowing finances the investment, then the costs of interest payments need to be factored in. Another worry is that this might lead to a call for buying out patents on other new drugs with a less-clear social return on investment. The solution to this, however, is to limit such investments to cures where the wedge between societal value and firm value is so large, that even with modest overpayment, patent buy-outs are a good deal for society.

We must make universal access to the cure for HCV and other diseases a top priority. There is no point in discovering medical cures if the majority of patients who would benefit cannot afford them. 

1. Mohd Hanafiah K, Groeger J, Flaxman AD, Wiersma ST. Global epidemiology of hepatitis C virus infection; new estimates of age-specific antibody to HCV and seroprevalence. 

. 2013;57(4):1333-1342. doi: 10.1002/hep.26141.

2. Lawitz E, Mangia A, Wyles D, et al. Sofosbuvir for previously untreated chronic hepatitis C infection. 

. 2013;368(20):1878-1887. doi: 10.1056/NEJMoa1214853.

3. Hill A, Khoo S, Fortunak J, Simmons B, Ford N. Minimum costs for producing hepatitis C direct-acting antivirals for use in large-scale treatment access programs in developing countries. 

2014;58(7):928-936. doi: 10.1093/cid/ciu012.

4. Gray E, O’Leary A, Kieran JA, Fogarty E, Dowling T, Norris S. Direct costs of interferonâ€based and interferonâ€free directâ€acting antiviral regimens for the treatment of chronic hepatitis C infection [published online March 21, 2016]. 

5. Trooskin SB, Reynolds H, and Kostman JR. Access to costly new hepatitis C drugs: medicine, money, and advocacy. 

. 2015;61(12):1825-1830. doi: 10.1093/cid/civ677.

6. Jung JK, Feldman R, Cheong C, Du P, Leslie D. Coverage for hepatitis C drugs in Medicare Part D. 

7. Karmarkar TD, Starner CI, Qiu Y, Tiberg K, Gleason PP. Sofosbuvir initial therapy abandonment and manufacturer coupons in a commercially insured population. 2016;In Press.

8. Younossi ZM, Park H, Gordon SC, et al. Real-world outcomes of ledipasvir/sofosbuvir in treatment-naïve patients with hepatitis C. 

9. Lakshmi S, Alcaide M, Palacio AM, et al. Improving HCV cure rates in HIV-coinfected patients—a real-world perspective. 

10. Moreno GA, Mulligan K, Huber C, et al. Costs and spillover effects of private insurers’ coverage of hepatitis C treatment. 

11. Jena AB, Stevens W, Sanchez Gonzalez Y, et al. The wider public heath value of HCV treatment accrued by liver transplant recipients. 

12. Linthicum MT, Sanchez Gonzalez Y, Mulligan K, et al. Value of expanding HCV screening and treatment policies in the United States. 

13. Clary R. It’s time to end insurance restrictions on life-saving hepatitis C treatments. 

14. Lakdawalla DN, Linthicum MT, Vanderpuye-Orgle J. Does patient cost sharing for HCV drugs make sense? 

15. Lakdawalla D, Sood N. Innovation and the welfare effects of public drug insurance. 

16. Digestive Disease Week. Curing hepatitis C could yield huge economic benefit. ScienceDaily website. www.sciencedaily.com/releases/2015/05/150517143506.htm. Published May, 17 2015. Accessed April 2016.&ensp;

Articles

FROM THE EDITORS

This special issue presents important new peer-reviewed research, covering issues ranging from access and the out-of-pocket costs of a treatment course, to the real-world consequences-both economic and clinical-of failing to treat.

A Way Out of the Dismal Arithmetic of Hepatitis C Treatment

Am J Manag Care. 2016;22(5 Spec Issue No. 6):SP188-SP190

The launch of novel direct-acting antivirals (DAAs) for the treatment of hepatitis C virus (HCV) in 2013 brought the ever-present tension among innovation, drug pricing, and patient access into the media and policy-making spotlight. Although the clinical value of these drugs is clear, their high cost raised protests from insurers and health systems concerned that treating HCV patients with these highly effective new drugs would bankrupt the healthcare system. As a result, some insurers have shifted the cost burden partly onto patients with HCV by placing these therapies on high-cost specialty drug tiers within their formularies. This approach might seem appealing and “equitable” at some level, but closer reflection reveals that it would expose patients with HCV to potentially significant financial burdens at a time when they are least able to cope with them. There are better options for sharing the costs of HCV treatment among beneficiaries.

Effects of Cost Sharing on Adherence in HCV

Opponents of high cost sharing commonly argue that it reduces patient adherence and leads to worse health outcomes. Indeed, there is considerable evidence to suggest that high levels of cost sharing have such effects in a wide variety of disease areas

: on average, 10% greater cost sharing reduces pharmaceutical spending by 2% to 6% but is also associated with lower initiation and adherence.

 However, this pattern does not appear to hold in the case of HCV therapies. Although adherence data for new DAAs remain scarce, we can assess the relationship between cost sharing and adherence to older, less-effective regimens, such as pegylated interferon (Peg-IFN) alone, Peg-IFN plus ribavirin (RBV), or so-called “triple therapy”—Peg-IFN plus RBV plus older generation DAAs boceprevir or telaprevir (BOC/TPV).

 illustrates the relationship between adherence and cost sharing for employer-insured patients with HCV who were prescribed older HCV regimens between 2004 and early 2014. Adherence is measured by the proportion of days covered (PDC), which measures the fraction of days on which the patient had the medication on hand—the lower the PDC, the more days of therapy patients are forced to miss. The degree of cost sharing is measured as the proportion of total drug costs, paid out of pocket, for each insurance plan and year in the data; it is broken into 4 quartiles: 1 (lowest cost-sharing plan-years) to 4 (highest).

The Figure also demonstrates how average adherence across plan-years varies across cost-sharing quartiles for different HCV regimens. It shows no economically or statistically meaningful effects of cost sharing on adherence; the only factor possibly affecting adherence appears to be regimen type, where patients may be slightly less adherent to regimens that include BOC/TPV. However, higher cost sharing does not seem to discourage adherence to earlier HCV regimens. Patients were willing to bear higher out-of-pocket costs for the older generation of drugs. Economic theory would suggest that they would be just as, if not more, willing to bear higher out-of-pocket costs for the newer, more effective and more tolerable generation of drugs.

Does High Cost Sharing for DAAs Make Sense?

In light of this, it would be tempting to conclude that high cost sharing for new DAAs is unlikely to discourage adherence. Although this might be the case, new DAAs are very expensive and likely to be considered specialty drugs for reimbursement purposes. Subsequently, patients’ actual out-of-pocket costs under high cost-sharing arrangements could be quite large, even with private insurance. High cost-sharing arrangements for these therapies would therefore impose a significant financial burden on patients who are already bearing the burden of a potentially life-threatening disease.

The point of insurance is to protect risk-averse consumers from bearing the full risk of large negative outcomes. Most individuals would prefer to pay a $1000 premium for homeowner’s insurance than live in the perpetual shadow of a 1% chance of a $100,000 loss. Even though the expected cost of both these arrangements is identical, risk-averse consumers would strictly prefer to buy actuarially fair insurance than to live with the risk of a sudden, large financial loss. The same logic applies to health insurance: consumers would rather pay an actuarially fair premium to cover services that they may or may not eventually use than to bear the risk of a large, but uncertain, negative outcome.

This is true even in the case of high cost-sharing arrangements. Suppose annual healthcare costs for a severely ill patient are $100,000 and that 1 of 10 premium-paying beneficiaries is severely ill in a given time period. Now, compare a cost-sharing arrangement that requires those who become severely ill to pay $10,000 of their healthcare costs out of pocket with one that increases premiums by $1000 for all beneficiaries. Both of these arrangements generate the same amount of total revenue to the insurer; however, most consumers would rather pay $1000 in premiums to avoid the 10% risk of suddenly losing $10,000 with the onset of a severe illness. To this point, prior research shows that consumers would rather pay $260 per month in additional premiums in order to avoid an expected cost of $100 per month in cost sharing on high-cost drugs.

This is not to say that cost sharing is never a useful tool or that premiums are always a preferable substitute for them, but the economic goal of cost sharing is to discourage inappropriate use of therapies that are not legitimately needed—this is not the issue with novel HCV therapies. New DAAs are highly appropriate and effective,

 curing HCV in more than 90% of patients.

 Other research in this special issue on HCV argues that even early stage patients with high functional status can benefit substantially over the long-term from DAA treatment.

Cost sharing appears even less efficient when one considers that treatment makes long term financial sense for society: the treatment of HCV today will lead to lower medical expenditures for covered individuals in later years. Combined with the health and longevity gains from treatment, this is likely to lead to a return on investment within 8 years.

 Curing more patients with HCV in the present also prevents more cases of infection in the future, and the reduced rate of transmission will accelerate the return on investment even further. Further, there is no good economic reason to discourage patients with HCV from using novel agents and every reason to protect them against financial risk.

Finally, high cost sharing for HCV is not only inefficient, it is also inequitable. The imposition of high out-of-pocket costs on highly effective drugs can have disastrous effects on patients’ financial well-being. These effects are seen clearly in other clinical areas: for example, patients with cancer in Washington state have more than double the risk of filing for bankruptcy compared with individuals without cancer.

 Furthermore, cancer’s financial burden is particularly high for low-income households, with over a quarter of personal bankruptcies among low-income households filed as a result of patients’ out-of-pocket medical expenses.

 The question of equity becomes particularly acute in the case of American patients with HCV, nearly one-third of whom live below the poverty line.

 Shifting costs onto the least financially secure members of society is tantamount to a regressive tax, with high potential penalties—in terms of illness and HCV transmission&mdash;for those unable to pay.

HCV treatment is expensive, but it is also extraordinarily effective. Although investing in it will impose real costs on society, those costs are justified by even larger downstream benefits. Insurers are struggling to find ways to afford the costs of treatment, but high cost-sharing arrangements that impose large out-of-pocket expenses on very sick individuals are sub-optimal solutions. Instead, modestly higher premiums for all beneficiaries can achieve the same financial goals for the insurer, without excessively burdening the sickest and least advantaged members of society. 

 Leonard D. Schaeffer Center for Health Policy & Economics, University of Southern California (DNL), Los Angeles, CA; Precision Health Economics (MTL, JV-O), Los Angeles, CA.

 Dr Vanderpuye-Orgle and Mr Linthicum are employed by Precision Health Economics (PHE), a healthcare consultancy to life science firms. Dr Lakdawalla is the chief strategy officer and owns equity in PHE.

 Concept and design (DNL, MTL, JV-O); acquisition of data (JV-O); analysis and interpretation of data (DNL, MTL, JV-O); drafting of the manuscript (DNL, MTL, JV-O); critical revision of the manuscript for important intellectual content (DNL, MTL, JV-O); statistical analysis (JV-O); and supervision (DNL).

 Darius N. Lakdawalla, PhD, USC Schaeffer Center, 635 Downey Way, Verna & Peter Dauterive Hall (VPD), Ste 210, Los Angeles, CA 90089-3333. E-mail: dlakdawa@healthpolicy.usc.edu.

1. Goldman DP, Joyce GF, Escarce JJ, et al. Pharmacy benefits and the use of drugs by the chronically ill. 

2. Goldman DP, Joyce GF, Zheng Y. Prescription drug cost sharing: associations with medication and medical utilization and spending and health. 

3. Romley JA, Sanchez Y, Penrod JR, Goldman DP. Survey results show that adults are willing to pay higher insurance premiums for generous coverage of specialty drugs. 

2012;31(4):683-690. doi: 10.1377/hlthaff.2011.1301.

4. Bhattacharya J. A way out of the dismal arithmetic of hepatitis C treatment. 

5. Afdhal N, Reddy KR, Nelson DR, et al; ION-2 Investigators. Ledipasvir and sofosbuvir for previously treated HCV genotype 1 infection. 

2014;370(16):1483-1493. doi: 10.1056/NEJMoa1316366.

6. Afdhal N, Zeuzem S, Kwo P, et al; ION-1 Investigators. Ledipasvir and sofosbuvir for untreated HCV genotype 1 infection. 

2014;370(20):1889-1898. doi: 10.1056/NEJMoa1402454.

7. Kowdley KV, Gordon SC, Reddy KR, et al; ION-3 Investigators. Ledipasvir and sofosbuvir for 8 or 12 weeks for chronic HCV without cirrhosis. 

2014;370(20):1879-1888. doi: 10.1056/NEJMoa1402355.

8. Moreno G, Mulligan K, Huber C, et al. Costs and spillover effects of private insurers’ coverage of hepatitis C treatment. 

9. Ramsey S, Blough D, Kirchhoff A, et al. Washington state cancer patients found to be at greater risk for bankruptcy than people without a cancer diagnosis. 

. 2013;32(6):1143-1152. doi: 10.1377/hlthaff.2012.1263.

10. Gross T, Notowidigdo MJ. Health insurance and the consumer bankruptcy decision: evidence from expansions of Medicaid. 

11. Denniston MM, Jiles RB, Drobeniuc J, et al. Chronic hepatitis C virus infection in the United States, National Health and Nutrition Examination Survey 2003 to 2010. 

2014;160(5):293-300. doi: 10.7326/M13-1133.&ensp;

COMMENTARY

Despite the high cost of novel hepatitis C treatments and patients' apparent willingness to bear part of it, high patient cost sharing is both inefficient and inequitable.

Does Patient Cost Sharing for HCV Drugs Make Sense?

To investigate the value of expanding screening and treatment for hepatitis C virus (HCV) infection in the United States.

We modeled HCV progression and transmission to analyze the costs and benefits of investment in screening and treatment over a 20-year time horizon. Population-level parameters were estimated using National Health and Nutrition Examination Survey data and published literature. We considered 3 screening scenarios that vary in terms of clinical guidelines and physician awareness of guidelines. For each screening scenario, we modeled 3 approaches to treatment, varying the fibrosis stage of treatment initiation. Net social value was the key model outcome, calculated as the value of benefits from improved quality-adjusted survival and reduced transmission minus screening, treatment, and medical costs.

Expanded screening policies generated the largest value to society. However, this value is constrained by the availability of treatment to diagnosed patients. Screening all individuals in the population generates $0.68 billion in social value if diagnosed patients are treated in fibrosis stages F3-F4 compared with $824 billion if all diagnosed patients in stages F0-F4 are treated. Moreover, increased screening generates cumulative net social value by year 8 to 9 under expanded treatment policies compared with 20 years if only patients in stages F3-F4 are treated.

Although increasing screening for HCV may generate some value to society, only when paired with expanded access to treatment at earlier disease stages will it produce considerable value. Such a “test and treat” strategy is likely to entail higher short-term costs but also yield the greatest social benefits.

Am J Manag Care. 2016;22(5 Spec Issue No. 6):SP227-SP235

We developed a discrete-time Markov model to simulate the effects of expanding screening for hepatitis C virus (HCV) infection and initiating treatment at different fibrosis stages. We compare screening and treatment policies in terms of net social value over a 20-year horizon.

Increased screening generates positive social value in 20 years, but this benefit is reduced without concurrent expansion of treatment. 

Investments in HCV screening and treatment are expected to “break even” from a social perspective after 20 to 22 years when treatment is limited to fibrosis stages F3-F4 and after only 8 to 9 years when treatment is expanded to include stages F0-F2.

Chronic infection with hepatitis C virus (HCV) is estimated to affect at least 3.5 million individuals in the United States,

 Chronic HCV infection can lead to hepatic damage, including cirrhosis and hepatocellular carcinoma, and is the most common cause of liver transplantation in the United States.

Because symptoms of HCV infection are usually absent or nonspecific until late stages of the disease, an estimated 50% to 75% of infected individuals are unaware of their HCV status and get tested only after significant symptoms develop.

 Prior research suggests that earlier identification and treatment of patients infected with HCV generates benefits for patients and society, but the potential social value of increased screening, whether alone or in combination with early treatment, is not well understood.

 Novel HCV regimens, including direct-acting antivirals (DAAs), have increased cure rates dramatically, which may affect the value of expanded screening.

 For example, rates of sustained virologic response (SVR) observed in clinical trials of DAA treatments generally exceed 98% for patients infected with genotype 1 HCV without cirrhosis or prior treatment failure.

Despite rapid innovation in HCV treatment, however, unmet need remains significant. Only 13% to 36% of patients diagnosed with chronic HCV infection have received treatment,

 and even fewer patients completed the treatment regimen and achieved SVR.

 Failures to screen, diagnose, and treat all contribute to this current state of affairs.

Broad consensus exists on the need for inclusive screening. In 2012, the CDC updated its guidelines and recommended expanding screening to include all individuals born between 1945 and 1965 (baby boomers)—a cohort comprising an estimated 75% of existing HCV infections.

 Similarly, the American Association for the Study of Liver Diseases (AASLD) and the Infectious Diseases Society of America (IDSA) updated their guidelines in 2015 to recommend one-time screening for asymptomatic baby boomers.

 Unfortunately, more than 40% of physicians are unaware of current guidelines,

 and many individuals infected with HCV may have limited contact with the healthcare system. For these and other reasons, HCV screening rates remain below recommended levels.

It remains unclear whether and to what extent expanded screening benefits society. All-oral DAA regimens present considerable up-front costs

; yet recent research suggests the value of their long-term health benefits is likely to be even higher.

 Screening can identify potentially treatable patients, with implications for both healthcare costs and health benefits. In this article, we explore whether and to what extent expanded screening policies provide net value to society and assess the net social value of varying levels of access to treatment after diagnosis.

Overview of the Markov Model of HCV Transmission and Progression

In this article, we present results of a discrete-time Markov simulation model (Microsoft Excel 2010/VBA, Microsoft Corporation, Redmond, Washington) that simulates the detection, treatment, and progression of populations susceptible to HCV infection, as well as associated costs and health benefits, under different screening and treatment policies. The model builds on previous work that simulates the effects of treatment policies (without screening) on population-level costs, health benefits, and disease dynamics.

The model tracks infected and uninfected individuals in 3 groups, stratified by risk of HCV exposure: a) people who inject drugs (PWID), b) HIV-positive men who have sex with men (MSM-HIV), and c) all other adults born before 1992, when systematic testing of the blood supply for HCV began (Other Adults). Of the last group, approximately 39% were baby boomers.

 The model further stratifies the infected population in each risk group by HCV genotypes 1, 2, and 3, which account for 70%, 16%, and 12% of the US population infected with HCV, respectively.

Once infected, individuals progress through disease states according to transition probabilities drawn from the literature (see 

). Undiagnosed patients face some probability of screening, which varies across the 3 scenarios described below. Diagnosed patients face a probability of treatment that varies according to 3 treatment policy scenarios. If successfully treated, cured patients return to the pool of susceptible, uninfected individuals and experience the same probability of reinfection as those without a previous infection.

The 3 HCV risk groups were modeled independently, such that individuals do not switch among risk groups and cannot infect individuals in a different risk group. Although patients are infected with only 1 genotype at a time, once cured they can be re-infected with any genotype. HCV transmission in the MSM-HIV and PWID risk groups is based on the number of infected individuals in each risk group and genotype, and is described in detail in the eAppendix.

 Outside of the PWID and MSM-HIV groups, the risk of HCV transmission is low.

 Therefore, we made the simplifying assumption of no further transmission of HCV in the Other Adults group.

Key model inputs included starting population size, transmission probabilities, and progression rates in each risk group; genotype and disease state at diagnosis; HCV treatment costs; nontreatment medical expenditures; screening costs; quality-adjusted life-year (QALY) utility weights; and mortality rates. Model parameters were obtained from the published literature or computed from National Health and Nutrition Examination Survey (NHANES) data for the years 2003 through 2012.

 All cost estimates were adjusted to 2015 US dollars, and all future costs and QALYs were discounted at 3% per year.

Base drug costs reflect wholesale acquisition costs as of December 2014. However, since treatment duration varies by genotype, this results in different treatment costs by genotype. All treatments considered are currently patent-protected and face price competition from other branded products. To account for branded competition, the model reduced treatment costs by 46% in years 2 to 20 of the simulation.

 Screening costs included the cost of an HCV antibody test (enzyme-linked immunoassay) and a level-1 outpatient visit.

 Medical expenditures for diagnosed patients were computed by disease state and diagnosis status.

QALY weights were assigned based on disease state and diagnosis status. We assumed that individuals diagnosed with HCV incur associated psychological costs; therefore, patients who are HCV-infected, but undiagnosed, have QALY weights 2% higher than their diagnosed and untreated counterparts.

 For details on model parameters, dynamics, and assumptions, see the eAppendix.

 The model explores 3 scenarios for the frequency and inclusiveness of screening in clinical practice (see 

). We used AASLD/IDSA screening guidelines to define screening practice,

 adjusting for 2 important realities. First, screening can occur only if a patient interacts with a healthcare provider. NHANES data were used to determine the annual rate at which patients received healthcare services. Second, patients might decline offered screening. In all scenarios, we assumed that 91% of those offered screening would accept it.

Real-world screening rates also depend on physician awareness of, and adherence to, screening guidelines. The baseline scenario (Current Screening) assumes that 58% of clinicians are aware of HCV screening guidelines based on data reported in the literature.

 To assess the effects of expanded screening on costs and patient outcomes, we considered 2 alternative scenarios: Physician Education explores the effect of increasing physician awareness of screening guidelines to 100%, with no change in the guidelines themselves, and Screen All assumes that, in addition to increasing physician awareness of guidelines to 100%, guidelines are expanded to provide one-time HCV screening to all individuals born before 1992. Because data for guideline adherence were not available, we made the simplifying assumption in all scenarios that all physicians aware of screening guidelines also adhere to them. In practice, physician adherence to clinical guidelines is likely to be imperfect

; however, assuming full adherence yields the maximum possible value that could be generated by screening. We examined the sensitivity of screening rates to physician adherence in the eAppendix.

The effect of HCV screening on patient health depends on whether a diagnosed patient is subsequently treated. To better understand the relationship between screening and treatment access, we varied the fibrosis stages at which treatment would be available to diagnosed individuals in each screening scenario. Using the METAVIR scoring system to categorize liver fibrosis stages from F0 (no fibrosis) to F4 (most severe)

we considered 3 levels of treatment access: a) treatment at fibrosis stages F3-F4, which reflects current practice

 and serves as the baseline; b) treatment at F2-F4; and c) treatment at F0-F4. We assumed that all screened and diagnosed individuals receive all-oral DAAs if they are insured (see eAppendix).

The key model output was net social value, defined as the difference between: a) the economic value of clinical benefits from improved quality-adjusted survival and reduced transmission, which is calculated as total QALYs multiplied by $150,000

; and b) total healthcare costs, measured as the sum of treatment costs, screening costs, and other medical expenditures. We reported results as changes relative to the baseline scenario (ie, Current Screening with treatment at F3-F4). Therefore, net social value is reported as the difference between a given alternative scenario and the baseline. We also examined the value of expanding screening while treatment remained constant. In this case, Current Screening serves as the baseline for comparing expanded screening scenarios.

In addition to net social value, we reported incremental cost-effectiveness ratios relative to the baseline. Also, since HCV treatment incurs short-term costs but generates long-term benefits, we calculated the break-even point (ie, the years required to switch from negative to positive cumulative net social value) for each screening and treatment combination. Cumulative social value and cost-effectiveness results are presented for a 20-year time horizon. For results at the 10-year time horizon, see the eAppendix.

Each parameter in our model is characterized by some degree of uncertainty. For example, estimates for disease and transmission dynamics vary in the literature. Additionally, our model includes a number of important assumptions that affect our results.

To test the sensitivity of our model to disease progression and transmission parameters, we conducted sensitivity tests of key model parameters within each scenario. For each key parameter, we varied the parameter across a range and report how the scenario’s value changes in percentage terms when using the upper and lower bounds of the range. We also examined several key assumptions, including physician adherence to screening guidelines, future reductions in treatment costs, and others. For details, see the eAppendix.

 reports the annual net value of screening scenarios stratified by treatment scenario. More inclusive screening policies involve net costs in the short term, but generate positive net value after 5 to 7 years. More comprehensive treatment policies cause inclusive screening policies to rise in value more quickly, but also make them more costly in the short run. Relative to Current Screening, annual net values in Screen All are approximately double those in Physician Education.

Costs and QALY gains used to calculate cumulative net social value over the 20-year time horizon are presented in 

Total cost is driven primarily by medical expenditures and treatment costs. In both expanded screening scenarios, medical expenditures increase under treatment at F3-F4. By contrast, savings from reduced medical expenditures exceed the costs of treatment in all scenarios under treatment at F0-F4.

Over a 20-year time horizon, Screen All generates the greatest cumulative net social value at all levels of treatment access compared with the baseline (see 

). In general, however, screening expansion has a relatively small effect on cumulative net social value, unless treatment is similarly expanded. Relative to Current Screening, Screen All generates a net gain of $0.68 billion under the most restrictive access to treatment (F3-F4) and Physician Education generates a net loss in social value of $1.76 billion. The relative gain from increased screening rises with more comprehensive access to treatment. Under treatment at F2-F4, Physician Education generates net social value of $421 billion and Screen All generates net social value of $464 billion over 20 years, relative to baseline. These gains increase to $752 billion (Physician Education) and $824 billion (Screen All) under more comprehensive treatment (F0-F4).

Under any given treatment strategy, Screen All is the highest-value screening strategy (see 

). With treatment at F2-F4, Screen All generates nearly twice the value generated by Physician Education over 20 years. The value of screening approximately doubles when treatment is expanded to F0-F4, under which Physician Education and Screen All generate $83.7 billion and $155.1 billion, respectively, in cumulative discounted social value. Broader treatment increases the costs and benefits by roughly the same proportion. Therefore, even though net social value doubles with wider treatment, incremental cost-effectiveness ratios (ICERs) for screening do not vary ($42,000/QALY for treatment at F2-F4 and $19,000/QALY for F0-F4).

All screening strategies are highly cost-effective after 20 years when combined with treatment at F2 or earlier (Table 3). In these expanded treatment scenarios, Screen All exhibits the highest ICER under treatment at F2-F4, at $6747/QALY gained, and 4 of the 6 scenarios are cost-saving. Expanded screening is less cost-effective when treatment is restricted to F3-F4, however, with ICERs reaching $163,933/QALY for Physician Education.

Varying screening policy has little impact on the number of years required to break even (see Table 3). The accompanying treatment scenario has a much larger effect. For example, with treatment restricted to F3-F4, Screen All breaks even in 20 years—just slightly earlier than Physician Education (22 years). With expansion of treatment to either F2-F4 or F0-F4, all screening scenarios break even in 8 or 9 years.

Sensitivity tests within screening/treatment combinations highlight 4 key drivers of uncertainty in our results: starting size of the total Other Adult population, QALY utility weights, discount rate, and economic value of QALYs. Combining the maximum and minimum values from these parameters’ ranges generates 16 permutations that allow us to approximate the upper and lower bounds on our results, given uncertainty in model parameters.

For scenarios with expanded treatment, net social value always remains positive after 20 years and all scenarios remain cost-effective or cost saving. For screening expansion under treatment at F3-F4, however, cumulative social value ranges from —$27 billion, at the minimum, to $69 billion at the maximum (both under the Screen All scenario); cost-effectiveness under treatment at F3-F4 ranges from $114,819 to $206,992/QALY gained. This result highlights the interdependence between screening and treatment policies. For detailed results and additional sensitivity analyses, see the eAppendix.

Both expanded screening and expanded treatment are valuable. However, they are each more valuable when used together. Screening is more effective when diagnosed patients are treated earlier, and treatment expansions generate greater benefits when more patients are diagnosed. Conversely, increasing screening without expanding treatment leads to minimal gains or net losses to society. Newly diagnosed patients derive less benefit and some may even be harmed by the knowledge of their HCV infection if they remain untreated.

For example, the strategy of expanding both screening and treatment breaks even after 8 to 9 years, but expanding screening alone takes 20 to 22 years to break even. The strong complementarity between screening and treatment policies remains over a wide range of cost estimates. Even under the most optimistic screening scenario sensitivities, expanding screening alone takes a minimum of 16 years to break even. One might see greater returns from screening alone, if diagnosed and untreated patients reduce risky transmission behaviors. We do not investigate this possibility, which should be considered in future research.

Our findings suggest that screening expansions are robustly cost-effective and socially valuable, but only when paired with expanded treatment. This is consistent with previous research suggesting that screening for HCV is cost-effective when paired with treatment, even when treating with more expensive DAAs; Rein et al (2012), for example, reported an ICER of $35,700/QALY saved by birth-cohort screening policies focused on the baby boomer population.

 A more recent study of novel DAAs suggest ICERs ranging from $24,921 to $72,169/QALY saved.

 At the 10-year time horizon, our results suggest similar levels of cost-effectiveness (see eAppendix).

The pursuit of both expanded screening and treatment for HCV is consistent with current trends in HIV management, where public health agencies and experts have increasingly supported a “test and treat” strategy, as the value of aggressive screening and early treatment for patients with HIV has become clear.

 Existing research, including this study, suggests that such a policy may be beneficial in HCV management as well.

Policy makers and payers in a fiscally constrained environment face a conundrum highlighted by our results. Expanding screening and treatment pays off in as few as 8 years, but the up-front costs are high in the scenarios examined. Because of patient turnover, private payers and state Medicaid systems may not retain patients long enough to directly benefit from their investments in HCV treatment. Furthermore, whereas the costs of screening and treatment are borne by insurers and other payers, only a small portion of the benefits accrue directly to them (in the form of reduced future medical costs).

 The vast majority of the benefits from treatment accrue to patients and society in longer lives and higher quality of life,

 potentially resulting in a “race to the bottom” in which public and private payers make decisions based on short-term costs alone.

Although Markov modeling as a tool for understanding chronic disease management policies is well established in the literature,

 the approach has limitations. First, as with any simulation, Markov models are not designed to generate predictions or forecasts. Similarly, as with all population-level studies, results from a Markov simulation cannot inform individual-level understanding of disease processes and outcomes.

 The results of our model should be approached as a guide for decision making rather than being predictive of real-world outcomes.

Second, each parameter carries a degree of uncertainty. We present sensitivity analyses and alternative model scenarios in the eAppendix in order to characterize this uncertainty. The model also assumes that parameter estimates are stable for the duration of the simulation and that this is a reasonable representation of HCV disease progression in the modeled risk groups.

Third, the model does not capture some important dynamics of the HCV epidemic. For example, the model does not account for the recent outbreak of HCV due to the increase in intravenous drug use among rural youth.

 In addition, the model does not capture the “treatment cascade” that occurs as patients are lost to follow-up between screening, treatment, and, ultimately, the achievement of SVR.

 We also lack concrete data on the extent to which physicians adhere to treatment guidelines.

 Our results are therefore an upper bound on the value of increased screening.

POLICY

Expanding screening for hepatitis C virus infection may generate substantial benefits for patients and society, but only when paired with expanded treatment policies.

Value of Expanding HCV Screening and Treatment Policies in the United States

Organs for transplantation are scarce, but new medical therapies can prevent organ failure and the need for transplants. We sought to describe the unique value created by treatments that spare organs from failure and thus conserve donated organs for transplant into others, using hepatitis C virus (HCV) as a case study. 

Using data on trends in chronic liver disease, liver disease progression, and liver transplant allocation models, as well as the effectiveness of new HCV treatments, we estimate the potential effects of systematic HCV screening and treatment on the demand for liver transplants in the United States. We estimate the spillover benefits to patients with all-cause liver disease in terms of increased availability of transplants and life-years gained.

We estimated that systematic HCV screening and treatment could spare 10,490 liver transplants to HCV-infected patients from 2015 to 2035. An estimated 7321 transplants would accrue to patients with end-stage liver disease without HCV and 3169 transplants to those with uncured HCV, providing approximately 52,700 and 22,800 additional life-years, respectively. 

Treatment advances for HCV have the potential to generate considerable spillover benefits to patients awaiting transplants for non—HCV-mediated liver failure. For other diseases in which organ transplants are in short supply, our study provides a novel pathway by which positive spillovers may accrue from treatments that prevent end-stage organ disease.

Am J Manag Care. 2016;22(5 Spec Issue No. 6):SP212-SP219

		Organs for transplantation are scarce, but innovative hepatitis C virus (HCV) therapies can prevent liver failure and the need for transplants.

		Systematic HCV screening and treatment would spare 10,490 livers for transplant over 20 years, with almost 70% of these livers benefitting patients with non—HCV-mediated liver failure.

		The increased availability of donor livers will lead to 52,700 additional life-years for patients without HCV infection and 22,800 life-years for patients with HCV, providing economic values of $7.9 billion and $3.5 billion, respectively.

		Such spillover benefits to untreated populations underscore the value of innovative therapies in preventing organ failure.

Medical treatments can offer value to society that extends beyond the patients who are directly treated. For example, vaccinations not only reduce the likelihood of infection in those who are vaccinated, they also reduce the spread of infection. Such positive spillovers generate considerable value.

 These spillovers may also result from the treatment of diseases that would otherwise lead to organ damage and transplantation. For example, improvements in nephropathy treatment within diabetes reduce the incidence of end-stage renal disease, thus sparing kidneys for use in patients with end-stage renal disease who do not have diabetes.

The availability of organs for transplantation is scarce. In the United States alone, more than 7000 individuals die awaiting organ transplantation each year.

 In this study, we explored this idea by applying it to recently introduced therapies for hepatitis C virus (HCV) infection. An estimated 3 million individuals in the United States are affected by chronic HCV, a condition associated with long-term injury to the liver and with complications including cirrhosis, hepatocellular carcinoma, and, ultimately, liver failure (see 

 The most common reason for liver transplantation in the United States is end-stage liver disease (ESLD), and currently, nearly 50% (14,000/29,000) of ESLD cases among transplant recipients are due to HCV.

Until recently, treatments for HCV were neither particularly effective nor well tolerated.

 However, newer HCV therapies suppress the virus in more than 90% of patients, making an effective cure of HCV highly likely for the majority of those affected.

 Curing patients of HCV obviates their need for future liver transplantation due to HCV, thus creating opportunities for transplantation into patients with other forms of ESLD.

Currently, only one-third of Americans who need liver transplants receive them,

 and shortages are expected to rise as the transplant waiting list continues to grow while the supply of organs remains flat.

 Obesity and the aging of the population are reducing the quality of deceased donor livers, while obesity-related liver disease is also increasing the demand for them.

Advances in treatment for hepatitis C virus (HCV) have the potential to generate considerable spillover benefits to patients awaiting transplants, especially among those with non—HCV-mediated liver failure.

Special Issue: HCV

A Way Out of the Dismal Arithmetic of Hepatitis C Treatment

Does Patient Cost Sharing for HCV Drugs Make Sense?

Value of Expanding HCV Screening and Treatment Policies in the United States

The Wider Public Health Value of HCV Treatment Accrued by Liver Transplant Recipients

Costs and Spillover Effects of Private Insurers' Coverage of Hepatitis C Treatment

Coverage for Hepatitis C Drugs in Medicare Part D

Real-World Outcomes of Ledipasvir/Sofosbuvir in Treatment-Naïve Patients With Hepatitis C

Sofosbuvir Initial Therapy Abandonment and Manufacturer Coupons in a Commercially Insured Population

Improving HCV Cure Rates in HIV-Coinfected Patients - A Real-World Perspective