Short-term budget affordability of hepatitis C treatments for state Medicaid programs

One of the key challenges in healthcare is that much of the spending is government-financed through Medicaid and Medicare, and therefore subject to short-term fiscal constraints, whereas the benefits from treatments and interventions accrue over much longer time horizons [1]. For instance, when the Congressional Budget Office produces budget projections meant to inform government planning decisions, these projections are conducted over a 10-year time horizon [2]. However, 10 years may only capture a fraction of the benefit of a treatment or other healthcare investment. As an example, in hepatitis C (HCV), progression of mild liver damage to severe or end-stage liver disease often takes place over several decades [3].

Short-term thinking is particularly acute when looking at policy decision-making at the state level where time horizons for budget impact assessments are 1 or 2 years. The outcome of evaluating healthcare investments on a time horizon too short to fully capture the return on investment can result in limiting patient access to valuable healthcare innovations. The limitations of short-term budget modeling can be seen in the case of HCV treatment when direct-acting antiviral (DAA) therapies first came to market. Facing a growing affected population and rising costs of care for HCV, many state Medicaid programs have restricted access to effective, but more costly, DAA therapies used to treat HCV infection [4]. While these therapies have been demonstrated to cure HCV infection in 90–100% of treated patients [5], a 2014 survey of state Medicaid fee-for-service (FFS) programs conducted by the National Viral Hepatitis Roundtable and the Center for Health Law and Policy Innovation of Harvard Law School found a number of treatment access restrictions based on disease severity, sobriety status, and prescriber eligibility [6]. In response, the Centers for Medicare and Medicaid Services (CMS) issued a letter to the states warning that these access restrictions may violate statutory requirements for Medicaid [4].

A follow-up survey found that 18 states had eliminated their disease severity restrictions as of mid-2017, a number that has since grown with states such as Colorado and Washington recently lifting restrictions [6]. Most states (40, according to the report), however, maintained sobriety requirements, ranging from screening and counseling to 12-month abstinence. Though treatment costs have decreased substantially from initial launch prices, state Medicaid programs that maintain access restrictions often indicate cost as the primary driver for limiting treatment [7].

State Medicaid policies for HCV treatment can have a big impact on public health, as HCV disproportionately impacts poor and vulnerable populations that are more likely to access healthcare through state Medicaid programs, including people who inject drugs (PWID) [8]. In fact, it has been estimated that the prevalence of HCV among Medicaid enrollees is 7.5 times higher than the prevalence in the commercially insured population [9]. Cases of new HCV infections have been growing rapidly over the past decade, and the CDC estimated that there were 41,200 new HCV infections in 2016 [10]. Rates of HCV transmission are particularly high among PWID; treating injection drug users can reduce the transmitting population and therefore future infections [11].

Recent research has demonstrated the nationally aggregated benefits and cost-savings of expanded access to DAA therapies for the Medicaid population [12, 13]. Further, the social value generated by treating HCV infection with DAA therapies has previously been quantified [14]. While some payers and policymakers recognize the benefits of DAAs and support expanded treatment [4], many state Medicaid programs, including North Carolina (NC) and Wisconsin (WI), continue to maintain sobriety requirements, while states like Oregon (OR) have recently lifted all restrictions.

In this study, we seek to demonstrate that states can expand DAA treatment access to a broader Medicaid population while balancing short-term budget concerns by quantifying the effect of broader treatment access strategies on the infected population, social value, and budget for the NC, OR, and WI state Medicaid programs.

The availability of effective DAAs for HCV infection has been life-altering for many patients. Prior to the advent of these therapies, treatment was burdensome and toxic, and the cure rate was inconsistent [32]. However, not all patients are able to access these treatments through state Medicaid programs due to a focus on very short-run (1–2 year) budget concerns. However, in keeping with the 10-year time horizon used by the Congressional Budget Office for budget projections [2], we found that all our simulated treatment access policies that expanded access were able to achieve break-even costs compared to Baseline within 4 to 8 years of model start.

The potential benefits of broad access to DAAs in the US was previously estimated by Van Nuys et al. [14]. This research extends those findings with a specific focus on three state Medicaid programs with restrictive access policies. The 2015 state Medicaid policies in NC, OR, and WI restricted treatment to patients with more severe disease and demonstrated sobriety (i.e., no current or recent injection drug use). The impact of more open treatment access policies on the size of the HCV population, HCV medical and treatment expenditures, overall social value, and state spending were modeled over a 10-year simulation period. In all three states, treating all patients with diagnosed HCV regardless of disease severity and sobriety status had the greatest impact on outcomes when compared to other modeled policy scenarios. Not only did removing access barriers substantially reduce the number of patients infected with HCV, it also decreased HCV-related health care expenditures over 10 years, with the initial investment in higher rates of treatment being offset in the first 5–6 years. The overall social value generated from increasing access broadly was substantial. Similar results, although of lesser magnitude, were found with the other treatment access policy scenarios, including treating those at earlier stages of disease and separately removing sobriety requirements. Finally, with the federal government matching a percentage of the spending for each state, the actual burden on states over time is further reduced.

Other recently published studies support broader HCV treatment access policies for state Medicaid programs. Chidi et al. (2016) used a Markov model to compare Medicaid policies to treat patients aged 45–55 with advanced HCV genotype 1 only (restricted access) to policies that treat both early-stage and advanced disease [12]. They assessed the impact of these treatment strategies on CMS as a whole, evaluating downstream clinical and cost effects of restrictive versus full access treatment on Medicare when those patients become eligible. Overall, they found that cases of hepatocellular carcinoma, liver transplants and mortality – all manifestations of severe disease – were substantially reduced with full access compared to the restrictive treatment strategy. Additionally, full access yielded both lower costs and higher QALYs compared to restrictive access. The authors also note that full access treatment is cost-effective and that the budget impact would break even with the restricted access strategy in 9–13 years. While the assumptions around treatment costs differ greatly between Chidi et al. and this research, the direction and magnitude of our findings were similar.

Younossi et al. (2017) also used a Markov model to evaluate economic and clinical outcomes for HCV patients with genotype 1 based on state-specific Medicaid treatment restrictions versus open treatment access over a lifetime horizon [13]. Patients were either: 1) untreated, and only received treatment upon aging into Medicare, or 2) were treated under a Medicaid open access scenario. Similar to Chidi et al., the results showed fewer cases of hepatocellular carcinoma, liver transplants, and cirrhosis, and reduced HCV-related mortality under the open treatment access strategy compared to the treatment restrictions scenario. This translated into an estimated $3.8 billion in overall healthcare cost savings over the lifetimes of the study cohort. While there were key differences between the approach and parameters used in Younossi et al. and the current research, including the time horizon modeled (lifetime versus 10-year), and treatment cost assumptions (annual WAC price increases versus dynamic real-world price reductions to treatment costs throughout model duration), the overall conclusions of the studies were the same: increasing access to DAAs for the Medicaid population ultimately results in lower costs and better outcomes over time.

Our study validates previous work that open access to HCV treatment reduces disease prevalence and is cost-saving in the long-term. We also make a unique case for removing HCV treatment access restrictions by conducting the analysis in three specific state Medicaid programs to compare the impact of access policies. Although several states have removed or reduced treatment access restrictions since 2015 [6, 33], there remain opportunities for other states to allow DAA treatment access to a broader HCV-infected population.

Our study had several limitations. First, the model is populated with literature-based parameters from the US, which may introduce uncertainty and imperfect measures into the analysis. Second, our intention was to model a few states most representative of the population and policies throughout the country, rather than to model the largest number of people infected with HCV. Although we did not model more populous states such as California or New York, we were able to capture variety in Medicaid policies and impacts of interventions that can inform policies in the more populous states. We would expect the more populous states to have larger upfront expenditures but similar break-even patterns as the benefit from reduced transmission would also be greater. Third, this model created artificial boundaries for population mixing, allowing each exposure group to only interact with itself. This does not reflect real-world realities, nor is it likely that a population within a state only remain in that state. However, these choices were deliberate and meant to simplify the assumptions for the model intended to serve as a simulation exercise of a few case examples. Fourth, the model does not incorporate the impact of behavioral changes, such as substance use disorder treatment programs, on disease transmission. Finally, we model only HCV-related health care expenditures. Costs of screening for HCV, costs of diagnostic and eligibility tests such as liver scans and tests for sobriety, and all other medical expenditures are not included in the model as the causality between treatment and the impact on other expenditures would have introduced additional assumptions requiring sufficient evidence. The implications for reducing transmission through treatment and removing sobriety restrictions on screening and diagnosis behavior merits further research.

Medicaid policies that expand screening and treatment for HCV have greater impact on reducing disease prevalence compared with those that expand screening or treatment alone. However, the greatest social value is generated by policies that expand treatment to PWIDs and all patients regardless of disease severity. Federal contributions to state Medicaid programs are also essential to sharing the initial investment from expanded access to treatment. Our findings suggest that states should consider revising their HCV treatment policies to include all individuals infected with HCV in order to achieve significant patient- and population-level benefits.

While increasing treatment access in Medicaid will raise short-term costs, it will also provide clear benefits by saving money and improving health within a 10-year window. Patients and taxpayers would benefit by considering these gains and viewing HCV treatment policies beyond 1–2 year windows.