Methods

Overview
We designed this interactive budget allocator to help understand the tradeoffs the State of Louisiana would need to make to fund treatment of Hepatitis C Virus (HCV) infection for individuals enrolled in the State’s Medicaid program, dually enrolled in Medicaid and Medicare, or who lack health insurance. The allocator allows users to explore reducing funding to some existing Louisiana programs to free up funds to treat HCV and estimates the consequences of doing so. The allocator focuses on a one year budget window to parallel how the state budget is created.

Treatment eligible population
The Louisiana Office of Public Health estimates that at least 73,000 Louisiana residents are infected with HCV.1 The allocator focuses on a large subset of these individuals: Louisiana’s Medicaid recipients, dually eligible Medicaid and Medicare recipients, and uninsured residents currently infected with HCV. State residents with Hepatitis C infection who have insurance coverage funded through another source are not included. A partial list of such individuals includes those who receive Medicare without Medicaid, veterans, active-duty military, and state and private employees covered by commercial insurance. The analysis also excludes prisoners, the group with the highest prevalence rate of HCV, as they are ineligible for Medicaid during their incarceration.

To estimate the subset of HCV-infected residents covered by Medicaid or uninsured, we weighted the currently infected population of 73,000 by the population distribution of the different payer types, adjusted for NHANES prevalence estimates by payer type as reported by Fitch et al.2,3 We also adjusted the prevalence rates estimated by Fitch et al for uninsured, Medicaid, and Medicaid and Medicare dual eligibles to account for shifts to publicly funded insurance following the ACA.

Treatment costs
The allocator focuses exclusively on drug costs for the state of Louisiana. We calculate these drug costs by assuming an average 11 week course of therapy, and applying Gilead’s 2016 4th quarter earnings call disclosure that Harvoni’s net price to Medicaid per 30 tablet bottle is less than $10,000 in states that grant unrestricted access.4 This yields a cost of $28,000 per Medicaid patient.

Drug costs that would be paid by the Federal government are not included in the allocator. We estimate that Louisiana’s current federal matching rate (FMAP) is 70%, taking into account a 63% matching rate for the proportion of beneficiaries on pre-ACA Medicaid programs, and 95% matching for the proportion who qualify through Medicaid expansion.5-7 Consequently, treatment costs to the State for Medicaid-insured patients were estimated to be on average $8,280 per patient.

Because uninsured patients do not have Medicaid coverage, we assume a treatment cost to the State of $41,000, based on overall net price to Gilead of less than $15,000 per 30 tablet bottle and assuming an average of eleven weeks of treatment.4

To be conservative, we focused solely on drug costs. We exclude from the analysis the costs associated with side effects of treatment, extended treatment duration due to adherence issues, and the costs of screening, and starting and monitoring treatment. We also disregard free goods, such as drug samples, or the 340B price, which is likely to be lower than Medicaid Best Price. We constrained the analysis to a year to mimic a budget exercise, but it is unlikely that this time would be sufficient to treat all eligible patients. Budgeting to treat a smaller population over time, although would save money in the short term, would incur other costs as it would dilute the potential for offsetting future cost savings from treatment such as achieved through a reduction in liver transplants or hospitalizations, and also fail to reduce as rapidly the rate of new infections.

Cost savings from treating infected individuals
The allocator gives a discount for expected savings that result from the treatment of HCV infection on the same one year time horizon. We used stage-specific estimates of annual patient costs before and after attaining sustained virologic remission (SVR) from a modeling study by Chahal et al,8 reweighting them according to the number of patients in each stage.9 This resulted in an average per patient per year cost saving of $605 for attaining SVR. We estimated that SVR would occur in 95% of treated patients, a midpoint estimate for treatment response across currently available HCV therapies for HCV genotype 1, the most common and difficult to treat strain.10 Of the savings for Medicaid recipients, approximately 70% would be returned to the federal government to offset the federal matching distribution, accounting for the higher FMAP rate for those covered under Medicaid expansion. The remaining 30% was included as a discount to the State. All savings for uninsured residents accrues to the state.

The allocator also excludes the costs of retreatment in patients who become re-infected, as well as variations in treatment costs by HCV genotype. It is also likely that fewer than 95% of patients will attain SVR, as some will be inconsistent in taking the medication, or suffer from other health conditions, such as HIV co-infection, that limits their treatment options or response to treatment.

Portion of Louisiana state budget that can be allocated
The budget allocator only allows for changes in the discretionary portion of the state general fund (SGF) for Fiscal Year 2016-2017, as other parts of the budget are statutory dedications and appropriations and cannot be reallocated without legislative action. Of the total Louisiana state budget for the year, $31.2 billion, the SGF accounts for $9.6 billion,11 $3.6 billion of which is discretionary. This discretionary portion is comprised of $1.9 billion for the Department of Health, $847 million for Higher Education, and $166 million for K through 12 education.12 The remaining $659 million is found in numerous other line items in the state budget that have been allocated in a proportional manner to align with the total amount of financing for each line item across the State budget13 presented by functional role – administrative, social services, and infrastructure.
Discretionary SGF funding for the Department of Health is further separated into direct Medicaid costs and administrative costs, where Medicaid accounts for 92% of the total.14

The exercise assumes the state budget currently meets its liabilities. But in truth Louisiana already faces a 2017 budget shortfall of $303 million.

Humanizing the impact of budget cuts
The allocator assumes that funding for HCV treatment would come at the expense of other state-funded programs and services. For the Department of Health and Education, we converted changes in funding into units of individuals affected by them, assuming actuarial equivalence between cutting scope of benefits and reducing the number of individuals receiving the benefit, as either approach can be used to reduce program spending. Medicaid enrollment by eligibility category was obtained from the Department of Health’s enrollment reports and reweighted by their cost intensity relative to the average adult Medicaid beneficiary in Louisiana.5,15 Public school and higher education enrollment was obtained from the Board of Regents and K through 12 reports, as were costs per student.16,17

Key assumptions and calculations
Total funds needed were based on the cost and total number of Louisiana residents with HCV who are enrolled in Medicaid or who are uninsured. To mimic the budget allocation exercise facing public officials, the analysis focuses on funding treatment in a single year time frame, and in parallel reduces benefits in other programs during that same year. Changing the exercise to one with a longer time horizon, such as five years, would have roughly the same aggregate impact on benefit reduction, it would just be spread over a longer period. Savings that usually come from deferring spending due to the time-value of money are offset in this case, as delaying treatment serves to increase the rate of new cases in the State from person to person transmission at an ambient rate of around 0.5% of prevalence, and also reduces the potential for future healthcare savings. The allocator requires that Medicaid and dually eligible beneficiaries are treated first; only after their treatment is fully funded are those without insurance treated.

Reallocation estimates
Reallocation of the discretionary portion of the SGF from Medicaid was calculated based on the number of enrollees covered by this funding, their annual cost relative to the average adult Medicaid recipient in Louisiana, and assuming in all other ways the benefits were actuarially equivalent. Categories of enrollment were combined if it was reasonable to assume that their annual costs were similar. For education, reallocation was similarly based on the number of students covered by the discretionary SGF. Reallocation from other line items was not based on enrollment or unit costs, primarily due to a lack of publicly available sources of information.

FAQ

What is the purpose of the Louisiana budget allocator and who developed it?
The Louisiana Budget Allocator was developed by the Drug Pricing Lab at Memorial Sloan Kettering Cancer Center at the request of, and with input from, the State of Louisiana. It was developed to evaluate tradeoffs facing policymakers who seek to fund treatment of individuals infected with HCV. The tool was developed by researchers at the Drug Pricing Lab, including Dr. Peter Bach, Anna Kaltenboeck, Val Pocus, Nancy Yu, Zach Helms, and Raina Jain. Jennifer Katzman provided insight and feedback on behalf of the state of Louisiana.

Why is it so difficult to find enough money to treat HCV in Louisiana?
The State faces a number of challenges. As is widely recognized, the cost of Hepatitis C treatment is extraordinarily high given the population burden of the condition. Meanwhile the State is one of the poorest in the country, has a relatively limited tax base, and has a thinly funded set of benefits at baseline.13 It also has a large number of individuals infected with HCV, not only among the uninsured and Medicaid population, but also among those incarcerated who are not included in the allocation exercise. Prevalence of HCV is believed to be lower in populations covered by commercial health plans.

Why does this model only consider one year? Wouldn’t it take longer to treat everyone?
This allocator was designed to evaluate the tradeoffs made in budget planning, a yearly activity. We make the optimal assumption that all patients could receive treatment within that year, contributing to virologic control at the population level and a lower infection rate. Realistically it would take longer to treat all Medicaid and uninsured patients, with the funding needed for treatment spread out over a number of years, and accordingly lower on a per-year basis. While delaying treatment over a number of years would reduce the cost of treatment in any particular year, the total amount of spending would be roughly similar. This is because whatever savings are achieved by delaying treatment are offset both by the additional healthcare costs of treating those infected for a longer duration, and the reality that new individuals acquire the infection from those left untreated at a rate of around .5% of the prevalent population.18 Thus the cost of treating the newly infected would also be borne in out years.

If Louisiana reallocated its discretionary funds to the treatment of Hepatitis C, would it reduce benefits in the manner we display?
We cannot be certain whether cuts to the programs the allocator allows would be effectuated through a reduction of the number of individuals receiving the benefits as we suggest, or a reduction in the scope of benefits each individual receives. Our best guess is that over time, budget pressure would lead to more stringent eligibility criteria that would reduce the numbers of beneficiaries, but in the short term it is more likely that funding for specific benefits and services would be scaled back for all enrollees. To ensure actuarial equivalence between the costs of cutting scope of benefits as compared to enrolled individuals, unit costs for individuals were estimated to include both direct expenditures (e.g., fees for medical services and prescription drugs, textbooks) as well as administrative costs.

What would it take to deliver HCV treatment to these patients and are those costs included?
Significant outreach and diagnostic resources would be required to reach the intended patient population. Our exercise does not account for the time and resources needed to identify eligible patients or guide their treatment. Such programs would come at an additional, up-front fixed cost, increasing the overall funding need. This makes our exercise conservative, as the amount by which other programs would need to be cut does not account for these additional services; adding them in would increase the amount required in the allocator.

Why aren’t cost savings from treating HCV higher? Isn’t treatment supposed to pay for itself through reduced future healthcare costs?
Our cost savings estimate, which accounts for the increase in costs as patients experience more progressive disease, is presented as a discount over one year to mimic a budgeting exercise. Savings would compound over time. Razavi et al estimated that HCV costs $64,490 (USD 2011) over the lifetime of an infected patient, much of which is attributable to the medical care of decompensated cirrhosis.19 But most patients do not progress to these stages, and progression even among those who do takes decades. Another source of savings not incorporated into the Allocator would come from the reduction of new HCV infections in the state over time due to a lowering of disease prevalence. Although the new infections are important clinically, they occur at a low enough rate that the savings from preventing them are orders of magnitude lower than the cost of achieving that public health objective.

Are there other costs not accounted for in the allocator?
The budget allocator does not consider the cost to the Federal government. Due to cost sharing with state Medicaid programs (FMAP), this is a considerable amount, ranging from 63-95% of HCV treatment costs for Medicaid enrollees. The allocator also excludes costs of retreatment in patients who become re-infected, as well as variations in treatment costs by HCV genotype. We also made the assumption that 95% of patients would attain SVR; in reality, this number is probably lower, as some patients will be inconsistent in taking the medication or suffer from other health conditions, such as HIV co-infection, that limits their treatment options or response to treatment.

Should all of these patients receive HCV treatment?
The HCV treatments now available are highly effective and tolerable, offering a significant improvement over past treatment options. Given the benefits of treating HCV, both to patients and their communities, guidelines now recommend treatment for nearly all patients.20

Sources

1 Louisiana Department of Health. Annual Infectious Disease Surveillance Reports. http://new.dhh.louisiana.gov/assets/oph/Center-PHCH/Center-CH/infectious-epi/Annuals/HepC_LaIDAnnual.pdf. Accessed April 4, 2017.

2 Fitch K, et al. Health Care Reform and Hepatitis C: A Convergence of Risk and Opportunity. http://us.milliman.com/uploadedFiles/insight/2013/convergence-of-risk-and-opportunity.pdf Accessed April 4, 2017.

3 US Census Bureau. 2010 and 2015 ACS data. https://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml. Accessed June 4, 2017.

4 Gilead investor relations. http://investors.gilead.com/phoenix.zhtml?c=69964&p=irol-EventDetails&EventId=5245324. Accessed March 30, 2017.

5 Louisiana Medicaid enrollment report. http://ldh.louisiana.gov/assets/medicaid/MedicaidEnrollmentReports/EnrollmentTrends/EnrollmentTrends-12.16.pdf. Accessed March 30, 2017.

6 CMS. Medicaid and CHIP FAQs: Newly eligible and expansion state FMAP. https://www.medicaid.gov/State-Resource-Center/FAQ-Medicaid-and-CHIP-Affordable-Care-Act-Implementation/Downloads/FAQs-by-Topic-Expansion-State-FMAP-2013.pdf. Accessed April 7, 2017.

7 Kaiser Family Foundation. Federal and state share of Medicaid spending. http://kff.org/medicaid/state-indicator/federalstate-share-of-spending/?currentTimeframe=0. Accessed January 20, 2017.

8 Chahal HS, et al. Cost-effectiveness of Early Treatment of Hepatitis C Virus Genotype 1 by Stage of Liver Fibrosis in a US Treatment-Naive Population. JAMA Intern Med. 2016;176(1):65-73.

9 Thein H-H, et al. Estimation of stage-specific fibrosis progression rates in chronic hepatitis C virus infection: a meta-analysis and meta-regression. Hepatology. 2008;48(2):418-431.

10 Chopra S, Muir AJ. Treatment regimens for chronic hepatitis C virus genotype 1 infection in adults. https://www.uptodate.com/contents/treatment-regimens-for-chronic-hepatitis-c-virus-genotype-1-infection-in-adults. Accessed January 10, 2017.

11 State of Louisiana. Budget Fiscal Year 2016-2017. http://www.doa.la.gov/opb/pub/FY17/StateBudget_FY17.pdf. Accessed March 28, 2017.

12 State of Louisiana. Discretionary and nondiscretionary funding. http://house.louisiana.gov/housefiscal/DOCS_OPERBDGT/Disc%20Non-Disc%20Flowchart2016.pdf. Accessed March 28, 2017.

13 US Census Bureau. Small area income and poverty estimates,. https://www.census.gov/did/www/saipe/data/interactive/saipe.html.

14 State of Louisiana. FY 12-17 Executive budget review: Department of health and hospitals. http://house.louisiana.gov/housefiscal/DOCS_APPBudgetMeetings2016/2016%20DHH%20Public.pdf. Accessed March 28, 2017.

15 Kaiser Family Foundation. Medicaid spending per enrollee. http://kff.org/medicaid/state-indicator/medicaid-spending-per-enrollee/?currentTimeframe=0&selectedRows=%7B%22nested%22:%7B%22louisiana%22:%7B%7D%7D%7D&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Accessed January 20, 2017.

16 Louisiana Board of Regents. The landscape of Louisiana higher education. http://www.regents.la.gov/assets/docs/Public_Affairs/BUDGET/LAHigherEdLandscapePowerPointDeck2016Updated6.2.pdf. Accessed March 28, 2017.

17 Louisiana Department of Education. Revenue and expenditure data. http://www.louisianabelieves.com/resources/library/fiscal-data. Accessed March 28, 2017.

18 Spach DH. HCV Epidemiology in the United States. http://www.hepatitisc.uw.edu/go/screening-diagnosis/epidemiology-us/core-concept/all. Accessed April 6, 2017.

19 Razavi H, et al. Chronic hepatitis C virus (HCV) disease burden and cost in the United States. Hepatology. 2013;56(6):2164–2170.

20 AASLD and IDSA. HCG Guidance: Recommendations for testing, managing, and treating Hepatitis C. http://www.hcvguidelines.org/. Accessed April 6, 2017.