708
Views
12
CrossRef citations to date
0
Altmetric
Hepatology

Modeling the fiscal costs and benefits of alternative treatment strategies in the United Kingdom for chronic hepatitis C

, &
Pages 19-26 | Received 06 Jul 2017, Accepted 18 Aug 2017, Published online: 05 Sep 2017

Abstract

Background: Hepatitis C (HCV) infection causes substantial direct health costs, but also impacts broader societal and governmental costs, such as tax revenue and social protection benefits. This study investigated the broader fiscal costs and benefits of curative interventions for chronic Hepatitis C (CHC) that allow individuals to avoid long-term HCV attributed health conditions.

Methods: A prospective cohort model, assessing the long-term fiscal consequences of policy decisions, was developed for HCV infected individuals, following the generational accounting analytic framework that combines age-specific lifetime gross taxes paid and governmental transfers received (i.e. healthcare and social support costs). The analysis assessed the burden of a theoretical cohort of untreated HCV infected patients with the alternative of treating these patients with a highly efficacious curative intervention (ledipasvir/sofosbuvir [LDV/SOF]). It also compared treating patients at all fibrosis stages (Stages F0–F4) compared to late treatment (Stage F4).

Results: Based on projected lifetime work activity and taxes paid, the treated cohort paid an additional £5,900 per patient compared to the untreated cohort. Lifetime government disability costs of £97,555 and £125,359 per patient for treated cohort vs no treatment cohort were estimated, respectively. Lifetime direct healthcare costs in the treated cohort were £32,235, compared to non-treated cohort of £26,424, with an incremental healthcare costs increase of £5,901 per patient. The benefit cost ratio (BCR) of total government benefits and savings relative to government treatment costs (including LDV/SOF) ranged from 1.8–5.6. Treating patients early resulted in 77% less disability costs, 43% lower healthcare costs, and 33% higher tax revenue.

Conclusion: The ability to cure Hepatitis C offers considerable fiscal benefits beyond direct medical costs and savings attributed to reduced disability costs, public allowances, and improved tax revenue. Changes in parameters, such as productivity, wage growth, and tax rates, can influence the conclusions described here.

Introduction

Hepatitis C infection is a global public health issue, and one of the main causes of chronic liver disease worldwide (with the global prevalence estimated to be >185 million (2.8%))Citation1. The disease is often asymptomatic in the acute phase, and typically progresses slowly, with 15–45% of infections spontaneously resolvingCitation2. The major disease burden comes from chronic infections (55–85%), potentially resulting in the development of significant liver disease, including cirrhosis and liver cancerCitation2. Low rates of early diagnosis due to the asymptomatic nature of this blood-borne virus often mean diagnosis occurs at more advanced stages, frequently decades after infectionCitation2. The World Health Organization (WHO) report worldwide deaths due to the hepatitis C virus (HCV) as 399,000 per year; it is, however, suspected that the actual number may be much higher due to under-diagnosisCitation2.

Estimates in the UK suggest ∼214,000 individuals (0.6%) are chronically infected with HCV, with >90% of infections being genotype 1 or 3Citation3. While it is recognized that both hospital episode statistics and deaths attributed to HCV are under-estimated, hospital admissions from HCV-related end stage liver disease (ESLD) and hepatocellular carcinoma (HCC) have nearly tripled in the UK and deaths more than doubled over the last decade (2003–2013)Citation3. This data highlights a growing burden from HCV, even with a reduced incidence, as patients infected decades ago reach the later stages of the disease, putting a substantial strain on the healthcare systemCitation4.

Highly efficacious and well-tolerated treatments have become available in recent years; however, many governments and healthcare providers have been hesitant to fully roll-out such advances to their full patient populations due to fears related to potential short-term cost impact and doubts about the longer-term cost-effectiveness of using these relatively expensive therapies. Previous analyses have largely been based on the direct healthcare costs of treatments and of progressive liver disease, but have generally not encompassed wider societal and governmental costs involved in treating or not treating. Such indirect costs are also a contributing factor to the disease’s overall economic burden, as employment rates are lower in HCV patients than in the general population, and overall work productivity among HCV patients is often impairedCitation5–7. A recent study from the RAND group assessed the societal costs that are associated with HCV in the UK, and estimated an annual productivity loss of £192 million pounds for 2015Citation8. Younossi et al.Citation9 showed that these losses can be mitigated by curative interventions. They calculated annual work productivity losses among treated and untreated HCV patients in the UK, and found a 17.9% improvement in work productivity in patients treated with LDV/SOF-based regimens, resulting in cost savings of €26 million annually in the UKCitation9.

Methods and data

In the context of healthcare resource allocation decisions, it is important to understand the discrete public investment in HCV therapies within a single cohort to assess how investments influence future government cash flows based on gross tax transfers and spending on public benefit. The objective of this analysis was to expand on prior research, explore the fiscal costs (i.e. lost tax revenue and spending on social benefit programs) and benefits (i.e. transfers attributed to HCV related morbidity and mortality in the UK, based on a modified generational account framework) of HCV curative therapiesCitation10,Citation11. Specifically, we will analyze HCV curative treatment compared to no treatment to estimate the magnitude of fiscal benefit for informing a broad range of public sector policy-makers regarding the cross-sectorial impact of HCV and investments in curative therapies. The analysis focuses on the average “per person” cost, which will enable local decision-makers to apply these findings to estimate budgetary impact and cost consequences.

The primary analysis was to compare fiscal costs and benefits of HCV with LDV/SOF vs no treatment for an average 40-year-old non-cirrhotic patient infected with HCV genotype 1. A second analysis of early use of LDV/SOF was performed to understand how early therapeutic intervention (i.e. treating non-cirrhotic (F0–F3) HCV patients at early disease stages) vs treating patients at later disease stages influences fiscal costs and benefits for government.

A single-cohort health-state Markov model was used to reflect the natural history of HCV. The Markov model used for performing the fiscal analysis was the same model employed for the UK cost-effectiveness analysis (CEA) evaluated by the National Institute for Health and Care Excellence (NICE) as part of the LDV/SOF technology appraisal (TA363), and has been previously describedCitation12. Sustained viral response of 97% with LDV/SOF was based on the pivotal efficacy studiesCitation13,Citation14. The original cost-effectiveness model captured the direct healthcare costs of the different HCV health states. Specifically, the CEA model estimated the lifetime direct medical costs of a single-cohort of HCV infected patients. HCV patients entered the model with a certain genotype (e.g. Genotype 1), at a certain age (e.g. age 40), and either as non-cirrhotic or cirrhotic. Subsequently, patients may progress to one of the following stages: (1) Sustained viral response (SVR); (2) cirrhosis with or without SVR; (3) decompensated cirrhosis; (4) hepatocellular carcinoma; (5) liver transplant; (6) post-liver transplant; or (6) death (HCV attributable or all-cause). The transition probabilities between health states are provided in Supplementary Appendix 1. Each of the above disease states were associated with direct medical costs.

For each HCV health state in the LDV/SOF cost-effectiveness model, a specific productivity profile that corresponds to certain labor market outcomes, tax paying potential, and receipt of social benefits were assigned. In this way, the LDV/SOF cost-effectiveness model was adjusted to capture lost productivity attributed to HCV morbidity and the resulting lost tax revenue. Lost productivity was attributed both to morbidity in more severe health states that impair work activity and decreased life expectancy. Furthermore, the CEA model was modified to capture social benefit provided by the government for people that are unable to live and work normally once they reach a severe HCV health state. The methods used for this analysis originated from the traditional human capital theory and the theory of generational accountingCitation10, originally developed to explore the cross-sectorial and inter-temporal impact of government policy on the value of current and future taxes a person is likely to pay net of transfer payments over their lifetime.

Direct and indirect (i.e. VAT) taxes paid by individuals were based on earnings and consumable spending, respectively; therefore, the starting point for estimating tax revenue was derived from age-specific earnings profilesCitation15. There are several peculiarities that relate to taxable incomes for HCV individuals and taxes paid. Individuals with HCV often earn lower wages than non-HCV individuals, have lower labour-force participation rates, and face higher unemployment rates; hence, adjustments to average population earnings were necessary. To reflect the income of a cohort with HCV patients, average general populations’ labour market outcomesCitation16 were adjusted based on observational studies from the literature which compared the labor market outcome of HCV patients to that of the general populationCitation5–7,Citation17,Citation18. Furthermore, published reductions in labor force participation rates were applied to late stages of HCV infection. The adjusted earnings data used in this analysis to calculate direct and indirect tax are illustrated in . Average old-age pensions and benefits were also derived from national statistical sourcesCitation19. To estimate the direct and indirect (i.e. VAT) taxes for people with HCV in the UK, direct taxes were linked to earnings and indirect taxes were linked to the age-specific consumption attributed to disposable income. The average direct tax paid on wages in the UK is 18.9%, whereas the average consumption or indirect tax rate applied to disposable income is 18.8%Citation20. Thus, the total tax burden in this analysis was ∼39%, which excludes national insurance contributions.

Table 1. Earnings, labor market outcomes for HCV, social protection costs, and healthcare costs by health state.

Individuals with advanced HCV states can become dependent on government social support programs as their health condition restricts their ability to work and live normally. The proportions of persons receiving social benefits in early and late stage disease were obtained from survey data for the UKCitation21. The cost for each social benefit received by people with HCV was identified in the Department for Work & Pensions, Benefit and Pension Rates reportCitation22. The reported frequency of benefits received amongst HCV-infected individuals was used to derive annual costs based on HCV health states. The estimated annual costs of social benefits for early stages (i.e. compensated cirrhosis) were estimated at ∼ £14,451 annually, whereas more advanced stages were associated with annual costs of £18,021. For non-cirrhotic HCV patients, an annual cost of £1,904 was assumed, which is similar to the average annual cost of social benefits received by the general population in the UK. The annual social benefit costs were applied at the various HCV health states, based on the proportion of people in each stage.

Based on an assumed starting age of 40, a cohort of n = 10,000 HCV patients entered the Markov model either as non-cirrhotic (F0–F3) or as cirrhotic (F4). Patients either receive treatment with LDV/SOF or no treatment. For each Markov cycle the HCV cohort progressed through the model, based on the transition probabilities to different health states. Based on the numbers of subjects in each health state, the annual health state-specific fiscal costs and benefits of HCV and HCV treatment were estimated in terms of annual gross tax revenue paid to the government, annual healthcare costs and social benefits, and disability costs per year paid by the government to HCV patients. In turn, the model calculated, for the lifetime of the cohort, the cumulative fiscal costs and benefits, discounted at 3.5%, of HCV and HCV treatments (EquationEquations 1–3). Future government costs and benefits were discounted at a rate of 3.5%, in line with the UK Treasury Green Book recommendations for assessing public funding of projectsCitation23. In addition to the costs described previously, old-age pension costs were calculated based on the average annual old-age pension and benefits in the UK multiplied by the number of patients surviving each year beyond the expected average age of retirement in the UK (i.e. 68 years of age). (1) (2) (3) where: m = number of Markov cycles; ns = number of patients in Markov-state s; Y = annual age-specific earnings (for age i) in Markov-state s; Tax = Average direct and indirect Tax burden; Ds = Annual social benefit transfers/disability costs in Markov-state s; and HCs = Annual healthcare costs in Markov-state s.

A sensitivity analysis was performed in the primary analysis comparing fiscal costs and benefits of HCV with LDV/SOF vs no treatment. The following parameters were varied to test the sensitivity of the primary analysis: Discount rate = 0–4%; Treatment cost = ±20%; disability cost = ±20%; wage productivity = ±20%; Earnings = ±20%.

Two different benefit cost ratios (BCR) were derived as follows using absolute values: where No Tx = usual care treatment costs; iD = incremental tax revenue; iTax = incremental tax revenue; and iHealth costs = the incremental cost increase switching from no treatment to LDV/SOF.

Results

Treatment vs no treatment

The disaggregated outputs from the HCV fiscal model have been independently presented to illustrate the impact of treatment vs no treatment scenarios on different government budgets. The costs of LDV/SOF and usual care were presented over the remaining lifetime of HCV individuals (). The total healthcare costs of treatment vs no treatment were £32,235 and £26,424, respectively, with a net healthcare cost of £5,901 with LDV/SOF compared to no treatment. Disability costs for treatment vs no treatment were £97,555 and £125,359, respectively, indicating that savings in disability costs with LDV/SOF of –£27,804 can be achieved. These costs are based on the proportions of subjects in different cirrhotic and non-cirrhotic health states and the public costs (non-health) associated with each disease state. A small increase in tax revenue per person of £4,324 is projected in the LDV/SOF treatment group compared to no treatment.

Table 2. Per person disaggregated discounted fiscal costs and benefits of HCV comparing no treatment with LDV/SOF.

The results of the analysis above were used to derive benefit-cost ratios (BCR) to project economic value in relation to LDV/SOF costs (). The BCRs were constructed using different combinations of costs to estimate the BCR, based on gross or net investment costs in LDV/SOF. The BCR(a) analysis that includes the gross treatment costs of LDV/SOF generates a ratio of 1.8, indicating that, for each £1 spent, £1.8 in fiscal benefit for government is generated. In contrast, the BCR analysis which considers the net costs of HCV treatment (LDV/SOF - usual care costs) generates a BCR(b) estimate of 5.4, suggesting that £1 spent offers £5.4 in fiscal benefit.

Sensitivity analysis

To assess the sensitivity of the results to the different parameters used, a deterministic, one-way sensitivity analysis of the BCR(a) was explored (). The sensitivity analysis results indicate the BCR is most sensitive to the discount rate. Changes in the treatment cost had an influence on the BCR, although in the variations assessed the BCR remained over 1, indicating that additional fiscal benefit was obtained in relation to investment costs in all scenarios tested. Earnings had a minor impact on the BCR findings, which reflects the observation that curative HCV treatment has limited impact on productivity and future earnings and tax revenues.

Figure 1. One-way sensitivity analysis for the BCR (treatment vs no treatment).

Figure 1. One-way sensitivity analysis for the BCR (treatment vs no treatment).

Early vs late treatment

Accounting for fiscal costs and benefits, early treatment (any fibrosis stage) offered healthcare cost savings of £24,747 compared to the late treatment (cirrhotic only) option. Furthermore, a net reduction in disability costs of –£119,602 was demonstrated with early intervention compared to late intervention (). The likely incremental improvement in gross tax revenue from early intervention was estimated at £49,829 over the lifetime of the cohort. Taking into consideration the absolute savings on health costs and disability, as well as future tax revenue gains, and excluding pensions, the net benefit of early intervention is estimated to be £194,178.

Table 3. Comparison of early vs late LDV/SOF therapy in non-cirrhotic Hep C population in the UK, based on cross-sectorial fiscal impact.

Discussion

From a governmental perspective, healthcare interventions that reduce morbidity and increase survival will have both positive and negative fiscal consequences for government. On the one hand, decreases in morbidity and mortality will increase the tax base, as more people are in the labor force generating tax revenues for the government whilst consuming less disease-specific healthcare and social benefits. On the other hand, more survivors are eligible to receive public benefits including healthcare costs, pensions, and other allowances, whilst consuming the healthcare resource required for the intervention, e.g. treatment for chronic Hepatitis C (CHC). In the case of CHC, wider fiscal costs and benefits to government are likely to extend significantly beyond healthcare costs. For instance, people with more advanced cirrhotic disease are often unable to live normal lifestyles and become increasingly reliant on government social support programsCitation21. Furthermore, reduced work activity translates into reduced tax revenue, both through reduced income taxes and lower disposable income, which lowers value added tax due to reduced consumption.

Central and local government have become increasingly aware of the importance of assessing the impacts of interventions on both health and social care, and this has been made even more pressing with the planned devolution of joint health and social budgets to local decision-makers within the UK and potentially elsewhereCitation24. The broader fiscal consequences of poor health in working aged adults on government are not trivial. Cross-sectorial studies in the UK indicated that ill-health in working aged adults reaches £62–76 billion annuallyCitation25. Of these costs, ∼ 89% (£57–65 billion) were attributable to foregone taxes and unemployment benefits, whereas healthcare spending, the focus of most cost-effectiveness analyses in the UK, represented ∼11% of government costs (£5–11 billion)Citation25. The results described here, and more generally applying a fiscal perspective to allocation decisions, can inform local or central government policy-makers about how changes in health will influence spending or savings in other public sectors.

The results described here are specific to the UK, where a generous welfare state exists, supported by high taxation rates. We might expect similar results would be obtained in other countries with comparable social welfare benefits, epidemiological profiles, and tax rates. However, without conducting the specific analyses for each country, caution should be exercised in directly translating these findings to other markets.

To translate spending on LDV/SOF into public benefit, we have generated two different cost ratios to demonstrate the fiscal benefit achievable in relation to each £1 of spending. The choice of which BCR to favor depends on whether one believes the analysis should consider only the treatment costs of the intervention arm (LDV/SOF in this case), giving a BCR of 1.8 or the incremental spending on healthcare costs, i.e. calculated by subtracting costs associated with “no treatment” from the costs associated with the intervention, giving a BCR of 5.4 in this case. In both scenarios, the fiscal benefit to government realized by treating patients with chronic HCV infection exceeds the fiscal costs.

These results underline the need to think more broadly about treatment benefits in relation to curative therapy costs—a sentiment also reflected by the World Health Organization consortium advocating the need to understand the broader economic benefits of vaccinationCitation26. Given that the government is the principal funder of healthcare in tax-based funding models such as in the UK, it is imperative that the full fiscal costs and benefits to the government of alternative treatment strategies are outlined. Benefits can also extend to private enterprise. Previous studies of large employers highlighted that HCV-infected individuals had increased sick leave, short- and long-term disability, and workers’ compensation with annual indirect costs of US$490 per yearCitation17. Furthermore, a recent survey of privately insured people with HCV noted the impact of the progressive nature of HCV infection on annual indirect work-loss, reporting costs for non-cirrhosis, compensated cirrhosis, and ESLD of US$742, US$1,449, and US$3,775 compared to non-HCV-infected individuals, respectivelyCitation27.

In the analysis described here we also illustrated the increased fiscal benefit that can be gained from early intervention (treating all patients, regardless of liver fibrosis level) vs delayed treatment (treatment commenced only when patients develop cirrhosis). Specifically, savings in healthcare (£24,747 per patient) and disability costs (£119,602 per patient) can be achieved from early intervention, with additional tax revenue gains of £49,829 per patient also achieved. Taking account of all fiscal costs and benefits, a strategy of treating all patients rather than waiting to commence treatment when patients become cirrhotic is estimated to deliver a fiscal benefit to government of ∼ £194,178 per patient (net) over the patient’s lifetime. While there may be pressure to delay therapy for non-cirrhotic people due to concerns around affordability, once disease progression occurs a proportion of people will progress to more severe health conditions, requiring more intensive healthcare interventions, with increased likelihood of withdrawing from the work force and relying increasingly on social benefits paid for by government.

Several weaknesses need to be taken into consideration in projecting lifetime fiscal costs and benefits, as described here. First, pension costs have not been included in our analysis. The reason being that survival from treatment will lead to higher pension costs in the long-term; however, we do not refuse lifesaving treatment of people simply because they will one day become pensioners. Furthermore, the analysis described here is based on long-term modeled projections, based on current tax rates, employment and wage rates, and growth. Small variations in any of these parameters could influence the conclusions described here. This was demonstrated by the sensitivity analysis, where changes to the discount rate could have an influence on the overall results. Additionally, the amount of social benefits being received by people with HCV that are unable to work was based on a small sample size. Improved estimates for government social benefits costs would enhance our understanding of fiscal transfer benefits received in more severe HCV states. The analysis described here does not take into consideration the potential heterogeneity of the HCV cohort. We have applied a simplifying assumption of an average HCV cohort based on reduced wage rates and estimates on government transfer benefits received. Variation in these parameters could influence the results. Furthermore, the results described here apply a monetization of health benefits from the government perspective, based on changes in tax receipts and transfer payments. Because no value is assigned to the value of being healthy from the perspective of individuals and society, the results here are an under-estimate of the overall benefits attributed to HCV elimination. Due to the absence of data on no treatment, the transition probabilities for no treatment are derived from the era of peginterferon therapy, which was the dominant strategy for many years. This would suggest that we over-estimate the effects of no treatment; hence, the incremental fiscal consequences described here are likely conservative.

Funding decisions in healthcare are multi-factorial, and are taken at many different levels of government and within health systems. The analysis described here is provocative, because it assigns future tax value to interventions made by public health systems. Even in those countries that aspire to allocate healthcare resources using welfare principles, there is a need for information regarding broader economic consequences of technologies, and there is a fiscal reality that needs to be taken into consideration. For example, the National Audit Office in the UK applies a similar method to the one described here for assessing healthcare interventions. Furthermore, in light of devolution of care in the UK, there is increasing demand from local policy-makers to understand the consequences of changes in health status.

Conclusions

Applying a fiscal analytic framework to investigate HCV infection and curative treatment reveals the cross-sectorial impact of the condition on different government sectors. We demonstrate here that investing in curative HCV therapy offers a broad range of fiscal benefits for government that includes reduced public spending on social programs and disability, and increases in tax revenue, in addition to cost savings within the health sector. Based on this fiscal analysis, governments can generate fiscal benefits between £1.8–£5.4 for every £1 invested in curative HCV treatment, with an estimated £4,324 increase in gross tax revenue. However, it is worth noting that the fiscal analysis described here remains an under-estimate of the broader economic benefits of curative HCV therapy, as it does not consider the broader macroeconomic consequences of people contributing to the economy and the likely multiplier effects achieved through improved health. We also illustrate that early intervention with curative HCV treatment (regardless of fibrosis level) offers improved economic gains compared to late intervention (treating only at F4 fibrosis), resulting in reduced disability and health spending, and generating important improvements in tax revenue. This means that the net financial/fiscal benefits to government of treating all patients with chronic HCV infection exceeds the net fiscal benefits achieved by waiting until patients become cirrhotic before commencing treatment.

Transparency

Declaration of funding

A research grant was provided by Gilead Sciences for conducting the work described here.

Declaration of financial/other relationships

The authors disclose no financial interests in the funding organization or the commercial operations of the funding organization. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Previous presentation

The results were presented in abstract form at the American Association for the Study of Liver Disease annual meeting in 2016.

Supplemental material

Supplementary_data.docx

Download MS Word (19 KB)

Acknowledgments

No assistance in the preparation of this article is to be declared.

References

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.