Cost-effectiveness analysis of a new paradigm to simplify testing, monitoring and treatment of hepatitis C virus in the United States

Abstract

The hepatitis C virus (HCV) testing and treatment pathway in the United States (US) includes a range of tests and appointments causing delays and loss to follow-up. We assessed the cost-effectiveness of simplifying the pathway using an economic model to estimate health outcomes, cost differences and incremental cost per quality-adjusted life year (QALY) and life year (LY) of the new paradigm compared to the other scenarios. The analysis compared three scenarios, one based on treatment guidelines, one based on real-world practice and a hypothetical scenario with a simplified pathway (“new paradigm”); these differed in testing and treatment process steps and times. The new paradigm resulted in cost reductions between $19,751 and $16,448, and excess QALYs between 0.42 and 0.70, suggesting that simplifying the US HCV patient pathway may be cost-effective and allows a quicker path to successful treatment and reduce the number of patients lost to follow-up.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12962-025-00622-y.

Background

HCV is a bloodborne virus which causes both acute and chronic hepatitis and may cause mild to severe illness lasting from a few weeks to a lifetime. Typically, acute HCV infections are asymptomatic, and in most cases, do not lead to life-threatening disease. Around a third (15–45%) will spontaneously clear the virus within approximately 6 months of infection, without treatment. The remaining 55–85% of persons will develop chronic infection. Untreated chronic HCV infection is associated with a risk of cirrhosis in the range of approximately 15–30%, within 20 years [1]. Each year, between 1% and 4% cirrhotic patients will develop hepatocellular carcinoma (HCC) [2]. An estimated 58 million individuals globally are chronically infected with CHC, with around 1.5 million new infections per year [1]. The World Health Organization (WHO) estimated that in 2019, approximately 290 000 people died worldwide from hepatitis C, most frequently from cirrhosis and hepatocellular carcinoma (liver cancer) [1].

In the United States, recent estimates published in 2023 from the National Health and Nutrition Examination Survey (NHANES) put the prevalence of HCV at 0.9%, or around 2.2 million people, with an estimated 1.4% prevalence in men compared to 0.5% in women in 2020 [3].

Recent years have seen a rise in HCV incidence and prevalence in the US, with approximately 107,000 new chronic HCV cases reported in 2021, equating to about 40 new cases per 100,000 people annually. In addition, there were around 5,000 new acute HCV cases, or 1.86 per 100,000 people. The CDC has noted a sharp increase in acute cases, doubling from 2014 to 2021 and rising by 5% from 2020 to 2021, partly due to a more accurate definition of acute cases introduced in 2020 [4]. However, the COVID-19 pandemic led to a significant decrease in HCV testing and treatment, exacerbating the impact on at-risk populations, including drug users, who faced increased drug use and reduced harm reduction services during social distancing [5–7].

Until recently, delivery of hepatitis C testing and treatment in many countries relied on specialist-led (usually by a hepatologist or gastroenterologist) care models in hospital settings to administer complex treatment. However, the wide adoption of short-course oral, curative pan-genotypic, direct acting antiviral (DAA) treatment regimens with few, if any, side-effects, means that testing, care, and treatment for persons with CHC can be provided by trained non-specialist doctors and nurses. Testing, care and treatment can also now be provided safely in primary care, harm reduction services and prisons where it is more accessible and convenient for patients [1]. Implementation of point-of-care testing in countries such as Australia and Canada has shown to be a successful strategy, one that may be cost-effective or even cost-saving in some at-risk populations [8, 9]. In line with WHO recommendations, the latest treatment guidelines in the US, published by the Infectious Diseases Society of America (IDSA) and the American Association for the Study of Liver Diseases (AASLD), now require less specialized care than before, enabled largely by the advent of pan-genotypic DAAs [10].

Studies conducted in recent years have shown the benefits of simplifying the treatment pathway of HCV treatment, with rapid treatment initiation and only minimal monitoring during treatment [11, 12]. Evidence also showed positive outcomes when treating recent drug users and acute HCV patients – two groups that traditionally have not been universally treated [13, 14]. Given the benefits of DAA treatment in acute patients, expanding treatment to this patient group is now considered an important part of HCV elimination, and thus are recommended by current HCV guidelines [10, 15].

These simplifications and the expansion of treatment have benefitted many HCV patients in the US. However, there remains a large proportion of patients whose patient journey still involves numerous process steps in the pathway associated with diagnosis and testing that they must go through before receiving treatment. This is important because early diagnosis and treatment can prevent both future complications in the infected individual as well as transmission of the virus.

Simplifying the pathway may also have implications to both costs and health outcomes in the HCV population by removing barriers in linkage to care which have been demonstrated to cause substantial loss to follow-up, especially in groups whose healthcare engagement levels are already low (i.e., young people) or otherwise limited (i.e., incarcerated people). The need for multiple appointments is also associated with delays and difficulties such as limited means of travel or lack of availability of appointments [16].

The objective of this analysis was to quantify the savings associated with treatment pathway simplification for the US HCV patient population. To do so, a previously developed economic model was adapted to the US setting, with a few modifications which will be discussed in the following Sect. [17].

Methods

The analysis utilized a series of linked Markov models to estimate the health outcomes and cost differences resulting from simplifying the treatment pathway for HCV patients in the US. The model was an adaptation of a previously published economic model, developed to estimate the efficiency gains resulting from two simplified pathways from diagnosis to treatment of chronic hepatitis C patients, from a societal perspective in Italy [17].

Patients entered the model following initial testing to confirm diagnosis, and only positive patients were included in the analysis. Patients were then followed through the diagnosis, treatment, and follow-up cascade to the point where cure was achieved. This cascade was based on current IDSA/AASLD treatment guidelines as was subsequently validated by US experts [10]. The cascade accounts for both primary care and specialist diagnosis, specialist disease staging/characterization, treatment approval, treatment initiation, and treatment follow-up.

The cascade varied by subgroup, given differences in access to care and proportion lost to follow-up (LTFU) in each step. In addition to the general US population, subgroups of interest considered in the model were people who inject drugs (PWID), men who have sex with men (MSM) and incarcerated people (INC). Patient characteristics at baseline, including proportion chronic and acute, HIV co-infection status, fibrosis stage and genotype distribution is presented in supplementary table S1.

The model structure is presented in Fig. 1.

Fig. 1.

Markov model structure. Fn = metavir stage; DCC = decompensate cirrhosis; EM = extra mortality; HCC = hepatocellular carcinoma; CHC = chronic hepatitis C virus; LT = liver transplant 1st year; PLT = liver transplant 2nd year+; SVR = sustained virologic response; Tx = treatment

In addition to chronic patients, acute patients were also considered in the base case analysis of the adapted model. This was implemented by first considering the proportion of acute patients out of the total population and applying a spontaneous clear rate. Patients whose disease cleared on their own did not enter the model. The remaining acute patients then entered the model and were either treated or followed the HCV natural history model trajectory, in the same way chronic patients did, while considering relevant patient characteristics such as their fibrosis stage and genotype. In the simplified pathway, acute patients were treated from the start, without any associated delay to establish whether spontaneous clearance was achieved. The modelling of acute patients is presented in Fig. 2.

Fig. 2.

Modelling of acute HCV patients. HCV = hepatitis C virus

In the model, three testing and treatment paradigms were compared. These paradigms differed in the testing and treatment process steps patients are required to go through, the time it takes for these steps to be completed, and the proportion of patients lost to follow-up during the entire pathway, from initial diagnosis through treatment completion. The three paradigms can be summarized as follows:

1. Status Quo (SQ): Testing and monitoring based on current IDSA/AASLD guidelines. Most patients require a burdensome array of tests and appointments before starting treatment. Many are lost to follow-up.

2. Real-World (RW): A simplified pathway for many patients, based on best real-world practice in the US. Fewer patients are lost to follow-up.

3. New Paradigm (NP): A hypothetical pathway where patients do not undergo testing and start treatment at first point of contact. The analysis assumes no patients are lost to follow-up.

An overview of differences in the three scenarios compared is presented in Table 1. Detailed descriptions of treatment step and LTFU inputs are presented in the supplementary materials.

Table 1.

Overview of the three scenarios compared in the model

Process step		Utilization by scenario			Process time in weeks
Diagnosis in primary care		SQ	RW	NP	SQ	RW	NP
1	Initial appointment to request anti-HCV and other tests	ALL	SOME	NONE	16	0	0
2	Patient performs anti-HCV in a local lab
3	Appointment to present positive anti-HCV test result
4	Patient undergoes confirmatory RNA tests
5	Referral to specialist/CoE (if ineligible for simplified treatment)
6	Waiting for specialist appointment
Diagnosis in specialist care or CoE (hospital context)		SQ	RW	NP	SQ	RW	NP
7	First specialist appointment. Additional tests prescribed	ALL	SOME	NONE	7	4	0
8	Patient undergoes laboratory tests
9	Second specialist consultation and prescription of additional tests, e.g., FibroScan
10	Patient undergoes FibroScan and/or biopsy
11	Waiting time for treatment approval (if any)
12	Patient undergoes virologic and microbiologic laboratory tests while on waiting list (if any).
13	Patient has regular appointments with specialist while on waiting list (if any).
Treatment		SQ	RW	NP	SQ	RW	NP
14	Treatment initiation appointment	ALL	SOME	NONE	25	12	12
15	In person / telehealth visit as needed.
16	Blood tests for cirrhotic patients.
17	Adverse event assessment for pts on treatment that includes ribavirin.
18	Patient performs lab tests at week 12 or later
Post-treatment follow-up		SQ	RW	NP	SQ	RW	NP
19	SVR: Ultrasound, endoscopic surveillance every 6 months	ALL	SOME	NONE
20	No SVR: evaluation for treatment by specialist
21	No SVR: disease progression assessment every 6 to 12 months

CoE = center of excellence; GP = general practitioner; HBV = hepatitis B virus; HCV = hepatitis C virus; NP = new paradigm; RNA = ribonucleic acid; SVR = sustained virologic response; SQ = status quo

The analysis adopted a US third-party payer perspective, with only direct costs considered. In the base case, patients were modelled through a lifetime horizon. Both health effects and costs were discounted at a 3% rate per year, with half cycle correction applied. Base case settings are summarized in Table 2.

Table 2.

Base case settings

Setting	Base case
Population
Selected population	All populations of interest combined (general population, PWID, incarcerated, MSM)
Total starting patient population	120,000
Age at model entry	49.5
Proportion with acute HCV	4.3%
Model setting
Time horizon	Lifetime
Discount rate – costs and health effects	3.0%

HCV = hepatitis C virus; MSM = men who have sex with men; PWID = people who inject drugs

Key model drivers and the robustness of results were tested through deterministic (DSA) and probabilistic sensitivity analyses (PSA). To further test the impact of key model inputs and assumptions, a number of alternative scenarios were also explored including alternative time horizons and discount rates, selected populations, alternative lost to follow-up and process time assumptions, and different inputs related to proportion of treated acute patients. A detailed description of sensitivity and scenario analyses are presented in the supplementary materials.

Health outcomes included total LYs and QALYs, as well as the number of advanced liver outcomes (compensated cirrhosis, decompensated cirrhosis, hepatocellular carcinoma, liver transplant, extra hepatic mortality). Cost outcomes included total costs and costs associated with testing, treatment, and disease management separately. The new paradigm was compared to the other two paradigms to derive incremental LYs, QALYs, costs, as well as its net mean benefit.

Model parameters

General model parameters included population inputs, testing and treatment process steps as well as times and number of patients lost to follow-up. Clinical inputs included treatment effectiveness, transition probabilities between health states, and health-related quality of life. Costs considered in the model included the cost of treatment, costs associated with each health state and cost of testing and follow-up. A detailed description of each of these model parameters is included within the supplementary materials.

Results

In the base case, the new paradigm was associated with the highest LY and QALY outcomes out of all three compared. Over a lifetime horizon, the NP was associated with discounted LYs of 19.87 vs. 18.43 in the SQ and 18.85 in the RW scenarios, respectively. Discounted QALYs were 16.34 in the NP vs. 15.02 and 15.35 in the SQ and RW scenarios, respectively. The NP also had the lowest overall costs of the scenarios, with a discounted total of $58,567 compared to $79,561 (SQ) and $77,428 (RW).

Given its lower costs and higher outcomes, the base case analysis showed the new paradigm of a simplified testing and treatment paradigm was a dominant strategy compared to both other scenarios. Aggregated base case results are presented in Table 3.

Table 3.

Base case aggregated results

	Status QUO	Real-world	New paradigm
Outcomes
LYs per person	18.43	18.85	19.87
QALYs per person	15.02	15.35	16.34
Costs
Total costs per person ($)	79,561	77,428	58,567
Cost-effectiveness – new paradigm vs. status quo
Incremental cost per LY ($)	DOMINANT
Incremental cost per QALY ($)	DOMINANT
Net mean benefit @ 100,000 WTP ($)	152,951
Cost-effectiveness – new paradigm vs. real-world
Incremental cost per LY ($)	DOMINANT
Incremental cost per QALY ($)	DOMINANT
Net mean benefit @ 100,000 WTP ($)	117,849

HCV: hepatitis C; QALY: quality adjusted life year; LY: life year; WTP; willingness to pay

Base case results

Base case results also pointed to a substantial reduction in advanced liver disease outcomes in the new paradigm. There was a large reduction in compensated and decompensated cirrhosis outcomes compared to both scenarios, and cases of other advanced liver outcomes decreased as well over the base case lifetime horizon. Overall, advanced liver disease outcomes decreased by 81.6% compared to the SQ and 80.3% compared to the RW scenarios, respectively. A breakdown of all advanced liver outcomes over patients expected lifetime horizon (from 49.5 to 100 years of age in this base case) is presented below in Table 4.

Table 4.

Base case advanced liver disease outcomes

	Status quo	Real-world	New paradigm
Advanced liver diseases
CC	14,082	13,636	659
DCC	3,106	2,471	107
HCC	2,996	2,933	2,139
LT	182	151	82
EM	1,638	1,394	1,061
Advanced outcomes – new paradigm vs. status quo
Difference in Total Advanced Liver Outcomes, n (%)	-17,956 (-81.6%)
Advanced outcomes – new paradigm vs. real-world
Difference in Total Advanced Liver Outcomes, n (%)	-16,536 (-80.3%)

CC = compensated cirrhosis; DCC = decompensate cirrhosis; EM = extra mortality; HCC = hepatocellular carcinoma; LT = liver transplant

In addition to the health outcomes described so far, overall costs also decreased with the new paradigm. Given the simplification of the pathway, testing costs were assumed to be zero for the new paradigm, with higher rates of patients being treated leading to an increase in treatment costs compared to the other scenarios. However, this increase was offset by the reduction of disease management costs. A breakdown of discounted total costs per person is presented in Table 5.

Table 5.

Base case cost outcomes

	Status quo	Real-world	New paradigm
Total costs per person ($), discounted
Total	79,561	77,428	58,567
Testing	1,465	1,049	0
Treatment	19,796	19,671	25,712
Disease Management Total	58,300	56,708	32,855
NC	2,687	973	793
CC	28,018	29,029	30,297
ESLD	27,595	26,707	1,765

ESLD = end-stage liver disease; NC = non-cirrhotic patients; CC = compensated cirrhosis

Sensitivity analyses

Deterministic and probabilistic sensitivity analysis, alternative scenarios

Deterministic sensitivity analysis suggested key model drivers were health state utilities and rates of sustained response. Probabilistic sensitivity analysis confirmed the robustness of results as all iterations gave incremental results with higher QALYs and lower costs. A detailed description of the DSA and PSA and their results are presented in the supplementary materials.

Scenario analysis results are presented in the supplementary materials. Throughout all scenarios explored, the new paradigm remained a dominant or cost-effective strategy compared to both comparator scenarios.

Discussion

Our analysis showed that simplifying the US HCV testing and treatment pathway would be associated with positive outcomes compared to current practice, both in terms of reduced costs and lower incidence of advanced liver disease cases.

It is worth highlighting once again that the new paradigm proposed in the current analysis represents a hypothetical best-case scenario of what can be achieved by reducing the time and steps needed in a typical HCV patient’s testing and treatment pathway in the US. Notably, the model relies on the assumption that the number of patients lost to follow-up can be reduced in the simplified pathway; this difference between the scenarios is a key driver of cost-savings and improved health outcomes. As such, the model is not intended to represent expected real-world gains; rather, it is an aspirational scenario to demonstrate the magnitude of benefits of simplifying the pathway.

Such simplifications are needed because, despite the availability of highly effective pangenotypic DAAs, many patients in the US and globally do not get the timely treatment they need in the absence of appropriate testing and linkage to care [18]. As a result of time and logistical barriers, some patients will have substandard rates of engagement, resulting in fewer patients achieving cure than the treatment would otherwise allow [18, 19]. Updated testing and treatment guidelines, such as the CDC recommending that testing should be performed at a single visit, represent welcome change but barriers remain [20].

Our study aligns with evidence published in recent years demonstrating the potential benefits of a simplified HCV pathway, both in the US and globally [16, 21]. The clinical benefits of such a pathway have been recently demonstrated in the phase 4 MINMON trial that enrolled 400 patients from a diverse population that included Brazil, South Africa, Thailand, Uganda, and the US [12]. In the trial, a minimal monitoring approach was utilized that consisted of the following: (1) no pre-treatment genotyping; (2) entire treatment course dispensed at entry; (3) no scheduled visits or laboratory monitoring; and (4) two points of remote contact (one at week 4 for adherence, and one at week 22, to schedule week 24 outcome assessment). Results showed that patients achieved comparable cure rates with similar safety than those who follow a standard treatment and monitoring course. The fact that these findings were consistent across the diverse population of the trial suggests that a simplified pathway is beneficial in low, middle, and high-income countries alike [12]. Loss-to-follow-up was also very low among patients, at less than 1% ¹².

In line with other recently published evidence, findings from our analysis also suggested that there are potential gains in health outcomes and costs when including certain patient groups in HCV treatment that may have been traditionally untreated or were at higher risk of being lost to follow-up. These include acute patients and PWIDs with recent drug use.

In the REACT study published in 2021, patients with recently acquired HCV were treated with either a short (6 weeks) or standard (12 weeks) course of sofosbuvir/velpatasvir [14]. While the short course treatment was not found to be as effective as the standard course, results from the standard 12-weeks arm clearly demonstrated that early treatment of infection is safe, feasible and effective [14]. Loss-to-follow-up was also low, at 3.2% and 5.3% on the short and standard arms, respectively [14].

Recent drug use has been traditionally a reason for excluding HCV patients from treatment. However, evidence suggests treating this patient group is also effective. In the international open label SIMPLIFY trial, 103 recent drug users were treated with a standard 12-week course of sofosbuvir/velpatasvir [13]. Results suggested that drug use prior or during treatment did not impact SVR rates [13].

Our results were also broadly comparable to the previously published Italian analysis which served as a basis for our model [17]. Results from that analysis suggested a reduction of total costs by 40.7%, in five years, with a best case of increased testing and treatment paradigm compared to the status quo. Advanced liver disease outcomes were 76.80% lower in the new paradigm of treatment, on average, than in the status quo [17].

Our study was associated with some limitations. Firstly, while early treatment of HCV has been shown to reduce disease transmission; this aspect of the disease was not considered in the model, potentially underestimating the beneficial effect of the simplified HCV pathway. Further studies that incorporate disease transmission may aid in understanding the additional benefits of early treatment in this context.

Additionally, even though the model aims to represent the current pathway in the US, some of the key inputs and assumptions are based on expert opinion rather than published data, introducing uncertainty to model inputs. However, our sensitivity and scenario analyses attempted to bound this uncertainty and suggest clearly that an improved patient pathway is likely cost-effective.

Finally, our model was not designed to account for the impacts of the COVID-19 pandemic on HCV testing and treatment, which, as previously described, were substantial. As the pandemic impacted the number of people being diagnosed and treated, the current treatment pathway may be further from ideal than what we assumed in our analysis – potentially causing it to underestimate the benefits of a simpler pathway.

In conclusion, our analysis demonstrated that substantial gains in health outcomes and cost reductions could be achieved by simplifying the HCV testing and treatment pathway in the US, in line with recent published evidence supporting the streamlining of the pathway. Enabling patients to receive treatment faster and with less potential for loss to follow-up may be a crucial part of HCV elimination efforts.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

RB and AIN developed the model with review from all authors. RB and AIN wrote the main manuscript text, prepared tables and figures and all supplementary information. All authors reviewed and approved the manuscript.

Funding

This study was funded by Gilead Sciences.

Data availability

All secondary data generated or analysed during this study are included in the published article or its supplementary materials. The economic model may be requested from the authors.

Declarations

Ethical approval

Not applicable.

Competing interests

DD has received consulting and speaking fees from Gilead Sciences. NR has received research funding and consulting fees from Gilead Sciences, AbbVie and Abbott. AA has received advisory and consulting fees from Gilead Sciences. AIN and RB are employees of Maple Health Group, LLC, which provides consulting services to Gilead Sciences.