Comparison of Markov Model and Dynamic Model in Economic Evaluations for Infectious Diseases: A case of Bictegravir/emtricitabine/tenofovir Treatment-native adults of HIV-1 in China

doi:10.21203/rs.3.rs-3934183/v1

Download PDF

Research Article

Comparison of Markov Model and Dynamic Model in Economic Evaluations for Infectious Diseases: A case of Bictegravir/emtricitabine/tenofovir Treatment-native adults of HIV-1 in China

https://doi.org/10.21203/rs.3.rs-3934183/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: This study aimed to compare the performances between Markov model and dynamic model in economic evaluations for antiretroviral therapy (ART) of HIV, using a case of bictegravir/emtricitabine/tenofovir (B/F/TAF) for treatment-native adults of HIV-1 infection in China.

Methods: A Markov model was used to simulate in detail the lifetime treatment of HIV among adult patients with ART with a cycle of one month. A dynamic model was used to consider the effects of ART on preventing transmission among all Chinese adults, with a time frame of 30 years and a cycleof one year. The primary outcomes were total costs, quality-adjusted life years (QALYs), and incremental cost-effectiveness ratio (ICER). Sensitivity analyses were conducted for result validation, and the model precision was tested using relative standard deviation (RSD).

Results: In Markov model and dynamic model, compared with dolutegravir/lamivudine (DTG/3TC), B/F/TAF incurred higher per-person expenses ($44,381.33 and $30.94 versus $42,160.13 and $29.28) but yielded superior QALYs (12.7788 and 17.9423 vs 12.6310 and 17.9420), resulting in higher ICER for Markov model compared to dynamic model (14,081.23 vs 6,524.03 USD/QALY). The robustness of the results was confirmed in uncertainty analyses, and Markov model exhibited a lower RSD.

Conclusion: B/F/TAF is deemed cost-effective in China according to both models, presenting a valuable treatment option despite its higher price in the era of optimized HIV care. The choice of the economic evaluation model influences the ICERs of ART, with dynamic model advantageous for incorporating externality and Markov model noted for its precision.

HIV

Markov

Dynamic Model

Bictegravir/emtricitabine/tenofovir

China

The gravity of the HIV epidemic in China persisted unabated until 2020, with a reported 1.05 million individuals (prevalence: 0.08%) afflicted by HIV, as disclosed by National Center for AIDS/STD Control and Prevention[1]. The advent of ART substantially curtailed the HIV-related mortality rate, aligning it closely with that of common chronic ailments[1]. The financial commitment from the Chinese government for HIV/AIDS antiviral treatment grew from $24 million in 2004 to a substantial $550 million in 2020[2]. This emphasizes the urgent need to carefully choose antiretroviral drugs that are both effective and cost-effective to strengthen the overall management of HIV.

The Markov model is widely used in economic evaluations of medicines, and ART drugs are no exception. In international studies, Markov models can simulate inevitable multi-line treatments in lifelong ART, along with the occurrence of clinical events such as tumors or opportunistic infections[3–5]. In China, economic evaluations of ART are relatively limited, with exploration being rather superficial. For instance, Yogesh's (2019) study focused on a short-term five-year horizon[6], and Min Li's (2023) lifelong simulation considered clinical events but still did not account for multi-line treatments[7]. Based on international research, Markov models could be more maturely employed for the economic assessment of ART in China. However, ART had the function of reducing transmission risks[8], surpassing the simulation capability of Markov models.

Therefore, there arises a need for a dynamic model to simulate HIV transmission and the impact of ART on transmission risk. HIV, as an infectious disease with external effects, not only affects the health and economy of infected individuals but also has repercussions on those around them[9]. Currently, economic evaluations using dynamic models for HIV interventions mainly focus on preventive measures, including screening, risk behavior interventions, and comprehensive strategies for prevention and treatment[10–12]. However, dynamic models for the economic evaluation of ART have not yet been developed. In this study, we aim to use both Markov and dynamic models for an economic evaluation of a chosen ART, comparing their performances and results disparities as a case study to support theoretical research in Pharmacoeconomics.

Guidelines from esteemed entities such as World Health Organization (WHO), Chinese Guidelines for Diagnosis and Treatment HIV/AIDS (2021), and U.S. Department of Health and Human Services (DHHS) all uniformly recommend a regimen comprising two nucleoside reverse transcriptase inhibitors (NRTIs) and a non-NRTI drug (core agent) for native treatment[13–15]. Among the core agents selected by ART, integrase inhibitors (INSTIs) are optimized in terms of high efficiency, low toxicity, and drug resistance compared with the previous non-nucleoside reverse transcriptase inhibitors (NNRTIs) and protease inhibitors (PIs)[16]. Bictegravir (BIC), an INSTI presented solely in the form of a fixed-dose compound (B/F/TAF), includes the other components Emtricitabine (FTC) and Tenofovir Alafenamide (TAF). B/F/TAF has emerged as the first-line favored primary treatment for HIV in numerous nations owing to its commendable viral inhibition rate, safety profile, and resistance characteristics[13–15, 17]. In China, four INSTIs—raltegravir (RAL), Elvitegravir (EVG), DTG, and BIC—are currently enlisted, with RAL and EVG being first-generation, and DTG and BIC classified as second-generation. BIC is particularly noteworthy for its enhanced antiviral efficacy against non-B subtypes of HIV-1, outperforming both RAL and EVG in this regard. Despite the similar in vitro inhibitory potency of BIC and DTG, BIC demonstrates structural robustness, leading to sustained effectiveness and decreased vulnerability to drug interactions[17]. Clinical trials have proved that B/F/TAF's efficacy in the initial treatment of HIV-1 is on par with the prevailing DTG-based ART regimen, with superior safety and resistance profiles and an enhanced patient medication experience[18, 19]. DTG/3TC happens to be the only fixed-dose compound containing DTG among the first-line preferred options in the Chinese guidelines[13]. Consequently, this study assessed the cost-effectiveness of B/F/TAF compared to DTG/3TC in China for treatment-naive adults with HIV-1, utilizing Markov and dynamic models to aid in clinical drug selection.

2.1 Study design

A cost-effectiveness evaluation was carried out to evaluate the two first-line strategies (B/FTAF versus DTG/3TC) from a social perspective. The primary outcomes were Quality-Adjusted Life Years (QALYs), total cost, and Incremental Cost-Effectiveness Ratio (ICER). The threshold of willingness to pay is 1–3 times per capita Gross Domestic Product (GDP), while China's per capita GDP in 2022 is $12741.11[20].

We developed two models to simulate the changes in the effectiveness and cost of ART for HIV. Firstly, a Markov model was employed to scrutinize the treatment process of HIV patients. Subsequently, on the foundation of clinical parameters provided by Markov, a dynamic model capable of simulating HIV transmission for Chinese adults has been established by incorporating population and transmission-related parameters[21]. According to the Chinese HIV treatment guidelines, both the control and intervention groups are eligible to switch to subsequent ART after the failure of first-line treatment[13]. In this study, the switch to treatment settings for both groups is the same. In China, since 2009, protease inhibitors, particularly lopinavir (LPV), have been used as antiviral therapy after first-line treatment failure[22]. Another protease inhibitor, darunavir (DRV), was used as the core agent in third-line treatment regimens in this study. The details of these models are described below.

2.2 Markov model

2.2.1 Model Structure and Assumptions

According to the data from China Disease Control and Prevention from 2011 to 2019, adults accounted for 96% of new HIV patients[23]. Therefore, we selected adult HIV patients(≥ 18 years of age) as the research subjects for this model. The model simulated the lifespan of HIV patients, adopting a one-month cycle aligned with the drug cycle[4].

The model consisted of five health states based on CD4 cell counts (state 5: more than 500 cells/mm³, state 4: 350–500 cells/ mm³, state 3: 350 − 200 cells/ mm³, state 2: 100–200 cells/mm³, and state 1: less than 100 cells/ mm³), which were derived from the immunological classification of HIV/AIDS by WHO and published literature[4, 24]. Further classification included native treatment, discontinuation in native treatment, switched treatment, and discontinuation in switched treatment. Infected persons in each state were subsequently divided into 20 categories, as illustrated in Fig. 1a. Moving to an adjacent state or remaining in the same state within one cycle depended on whether drug discontinuation occurred and changed in CD4 counts. In the event of drug discontinuation failed viral suppression, the CD4 levels in some patients decreased. Conversely, in cases where no drug discontinuation occurred, CD4 levels increased in some patients.

Assumptions were made in this model. Patients in the B/F/TAF group started with B/F/TAF as the preferred first-line treatment, transitioning to an alternative DTG/3TC group if virus suppression isn't achieved. Similarly, those in the DTG/3TC group began with DTG/3TC, opting for an alternative B/F/TAF regimen[4].

2.2.2 Model parameters

Clinical parameters

The cohort was set up with 100,000 HIV patients (≥ 18 years of age). Some Chinese studies have found that the distribution of CD4 cell counts in newly diagnosed HIV-positive patients is uniform at the boundaries of 500, 350, and 200[25–28]. Therefore, the starting five health states (from higher to lower CD4 counts) for the patient cohort were distributed assuming CD4 counts of 25%, 25%, 25%, 12.5%, and 12.5%.

ART efficacy was measured by viral suppression rate (< 50 copies/mL) and immune response (average CD4 increase). These efficacy parameters for B/F/TAF were derived from the GS-US-380-1489 trial[18, 29], and for DTG/TAF, it originated from the GEMINI-1 (NCT02831673) and GEMINI-2 (NCT02831764) trials[30, 31], with data reported for both 0–48 weeks and 48–96 weeks. To facilitate the implementation of this study, efficacy parameters for second-line or third-line treatments were only reported for the 0–96 weeks. The pertinent data are meticulously presented in Table 1. A subtle transition in CD4 count was anticipated after early ART-driven CD4 rise [4]. Consequently, it was assumed that CD4 count remained unchanged after 96 weeks of all ART regimens.

Discontinuation of ART was classified as virological (drug resistance) and non-virological (mainly adverse reactions) reasons. The discontinuation rates for each ART regimen were derived from relevant literature, including trial data[4, 5, 18, 29–33], with specific details presented in Table 1. During the period of discontinuation, the CD4 cell counts of HIV patients decreased by 4.3 cells/mm³ each month[34].

Mortality included age-varying natural mortality and HIV-related mortality. HIV patients with lower CD4 count levels faced higher mortality rates. These data were obtained or calculated based on the literature and are presented in detail in Table 1[35–37].

Additional clinical events comprised AIDS-defining events (ADEs) and adverse reactions (AEs) associated with various ART regimens, incurring disutility and cost. Specifically, there are 27 types of ADEs, including inflammation, infection, tumors, and cancers. Diverse statuses correlated with distinct rates of ADEs[38–40]. Most ART drugs manifested a diverse array of AEs, including anemia, nausea, gastrointestinal effects, and others. The incidence of adverse reactions for each ART regimen in this study was derived from relevant trials and other literature, as shown in Table 1[4, 41–43].

Costs

As this study adopted a societal perspective, costs were incorporated into direct medical costs, direct non-medical costs, and indirect costs. All costs inflated to 2022 values and are calculated based on the average exchange rate in 2022 (1 USD = 6.7261 RMB) [20].

Direct medical costs were considered in the following: (1) expenditures on ART regimens, (2) examination costs, (3) costs associated with adverse reactions to ART regimens, and (4) administrative expenses for ADEs. Medicine prices were sourced from China Pharmaceutical Information Database[44]. The second-line cost included the prices of the individual components LPV/r, TDF (tenofovir disoproxil fumarate), and 3TC (lamivudine). The third-line cost comprised the prices of the combination DRV/r, DTG, TDF, and 3TC. Examination costs, involving both HIV and CD4 cell examinations, were derived from official prices provided by Jiangsu Province (in China)[45]. Costs related to ADEs and AEs treatment were meticulously calculated based on the likelihood of clinical manifestation. The weighted average cost per month for each occurrence of ADEs, considering the proportional representation and individual cost of each ADE type, was determined to be $7272.00[39, 46, 47]. Additionally, the monthly costs per occurrence of AEs for each ART regimen were computed following a similar methodology[18, 31, 41, 47–50]. The detailed classification and costs of ADEs and AEs were conveniently presented in Table 1, with further elaboration available in Table S2 and Table S3. The direct non-medical expenses comprised four categories, taking into account transportation costs for patients and their relatives traveling to outpatient or inpatient clinics. These costs were gleaned from an empirical study that involved patient questionnaires and expert interviews[51]. Furthermore, indirect costs were considered productivity losses and associated expenses using relevant Chinese data[6].

Utilities

HIV patients exhibited varying utility levels across different health states based on CD4 count[4], with utility decreasing as the CD4 count declined. The occurrence of AEs[52] and ADEs would reduce utility[39, 46, 47]. These data were shown in Table 1.

Discount rate

Both costs and outcomes were discounted annually at a rate of 5%, in accordance with China Guidelines for Pharmacoeconomic Evaluation[53].

2.2.3 Model verification

Clinical experts meticulously reviewed the model's accuracy, thoroughly assessing the reliability of the underlying algorithm for a comprehensive evaluation.

2.3 Dynamic model

2.3.1 Model Structure and Assumptions

We developed a dynamic model to simulate HIV transmission among all adults in China, stratifying them into five distinct subgroups based on their risk of transmission: normal males, normal females, men who have sex with men (MSM), male injecting drug users (MIDU), and female injecting drug users (FIDU)[13]. The research time frame was 30 years, spanning from 2015 to 2044, enabling a comprehensive examination of the outcomes of ART interventions. We adopted a one-year cycle aligned with the behavior cycle for the simulation.

The foundation for constructing this dynamic model relied on the classic Susceptible-Infectious-Removed (SIR) model commonly employed in infectious disease research and published literature[10]. The SIR model considered that susceptible individuals could become infected with the virus from HIV patients, with different risks of infection among various subgroups. Specifically, it was assumed that normal males and females could only be infected through heterosexual intercourse, MSM through homosexual behavior, and injecting drug users through both heterosexual and bloodborne transmissions. Therefore, moving to an adjacent state or remaining in the same state within one cycle depended not only on whether drug discontinuation occurred and changed in CD4 count but also on the impact of ART treatment on infection risk. Some HIV patients were diagnosed and received antiviral treatment, which reduced new incidence due to the decreased risk of transmission. However, in cases where HIV suppression failed due to drug discontinuation, the risk of transmission did not decrease. The dynamic model simulating the progression of HIV disease and HIV transmission is illustrated in Fig. 1b.

2.3.2 Model parameters

Demographic and transmission-related parameters

We have collated the total number of individuals and the total number of infected individuals for the five subgroups from 2015 to 2019. These data were sourced from China Statistical Yearbook[54], the international homoerotic association[55], National Center For AIDS/STD Control And Prevention[23, 56–63], China Anti-Drug Network, and other relevant sources[64, 65]. The data from the year 2015, serving as the initial values, is presented in Table 2. The infection coefficient, denoted as β, represented the ratio of the new incidence rate to the infection rate in the model. The non-linear least square method was used to fit the β of each subgroup with the data from these 5 years.

Some studies have confirmed that patients almost did not have the risk of transmission when treated with ART to control their HIV viral load to a lower level. The study set a 100% transmission risk reduction rate for patients effectively suppressing HIV. This decision was based on using 50 copies/mL as the standard for effective virus suppression in this work, given that it represented a very low viral load level[8]. This study considered the probability of examination and treatment to be 74.93%, based on data from Guangxi Province (in China) in 2021[66].

Moreover, a natural growth rate of 1.1963% was factored into this study[54]. The natural mortality rate of each subgroup was considered as follows: normal males, normal females, and MSM are normal populations, and the average natural mortality rate from 2015 to 2019 was 0.7068%[54]. For injecting drug users, MIDU was 7.18%, and FIDU was 5.48%[67].

Other parameters

The clinical parameters, costs, utilities, and discount rate required for the dynamic model were based on the data from the Markov model. These data were converted to yearly units when utilized in the dynamic model.

2.3.3 Model verification

The HIV infection numbers for the five subgroups and the overall HIV infection numbers from 2015 to 2019 were predicted by a dynamic model in this study under the assumption of no interventions. The model's accuracy was rigorously confirmed by comparing predicted data to actual epidemic figures using the coefficient of determination, R². Six datasets (Fig. S2) demonstrated strong congruence, with R² values remarkably close to unity (0.9078 to 0.9998).

2.4 Sensitivity analyses

The stability of the model was assessed through one-way sensitivity analysis (OWSA) and probability sensitivity analysis (PSA) on all parameters. The parameter ranges for conducting OWSA were set at ± 5% or ± 10% of the baseline values. Probability parameters and cost parameters for PSA were set to follow beta and gamma distributions. Subsequently, PSA was performed using 10,000 Monte Carlo simulations to illustrate the distribution of the results. Additionally, a cost-effectiveness acceptability curve was derived to represent the proportion of simulations in which the ART regimen B/F/TAF was deemed cost-effective relative to its comparator.

3.1 Base case

The base-case analysis of the Markov model showed that, compared to the DTG/3TC group, the B/F/TAF group gained 0.1577 extra QALYs for an extra cost of $2,221.21, resulting in an ICER of $14,081.23 per QALY gained per patient (1.11 times the GDP). However, the base-case analysis of the dynamic model showed that the B/F/TAF group was associated with a 30-year cost of $30.94 and expected QALYs of 17.9423, while the DTG/3TC group incurred a 30-year cost of $29.28 and expected QALYs of 17.9420 per Chinese adult. Resultantly, this led to an ICER of $6,524.03 per QALY gained per Chinese adult (0.50 times the GDP). The ICER of the dynamic model was lower than that of the Markov model (Table 3).

The cost of medications accounted for over half of the total expenditures, while costs associated with examinations, ADEs, AEs, direct non-medical costs, and indirect costs were relatively minor and showed similar proportions across the two model groups (Table 3). In both models, the B/F/TAF group significantly alleviated the health burden. In the Markov model, compared to the DTG/3TC group, the B/F/TAF group collectively reduced the occurrences of 622 HIV-related deaths and 828 ADEs. In the dynamic model, compared to the DTG/3TC group, the B/F/TAF group reduced the occurrences of 57,532 new HIV infections (five subgroups from 2015 to 2044 are detailed in Fig. S3), 28,296 HIV-related deaths, and 65,946 ADEs.

3.2 Sensitive analyses

In OWSA, we found that the ART effect (viral inhibition rate and immune response) and cost of B/F/TAF and DTG/3TC were the parameters that most influenced the results (Fig. 2a, b). However, the fluctuations of all outcomes from OWSA fell within the acceptable range of WTP.

PSA indicated a high degree of confidence in the results. At a willingness-to-pay threshold of 3 times the GDP, the probabilities of cost-effectiveness for B/F/TAF in the Markov model and dynamic model were 95.33% and 94.16%, respectively (Fig. 2c, d). The calculation of 10,000 ICERs from PSA revealed RSDs (Relative Standard Deviations) of 16.6 for the Markov model and 26.81 for the dynamic model. This suggests that in the simulation, the Markov model yields relatively stable results compared to the dynamic model.

Through the examination of the cost-effectiveness acceptability curves, it was determined that when the willingness-to-pay threshold was set at the per capita GDP, the probabilities of B/F/TAF being considered acceptable were 41.89% in the Markov model and 77.95% in the dynamic model, respectively. If the threshold was set at double the per capita GDP, these probabilities reached 87.08% and 91.21% in the Markov model and dynamic model, respectively (Fig. 2e).

B/F/TAF has been included as first-line HIV treatment guidelines in China, the United States, and WHO due to its outstanding clinical efficacy. However, its relatively higher cost has limited its widespread clinical application. Therefore, a pharmacoeconomic evaluation is necessary to determine if the higher medication cost leads to greater health benefits. This study aims to evaluate the cost-effectiveness of B/F/TAF in comparison to DTG/3TC for treatment-native adults of HIV-1 Infection in China, utilizing Markov and dynamic models. When compared to DTG/3TC, the ICER of B/F/TAF was $14,081.23 and $6,524.03 per QALY in the Markov model and dynamic model, respectively. We attempted to analyze the reasons supporting the cost-effectiveness of B/F/TAF. Firstly, B/F/TAF demonstrated higher clinical effectiveness than DTG/3TC, resulting in a reduction in ADEs occurrence, HIV-related deaths, and new HIV cases. These gained extended beyond individual health improvements to encompass savings in direct non-medical and indirect costs, coupled with a decrease in the likelihood of transmission. Secondly, as HIV is a chronic infectious disease, its treatment costs are intricate, with antiretroviral drugs constituting only around 50% of the overall costs, and first-line drugs comprising merely 20%. Consequently, the price advantage of DTG/3TC lacked significant clinical and macro-application impact.

The results from the two models showed a significant disparity in the ICER, with the dynamic model yielding more economically favorable outcomes for the B/F/TAF group compared to the Markov model. This difference was attributed to the dynamic model's enhanced consideration of externalities, explained specifically as follows. Firstly, the B/F/TAF group had fewer newly diagnosed HIV cases compared to the DTG/3TC group. Secondly, by reducing the transmission risk for a greater number of individuals infected with HIV, the B/F/TAF group indirectly protected the susceptible population. Therefore, compared to the Markov model, the dynamic model took into account more health benefits and cost savings associated with avoiding transmission risks, resulting in a smaller ICER. When economically evaluating ART with slight efficacy differences, like B/F/TAF and DTG/3TC, the dynamic model tends to magnify outcomes as opposed to the Markov model.

The two models exhibited fundamental differences, particularly in their roles within economic evaluation and their emphasis on algorithmic aspects. The Markov model simulated the details of ART for treating HIV, including switching treatment plans. In clinical drug selection, patient demands took precedence, making a Markov model based on patient cohorts more aligned with actual needs. In contrast, the dynamic model took into account transmission factors and assessed the preventive effects of ART, offering improved guidance for HIV prevention and control. A dynamic model, which thoroughly incorporates both treatment and transmission risk prevention, exhibits greater theoretical robustness in the economic assessment of various ART options. From an algorithmic perspective, the dynamic model, in addition to incorporating clinical, cost, and utility parameters from Markov, required population parameters and transmission-related parameters for five subgroups. This also precisely explains why the dynamic model has lower precision compared to the Markov model, resulting from diverse parameter types and more complex algorithmic processes. While the Markov model carefully simulated the course of ART treatment for HIV, this aspect of the dynamic model was simpler. Specifically, the Markov model carefully considered the distinct effects and costs of second and third-line treatments, while the dynamic model used average data of them. Moreover, regarding AEs, the Markov model addressed a period after changing medication, whereas the dynamic model handled AEs for each year.

We have included international findings for comparative analysis with our results. In an economic evaluation of DTG-based regimens in Ethiopia, 15.3 QALYs per patient (≥ 18 years of age ) over a lifetime were obtained compared to EFV-based regimens (vs 14.7 QALYs)[4]. The Ethiopian study reported slightly higher results than the QALYs obtained in our Markov model, mainly due to the inclusion of disutility associated with ADEs and AEs, which was not considered in the Ethiopian study. In an economic evaluation of DTG-based treatment regimens in the United States, compared with EFV-based regimens, the cost was calculated from the perspective of the US payer: 88.20% for medicines (vs 88.42%), 8.39% for tests (vs 8.18%), and 3.41% for events (vs 3.40%)[39]. The order of cost distribution in our study, from highest to lowest, was the cost of medicine, the cost of tests, and the cost of event management.

At present, HIV/AIDS treatment in China has entered the era of "Troika" optimization, ensuring the provision of low-end (free drugs), mid-end (Medicare drugs), and high-end (self-financed drugs) options for patients[68]. This setup allows individuals to select treatments aligned with their preferences. For example, economically well-off HIV patients may select pricier yet highly effective and durable drugs, exemplifying the market positioning of B/F/TAF. With the introduction of innovative antiretroviral therapies and the evolution of China's economic landscape, the HIV treatment environment has gradually been optimized. Consequently, patients may increasingly prefer B/F/TAF despite higher costs, gaining wider acceptance in the evolving HIV treatment landscape.

However, it's crucial to recognize limitations. Firstly, achieving precise control over the clinical trial effectiveness of both interventions is nearly impossible. There are no direct trials establishing efficacy, and the support from network meta-analysis for indirect clinical evidence is insufficient. Secondly, to ensure the smooth progress of our study, individuals discontinuing first-line treatment transitioned entirely to second-line therapy, and those discontinuing second-line moved directly to third-line therapy. Some practical aspects, such as adherence and patient choice, are overlooked in this setup. Lastly, the lack of Chinese research on certain parameters, such as utility values, compelled us to adopt foreign parameters for substitution. Despite introducing uncertainty in long-term extrapolation, we addressed it with sensitivity analyses in the study.

If these data become available in the future, our study will be supplemented. Furthermore, the limitation of our study lies in that the results are only applicable to the cohort of Chinese patients.

Our economic evaluation of B/F/TAF provided supplementary evidence for its recommendation in the Chinese HIV treatment guidelines, supporting its widespread adoption. In the application of methodology, the Markov model was precise and detailed, whereas the dynamic model was sensitive and comprehensive. We should consider the impact of both models on the results of ART economic evaluation.

B/F/TAF: bictegravir/emtricitabine/tenofovir;

DTG/3TC: dolutegravir/lamivudine;

MSM: men who have sex with men;

MIDU: male injecting drug users;

FIDU: female injecting drug users;

ART: antiretroviral therapy；

ADEs: AIDS-defining events;

AEs: adverse reactions;

QALYs: quality-adjusted life-years;

ICER: incremental cost-effectiveness ratio;

LY: life year;

WTP: willing to pay;

PSA: probability sensitivity analysis;

OWSA: one-way sensitivity analysis;

RSD: Relative Standard Deviation.

Consent for publication: Not applicable.

Availability of data and material: The datasets generated or analyzed during the current study are available in the following repositories:

1.National Center For AIDS/STD Control And Prevention

Website: https://ncaids.chinacdc.cn/

2.Menet - China Pharmaceutical Information Database

Website: https://www.menet.com.cn/

3.China Statistical Yearbook

Website: http://www.stats.gov.cn/sj/ndsj/

Competing interests: All authors declare no competing interests.

Funding: This work was supported by National Innovation and Entrepreneurship Training Program for Undergraduate（No.202310316065Y）and Full-value driven decision-making support platform (DSP) and management mechanism for rapid vaccine R&D, efficient trial design and scientific regulation: from a public administration perspective (2023YFVA1002).

Ethics approval and consent to participate: Not applicable.

Author contributions: Wenxi Tang conceived the project. Di Zhang, Hao Li, Hongliu Zhang, Xin Jiang, and Ruihua Wang collected the data, with Wenjuan Wang, Dachuang Zhou, and Kejia Zhou involved in the verification of the collected data. Wenjuan Wang and Dachuang Zhou developed the model and conducted the analysis. Wenjuan Wang prepared the manuscript. Xi Wang reviewed the manuscript. All authors provided critical feedback on the analysis, and they reviewed and approved the submitted manuscript.

HE N. New progress in HIV/AIDS epidemiology in China. Chinese Journal of Disease Control and Prevention. 2021;25(12):1365-8.
Zhang F, Yan Z, Ye M, Decai Z, Wisdom B, Yasong W, et al. Progress and achievements of free HIV antiviral treatment in China. Chinese Journal of AIDS & STD. 2022;28(01):6-9.
Juday T, Correll T, Anene A, Broder MS, Ortendahl J, Bentley T. Cost-effectiveness of the once-daily efavirenz/emtricitabine/tenofovir tablet compared with the once-daily elvitegravir/cobicistat/emtricitabine/tenofovir tablet as first-line antiretroviral therapy in HIV-infected adults in the US. CLINICOECONOMIC OUTC. 2013 2013/1/20;5:437-45. 10.2147/CEOR.S47486
Belay YB, Ali EE, Chung KY, Gebretekle GB, Sander B. Cost-Utility Analysis of Dolutegravir- Versus Efavirenz-Based Regimens as a First-Line Treatment in Adult HIV/AIDS Patients in Ethiopia. PharmacoEconomics - open. 2021 2021/12/1;5(4):655-64. 10.1007/s41669-021-00275-6
Butler K, Hayward O, Jacob I, Anderson SJ, Punekar YS, Evitt LA, et al. Cost-effectiveness and budget impact of dolutegravir/lamivudine for treatment of human immunodeficiency virus (HIV-1) infection in the United States. Journal of Managed Care and Specialty Pharmacy. 2021 2021/1/1;27(7):891-903. 10.18553/JMCP.2021.27.7.891
Punekar YS, Guo N, Tremblay G, Piercy J, Holbrook T, Young B. Improving access to antiretrovirals in China: economic analyses of dolutegravir in HIV-1 patients. COST EFFECT RESOUR A. 2019 2019/1/20;17:26. 10.1186/s12962-019-0195-2
Li M, Cao Y, Huang H, Qin G, Chu M, Zou M, et al. Cost-effectiveness analysis of antiretroviral drugs for treatment-naive HIV infection in China. BMC PUBLIC HEALTH. 2023 2023/11/13;23(1):2228. 10.1186/s12889-023-17052-1
Broyles LN, Luo R, Boeras D, Vojnov L. The risk of sexual transmission of HIV in individuals with low-level HIV viraemia: a systematic review. LANCET. 2023 2023/8/5;402(10400):464-71. 10.1016/S0140-6736(23)00877-2
Gersovitz M, Hammer JS. Infectious Diseases, Public Policy, and the Marriage of Economics and Epidemiology. 2003;18(2):129-57. 10.1093/wbro/lkg011
Jing W. projecting the effects of HIV/AIDS prevention and intervention strategics in China using a dynamic model [Ph.D]: Chinese Center for Disease Control and Prevention; 2015.
Yaylali E, Erdogan ZM, Calisir F, Gokengin D, Korten V, Tabak F, et al. Modeling the future of HIV in Turkey: Cost-effectiveness analysis of improving testing and diagnosis. PLOS ONE. 2023 2023/1/20;18(6):e286254. 10.1371/journal.pone.0286254
Krebs E, Zang X, Enns B, Min JE, Behrends CN, Del RC, et al. Ending the HIV Epidemic Among Persons Who Inject Drugs: A Cost-Effectiveness Analysis in Six US Cities. J INFECT DIS. 2020 2020/9/2;222(Suppl 5):S301-11. 10.1093/infdis/jiaa130
Acquired Immunodeficiency Syndrome and Hepatitis C Professional Group, Society of Infectious Diseases, Chinese Medical Association, Chinese Center for Disease Control and Prevention. Chinese Guidelines for Diagnosis and Treatment of Human Immunodeficiency Virus Infection /Acquired Immunodeficiency Syndrome ( 2021 edition). Medical Journal of Peking Union Medical College Hospital. 2022;13(02):203-26.
Kaplan JE, Benson C, Holmes KK, Brooks JT, Pau A, Masur H. Guidelines for prevention and treatment of opportunistic infections in HIV-infected adults and adolescents: recommendations from CDC, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. MMWR RECOMM REP. 2009 2009/4/10;58(RR-4):1-207, quiz CE1-4.
World Health Organization. Consolidated Guidelines on HIV Prevention, Testing, Treatment, Service Delivery and Monitoring: Recommendations for a Public Health Approach. Geneva; 2021.
Xueyun W, Yinzhong S. Recent advances in antiretroviral therapy for AIDS. Infectious Disease Information. 2019;32(01):81-7.
Jie B, Chao Z. Application and recent research progress of Bictegravir/Emtricitabine/Tenofovir Alafenamide in the treatment of acquired immunodeficiency syndrome. Clinical Medication Journal. 2021;19(03):20-5.
Gallant J, Lazzarin A, Mills A, Orkin C, Podzamczer D, Tebas P, et al. Bictegravir, emtricitabine, and tenofovir alafenamide versus dolutegravir, abacavir, and lamivudine for initial treatment of HIV-1 infection (GS-US-380-1489): a double-blind, multicentre, phase 3, randomised controlled non-inferiority trial. LANCET. 2017 2017/11/4;390(10107):2063-72. 10.1016/S0140-6736(17)32299-7
Sax PE, Pozniak A, Montes ML, Koenig E, DeJesus E, Stellbrink HJ, et al. Coformulated bictegravir, emtricitabine, and tenofovir alafenamide versus dolutegravir with emtricitabine and tenofovir alafenamide, for initial treatment of HIV-1 infection (GS-US-380-1490): a randomised, double-blind, multicentre, phase 3, non-inferiority trial. LANCET. 2017 2017/11/4;390(10107):2073-82. 10.1016/S0140-6736(17)32340-1
National Bureau Of Statistics. Statistical Bulletin of the People's Republic of China on National Economic and Social Development for 2022. 2023. http://www.stats.gov.cn/sj/zxfb/202302/t20230228_1919011.html?eqid=a33d23500000415c00000003642fe3df. Accessed 12 Dec 2023
Sun X, Xiao Y, Peng Z, Wang N. Modelling HIV/AIDS epidemic among men who have sex with men in China. BIOMED RES INT. 2013 2013/1/20;2013:413260. 10.1155/2013/413260
Dongli W, Yanke A, Jun Y, Xiuxia M, Huijun G, Qianlei X, et al. Prevalence and influencing factors of immunological no-responese in HIV/AIDS patients with second-line antiviral therapy. Chinese General Practice. 2022;25(30):3789-95.
National Center For AIDS/STD Control And Prevention, China CDC. 2023. https://ncaids.chinacdc.cn/. Accessed 01 Oct 2023
WHO. WHO case definitions of HIV for surveillance and revised clinical staging and immunological classification of HIV-related disease in adults and children. 2007. https://iris.who.int/handle/10665/43699. Accessed 01 Otc 2023
Aiwang W, Mingrun Y, Zhilong M, Xiang-bin L. Analysis of initial CD4+T lymphocyte counts of newly indentified HIV/ADIS Patients in Taizhou City. Jiangsu Journal of Preventive Medicine. 2016;27(04):400-2. 10.13668/j.issn.1006-9070.2016.04.006
Haijuan L. Analysis of initial CD4+T lymphocyte counts of 131 patients with HIV infection. Modern Disease Control and Prevention. 2018;29(05):354-6.
Xiying L, Zheng L, Xuelei Z. Analysis of CD4~+T lymphocyte detection results in 117 newly confirmed HIV infected patients. Modern Disease Control and Prevention. 2020;31(06):436-8. 10.13515/j.cnki.hnjpm.1006-8414.2020.06.011
Yang L, Ji Z, Jing-rong Y, Hong-yan L. the first CD4+T count test of newly detected HIV/AIDS cases of Beijing in 2013. Chinese Journal of AIDS & STD. 2015;21(02):103-6.
Wohl DA, Yazdanpanah Y, Baumgarten A, Clarke A, Thompson MA, Brinson C, et al. Bictegravir combined with emtricitabine and tenofovir alafenamide versus dolutegravir, abacavir, and lamivudine for initial treatment of HIV-1 infection: week 96 results from a randomised, double-blind, multicentre, phase 3, non-inferiority trial. LANCET HIV. 2019 2019/6/1;6(6):e355-63. 10.1016/S2352-3018(19)30077-3
Cahn P, Madero JS, Arribas JR, Antinori A, Ortiz R, Clarke AE, et al. Durable Efficacy of Dolutegravir Plus Lamivudine in Antiretroviral Treatment-Naive Adults With HIV-1 Infection: 96-Week Results From the GEMINI-1 and GEMINI-2 Randomized Clinical Trials. JAIDS-J ACQ IMM DEF. 2020 2020/3/1;83(3):310-8. 10.1097/QAI.0000000000002275
Cahn P, Madero JS, Arribas JR, Antinori A, Ortiz R, Clarke AE, et al. Dolutegravir plus lamivudine versus dolutegravir plus tenofovir disoproxil fumarate and emtricitabine in antiretroviral-naive adults with HIV-1 infection (GEMINI-1 and GEMINI-2): week 48 results from two multicentre, double-blind, randomised, non-inferiority, phase 3 trials. LANCET. 2019 2019/1/12;393(10167):143-55. 10.1016/S0140-6736(18)32462-0
Huang X, Xu L, Sun L, Gao G, Cai W, Liu Y, et al. Six-Year Immunologic Recovery and Virological Suppression of HIV Patients on LPV/r-Based Second-Line Antiretroviral Treatment: A Multi-Center Real-World Cohort Study in China. FRONT PHARMACOL. 2019 2019/1/20;10:1455. 10.3389/fphar.2019.01455
Avihingsanon A, Hughes MD, Salata R, Godfrey C, McCarthy C, Mugyenyi P, et al. Third-line antiretroviral therapy, including raltegravir (RAL), darunavir (DRV/r) and/or etravirine (ETR), is well tolerated and achieves durable virologic suppression over 144 weeks in resource-limited settings: ACTG A5288 strategy trial. J INT AIDS SOC. 2022 2022/6/1;25(6):e25905. 10.1002/jia2.25905
Patrikar S, Basannar DR, Bhatti VK, Kotwal A, Gupta RM, Grewal RS. Rate of decline in CD4 count in HIV patients not on antiretroviral therapy. Med J Armed Forces India. 2014 2014/4/1;70(2):134-8. 10.1016/j.mjafi.2013.08.005
Girouard MP, Sax PE, Parker RA, Taiwo B, Freedberg KA, Gulick RM, et al. The Cost-effectiveness and Budget Impact of 2-Drug Dolutegravir-Lamivudine Regimens for the Treatment of HIV Infection in the United States. CLIN INFECT DIS. 2016 2016/3/15;62(6):784-91. 10.1093/cid/civ981
Simei L, Shaoxun L, Yue Z, Yuan Y, Delei D, Yuanmei H, et al. Health economics evaluation of comprehensive intervention of HIV/AIDs among MSM in Shenzhen. Chinese Journal of AIDS & STD. 2023;29(07):761-7. 10.13419/j.cnki.aids.2023.07.06
Yanhong G, Chunjuan Z, Liqiang X. Analysis of AIDS death cases in Changshu City from 2000 to 2020. Jiangsu Journal of Preventive Medicine. 2022;33(03):306-7. 10.13668/j.issn.1006-9070.2022.03.018
Ke Y, Shenghua H, Nan G, Ruifeng Z, Lin C. The correlation between opportunistic infections and CD4+T lymphocytes in HIV/AIDS patients. Chinese Journal of Microecology. 2019;31(02):167-70.
Peng S, Tafazzoli A, Dorman E, Rosenblatt L, Villasis-Keever A, Sorensen S. Cost-effectiveness of DTG + ABC/3TC versus EFV/TDF/FTC for first-line treatment of HIV-1 in the United States. J MED ECON. 2015 2015/1/20;18(10):763-76. 10.3111/13696998.2015.1046878
Pudong, Kunli W, Junhua S, Songqin L, Hong-wei L, Yaling W. Correlation between level of CD4 Cells, viral load and oral Oral Candidiassis in AIDS patients. Journal of Kunming Medical University. 2013;34(06):141-3.
An L, Jianwei L, Leaf P, Ying S, Liang H, Hongli M, et al. Clinical observation of efficacy and safety of second-line antiviral therapy of HIV/AIDS. Chinese Journal of AIDS & STD. 2015;21(06):450-2. 10.13419/j.cnki.aids.2015.06.02
GILEAD. Gilead Presents Real-World Evidence Reinforcing the Use of Biktarvy® for the Treatment of People Living With HIV With a Range of Comorbidities. 2022. https://www.gileadchina.cn/-/media/gilead-china/pdfs/news-and-press/press-releases/20221026-cn.pdf. Accessed 01 Oct 2023
Osiyemi O, De Wit S, Ajana F, Bisshop F, Portilla J, Routy JP, et al. Efficacy and Safety of Switching to Dolutegravir/Lamivudine Versus Continuing a Tenofovir Alafenamide-Based 3- or 4-Drug Regimen for Maintenance of Virologic Suppression in Adults Living With Human Immunodeficiency Virus Type 1: Results Through Week 144 From the Phase 3, Noninferiority TANGO Randomized Trial. CLIN INFECT DIS. 2022 2022/9/29;75(6):975-86. 10.1093/cid/ciac036
Guangzhou Biaodian Medical Information Co.,Ltd, National Medicine products Administration Institute of Medicine Economics. Menet - China Pharmaceutical Information Database. 2023. https://www.menet.com.cn/. Accessed 01 Oct 2023
Burea JPP. Notice of Provincial Bureau of Prices, Provincial Health and Family Planning Commission, Provincial Department of Human Resources and Social Security, Provincial Department of Finance on the Implementation Plan for the Comprehensive Reform of Pharmaceutical Prices in Public Hospitals at the provincial (ministerial) Level in Ning (Su Biyi (2015) No. 284). 2015. http://www.changzhou.gov.cn/ns_news/791504063005455. Accessed 01 Oct 2023
world development indicators. 2023. https://databank.worldbank.org/reports.aspx?dsid=2&series=PA.NUS.PPP. Accessed 01 Oct 2023
Simpson KN, Pei PP, Möller J, Baran RW, Dietz B, Woodward W, et al. Lopinavir/ritonavir versus darunavir plus ritonavir for HIV infection: A cost-effectiveness analysis for the United States. PHARMACOECONOMICS. 2013 2013/1/1;31(5):427-44. 10.1007/s40273-013-0048-3
Dai L, Su B, Liu A, Zhang H, Wu H, Zhang T, et al. Adverse events in Chinese human immunodeficiency virus (HIV) patients receiving first line antiretroviral therapy. BMC INFECT DIS. 2020 2020/2/19;20(1):158. 10.1186/s12879-020-4878-2
Society EAC. EACS Guidelines（Version10.1）. 2022. https：//www.eacsociety.org/files/guidelines-10.1_30032021_1.pdf. Accessed 01 Oct 2023
Spinner CD, Kümmerle T, Schneider J, Cordes C, Heiken H, Stellbrink HJ, et al. Efficacy and Safety of Switching to Dolutegravir With Boosted Darunavir in Virologically Suppressed Adults With HIV-1: A Randomized, Open-Label, Multicenter, Phase 3, Noninferiority Trial: The DUALIS Study. OPEN FORUM INFECT DI. 2020 2020/9/1;7(9):ofaa356. 10.1093/ofid/ofaa356
Zhang X. Economics evaluation for prevention mother-to-child transmission of human immunodeficiency (HIV) in China [Master]: Chinese Center for Disease Control and Prevention; 2016.
Restelli U, Rizzardini G, Antinori A, Lazzarin A, Bonfanti M, Bonfanti P, et al. Cost-effectiveness analysis of dolutegravir plus backbone compared with raltegravir plus backbone, darunavir+ritonavir plus backbone and efavirenz/tenofovir/emtricitabine in treatment naïve and experienced HIV-positive patients. THER CLIN RISK MANAG. 2017 2017/1/20;13:787-97. 10.2147/TCRM.S135972
Peking University, Fudan University, Tianjin University, China Pharmaceutical University, Ministry Of Human Resources And Social Security MHRSS, Hospital, PLA. China Guidelines for Pharmacoeconomic Evaluations (2020). 2020. https://www.ispor.org/heor-resources/more-heor-resources/pharmacoeconomic-guidelines/pe-guideline-detail/china-mainland. Accessed 01 Oct 2023
National Bureau Of Statistics. China Statistical Yearbook. 2023. http://www.stats.gov.cn/sj/ndsj/. Accessed2023
The International Lesbian, Gay, Bisexual, Trans and Intersex Association. ILGA WORLD. 2023. https://database.ilga.org/en. Accessed 01 Oct 2023
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic and main response on Control And Prevention in China in December 2013. Chinese Journal of AIDS & STD. 2013;19(02):85. 10.13419/j.cnki.aids.2013.02.025
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic and main response on Control And Prevention in China in December 2012. Chinese Journal of AIDS & STD. 2013;19(02):85. 10.13419/j.cnki.aids.2013.02.025
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic and main response on Control And Prevention in China in December 2014. Chinese Journal of AIDS & STD. 2014;20(02):75. 10.13419/j.cnki.aids.2014.02.004
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic and main response on Control And Prevention in China in December 2014. Chinese Journal of AIDS & STD. 2015;21(02):87. 10.13419/j.cnki.aids.2015.02.01
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic and main response on Control And Prevention in China in December 2015. Chinese Journal of AIDS & STD. 2016;22(02):69. 10.13419/j.cnki.aids.2016.02.01
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic in China in December 2016. Chinese Journal of AIDS & STD. 2017;23(02):93. 10.13419/j.cnki.aids.2017.02.01
National Center For AIDS/STD Control And Prevention, China CDC. Update on the AIDS/STD epidemic in China in December 2017. Chinese Journal of AIDS & STD. 2018;24(02):111. 10.13419/j.cnki.aids.2018.02.01
Guangming Daily. China's AIDS epidemic continues to be at a low epidemic level. 2019. https://www.gov.cn/xinwen/2019-12/01/content_5457304.htm. Accessed 10 Oct 2023
Min S, Ye L, Yu-song F, Ran Z, Yang X, Xin L. analysis of behavior characteristics and HIV infection status among drug users in Jinniu district of Chengdu city from 2009 to 2020. Occupational and Health. 2022;38(01):78-83.
Office Of China National Narcotics Control Commission. China Anti-Drug Network. 2023. http://www.nncc626.com/gjjdb/bg.htm. Accessed 10 Oct 2023
Health Commission Of Chengde. The rate of HIV/AIDS antiretroviral treatment in Guangxi was over 90%, and the success rate of treatment was over 95%. 2022. https://wjw.chengde.gov.cn/art/2022/1/27/art_464_830778.html. Accessed 10 Oct 2023
Jin Y, Chang C, Chen FF, Qin Q, Tang H. Survival analysis since diagnosis of HIV-positive injecting drug users aged 15 years and above in China. Chinese Journal of Epidemiology. 2022(06):860-4.
Yaozu H, Xiaoting C, Jingliang C, Pengle G, Linghua L. The Changes and progress of antiviral drugs for acquired immune deficiency syndrome in China. Guangdong Medicial Journal. 2023;44(02):157-60.

Tables 1 to 3 are available in the Supplementary Files section.

No competing interests reported.

Tables.docx
AdditionalFile.docx
Supplementary Information The online version contains supplementary material available at https://doi.org.

Download PDF

Version 1

posted

You are reading this latest preprint version

Comparison of Markov Model and Dynamic Model in Economic Evaluations for Infectious Diseases: A case of Bictegravir/emtricitabine/tenofovir Treatment-native adults of HIV-1 in China

Status:

Version 1

Abstract

Figures

1. Background

2. Methods

2.1 Study design

2.2 Markov model

2.2.1 Model Structure and Assumptions

2.2.2 Model parameters

2.2.3 Model verification

2.3 Dynamic model

2.3.1 Model Structure and Assumptions

2.3.2 Model parameters

2.3.3 Model verification

2.4 Sensitivity analyses

3. Results

3.1 Base case

3.2 Sensitive analyses

4. Discussion

5. Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1