Cost-utility analysis of repetitive transcranial magnetic stimulation and transcranial direct current stimulation as adjunctive therapies in Chinese patients with major depressive disorder

Abstract

Objective

Major depressive disorder (MDD) imposes a substantial burden in China, with many patients showing an inadequate response to first-line drugs. While repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) show promise, their economic value remains uncertain. This study aims to evaluate the cost-effectiveness of rTMS, tDCS, and their combination as adjunctive therapies for MDD in China.

Methods

A 4-week Markov model used efficacy data from a Phase II trial (ChiCTR2100052122). Four strategies were compared: drug therapy alone, drug plus rTMS, drug plus tDCS, and combined therapy (rTMS+tDCS+drug). Direct and indirect costs were recorded weekly. Outcomes were quality-adjusted life years (QALYs), and incremental cost-effectiveness ratios (ICERs) were calculated. The willingness-to-pay (WTP) threshold was 1–3 times China’s 2024 GDP per capita (¥95,749 - ¥287,247 per QALY). Deterministic and scenario analyses tested robustness and longer treatment durations.

Results

All combination therapies outperformed drug therapy alone. The ICERs for rTMS, tDCS, and combined therapy were ¥113,174/QALY, ¥130,194/QALY, and ¥177,569/QALY, respectively - all below the WTP threshold. Compared with rTMS, tDCS showed better cost-effectiveness. The results were robust in sensitivity analyses, and tDCS’s economic advantage grew with longer treatment duration.

Conclusions

In the Chinese healthcare context, both rTMS and tDCS as adjuncts to pharmacotherapy are cost-effective, with tDCS showing a better cost-effectiveness ratio and increasingly prominent benefits in long-term treatment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-026-08240-5.

Background

Major depressive disorder (MDD), one of the most burdensome mental disorders globally, is estimated to affect over 95 million people in China. It not only severely impairs patients’ social functioning and quality of life but also continuously increases healthcare resource utilization and socioeconomic costs [1, 2]. Current first-line treatments typically involve selective serotonin reuptake inhibitors or serotonin-norepinephrine reuptake inhibitors. However, the remission rate remains below 50%, and issues such as delayed onset of action, high risk of relapse, and poor tolerability lead many patients to develop treatment-resistant depression [3], further exacerbating long-term treatment needs and economic burdens. This reality has driven clinicians and researchers to explore more effective and tolerable adjunctive treatment strategies to improve symptom control and reduce relapse [4].

Repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) are two non-invasive brain stimulation techniques that have gained considerable attention in the treatment of depression in recent years [5]. rTMS uses magnetic fields to induce electrical currents in the cerebral cortex, modulating neural network activity involved in emotion regulation [6], while tDCS applies low-intensity direct current to alter cortical neuronal excitability, thereby promoting neuroplasticity and improving mood and cognitive function [7–9]. Numerous clinical trials have demonstrated that both techniques, when used alongside pharmacotherapy, can further improve response and remission rates with favorable safety profiles [10–13]. The Phase II randomized controlled trial by Zhou D et al. (2024) (Registration No. ChiCTR2100052122; October 19, 2021), which compared the efficacy of rTMS, tDCS, and their combination as adjunctive therapies to antidepressants [14], reported a markedly higher response rate of 83.3% for rTMS+tDCS over a 4-week period in Chinese patients, compared to 71.7% for rTMS alone and 61.7% for tDCS alone—all significantly superior to the 55.0% response rate with pharmacotherapy alone, with a favorable safety profile.

Although both rTMS and tDCS demonstrate notable clinical efficacy and safety, existing pharmacoeconomic literature has primarily focused on the cost-effectiveness of rTMS as an adjunctive treatment, generally affirming its favorable cost-utility profile [15–17]. In contrast, economic evidence for tDCS and its combination with rTMS remains scarce. Different treatment modalities may lead to varying health outcomes and cost structures, and the absence of comprehensive analysis based on Chinese healthcare costs, social expenditures, and health utility values limits informed decision-making regarding the integration of these technologies into clinical pathways and health insurance policies.

Therefore, this study utilized clinical data from the ChiCTR2100052122 randomized controlled trial to develop a Markov model evaluating MDD treatment in China [14]. From a societal perspective, we compared the cost-effectiveness of four strategies during the acute intervention phase: drug therapy alone, rTMS plus drugs, tDCS plus drugs, and rTMS combined with tDCS plus drugs. The aim is to provide quantitative evidence for developing cost-effective treatment protocols and to support resource allocation and policy-making in the field of mental health.

Methods

Overview

We used TreeAge Pro software (version 2022, https://www.treeage.com/) for mathematical modeling. This study adopted a societal perspective to develop a Markov model evaluating the cost-effectiveness of four treatment strategies for adult MDD patients in China, covering various combinations of rTMS, tDCS, and antidepressants. The cycle length of the model was set to one week, with a total time horizon of four weeks, to capture short-term treatment responses and disease fluctuations characteristic of non-invasive brain stimulation techniques [18]. The four policies evaluated in the study are as follows (hereinafter uniformly referred to as the combined therapy group, rTMS group, tDCS group, and drug group):

1. Combined therapy group: rTMS + tDCS + Drug

2. rTMS + Drug

3. tDCS + Drug

4. Drug

Ethical approval was not required for this study as it solely utilized existing, non-identifiable data, including clinical efficacy data from Zhou D et al. [14]. Cost and utility data were obtained from published literature and public data platforms. All analyses strictly adhered to the relevant requirements of the China Guidelines for Pharmacoeconomic Evaluations [19].

Model construction

This study develops a disease progression model based on a Markov framework to simulate the health state transitions of MDD patients under four treatment regimens, aiming to evaluate their clinical and economic outcomes, as illustrated in Fig. 1. The model includes three mutually exclusive health states: MDD, defined as meeting clinical diagnostic criteria with a Hamilton Depression Rating Scale-24 items (HDRS-24) score > 20; remission, defined as a HDRS-24 total score ≤ 9; and death. All patients begin in the MDD state and may transition between states each cycle based on predefined transition probabilities. Patients who do not achieve remission remain in the MDD state, while those who meet remission criteria enter the remission state. However, patients in remission face the risk of relapse and may return to the MDD state. It should be noted that patients in any health state may transition to death due to suicide related to depression, worsening comorbidities, or natural causes.

Fig. 1.

Markov Model Diagram. (A) Markov model structure. This model is used to compare four treatment strategies for patients with MDD. (B) Health state network. The diagram shows possible transitions between the three health states: MDD (Major Depressive Disorder), Remission, and Death

The model uses a cycle length of one week. This is primarily based on the common clinical practice of “once daily, five times per week” used for both rTMS and tDCS during acute phase treatment. A one week cycle corresponds to a complete treatment unit and reflects actual clinical routines. It also matches standard outpatient follow-up schedules and the typical timing of assessments with tools like the HDRS-24. The total time horizon for the model is set at four weeks. This setting is based on the FDA recommended treatment period of 4 to 6 weeks (20 to 30 sessions). This ensures the model covers the complete acute treatment course of non-invasive brain stimulation as an add-on therapy and remains consistent with existing clinical evidence [20, 21]. Moreover, the ChiCTR2100052122 trial reported cumulative remission rates at both two weeks and four weeks; thus, setting the model endpoint at 4 weeks, primarily based on the observed period of the included randomized controlled trial, ensures direct use of the original efficacy data and reduces uncertainty from extrapolation.

Transition probabilities and effectiveness

Transition probabilities for various events were standardized to weekly probabilities using the following formula:

All probabilities were converted to weekly rates, where P represents the probability of an event occurring within time t, r is the event rate per unit time, and t is the time span (one week in this study). This standardization method ensures that clinical data from different studies and time scales have a consistent measurement basis in the model. The transition probabilities calculated using this formula include the probability of moving from the MDD health state to Remission, the probability of relapse from Remission back to MDD, and the probability of disease-related death from both the MDD and Remission health states.

The specific data sources are as follows. The original data for the transition probability from MDD to Remission were extracted from a phase II randomized controlled trial by Zhou D et al. (2024) (registration number ChiCTR2100052122). This study evaluated the efficacy of all four required treatment strategies and reported cumulative remission rates both at two weeks of treatment and at the four-week follow-up [14]. Since our model uses a one-week Markov cycle, the original multi-time-point remission data were converted into a weekly probability using the formula. This conversion ensures the data align with the dynamic changes of short-term intervention effects.

The relapse probability from Remission to the MDD state was estimated based on the natural disease course study of MDD by Hua-Ning Wang et al. [22], which provided robust estimates based on relapse risk data from patients in remission over 52 weeks. The all-cause mortality probabilities for patients in the MDD state and remission state were cited from two sources: the epidemiological analysis of all-cause mortality in Chinese MDD patients by Jifei Wang et al. (2024), and the clinical trial on drug efficacy by A John Rush et al. (2006) [23, 24]. Mortality rates were assumed to be consistent for all patients in the same health state (i.e., uniform mortality within the MDD state and uniform mortality within the remission state). All raw data were converted to weekly probabilities using the aforementioned formula upon incorporation into the model. Table S1 provides these data in detail.

Cost and utility

This study adopted a societal perspective to systematically evaluate the economic burden and health outcomes of the four treatment strategies. Direct medical costs included expenses for neurostimulation therapy, antidepressant medications, and registration fees; indirect social costs covered transportation expenses incurred by patients for medical visits and productivity losses due to sick leave. Excluded cost categories included productivity loss of patients’ family members (due to lack of reliable data), long-term care costs (not applicable to a 4‑week acute phase), and equipment depreciation and facility fees (already included in the treatment costs).

Data for neurostimulation therapy costs and registration fees were sourced from the official website of Beijing Anding Hospital. As a key supporting institution of the National Center for Mental Disorders, its fee schedule represents, to some extent, the pricing level of large tertiary psychiatric hospitals in China. Medication costs were calculated based on price information from the National Medical Insurance Service Platform and the baseline medication regimens of patients in the four groups from the ChiCTR2100052122 trial (Table S2), ensuring that cost estimation aligned with clinical medication patterns. Transportation costs were referenced from the study by Guoping Wu et al. (2024) on the cost of illness in hospitalized patients with depression in China [25]. Productivity losses were estimated based on the per capita wage income and employment rate for the first half of 2025 published by the National Bureau of Statistics. It is important to note that all costs were denominated in 2024 Chinese Yuan (CNY), and due to the model’s time horizon being only 4 weeks (less than one year), no discounting was applied, in accordance with the China Guidelines for Pharmacoeconomic Evaluations (2020). Detailed data are shown in Tables S3-S4.

Health outcomes were quantified using quality-adjusted life-years (QALYs), calculated based on the utility weights of each health state in the Markov model and the time patients spent in the corresponding states. The utility values in this model were all obtained from published literature [26], sourced from clinical trials and meta-analyses, etc. Detailed data are provided in Table S5.

Statistical analysis

Base-case ICERs

In the base-case analysis, this study compared the cost-effectiveness of four treatment strategies over a four-week period, using the incremental cost-effectiveness ratio (ICER) as the primary evaluation metric. The ICER was calculated as the ratio of incremental cost to incremental QALYs between two strategies. The final step involved comparing the ICER of competing strategies against a predefined willingness-to-pay (WTP) threshold to determine their relative cost-effectiveness. First, using the drug‑alone group as the common reference, the ICERs of the rTMS, tDCS, and combined therapy groups relative to the drug group were calculated. Second, all strategies were ordered by increasing cost, and stepwise ICERs between adjacent strategies were calculated, assessing absolute or extended dominance. A strategy is absolutely dominated if it is more costly and less effective than an alternative; extendedly dominated if it is less effective and less cost‑effective than a more expensive strategy. Dominated strategies are excluded, and the remaining form the cost‑effectiveness frontier. Since China lacks an explicit WTP threshold for depression-related outcomes, this study referred to the “China Guidelines for Pharmacoeconomic Evaluations (2020)” and set the WTP threshold at 1–3 times the 2024 per capita GDP (CNY 95,749 − 287,247). A strategy was considered cost-effective if its ICER was below this threshold; otherwise, it was deemed not economically advantageous.

Sensitivity analysis

DSA

Two types of sensitivity analyses were conducted to assess and validate the robustness of the cost-effectiveness results. Deterministic sensitivity analysis (DSA) was performed to evaluate the impact of uncertainty in various model parameters on the ICER. For parameters lacking 95% confidence intervals (e.g., some utility values, costs, and clinical effectiveness measures), a variation range of ± 25% from the base-case value was applied. The results of the DSA were presented in a tornado diagram, ranking the parameters according to their degree of influence on the ICER to identify the factors most sensitive to the results.

PSA

To account for joint parameter uncertainty, a probabilistic sensitivity analysis (PSA) was conducted using 1000 Monte Carlo simulations. Parameters were sampled from appropriate distributions—costs from a Gamma distribution, and probabilities and effects from a Beta distribution. The results of the PSA were presented as a cost-effectiveness acceptability curve, showing the probability of each strategy being considered cost-effective at different WTP thresholds.

Scenario analysis

Based on the recommendations for intervention frequency in the acute and consolidation phases from the Chinese Guidelines for the Prevention and Treatment of Depressive Disorders (Second Edition), this study extended the total model time horizon to 3 months and 6 months in scenario analyses. This simulated consolidation treatment regimens of twice per week (1–3 months) and once per week (3–6 months) after the acute phase, to assess the cost differences arising from changes in treatment frequency over the long term and their impact on cost-effectiveness results, thereby providing references for optimal resource allocation across different treatment stages.

Results

Base-case analysis

In the base-case analysis, using the drug-alone group as the common reference, the other three treatment strategies all demonstrated significant cost-effectiveness advantages, as detailed in Table 1. Relative to drug therapy alone, the total costs for the rTMS group and tDCS group increased to CNY 3,132.66 and CNY 4,431.69, respectively, with corresponding QALYs increasing to 0.05787 and 0.06682. The calculated ICERs were CNY 113,173.77 per QALY and CNY 130,193.83 per QALY, both significantly below the pre-defined WTP threshold. Furthermore, although the combined therapy group achieved the highest QALY value (0.06732), its incremental cost was CNY 3,084.38, resulting in an ICER of CNY 177,569.37 per QALY. While still below the threshold, its economic advantage was less prominent compared to the strategies combining a single neuromodulation technique with drugs.

Table 1.

ICERs of each adjunctive neuromodulation strategy versus drug alone

ICERs(vs. drug)
Strategy	Cost, CNY	QALYs	Incremental		ICER,ΔCNY/QALY
			Cost,ΔCNY	QALYs
Drug	2,235.32	0.04995	-	-	-
rTMS+Drug	3,132.66	0.05787	897.34	0.00793	113,173.77
tDCS+Drug	4,431.69	0.06682	2,196.37	0.01687	130,193.83
Combined therapy	5,319.70	0.06732	3,084.38	0.01737	177,569.37

QALY, quality-adjusted life-years; ICER, incremental cost-effectiveness ratios

To further clarify the relative economic advantages among the neuromodulation strategies, all strategies were ordered by increasing total cost, and stepwise ICERs between adjacent strategies were calculated sequentially, as shown in Table 2. The results further revealed that among the three interventions involving neuromodulation, the tDCS group showed a slight improvement in incremental QALYs compared to the rTMS group (increasing from 0.05787 to 0.06682), with a corresponding ICER of CNY 145,153.94 per QALY, which remained within the cost-effective range. Additionally, although the combined therapy group showed a marginal QALY improvement compared to the tDCS plus drug group (from 0.06682 to 0.06732), it was accompanied by a substantial cost increase, leading to an ICER of CNY 1,785,821.36 per QALY, exceeding the WTP threshold and significantly weakening its cost-effectiveness advantage. These results suggest that in resource-limited settings, the tDCS plus drug strategy offers a better balance between health benefits and cost control, potentially representing a more cost-effective option.

Table 2.

Stepwise ICERs of the four strategies ordered by increasing total cost

Stepwise ICERs comparison
Strategy	Cost, CNY	QALYs	Incremental		ICER,ΔCNY/QALY
			Cost,ΔCNY	QALYs
Drug	2,235.32	0.04995	-	-	-
rTMS+Drug	3,132.66	0.05787	897.34	0.00793	113,173.77
tDCS+Drug	4,431.69	0.06682	1,299.03	0.00895	145,153.94
Combined therapy	5,319.70	0.06732	888.01	0.00050	1,785,821.36

Sensitivity analysis

DSA results

Because tDCS remains exploratory in routine Chinese practice, we focused the DSA on comparing the rTMS group and tDCS group. Analysis of the top ten parameters with the greatest influence on the ICER (see Fig. 2) showed that within the comparative framework of the two groups, the cost of neurostimulation therapy in the tDCS group was the most sensitive parameter affecting the ICER. It was also significantly influenced by the utility value for the remission state in the rTMS group and the cost of neurostimulation therapy in the rTMS group. Even when these parameters varied within a broad and plausible range, the ICERs of all intervention strategies remained below the pre-set WTP threshold, indicating that the base-case analysis conclusions are highly robust within reasonable parameter fluctuations.

Fig. 2.

Shows the tornado plot of the unidirectional sensitivity analysis of the ICER from the tDCS group to the rTMS group. ICER, incremental cost-effectiveness ratios; WTP, willingness-to-pay; QALY, quality-adjusted life-year; rTMS group, rTMS plus Drug; tDCS group, tDCS plus Drug

PSA results

The PSA results, presented via the CEAC, showed that when the WTP threshold was low, drug therapy was more economical for patients; however, as the WTP level increased, the probability of the tDCS group being cost-effective gradually became more prominent, reinforcing its potential advantage in cost-effectiveness. For details, see Fig. 3.

Fig. 3.

Shows the cost-effectiveness acceptability curves of the four treatment strategies

The incremental cost-effectiveness scatter plot (Fig. 4) showed that most simulated results fell in the first quadrant, indicating that the tDCS group had higher efficacy but also higher costs. The majority of data points were located below the WTP threshold line, further supporting the cost-effectiveness of tDCS.

Fig. 4.

Incremental cost-effectiveness scatter plot for probabilistic sensitivity analysis

Scenario analysis

When the model time horizon was extended to 3 months or 6 months, all strategies involving neuromodulation showed a decreasing trend in ICER, and the results remained consistent with the base-case findings, indicating that all were cost-effective. Notably, when comparing the tDCS group to the rTMS group, the cost-effectiveness advantage of tDCS became more pronounced over time, with the ICER decreasing from CNY 113,993.80 per QALY to CNY 81,462.18 per QALY, well below the pre-set WTP threshold. In summary, the extended time horizon scenario analyses further validated the robustness and relative advantage of tDCS in terms of cost-effectiveness. For details, see Tables S6–S7.

Discussion

This study is the first to systematically evaluate the cost-effectiveness of rTMS, tDCS, and their combined use with antidepressants for patients with major depressive disorder in China, drawing upon randomized controlled trial efficacy data and localized cost parameters. The objective was to generate evidence to inform the optimization of clinical treatment pathways and the rational allocation of healthcare resources. Baseline analyses indicated that all three neurostimulation-augmented strategies yielded greater QALY gains compared with antidepressant therapy alone, thereby demonstrating a degree of cost-utility advantage. Moreover, stepwise ICER comparisons suggested that, under resource-constrained conditions, tDCS combined with pharmacotherapy may represent the most economically favorable option.

The DSA results further validated the robustness of the base-case analysis. Across varying parameter ranges—including utility values, cost parameters, and transition probabilities—the ICER consistently remained below the WTP threshold, indicating high reliability of the study findings. Among all parameters, the utility value for the remission state in the rTMS group, the cost of tDCS treatment, and the utility value for the MDD state in the rTMS group had the most significant impact on the ICER. It should be noted, however, that clinical utility values, which reflect the effectiveness of treatment strategies, are objective parameters that are difficult to modify through clinical or policy interventions. In contrast, the treatment cost of tDCS represents a key adjustable factor influencing its economic outcomes. Currently, tDCS is not widely covered by medical insurance in China, and patients usually bear the expenses by themselves, which limits its clinical application due to the relatively high cost per session [27]. As a promising non-invasive brain stimulation strategy, further efforts are needed to establish standardized and optimal stimulation protocols in clinical practice [28]. Nonetheless, domestic procurement data indicate that tDCS devices are generally simpler and more economical than rTMS systems, and offer advantages such as rapid onset and lower costs [29, 30].

Previous epidemiological studies in China have demonstrated that depression is more prevalent among women, unemployed individuals, those with disrupted marital status, and people with social dysfunction — groups that often experience socioeconomic disadvantages [31, 32].For this reason, we adopted a wider WTP range of 1–3 times per capita GDP to better reflect heterogeneous affordability across subgroups. Notably, our PSA results indicated that the tDCS plus drug therapy strategy still maintained a considerable probability of being cost-effective even at the lower bound of the WTP threshold (1× per capita GDP). This suggests that the economic advantage of combined neuromodulation therapy is reasonably robust across varying levels of willingness to pay, although its relative advantage becomes more pronounced as the threshold increases.

Scenario analysis results further supplemented the base-case findings. When the model time horizon was extended to 3 months and 6 months to simulate consolidation-phase interventions, all adjunctive strategies showed increased QALYs, and the advantage of the tDCS group in incremental QALYs became more pronounced, with its ICER consistently remaining below the WTP threshold. This indicates that tDCS is not only cost-effective in the acute phase but also demonstrates stronger economic efficiency in long-term consolidation treatment, supporting its inclusion in consolidation-phase interventions in clinical practice, especially for patients with inadequate drug response or high relapse risk [5, 33, 34].

Our model was specifically developed for the outpatient acute-phase treatment setting, complementing the existing evidence base for severe depression care that includes hospitalization, electroconvulsive therapy (ECT), and intensive multidisciplinary interventions. This study provides targeted economic evidence supporting the value of rTMS and tDCS as effective adjunctive treatment options for outpatients with inadequate response to pharmacotherapy or those unsuitable for ECT [35], thereby offering a clinically meaningful economic reference for individualized depression management in routine outpatient practice.

This study has several notable strengths: First, as the first pharmacoeconomic study in China based on RCT data to systematically evaluate rTMS, tDCS, and their combination as adjunctive treatments for MDD, it addresses the gap in localized economic evaluations of non-invasive neuromodulation technologies and provides direct and critical evidence on the cost-effectiveness of tDCS. Second, although the model time horizon focused on the acute phase, the construction of a reasonable Markov model that integrated direct medical costs and indirect social costs from a societal perspective effectively enhanced the relevance of the evaluation results to real-world healthcare decision-making, increasing their practical reference value. Third, in addressing uncertainty, in addition to conducting deterministic sensitivity analysis (DSA) and probabilistic sensitivity analysis (PSA), multi-scenario analyses were performed to examine the impact of changes in treatment frequency during the consolidation phase, thereby multi-dimensionally verifying and strengthening the robustness and generalizability of the conclusions.

Nevertheless, several limitations should be acknowledged. First, the utility values came from non‑Chinese EQ‑5D data, which may not fully reflect Chinese health preferences; however, sensitivity analyses showed that variations in these utilities did not materially affect the ICERs. Second, the three‑state model (MDD, remission, death) did not capture partial response or treatment discontinuation, but this structure was chosen based on available evidence to ensure model stability. Third, cost parameters were drawn from a single institution, so their national representativeness should be interpreted wit caution—yet sensitivity analyses again confirmed robustness. Finally, the 4‑week horizon does not fully reflect the chronic nature of MDD; scenario analyses over 3 and 6 months partly address this issue, though longer‑term studies are still warranted.

In summary, within the Chinese healthcare system, rTMS, tDCS, and their combination as adjunctive therapies to pharmacotherapy for MDD all demonstrate certain cost-effectiveness advantages compared to drug therapy alone. Among these, tDCS combined with drugs not only shows significant economic efficiency but also exhibits more prominent benefits when the treatment duration is extended, providing clear direction for implementing cost-effective adjunctive treatment strategies.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

MDD

Major Depressive Disorder

rTMS

Repetitive Transcranial Magnetic Stimulation

tDCS

Transcranial Direct Current Stimulation

QALY

Quality-Adjusted Life Year

ICER

Incremental Cost-Effectiveness Ratio

WTP

Willingness-To-Pay

CNY

Chinese Yuan

Author contributions

W.L. conceived and designed the study, performed the literature search and data extraction, and drafted the initial manuscript. Q.L. contributed to the study design, conducted data analysis and interpretation, and critically revised the manuscript. Both W.L. and Q.L. contributed equally to this work and should be considered co-first authors. All authors read and approved the final manuscript.

Funding

China Medical Board (CMB) (No.19–343).

Data availability

The data analysed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Ethical approval and consent to participate are not applicable because this study did not include any patient-level data. The health economic model was populated with data from the literature and simulation. Therefore, according to the China Guidelines for Pharmacoeconomic Evaluations (2020), such modeling studies do not require separate ethical approval or informed consent.All methods were performed in accordance with the principles of the Declaration of Helsinki and the national guidelines for conducting economic evaluations, as outlined here: https://www.cpa.org.cn/cpadmn/attached/file/20201203/1606977380634185.pdf.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.