Cost-effectiveness of dienogest compared to gonadotropin releasing hormone agonists for the management of endometriosis in Vietnam

The cost-effectiveness analysis was conducted from the healthcare payer perspective. Therefore, only medical costs were included in the model, including drug, outpatient visiting, examination fees, and surgery costs. Drug costs were based on publicly available 2023 drug procurement data from the Drug Administration of Vietnam. Dienogest (2 mg/day) was compared to triptorelin (3.75 mg intramuscular injection every 28 days), goserelin (3.6 mg subcutaneous injection every 28 days), and leuprorelin (3.75 mg subcutaneous injection every 28 days) for six-month treatment period. The cost of laparoscopic surgery was derived from micro-costing, including laboratory tests, imaging diagnostics, medication, surgical procedure, medical supplies, and hospital stay. All costs were converted to US dollars at the average exchange rate in 2023 (1 USD = 23,836 VND) [ 17 ]. Table 1 Input parameters for cost-effectiveness models Parameter Value One-way a Distribution Reference Surgery Percent of laparoscopic surgery 0.04523 0.04071–0.04975 Triangular [ 18 ] & Supplementary materials Recurrence of dysmenorrhea after surgery 0.05426 0.04883–0.05969 Triangular [ 19 ] & Supplementary materials Recurrence of pelvic pain after surgery 0.16934 0.15241–0.18627 Triangular [ 20 ] & Supplementary materials Mortality due to laparoscopic surgery 0.00003 Invariant Uniform [ 8 ] Mortality due to any cause 0.000204 Invariant Triangular [ 21 ]& Supplementary materials Dysmenorrhea Percentage of patients pain-free after 6 months of dienogest treatment 0.82 0.74–0.90 Triangular Supplementary materials Percentage of patients pain-free after 6 months of leuprorelin treatment 0.90 0.81–0.99 Beta [ 12 ] Percentage of patients pain-free after 6 months of triptorelin treatment 0.95 0.86-1.00 Triangular [ 22 ] Recurrence of pain after dienogest treatment 0.01377 0.01239–0.01515 Triangular Supplementary materials Recurrence of pain after leuprorelin treatment 0.07884 0.07092–0.08668 Triangular Supplementary materials Recurrence of pain after triptorelin treatment 0.23646 0.21281–0.26011 Triangular Supplementary materials Pelvic pain Percentage of patients pain-free after 6 months of leuprorelin treatment 0.6 0.36–0.44 Beta [ 12 ] Percentage of patients pain-free after 6 months of dienogest treatment 0.68 0.61–0.75 Beta Supplementary materials Percentage of patients pain-free after 6 months of triptorelin treatment 0.68 0.61–0.75 Beta Supplementary materials Percentage of patients pain-free after 6 months of goserelin treatment 0.67 0.60- 0.737 Beta [ 23 ] Recurrence of pain after dienogest treatment 0.01377 0.01239–0.01515 Triangular Supplementary materials Recurrence of pain after leuprorelin treatment 0.07884 0.07092–0.08668 Triangular Supplementary materials Recurrence of pain after triptorelin treatment 0.0424 0.03816–0.04664 Triangular Supplementary materials Recurrence rate of pain after goserelin treatment 0.1833 0.16497–0.20163 Triangular Supplementary materials Utility Patients with dysmenorrhea 0.720 0.65–0.79 Supplementary materials, Table S2 Patients relieved from dysmenorrhea 0.905 0.815–0.996 Uniform [ 24 ] Patients with pelvic pain 0.63 0.57–0.69 Supplementary materials, Table S2 Patients relieved from pelvic pain 0.905 0.815–0.996 Uniform [ 24 ] Costs (USD) Cost of 6-month drug treatment [ 25 ] Dienogest 323.2 290.8–355.5 Normal Leuprorelin 680.5 612.5–748.6 Normal Goserelin 646.5 581.8-711.1 Normal Triptorelin 643.6 579.3–708.0 Normal Outpatient visiting 1.8 Invariant Invariant Circular 21-2023/TT-BYT, Circular 22-2023/TT-BYT Transvaginal ultrasound 8.3 Invariant Invariant Circular 21-2023/TT-BYT, Circular 22-2023/TT-BYT Injection 0.6 Invariant Invariant Circular 21-2023/TT-BYT, Circular 22-2023/TT-BYT Endometriosis surgery 406.6 365.9–447.3 a Normal Micro-costing a 10% of base case Input parameters for cost-effectiveness models Supplementary materials, Table S2 Supplementary materials, Table S2 a 10% of base case The primary outcome measurements were total costs, total QALys, and incremental cost-effectiveness ratio (ICER). ICER is defined as incremental costs divided by incremental QALYs. Dienogest was considered cost-effective is the ICER less than the cost-effectiveness threshold of one time the gross domestic product (GDP) per capita of Vietnam in 2023, equivalent to $4,275 per additional QALY [ 26 ]. Dienogest was considered dominant if the total costs were lower and QALYs were higher than other treatments. A 3% annual discount rate was applied to costs and QALYs, in line with the Vietnam guidelines for pharmacoeconomic evaluations. We reported this study following the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement [ 27 ].

The study adopted the age-specific mortality rate for women aged 30–35 as main population in endometriosis typically fell within the age range of 30–35 years. Transition probabilities were adjusted to reflect the three-month cycle length of the Markov model, with details presented in Supplementary materials. The mortality rate associated with laparoscopic surgery was derived from the study by Chapron et al. [ 8 ]

This study employed a hybrid decision tree and Markov model to estimate the total costs and total quality-adjusted life years (QALYs) associated with the use of dienogest and GnRH-a drugs (triptorelin, leuprorelin, goserelin) in endometriosis patients with dysmenorrhea and pelvic pain, constructing separate sub-models for each condition (Fig.  1 A). Patients included in the model were women of reproductive age diagnosed with endometriosis and presenting with either dysmenorrhea or pelvic pain, eligible for medical therapy with dienogest or a GnRH-a. Exclusion criteria included pregnancy or breastfeeding, amenorrhea within three months prior to treatment, a primary indication for surgical treatment of endometriosis, or prior use of GnRH-a within six months. The decision tree was used to represent the initial six-month treatment phase, during which patients receiving dienogest or a GnRH-a could either achieve pain relief or continue to experience pain. Following this phase, patients entered the Markov model, which was designed to reflect the chronic and recurrent nature of endometriosis-related pain. In accordance with endometriosis treatment guidelines, patients who do not respond to medical treatment may undergo surgical intervention. The Markov component included health states of maintained pain relief, pain recurrence, post-surgery, and death. Surgical patients were also at risk of surgery-related mortality. Patients cycled through these states every three months. Those who initially achieved pain relief could remain pain-free, relapse, or die from any cause, while patients with recurrent pain could undergo surgery (Fig.  1 B). The Markov model used a three-month cycle, and ran for a total duration of two years to capture the long-term effects of treatment. Healthcare utilization was calculated across the 2-year time horizon based on the health states patients occupied in each cycle. Total expected costs per patient were calculated by combining costs of treatments (dienogest or GnRH-a drugs) and surgical costs. Fig. 1 ( A ) Cost-effectiveness model; ( B ) Markov model M: Markov ( A ) Cost-effectiveness model; ( B ) Markov model M: Markov

For dysmenorrhea, dienogest had the lowest total cost ($363.3) and highest QALYs (1.74). Compared with dienogest, triptorelin ($739.3; 1.62 QALYs; ICER −$3,292.2/QALY) and leuprorelin ($744.0; 1.70 QALYs; ICER −$11,454.2/QALY) were more costly and less effective. Similarly, for pelvic pain, dienogest ($381.5; 1.64 QALYs) dominated triptorelin ($720.4; 1.60 QALYs; ICER −$10,918.5/QALY), leuprorelin ($773.4; 1.54 QALYs; ICER −$4,299.5/QALY), and goserelin ($753.1; 1.49 QALYs; ICER −$2,608.7/QALY). Table 2 Cost-effectiveness of dienogest and GnRHa in endometriosis Drug Total Cost ($) Δcosts vs. dienogest($) Total QALYs ΔQALYs vs. dienogest ICER ($/QALY) Treatments for dysmenorrhea Dienogest (reference) 363.3 0.00 1.74 0.00 Triptorelin 739.3 376.0 1.62 −0.11 −3,292.2 Leuprorelin 744.0 380.7 1.70 −0.03 −11,454.2 Treatments for pelvic pain Dienogest (reference) 381.5 0.00 1.64 0.00 Triptorelin 720.4 338.9 1.60 −0.04 −10,918.5 Leuprorelin 773.4 391.9 1.54 −0.09 −4,299.5 Goserelin 753.1 371.6 1.49 −0.14 −2,608.7 Dienogest is the new treatment and used as the reference; ICERs were calculated in TreeAge using full-precision model outputs. Minor differences may occur compared to ΔQALYs/Δcosts Cost-effectiveness of dienogest and GnRHa in endometriosis Dienogest is the new treatment and used as the reference; ICERs were calculated in TreeAge using full-precision model outputs. Minor differences may occur compared to ΔQALYs/Δcosts A one-way sensitivity analysis was conducted to assess key parameters influencing the ICER of dienogest compared to GnRH-a drugs for dysmenorrhea. The results highlighted that the probability of dysmenorrhea resolution after six months of treatment and utility values of dysmenorrhea were the most influential factors affecting ICER (Supplementary materials, Figures S4-S5). PSA demonstrated that dienogest has 99.9% probability of being cost-effective at a willingness-to-pay threshold of one GDP per capita when compared to leuprorelin and triptorelin (Fig.  2 ). Fig. 2 Sensitivity analyses in dysmenorrhea. (A) Cost-effectiveness acceptability curve at different willingness-to-pay thresholds; (B) Cost-effectiveness plane of dienogest vs. triptorelin in dysmenorrhea; (C) Cost-effectiveness plane of dienogest vs. leuprorelin in dysmenorrhea Sensitivity analyses in dysmenorrhea. (A) Cost-effectiveness acceptability curve at different willingness-to-pay thresholds; (B) Cost-effectiveness plane of dienogest vs. triptorelin in dysmenorrhea; (C) Cost-effectiveness plane of dienogest vs. leuprorelin in dysmenorrhea The cost-effectiveness results comparing dienogest with goserelin, leuprorelin, and triptorelin for pelvic pain management are presented in Table  2 . Dienogest remained the dominant strategy, incurring lower costs ($339 - $392 less) while generating more QALYs (0.04–0.14 additional QALYs) compared to GnRH-a drugs. One-way sensitivity analysis results indicated that the probability of pain resolution after treatment with dienogest or GnRH-a drugs at six months, drug costs, and utility were the most influential parameters affecting ICER (Supplementary materials, Figures S6-S8). The results of PSA showed that dienogest has the highest probability (> 99%) of cost-effectiveness for pelvic pain treatment compared to GnRH-a (Fig.  3 ). Fig. 3 Sensitivity analyses in pelvic pain. (A) Cost-effectiveness acceptability curve at different willingness-to-pay thresholds; (B) Cost-effectiveness plane of dienogest vs. triptorelin in pelvic pain; (C) Cost-effectiveness plane of dienogest vs. leuprorelin in pelvic pain. (D) Cost-effectiveness plane of dienogest vs. goserelin in pelvic pain Sensitivity analyses in pelvic pain. (A) Cost-effectiveness acceptability curve at different willingness-to-pay thresholds; (B) Cost-effectiveness plane of dienogest vs. triptorelin in pelvic pain; (C) Cost-effectiveness plane of dienogest vs. leuprorelin in pelvic pain. (D) Cost-effectiveness plane of dienogest vs. goserelin in pelvic pain

Meta-analyses from the published systematic review [ 9 ] were performed to estimate percentage of patients with pain-free (dysmenorrhea and pelvic pain) after six-month treatments. The systematic review identified four relevant studies [ 10 – 13 ] examining dysmenorrhea reduction and pelvic pain after 6 months of dienogest treatment while two trials [ 14 , 15 ] informed the estimates for triptorelin. Efficacy of leuprorelin in managing dysmenorrhea and pelvic pain in endometriosis was derived from the previous publication [ 12 ]. However, due to a lack of data on dysmenorrhea relief rates after six months of goserelin use, a cost-effectiveness analysis between dienogest and goserelin for dysmenorrhea management was not feasible. The pooled percentage of patients pain-free after six months was 0.82 in dienogest treatment group, 0.90 in leuprorelin treatment, and 0.95 in triptorelin treatment for dysmenorrhea. The respective values were 0.68 in dienogest treatment, 0.60 in leuprorelin treatment, 0.68 in triptorelin treatment, and 0.67 in goserelin treatment for pelvic pain (Table  1 , (Supplementary materials, Figures S1-S3). Transition probabilities are detailed in Supplementary materials Table S1, with adjustments made for the three-month cycle length. Utility values are extrapolated from SF-36-derived estimates published by Ara et al., [ 16 ] with calculation details available in Supplementary materials, Table S2.

The results indicated that dienogest was a dominant treatment option for dysmenorrhea and pelvic pain caused by endometriosis compared to GnRH-a drugs such as triptorelin and leuprorelin (better health outcomes and lower total costs). This advantage persisted across both base-case and sensitivity analyses, showing the robustness of these findings. Additionally, the oral administration of dienogest presents a further advantage over the injectable forms of GnRH-a drugs, enhancing its clinical and practical appeal. The National Health Insurance scheme covers four medications for endometriosis treatment, including danazol, triptorelin, leuprorelin, and goserelin in Vietnam. However, danazol is rarely used due to its side effects, leaving GnRH-a drugs as the most effective treatment options. While contraceptives and NSAIDs are often used initially, their off-label status led to their exclusion from this model. Therefore, the study evaluated dienogest’s cost-effectiveness in comparison to these reimbursed therapies. Our cost-effectiveness study based on systematic reviews and meta-analyses to ensure a high level of available evidence. Using a health economic modeling approach, this study builds upon prior research by incorporating surgical intervention after failed medical therapy, making it a more comprehensive representation of treatment pathways and aligning with Vietnam’s national treatment guidelines. Furthermore, the model captures the two most common symptoms of endometriosis, dysmenorrhea and pelvic pain. Endometriosis is a common gynecological condition but economic evaluations of treatment options remain limited, relying on multiple assumptions. To date, only one cost-effectiveness study compared dienogest and GnRHa (goserelin) after surgery in endometriosis patients in China [ 28 ]. This study found that dienogest increased QALYs by 0.02 while reducing costs by 7,274 yuan. Additionally, probabilistic sensitivity analysis in that study demonstrated that dienogest was cost-effective with 100% certainty at three times China’s GDP per capita. These findings support our conclusion that dienogest is a dominant treatment option over GnRH-a drugs from a healthcare payer perspective in managing endometriosis-related pain. Moreover, while GnRH-a therapies such as triptorelin, leuprorelin, and goserelin are effective in relieving pain, they are associated with significant hypoestrogenic side effects, including hot flushes, decreased bone mineral density, mood disturbances, and vaginal dryness. These adverse events not only impact patients’ quality of life but may also reduce adherence, especially in long-term use. In contrast, dienogest has a more favorable safety profile, with fewer hormonal side effects, which can contribute to better tolerability and sustained use. In addition, the injectable administration of GnRH-a drugs requires regular clinic visits for intramuscular or subcutaneous injections, posing a barrier to access, particularly in rural or resource-limited settings. Dienogest, as a once-daily oral tablet, offers greater convenience and autonomy for patients, potentially improving treatment adherence and satisfaction. Our study has several limitations. First, the model does not account for the transition to alternative medical treatments after initial therapy failure due to limited data availability and the variability of clinical decision-making, which is often guided by physician experience rather than standardized protocols. Second, we did not incorporate a disutility penalty for adverse events due to a lack of data, as these events were generally mild and resolved upon treatment discontinuation. Additionally, our cost-effectiveness evaluation was confined to a two-year time horizon, which may not fully capture long-term endometriosis recurrence and pain relapse. Future research with extended follow-up data would allow longer time horizons to be modelled more reliably. Research on pain recurrence after treatment discontinuation is scarce, with small sample sizes limiting the reliability of existing estimates. The study also relies on data from various sources, leading to potential inconsistencies in patient populations—a common challenge in economic modeling studies. Lastly, our study only considered direct medicine and surgery costs and did not account for readmission, adverse event costs, or indirect costs such as lost productivity and transportation expenses. Given that the economic burden of endometriosis largely stems from indirect costs [ 29 , 30 ]future research should adopt a broader societal perspective to provide a more comprehensive assessment of the economic impact of different treatment strategies. Given the favorable cost-effectiveness profile of dienogest, particularly its dominance over GnRH-a therapies, policymakers and payers in Vietnam may consider including dienogest in the national health insurance reimbursement list. Such inclusion could improve accessibility, reduce financial burden, and promote adherence among patients with endometriosis.

This study provided compelling evidence that dienogest was a cost-effective alternative to GnRH-a drugs for treating dysmenorrhea and pelvic pain due to endometriosis. By reducing treatment costs and improving patient quality of life, dienogest emerges as the preferred therapeutic option under Vietnam’s healthcare payer perspective. The study employed a Markov model to simulate treatment outcomes and compared dienogest with multiple GnRH-a therapies, proving a broad comparison of current treatment options. Sensitivity analyses and scenario analyses were performed to explore uncertainty of input parameters and assess the robustness of results. Long-term outcomes required modeling assumptions beyond the available data. Some model inputs were derived from international sources, and the analysis excluded indirect costs such as productivity losses, potentially underestimating the total economic impact. The study employed a Markov model to simulate treatment outcomes and compared dienogest with multiple GnRH-a therapies, proving a broad comparison of current treatment options. Sensitivity analyses and scenario analyses were performed to explore uncertainty of input parameters and assess the robustness of results. Long-term outcomes required modeling assumptions beyond the available data. Some model inputs were derived from international sources, and the analysis excluded indirect costs such as productivity losses, potentially underestimating the total economic impact.

Given that input parameters are primarily sourced from literature and previous studies, a one-way sensitivity analysis was conducted to assess the uncertainty of individual parameters and their impact on the ICER. The analysis examined parameter variations of ± 10% relative to baseline values, with the most influential parameters presented in a Tornado diagram. A probabilistic sensitivity analysis (PSA) was conducted by varying input parameters according to predefined probability distributions (Supplementary materials, Table S3). Monte Carlo simulations ( n  = 10,000) assessed the likelihood of achieving cost-effectiveness at different willingness-to-pay thresholds.

Endometriosis is a chronic gynecological condition defined as the presence of endometrium-like tissue outside the uterus cavity, affecting approximately 10% of women of reproductive age [ 1 ]. Endometriosis is associated with a high prevalence of dysmenorrhea (40-60%), subfertility (21-74%), and/or pelvic pain (71-87%), significantly impacting patients’ quality of life [ 2 ]. Beyond its physical manifestations, endometriosis also has substantial psychological consequences, contributing to depression, anxiety, and impaired social relationships [ 3 ]. Laparoscopic surgery is a common treatment option; however, due to concerns about recurrence and the need for repeated procedures, many patients prefer non-surgical alternatives for long-term disease management. Several hormonal therapies, including oral contraceptives, progestins, and gonadotropin-releasing hormone analogs (GnRHa), have been proposed for the treatment of endometriosis-related pain. Among these, GnRHas are highly effective in pain relief but are associated with significant hypoestrogenic side effect [ 4 ]. Dienogest is a synthetic progestogen with a dual pharmacological profile, combining properties of 19-norprogestogens and 17a-hydroxyprogesterone derivatives [ 5 ]. Dienogest has demonstrated comparable efficacy to GnRHa in reducing endometriosis-related pain while causing fewer hypoestrogenic side effects [ 6 , 7 ] Despite its clinical benefits, the cost-effectiveness of dienogest compared to GnRHa remains an essential consideration, particularly in resource-limited settings such as Vietnam. Therefore, the study aimed to assess the cost-effectiveness of dienogest compared to GnRHa in the management of endometriosis in Vietnam.

Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2