Cost-effectiveness and budget impact analysis of a patient visit support system for blindness reduction in Japanese patients with glaucoma

Abstract

Aim

To evaluate the cost-effectiveness (CE) and budget impact (BI) of introducing a patient visit support system (ACT Pack) along with standard of care (SoC), in glaucoma treatment in Japan.

Methods

A Markov model was designed to estimate the CE and BI of introducing the support system from Japanese payer and governmental perspective, respectively. Inputted data for CE and BI analysis were referred from published literature and based on medical specialists’ inputs. Base case scenario for CE considered the support system cost of 30,000 yen per patient per year and a time horizon of 45 years. BI analysis compared the financial impact due to introduction of support system with SoC compared to SoC alone scenario on Japanese healthcare system with a time horizon of 20 years.

Results

The base case of CE analysis showed the incremental cost per quality-adjusted life years (QALYs) gained with the support system was 3,241,729 yen/QALY (29,470 USD/QALY). The sensitivity analysis showed that the probability of this support system being cost-effective at a threshold of 5 million yen/QALY (45,455 USD/QALY) was 53.26%. Blindness reduction after introduction of this support system was 8.68%. The BI analysis showed that the introduction of support system will lead to a cumulative cost savings of 1,132 billion yen (10 billion US dollars) for Japanese healthcare system over 20 years of time horizon.

Limitation

Due to paucity of similar comparative studies, some assumptions were made based on medical specialists’ inputs. Death status was not considered in the analysis.

Conclusion

Introduction of this support system with SoC is cost-effective and will lead to blindness reduction in Japanese patients with glaucoma. Over a 20 year period, it will lead to an overall cost savings of 1,132 billion yen (10 billion US dollars) for the Japanese healthcare system.

Keywords:

• Glaucoma
• cost-effectiveness
• budget impact
• blindness reduction
• Japan
• patient visit support system
• adherence

JEL CLASSIFICATION CODES:

• D61
• H51
• I00
• I15

Introduction

Glaucoma is one of the leading causes of acquired irreversible blindness worldwide¹. It is a chronic ocular disease broadly classified into open-angle glaucoma (OAG) and angle-closure glaucoma^1,2. The OAG is further categorized into primary OAG (POAG) and normal-tension glaucoma (NTG)². Global burden of glaucoma for population aged 40–80 years is 3.5%, i.e. approximately 64.3 million people³. A cross-sectional epidemiologic study reported a prevalence of 3.9% for POAG in Japanese population older than 40 years⁴. The study further reported the prevalence of NTG to be 3.6%, which was comparatively higher when compared to the worldwide statistics for NTG prevalence i.e. 0.17 − 0.67%^4,5.

Visual impairment due to glaucoma exerts significant economic consequences on both society and the payer. The available evidence from Japan reported that visual impairment accounted for 8785 yen billion of the economy, equivalent to 1.7% of country’s gross domestic product⁶. Direct health system costs and other financial costs (i.e. productivity losses, caretakers' burden, and efficiency losses from welfare payments and taxes) were 1338 billion yen (11 billion US dollars) and 1583 billion yen (13 billion US dollars), respectively⁶. The Japan Ophthalmologists Association reported visual disorder to account for a total cost of 2.9 trillion yen per year⁷.

Treatment options for glaucoma include medications, laser treatment and surgery. The therapeutic modality is selected on the basis of patient’s disease stage and type. It is also to consider the efficacy and adverse event, age, disease severity, treatment persistence and economic burden to select the treatment option. The first line treatment is medications to reduce eye pressure in POAG and NTG. Whilst laser treatment or surgery are considered for PACG as first line treatment². Being an asymptomatic disease in its early stages, low adherence to anti-glaucoma ophthalmic medications is frequently observed in patients with glaucoma⁸. Previously published studies have also reported low persistence to glaucoma medications among the patients^9,10. Poor treatment adherence can, in turn, increase the risk of progression of visual field defects and risk of blindness. Treatment adherence takes into account not only the medication adherence but also the continuous visits to an ophthalmic institute to obtain the medication (i.e. persistent patient visits) and the skill to instill one drop into the targeted eye (i.e. technical accuracy). Owing to the lack of a prescription refilling system in Japan, patients lose access to their glaucoma treatment if they fail to visit a medical institute. This may make patient dropout the main reason of non-optimal treatment adherence in Japan¹¹.

To achieve optimal adherence in patients with glaucoma, a comprehensive patient visit support system (ACT Pack) was developed. This support system is a package of patient support system that consists of (1) materials/tools which enables patients to ascertain their conditions and schedule for treatment, (2) video materials which help them understand their disease and risks better, and (3) appointment and reminder tools to encourage them to visit to clinic or hospital regularly. This support system was designed to be used as a communication tool between physicians or healthcare professionals and patients. The use of this support system is expected to maintain and improve patients' motivation to visit the hospital. Real-world evidence showed that introduction of this support system improved the persistence of patient visits by approximately 50–90%¹². However, the economic impact of this support system has not been studied in Japan. Thus, our study aimed to evaluate the cost-effectiveness (CE) and budget impact (BI) of introducing this support system with standard of care (SoC) in the treatment of Japanese patients with glaucoma.

Methods

Model structure and perspectives

A Markov model implemented in Python (version 3.7, Python Software Foundation, Beaverton, USA) was performed to estimate CE of introducing the support system with SoC in the treatment of Japanese patients with glaucoma. Incremental CE ratio (ICER) for introducing this support system with SoC versus SoC alone was assessed and the model was applied to Japanese clinical treatment algorithm. Blindness reduction was also calculated, which was defined as the ratio of absolute risk reduction and population with blindness without the support system. Model inputs were validated by medical specialists from Japan through face-to-face interviews.

The CE model comprised of six transition probabilities using severity and treatment. Severity consisted of three disease progression rates (i.e. slow, moderate and fast) which were derived from the severity of Japanese patient with glaucoma. Furthermore, treatment was composed of two disease progression rates reflected from with- or without-treatment to prevent visual field defect site (Figure 1). A patient could transit through five mutually exclusive health states (i.e. early, mild, moderate, severe and blindness), and also within a health state (treatment/no treatment). Transition probability on each health states was dependent on whether the health state was severe and the treatment for glaucoma was continued.

Figure 1. An overview of the model structure, used for the cost-effectiveness analysis. Glaucoma was assumed to develop from early state to blindness. In each state categorized by severity of glaucoma, treatment adherence like non-treatment was taken into account, which impacted the progression or blindness rate.

The BI analysis assessed overall financial impact of introducing the support system for the treatment of patients with glaucoma receiving SoC. The analysis compared the budget for two scenarios: support system with SoC and SoC alone.

The base case analysis for CE and BI were conducted from Japanese payer and governmental perspective, respectively. Direct medical costs were considered only for CE analysis. In BI analysis, in addition to direct medical costs, indirect medical costs such as social welfare cost and dropping tax revenues according to productivity loss were also considered. Though inclusion of indirect medical costs in BI analysis is not desirable because these costs are often non-relevant to the budget holder, except for few cases¹³. However, qualifying under such exceptional cases, our study included these costs as most of the medical and care costs due to visual detection and blindness by glaucoma are covered through social insurance and taxes in Japan. Moreover, the indirect costs such as the productivity loss related to blindness which exerts a negative impact on Japan's finances has been clearly described in the report by Japan Ophthalmologists Association⁷.

For the base case scenario, support system cost of 30,000 yen per patient per year was considered. The Japan specific CE willingness-to-pay (WTP) threshold was 5 million yen per quality-adjusted life year (QALY)^14,15.

Study population, comparators and time horizon

Patients with glaucoma aged 40 years or older, with NTG and/or POAG, and without any treatment history were considered for the analysis. They were divided into three distinct groups based on their disease progression rate: fast, moderate and slow with a percent proportion of 10, 18, and 72%, respectively. The analysis based its comparison on SoC for treatment of glaucoma¹⁶. Base population for BI analysis was 5% of Japanese population aged 40 years, i.e. about 4,000,000.

Given the chronic nature of the disease, a time horizon of 45 years for CE and 20 years for BI analyses were considered, with a cycle length of one year for both the analyses. The time horizon for CE analysis was set based on Japanese life expectancy at 45 years of base-line¹⁷. For BI analysis, the time horizon of 20 year was decided to estimate the BI at the year of 2040 because Japanese government has been trying to address the issue expected in 2040 which becomes not only the peak of medical and care cost expense but also shortage of required healthcare provider¹⁸. Cost and QALYs were discounted at a rate of 2% per annum based on the Japanese guidelines¹⁹.

Model validation

The prevalence rate of blindness in our study were compared for reliability with the rates observed in a population-based study of Olmsted County residents, Minnesota and the Advanced Glaucoma Investigation Study (AGIS)^20,21. The blindness rates reported in Olmsted County study were 13.5% (95% CI 8.8–17.9) at an average age of 81.2 ± 10.3 years and 7.3–18.5% in the AGIS. Additionally, a recent study by Casey et al. which conducted a model analysis related to therapeutic adherence in glaucoma, was further considered to compare the blindness rates with our results²².

Model inputs

Clinical

Clinical inputs used in the CE analysis were disease progression and blindness rate by glaucoma, prevalence and cessation, and utility. Table 1 describes the model parameters. The base case transition probabilities between glaucoma severities for patients receiving SoC with/without the support system were extracted from the literature and assumed based on inputs received from the medical specialists. Transition probabilities were derived from the mean deviation (MD) and MD slope reported in published literature. MD is known as an average of overall deviation values from sensitivity expected for the patient’s age. MD is available to estimate the severity and extent of visual field defects as a simplified descriptor value²³. MD slope was used to confirm the progression of glaucoma in previous studies^24,25. The rates derived from the specified MD slope in three distinct disease progression groups were applied to the four health states before blindness^24,26. While, the progression rate for blindness was obtained from previously published literature^27,28. Chang et al. classified the MD for mild, moderate and severe visual field defects as > −6 dB, ≤ −6 dB but ≥ −12 dB and < −12 dB, respectively²⁹. In our model, the MD started from zero and decreased to minus value based on the derived disease progression rates. Once the MD value exceeded the threshold, there was transition from one health state to the subsequent health state. The MD threshold was −0.5 dB for early to mild state, −6 dB for mild to moderate state, and −12 dB for moderate-to-severe state. For insistence, when the patient who is with fast progression rate and under treatment, the decrement of MD slope is 0.49 dB per every year. The decrement of MD is cumulated yearly. If the MD of the patient in mild state is exceeded −6 dB, the patient proceeds from mild to moderate state. The detection rates of disease progression were established from the Japan Glaucoma Society and based on medical specialists’ opinion.

Table 1. Base case values for clinical, quality of life and cost parameters included in the study model.

Model parameters	Classifications	Base case values	References
Patient proportion
Progression	Fast	10.0%	^16
	Moderate	18.0%
	Slow	72.0%
Valuables for progression (yearly)
MD slope under treatment	Fast	0.49 dB	^24
	Moderate	0.41 dB
	Slow	0.35 dB
MD slope under non-treatment	Fast	0.82 dB	^24^,^26
	Moderate	0.68 dB
	Slow	0.58 dB
Probability to blindness (yearly)
During treatment		1.1%	^27
During non-treatment		2.2%	^27^,^28
Values for detection (yearly)
Detection rate	Early	3.7%	Japan Glaucoma Society
	Mild	10.0%	Medical specialists’ assumption
	Moderate	50.0%
	Severe	99.0%
Values for drop out (yearly)
Drop out under patient visit support system	Early	5.8%	^12
	Mild	5.8%
	Moderate	5.8%
	Severe	0.0%	Medical specialists’ assumption
Drop out under non-patient visit support system	Early	40.0%	^9
	Mild	40.0%
	Moderate	11.6%	Medical specialists’ assumption,^22
	Severe	0.0%	Medical specialists’ assumption
Utilities
Utility in each state	Early	1.00	Medical specialists’ assumption,^24^,^29^,^32
	Mild	0.99
	Moderate	0.94
	Severe	0.65
	Blindness	0.54
Cost (yearly)
Standard of care in each state	Early stage	¥40,000	Medical specialists’ assumption
	Mild	¥65,000
	Moderate	¥90,000
	Severe	¥90,000
	Blindness	¥0
Annual disability basic pension in each state	Severe	¥780,100	^30
	Blindness	¥975,125
Annual disability employee's pension in each state	Moderate	¥589,900
	Severe	¥890,000
	Blindness	¥1,060,000
Care cost		¥94,080	^7
In-kind benefit		¥139,200
Community care		¥3,827,040
Average salary		¥4,407,000	^31
Income tax		¥453,900

Abbreviations. dB, Decibel; MD, mean deviation.

Patient dropout rates without the support system in early and mild glaucoma stages were derived from Kashiwagi et al. (2014)⁹. The rates were reported for newly diagnosed patients with glaucoma and who were initiating treatment with anti-glaucoma topical medication. The rates for moderate and severe stages were based on medical specialists’ inputs and Leske et al. (2003)²⁸. Dropout rates among those with the support system were referred from the support system real-world study for early, mild and moderate stages, and assumed based on medical specialists’ inputs for severe stage. The dropout from the patient visit to take the general treatment for glaucoma like medication, laser treatment or surgery is considered based on the study to confirm the improvement of patient visit to clinic or hospital by the support system¹².

Cost

Cost inputs considered in the model included costs associated with glaucoma treatments (test, visiting and service cost), welfare cost (annual disability pension, care cost, in-kind benefit and community care) and other cost (average salary and income tax). The treatment cost was based on inputs from medical specialists, while welfare cost and other cost were derived from Japan pension service³⁰, Japan Ophthalmologists Association (2009)⁷, and National tax agency³¹, respectively. Additionally, in the BI analysis, social welfare cost and dropping tax revenues according to productivity loss caused by visual field defect were included. 1 US dollar was converted to 110 Japanese yen.

Utility scores

The utility scores for patients with glaucoma depends on the impact caused by visual field defect. The input across five health states of glaucoma were derived from Rein (2007)³². In the study, Rein developed a function to describe utility losses associated with visual field losses for CE studies. As a result, the best fitting line for the Snellen minimum angle of resolution (MAR)-based estimates yielded the function “Utilities = 0.98991 + 0.0022 · dBs − 0.00080518 · dBs²”, wherein dBs were expressed as an absolute value. Regardless of the three function estimates used, MAR-based estimation resulted in the most conservative estimate of utility losses resulting from visual field losses. Using this Snellen MAR best fitting line, the utility scores in each health state were calculated in our study. The representative MD for the health states were estimated from previous studies and medical specialists’ assumption^24,29,33, and defined as -3dB, -9dB, −22 dB and −25 dB for mild, moderate, severe and blindness states, respectively. Early state was considered as the perfect health state, with a utility score of 1. Calculated utility scores were confirmed by the medical specialists.

Sensitivity analysis

Deterministic and probabilistic sensitivity analysis were performed. In the deterministic sensitivity analysis, a one-way sensitivity analysis was performed to assess the impact of different parameters on the CE analysis results. The changed parameters were dropout rates (±20%), MD slope (± 20%), treatment cost (± 20%), blindness rates (±20%), detection rates (±20%), cost of support system (10,000–50,000 yen) and discount rate (0–4%)¹⁹. Moreover, to check the robustness of the CE results, a Monte Carlo multivariate sensitivity analysis was performed. A total of 10,000 iterations were used in the simulation. Since the south-west (SW) quadrant is not considered to provide in clinical setting, probabilistic sensitivity analysis was performed²⁷ by excluding the region of SW quadrant due to being lower than the WTP threshold³⁴.

Result

Model validation

Considering 40 years as the age for glaucoma onset, prevalence of blindness using this support system were 0.29% after 10 years from the age of onset, 2.1% after 20 years from age of onset, 5.67% after 30 years from the age of onset, 10.63% after 40 years from the age of onset and 13.48% after 45 years from the age of onset, respectively. Upon comparing the scenarios of SoC with and without the support system, the blindness reduction rate upon introduction of the support system with SoC was 8.68%.

Compared results from our study with two epidemiological studies, which the blindness rates reported in Olmsted County study were 13.5% (95% CI 8.8–17.9) at an average age of 81.2 ± 10.3 years and 7.3–18.5% in the AGIS^20,21. In our study, the prevalence of blindness at 80 years of age was 10.63%, which was very much in line with the values reported in aforesaid studies.

Cost-effectiveness analysis

The ICER of the CE analysis was compared to the WTP threshold of 5 million yen per QALY (45,455 USD/QALY) specific to Japan. Table 2 describes the CE of introducing the support system in Japan. For the base case, with the support system cost of 30,000 yen per year for a patient with glaucoma, the ICER was 3,241,729 yen/QALY (29,470 USD/QALY). Upon comparison with Japanese WTP threshold, the support system was found to be cost effective. The ICER continued to remain below the WTP threshold in Japan till a support system cost of 50,000 yen per patient per year (ICER was 4,803,051 yen/QALY or 43,664 USD/QALY).

Table 2. Results of the cost-effectiveness analysis.

Glaucoma patient visit support system cost per patient	¥10,000/year	¥20,000/year	¥30,000/year	¥40,000/year	¥50,000/year
Incremental costs (JPY)	428,923	628,186	827,449	1,026,713	1,225,976
Incremental QALYs	0.26
ICER (JPY/QALY)	1,680,406	2,461,067	3,241,729	4,022,390	4,803,051
Blindness reduction	8.68%
Sensitivity analysis*(Monte Carlo simulation) (%)	56.33	55.09	53.26	52.66%	51.05

Abbreviations. ICER, incremental cost-effectiveness ratio.

*The probability of this support system being cost-effective at the threshold of 5 million yen/QALY.

Sensitivity analysis

Figure 2 describes the result of one-way sensitivity analysis. MD slope without glaucoma treatment had the greatest impact on the ICER in the model, with ICER variation ranging from 2,355,099 yen/QALY to 5,915,224 yen/QALY (21,410 USD/QALY to 53,775 USD/QALY). This was followed by cost of the support system, MD slope with glaucoma treatment and discount rate. Dropout without and with the support system were found to be the fifth and sixth most important parameters to impact the model.

Figure 2. Tornado diagram showing one-way sensitivity analysis targeting ICER.

The probabilistic sensitivity analysis using Monte Carlo simulation showed that the support system cost of 30,000 yen per year for a patient yielded a probability of 53.26% for the support system of being cost-effective compared to SoC only scenario at WTP threshold of 5 million yen/QALY (45,455 USD/QALY) (Figure 3). The probability of being cost-effective remained above 50% until the cost of the support system was 50,000 yen per patient per year (Table 2).

Figure 3. Monte Carlo simulation for probabilistic sensitivity analysis results.

Budget impact analysis

The BI analysis showed that introduction of the support system would lead to a cumulative cost saving trend from the 16th year onwards, with a cost saving of 92,923 million yen (844 million US dollars). With the longer time horizon, the results revealed more cost saving of 1,132 billion yen (10 billion US dollars) over 20 years of time horizon from Japanese healthcare perspective (Figure 4). Comparison of yearly budgetary difference rate showed that the annually cost savings initiated from 12th year (−7%), and the trend continued thereafter till the 20th year of introducing the support system.

Figure 4. Budget impact analysis results. Budget impact shows cumulative values. Difference rate depicts absolute budgetary difference rate in the particular year compared to the BI of the previous year.

Discussion

The results showed that introducing the support system with SoC was cost-effective in the treatment of Japanese patients with glaucoma, with an ICER of 3,241,729 yen/QALY (29,470 USD/QALY). In the CE analysis, if the support system cost per patient per year was increased to 50,000 yen, the ICER was estimated to be 4,803,051 yen/QALY (43,664 USD/QALY) with >50% probability of being cost-effective compared to SoC alone scenario. Blindness reduction post introduction of the support system along with the SoC, compared to SoC alone scenario, was 8.68%.

The sensitivity analysis findings from our study showed that the cost savings would mainly be attributed by MD slope and cost of the support system. The findings also showed that the dropout without the support system parameter was found to have a greater impact over the model than the dropout with the support system. Considering the observation that dropouts can increase the overall cost, the continuation of hospital visits will be an important factor directly related to the glaucoma treatment and its subsequent impact on MD slope. This corroborates with the real-world evidence that the introduction of this support system prevents patient visit dropouts and improves the cumulative persistence rate of patient visits, thereby complementing the its CE shown in this economic analysis¹². As the limitation, the parameter for utility was not confirmed because their parameters were generated by the inherent values of MD defined in each state and the formula to calculate QALYs.

In terms of BI, introduction of this support system across Japan is expected to lead a cost saving over a 20 years period. The long-time horizon was used to estimate the BI effectively since glaucoma is a chronic disease with slow progression³⁵. Our study showed, the cumulative cost saving trend is expected to initiate from the 16th year onwards, with a yearly budgetary cost saving difference rate from the 12th year onwards.

As scenario analysis, two scenarios were performed. The time horizon of CE analysis was changed from 45 to 70 years to confirm the CE in longer life time horizon in the first scenario analysis. When the scenario was conducted, the general mortality was also considered to adjust the death caused by longer time horizon. The mortality was reflected from Japanese mortality¹⁷. For the second scenario analysis, the base case result of BI analysis was investigated when the indirect cost is excluded from the cost items. The indirect cost items are dropping tax revenues according to productivity loss and welfare cost (income tax derived from average salary, pension, care cost, in-kind benefit and community care).

In the first scenario analysis, the ICER was 3.45 million yen/QALY (31,385 USD/QALY) which was still less than the threshold in Japan. The result was similar to the base case result. The result indicated the CE was good as well even if the usage of the support system continued for longer time horizon to prevent from treatment dropout. In the second scenario analysis, the result indicated no-cost saving at 20th year. However, this result is one of scenario analysis, the welfare given by the reduction of blindness is contributed to the patient with glaucoma.

Non-adherence to the glaucoma medication has been identified as a major challenge by the physicians worldwide³⁶. Poor adherence not only leads to disease progression and deterioration of patient’s QoL but also exerts significant economic burden^37,38. Moreover, Sleath et al. shows the importance to provide how to administrator glaucoma medications and improve the self-efficacy to glaucoma medication adherence³⁹. Previously published literature showed that increasing the treatment adherence can lead to a reduction in overall economic burden on the healthcare system^22,40. In recent study by Casey et al., non-adherent patients to glaucoma medication were found to have a mean loss of 0.34 QALYs compared to the adherent patients, resulting in a CE ratio of 29,600 USD/QALY gained²². Hence, acknowledging the economic aspects associated with treatment adherence, we evaluated the CE and BI of introducing patient visit support system developed to enhance persistence and adherence to glaucoma treatment.

To ensure the reliability of blindness prevalence rate in the study, results from our study were compared with two epidemiological studies^20,21. In our study, the prevalence of blindness at 80 years of age was 10.63%, which was very much in line with the values reported in aforesaid studies^20,21. Recent study by Casey et al., also estimated the prevalence of blindness (single eye or bilateral) by alive cohort with glaucoma medication adherence i.e. 0% at 60 years of age; 1.2% at 70 years of age; 5.4% at 80 years of age; and 19.0% at 90 years of age. Overall, the blindness prevalence rates from our study were found to be similar with the estimates from Casey et al. The blindness rates from our study differed from rates reported by Casey et al. by few percentages for the initial 10–30 years from the age of onset. The blindness rate at the age of 85 years in our study was 13.48%, which was within the blindness rate reported by Casey et al.²².

To the best of our knowledge, our analysis is the first attempt in reporting the cost-effectiveness and budgetary impact for a tool targeted to prevent visit dropout of patients with glaucoma in Japan. Despite the novelty of this analysis, we acknowledge certain limitations to our study. Due to paucity of similar comparative studies, some of the clinical and cost inputs in the model were based on inputs received from medical specialists. Their assumptions lead the remains of uncertainty which is about 50% of the probability for SoC with the support system of being cost-effective compared to SoC only scenario at WTP threshold of 5 million yen/QALY (45,455 USD/QALY) in Monte Carlo simulation. Moreover, the most of results may not be extrapolated to other countries because the efficacy of the support system was based on Japanese patient and other parameters were applied to Japanese situation.

It should be worth noting that Japan has no prescription refilling system, and hence, the readers should be cautious while drawing direct comparisons on the applicability of this support system in other countries which may not have a similar prescription refilling system. Primary angle-closure glaucoma (PACG) is also a possible cause for blindness with a high probability. However, the disease progression and blindness in PACG is sudden when compared to POAG³⁵. Additionally, the estimated prevalence of PACG in Japanese population aged 40 years or older was 0.5% (95% CI, 0.4 − 0.9%), which is lower than the estimates reported for NTG and POAG¹⁶. Therefore, the progression rate in this study was estimated using NTG and POAG types only to focus on large population in Japan.

Despite of the aforesaid limitations, especially being applied to Japanese patient with glaucoma, the study will add to the existing evidence on non-adherence aspect in glaucoma from an economical perspective and can be considered as a relevant reference for future researchers and healthcare systems.

Conclusions

The introduction of this support system along with the SoC is found to be cost-effective (ICER 3,241,729 yen/QALY (29,470 USD/QALY) compared to SoC alone scenario for the treatment of Japanese patients with glaucoma. It is expected to lead an overall cost saving of 1,132 billion yen (10 billion US dollars) for the Japanese healthcare system after 20 years of its introduction and had a blindness reduction of 8.68%. This study should serve as a fruitful evidence in the decision-making for policy makers, budget formulators and health advisers while choosing the optimum treatment option for patients with glaucoma.

Transparency

Declaration of funding

Sponsorship and article processing charges for this study were funded by Santen Pharmaceutical Co., Ltd., Osaka, Japan.

Declaration of financial/other interests

MY and TN have received lecture fees from Santen Pharmaceutical Co., Ltd. and are advisors of Santen Pharmaceutical Co., Ltd.

YT is an employee of Santen Pharmaceutical Co., Ltd.

The peer reviewers on this manuscript have received an honorarium from JME for their review work. In addition, a reviewer on this manuscript has disclosed that they have a financial interest in Glaukos (In 2019 it was announced that Santen’s U.S. subsidiary, Santen Inc., entered into a multi-year agreement whereby Glaukos will become the exclusive distributor of the MicroShunt solely in the U.S. market. This agreement does not represent an agreement outside the US. This study was sponsored by Santen Inc). The reviewers have no other relevant financial relationships or otherwise to disclose.

Author contributions

All named authors meet the ICMJE criteria for authorship for this manuscript, take responsibility for the integrity of the work as a whole, and have given final approval for this version to be published.

MY, TN, and YT conceived and designed the study and constructed the model and reviewed the manuscript. HM, and SWK reviewed the model and analyzed the data.

HM, SWK, and YT wrote the manuscript.

Acknowledgements

The medical writing for the manuscript was supported by Yukti Singh and Pragya Rai from IQVIA, India.