Cost-effectiveness of adding a Helicobacter pylori antibody test to the upper gastrointestinal series in gastric cancer screening at the workplace

Abstract

Objective: Helicobacter pylori infections increase gastric cancer risk. Detecting and eradicating Helicobacter pylori infections and implementing a follow-up strategy should be considered by occupational health practitioners. This study aimed to evaluate the cost-effectiveness of adding an H. pylori antibody (HPA) test to current gastric cancer screening using upper gastrointestinal series (UGI) at the workplace in Japan. Methods: The data of Japanese people aged ≥40 years were collected from PubMed and evaluated in two cohorts: UGI (X-ray examination)+HPA test and UGI only. A Markov model was used for the cost-effectiveness analysis of the UGI+HPA test and UGI-only cohorts. The main outcomes were cost, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs). The impact of uncertainty was assessed using one-way sensitivity analyses (OWSA) and probabilistic sensitivity analyses (PSA). Results: A base-case analysis showed that the UGI+HPA test strategy was less costly (−US＄1,039 and −US＄1,496) and more effective (0.415 and 0.437 QALYs) than the UGI-only strategy in the 40- and 50-year-old groups, respectively. The UGI+HPA test strategy was assessed as a dominant strategy. In the OWSA, the tornado diagram showed negative expected costs and positive QALY gains within the established ranges for all parameters. In the PSA, more than 95% of the simulations demonstrated ICER <5 million yen (US＄51,674; US＄1=96.76 yen)/QALY. Conclusions: This modeling study suggests that gastric cancer screening using UGI+HPA test followed by eradication and annual opportunistic screening, compared with UGI only, resulted in lower costs and greater QALY gains for both 40- and 50-year-old groups at the workplace.

Introduction

Gastric cancer was the sixth leading cause of cancer death globally in 2020¹⁾. In the same year, gastric cancer was ranked as the second and fifth most common cause of cancer-related mortality in men and women in Japan, respectively²⁾. A meta-analysis of randomized controlled trials for estimating the effect of Helicobacter pylori (H. pylori) infection on gastric cancer showed that screening and eradication of asymptomatic infections in populations at high risk of gastric cancer reduced the incidence of gastric cancer³⁾. A systematic review also revealed that screening for H. pylori infection was more cost-effective than not based on reports of 13 out of 14 papers analyzed⁴⁾. Therefore, the risk of H. pylori infection and its follow-up strategy should be considered by personnel who conduct screenings, such as occupational health practitioners.

Currently, population-based gastric cancer screening^5,6,7) at the workplace focuses on upper gastrointestinal series (UGI; X-ray examination), and the risk of H. pylori is not adequately addressed. Follow-up observation (opportunistic screening)^5,6,7) with esophagogastroduodenoscopy (EGD) is recommended for those with abnormal findings or prior H. pylori infection^8,9). Studies have shown that the risk of gastric cancer decreases with H. pylori eradication^3,10,11). However, the risk of developing gastroesophageal reflux disease and Barrett’s carcinoma, among others, remains¹²⁾. The strategy of screening for and eradicating H. pylori along with follow-up observations (opportunistic screening)⁸⁾ is based on the higher prevalence of gastric cancer among those who have had an infection relative to those who have not. Gastric cancer screening is not legally required in Japan; however, it is voluntarily implemented by insurers and businesses as part of welfare programs. The medical screening can be performed using different methods and in varying age groups. Thus, occupational health practitioners need to consider how to implement gastric cancer screening and follow-up, as well as the cost-effectiveness and long-term outlook^13,14) for cancer screening as healthcare policy. In Japan, the cost-effectiveness evaluation system started in the 2019 fiscal year, and cost-effectiveness analyses have been conducted for medical technologies (ie, pharmaceuticals and medical devices) according to the guideline¹³⁾ issued by the Center for Outcomes Research and Economic Evaluation for Health (CORE2Health). The guideline recommends quality-adjusted life years (QALYs) as the outcome measure. This study evaluated the cost-effectiveness of gastric cancer screening conducted under the current occupational health system, from the insurer’s perspective, including occupational health practitioners.

Cost-effectiveness can be determined using simulation modeling^{15,16,17,18,19)}. An H. pylori antibody (HPA) test screening followed by eradication treatment for employees in Japan has been shown to be more cost-effective than no screening¹⁵⁾, and an HPA test followed by eradication is more cost-effective than UGI only and endoscopy only for gastric cancer¹⁶⁾. However, for a specific on-site practice, the addition of HPA to the current gastric cancer screening has not been evaluated. Adding an HPA test increases the number of participants with abnormalities, treatments, opportunistic follow-up visits, and screenings, and raises concerns about increased costs; hence, it is important to consider cost-effectiveness. Kowada’s previous study^15,16) did not include post-eradication follow-up in the model. In this study, since follow-up observation post-eradication is strongly promoted in the on-site occupational health field, it was incorporated into the model. The model also factored in productivity losses from the employer’s perspective. Therefore, using a Markov model, this study aimed to investigate the cost-effectiveness of gastric cancer screening at the workplace using the UGI+HPA test relative to UGI only.

Methods

Model structure

Economic evaluations of medical technology are generally conducted as data integration-type research using model analysis. Model analysis is used to predict and evaluate the long-term cost-effectiveness of treatment by simulating disease progression, prognosis, and costs and effects associated with treatment^13,14). The data of Japanese people aged ≥40 years were collected from PubMed and used in the study. This was based on the recommendation of the Manual on Cancer Screening in the Workplace (Ministry of Health, Labour and Welfare, March 2018)⁵⁾ that “gastric cancer screening is indicated for those aged 50 and older; however, for the time being, UGI (X-ray examination) may also be performed for those aged 40 and older.” The participants were divided into two hypothetical cohorts: one cohort that underwent UGI and an H. pylori antibody testing (UGI+HPA test) and another cohort that underwent UGI only. In the UGI+HPA test cohort, only the participants who were aged 40 and 50 years underwent the HPA test. Figure 1A shows a decision tree at initial screenings and subsequent post-screening treatments in the two cohorts, and each branch enters its clinical state, followed up by appropriate annual screening. Figure 1B shows an overview of the simplified Markov model structure consisting of five clinical states: annual population screening is conducted for no gastric cancer and no H. pylori infection. No gastric cancer and successful H. pylori eradication, no gastric cancer and unsuccessful H. pylori eradication, and post-gastric cancer are followed by annual opportunistic screening and death.

Fig. 1. Decision tree and Markov model. (A) Simplified decision tree. (B) Schematic depiction of cohort simulation in a state-transition Markov model. UGI, upper gastrointestinal series; HPA, Helicobacter pylori antibody; GC, gastric cancer; post GC, post-gastric cancer; Pop, population-based screening per the guidelines of Japan’s National Cancer Center; Opp, opportunistic screening during the annual medical examination. An oval node with its state represents a Markov node.

In both groups, an EGD was performed for participants with abnormalities in the UGI or those with a positive HPA test. Gastric cancer management included treatment with endoscopic mucosal resection and routine surgeries, follow-up, and additional treatment depending on the cancer stage, as determined by an attending physician. Eradication treatment for current H. pylori infection was provided up to two times in the same year, and post-eradication and post-infection patients were followed-up with annual opportunistic screening, regardless of whether the eradication was successful or unsuccessful. Participants with no abnormalities continued to undergo annual population-based UGI screening for the duration of their lives.

Model parameters

The total cost, total effectiveness, and QALYs were calculated using values obtained from a PubMed search from January 1, 2000, to September 7, 2021. The search query used in PubMed was as follows: ((“helicobacter pylori”[MeSH Terms] OR (“helicobacter”[All Fields] AND “pylori”[All Fields]) OR “helicobacter pylori”[All Fields]) AND (“stomach neoplasms”[MeSH Terms] OR (“stomach”[All Fields] AND “neoplasms”[All Fields]) OR “stomach neoplasms”[All Fields] OR (“gastric”[All Fields] AND “cancer”[All Fields]) OR “gastric cancer”[All Fields]) AND (“Simulation”[Journal] OR “simulation”[All Fields]) AND model[All Fields]). Literature relevant to the research objectives and its citations were identified. Other methods included relevant journal articles, conference proceedings, and update searches. Variables and ranges were reviewed against the previous literature and further PubMed searches were conducted to obtain data relevant to the present study and to determine whether they could be adopted or not (Table 1)^{11,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38)}.

Table 1. Variables for one-way and probabilistic sensitivity analyses

Variable	Baseline value	OWSA			PSA
		Low	High	Range	Distribution	Reference
Probabilities
Incidence of gastric cancer in individuals aged 40–45 years	0.000103	0.000097	0.000259	Low: 95% CI High: 250%	β	21) L, H: Assump
Incidence of gastric cancer in individuals aged 50–55 years	0.000312	0.0003	0.000780	Low: 95% CI High: 250%	β	21) L, H: Assump
Sensitivity of radiography for detection of gastric cancer	0.855	0.637	0.97	95% CI	β	22)
Specificity of radiography for detection of gastric cancer	0.891	0.885	0.896	95% CI	β	22)
Sensitivity of HPA	0.93	0.87	0.96	95% CI	β	23)
Specificity of HPA	0.79	0.5	0.98	L, H: Assump	β	23,24)
Sensitivity of endoscopy for detection of gastric cancer	0.954	0.842	0.994	95% CI	β	22)
Specificity of endoscopy for detection of gastric cancer	0.888	0.883	0.892	95% CI	β	22)
Prevalence of H. pylori infection in individuals aged 40 years	0.237	0.225	0.25	95% CI	β	25)
Prevalence of H. pylori infection in individuals aged 50 years	0.337	0.327	0.346	95% CI	β	25)
Prevalence of gastric cancer without H. pylori infection	0.01	0.0066	0.14	Baseline value: average of 4 documents L, H27)	β	26,27,28,29) L, H27)
Prevalence of gastric cancer with H. pylori infection	0.045	0.0402	0.0498	Calculated from11) Japan	β	11)Japan
Odds ratio of gastric cancer development after successful H. pylori eradication	0.39	0.31	0.49	95% CI	Log-normal	11)
Success rate of first-line H. pylori eradication therapy	0.926	0.892	0.952	95% CI	β	30)
Success rate of second-line H. pylori eradication therapy	0.98	0.894	0.999	95% CI	β	30)
Gastric cancer death rate	0.034	0.0333	0.0347	95% CI	β	31)
Rate of population-based screening for gastric cancer (after initial follow-up)	0.497	0.4	0.9	L, H: Assump	β	32) experienced
Utility
Healthy (H. pylori-negative)	1	0.87	1	Assump	NA
H. pylori-positive	0.83	0.80	0.86	95% CI	β	33,34)
Gastric cancer a	0.68	0.55	0.81	95% CI	β	19,33,37)
After gastric cancer surgery a	0.76	0.63	0.88	95% CI	β	19,33,37)
Successful H. pylori eradication	0.87	0.84	0.9	Assump	β
Unsuccessful H. pylori eradication	0.79	0.77	0.81	95% CI	β	33)
Death	0				NA
Costs, US＄ (US＄1=\96.76)
UGI and HPA	52.5	42.0	63.0	80%–120%	γ	35,36,37) experienced
Endoscopy	155.0	124.0	186.0	80%–120%	γ
UGI	43.4	34.7	52.1	80%–120%	γ
First-line H. pylori eradication therapy	217.0	173.6	260.4	80%–120%	γ
Second-line H. pylori eradication therapy	206.7	165.4	248.0	80%–120%	γ
Cost of initial treatment of gastric cancer b	12,842	6,421	25,683	50%–200%	γ
Cost of follow-up treatment of gastric cancer b	2,487	1,243	4,974	50%–200%	γ
Cost of productivity loss for 4 hours	100.1	50.0	200.2	50%–200%	γ	38)

Assump, assumption; CI, confidence interval; H, high; HPA, H. pylori antibody; L, low; OWSA, one-way sensitivity analysis; PSA, probabilistic sensitivity analysis; UGI, upper gastrointestinal series.

^a   Utility as the weighted average by percentage³⁷⁾ of the utility of stages I, II, III, and IV^19,33) at the time of gastric cancer detection in the first year.

^b   Cost as the weighted average by percentage³⁷⁾ of the cost of stages I, II, III, and IV^19,35) at the time of gastric cancer detection in the first year.

Transition probability

The incidence of gastric cancer in individuals aged 40–45 and 50–55 years, prevalence of H. pylori infection in individuals aged 40 and 50 years, prevalence of gastric cancer with and without H. pylori infection, success rate of first- and second-line H. pylori eradication therapy, and gastric cancer death rate were derived from the published literature^{11,21,25,26,27,28,29,30,31)}. The mortality rate due to other causes was derived from life tables. The sensitivity and specificity of radiography for detecting gastric cancer, endoscopy for detection of gastric cancer, and HPA test were estimated from the published literature^22,23,24). The rate of population-based screening for gastric cancer (after initial follow-up) was estimated from the published literature³²⁾ and experienced data. The odds ratio (OR)¹¹⁾ of gastric cancer development after successful H. pylori eradication is not adequate for the estimation of transition probability. However, as the incidence of gastric cancer is considered sufficiently low, the OR in the literature was used in this study as an approximate value for the relative risk (RR).

Utility

Health state utilities were obtained from the literature^19,33,34,37). The utilities of gastric cancer and “after gastric cancer surgery” were calculated using the utility weighted average based on the percentage³⁷⁾ utilities of stages I, II, III, and IV^19,33) at the time of gastric cancer detection. The utilities of “gastric cancer” and “after gastric cancer surgery” were for the first and second years or later, respectively.

Cost

Costs were obtained from the literature and data^35,36,37,38) and were calculated based on the Japanese national fee schedule³⁶⁾. Subsequently, the currency was converted to United States dollars (US＄) using the Organization for Economic Co-operation and Development purchasing power parity rate of 2021 (US＄1=¥96.76)³⁹⁾. Public willingness-to-pay (WTP) was set at 5 million yen/QALY, which is considered cost-effective (no price adjustment required) under the Japanese cost-effectiveness evaluation system²⁰⁾. The cost of annual population screening is based on the cost of the UGI. The cost of annual opportunistic screening is based on the cost of the endoscopy. “Cost of initial treatment of gastric cancer” and “cost of follow-up treatment of gastric cancer” were calculated using the cost-weighted average by percentage³⁷⁾ of the cost of stages I, II, III, and IV^19,33) at the time of gastric cancer detection. The costs of “gastric cancer” and “after gastric cancer surgery” were for the first and second years or later, respectively.

Conditions for analysis

Analyses were performed from the perspective of the insurer and that of the occupational health professional. This approach was influenced by the guidelines published by CORE2Health, which consider the “Public health payer’s perspective” as the standard in the analysis perspective section (2 Analysis perspective)¹³⁾. To be clear, occupational physicians are uniquely positioned to bridge the gap between pre-employment community health, occupational health, and post-employment community health within the context of lifelong community health (universal healthcare). Regarding productivity loss, the concept is that, even if an individual is unable to work due to subsequent post-screening treatments, someone else will step in and the productivity might not be lost in the long term. Hence, the outcomes considering productivity loss from the employer’s (occupational physician’s) viewpoint (limited societal perspective) are provided as a reference. Productivity loss was estimated by multiplying the average wage across all industries, age groups, and genders by the number of hours of lost productivity. Specifically, ages and genders were estimated based on data from the 2021 Basic Survey on Wage Structure³⁸⁾. This calculation accounted for workers aged 40 years and older, assuming an experience-adjusted index of 20 years, an hourly wage of US＄25.0, and a loss of 4 hours. The analysis was conducted over the individual’s lifetime, applying a discount rate of 2%¹³⁾.

Sensitivity analysis

One-way sensitivity analyses (OWSA) were performed for the transition probabilities, test characteristics, costs, and utilities using the values derived from the literature, and the ranges from low to high values were determined, as shown in Table 1. The costs of initial treatment of gastric cancer, follow-up treatment of gastric cancer, and productivity loss were used with a range of 50–200%; for other costs, a range of 80–120% was used. These ranges were obtained from previous publications^{9,15,16,18,19,38)} and based on empirical costs. The results of OWSA from the employer’s perspective were summarized in tornado diagrams. A probabilistic sensitivity analysis (PSA) was also performed using Monte Carlo simulation for transition probabilities, test characteristics, costs, and utilities. In the PSA, β distributions were applied for the parameters acquired from raw data for probabilities and utilities, except OR. A log-normal distribution was applied to the OR, and γ distributions were applied to the costs. The results of 10,000 trials for PSA from the employer’s perspective are shown with scatterplots for 40-year-old participants. The proportion of trial results for both 40- and 50-year-old participants was calculated in six quadrants made up of four quadrants and the threshold of the WTP of 5 million yen/QALY.

As this was a simulation modeling study using values from literature review and demographic data for Japan, ethics approval was not required. TreeAge Pro Healthcare 2022 (TreeAge Software Inc., Williamstown, MA, USA) was used for the analytical calculations.

Results

Search results and number of documents used

A PubMed search on September 7, 2021 identified 40 documents, and we proceeded based on 16 documents relevant to the study objectives and their citations. Other methods included reviewing the Journal of the Japanese Society of Internal Medicine, the Journal of the Japanese Medical Association, and PubMed search results, resulting in more than 150 documents being considered. For variables and ranges, as far as possible, we adopted Japanese official statistics (government and community health survey results), the latest data, and reliable journal articles for statistical values; for utilities, we reviewed 10 documents and adopted the results calculated using the same evaluation method as the highly versatile EQ-5D as much as possible. Ultimately, the 20 documents shown in Table 1^{11,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38)} were used to calculate the results using the model.

Base-case analysis

The outcomes for the UGI+HPA test and UGI only in the groups of participants aged 40 and 50 years from both the insurer’s and the employer’s perspectives in the base-case analysis are summarized in Table 2. The UGI+HPA test strategy was better than the UGI-only strategy in both the 40- (US＄6,150 and 28.199 QALYs vs. US＄7,189 and 27.784 QALYs) and the 50-year-old groups (US＄7,166 and 23.773 QALYs vs. US＄8,661 and 23.335 QALYs) from the insurer’s perspective and in both the 40- (US＄6,195 and 28.199 QALYs vs. US＄7,200 and 27.784 QALYs) and the 50-year-old groups (US＄7,217 and 23.773 QALYs vs. US＄8,672 and 23.335 QALYs) from the employer’s perspective. The UGI+HPA test strategy was less costly (−US＄1,039 and −US＄1,496) and had more benefits (0.415 and 0.437 QALYs) than the UGI-only strategy in the 40- and 50-year-old groups, respectively, from the insurer’s perspective, and was also less costly (−US＄994 and −US＄1,455) and had more benefits (0.415 and 0.437 QALYs, respectively) from the employer’s perspective. The per capita cost savings and QALY gains of the UGI+HPA test strategy were higher for the 50-year-old group than for the 40-year-old group from both perspectives.

Table 2. Results of base-case analyses

The insurer’s perspective
Age group	Strategy	Cost (US＄)	Effectiveness (QALYs)	Incremental cost (US＄)	Incremental QALYs	ICER (US＄/QALY gained)
40 years	UGI+HPA test	6,150	28.199	−1,039	0.415	Dominant
	UGI only	7,190	27.784
50 years	UGI+HPA test	7,166	23.773	−1,496	0.437	Dominant
	UGI only	8,661	23.335
The employer’s (occupational health practitioner’s) perspective, taking productivity loss into account
40 years	UGI+HPA test	6,195	28.199	−994	0.415	Dominant
	UGI only	7,200	27.784
50 years	UGI+HPA test	7,217	23.773	−1,455	0.437	Dominant
	UGI only	8,672	23.335

HPA, Helicobacter pylori antibody; ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life-year; UGI, upper gastrointestinal series.

Sensitivity analysis

The results of the OWSA for the top 10 variables from the insurer’s perspective are presented as tornado diagrams in Figure 2A for the 40-year-old group and in Figure 2B for the 50-year-old group. Both age groups had negative expected costs and positive QALY gains within the established ranges for all parameters. In both populations, the factor with the greatest impact on the incremental cost-effectiveness ratio (ICER) was the cost of follow-up treatment of gastric cancer, followed by the cost of initial treatment of gastric cancer. Compared with other parameters, the impact of the OR for gastric cancer development after successful H. pylori eradication and the rate of population-based screening for gastric cancer on ICER were greater in the 40-year-old population. From the employer’s perspective, the results of the OWSA for the top 10 variables are the same for both age groups. Scatterplots of incremental cost and incremental effectiveness (sampling 10,000 trials) for the UGI+HPA test cohort relative to those of the UGI-only cohort in the 40-year-old population from the insurer’s perspective are shown in Figure 3. The dashed line represents WTP 5 million yen (US＄51,674)/QALY. In both the 40- and 50-year-old populations, 99% of the simulations demonstrated ICER of <5 million yen (US＄51,674)/QALY. From the employer’s perspective, the results were similar: in both the 40- and 50-year-old populations, 99% of the simulations demonstrated ICER of <5 million yen (US＄51,674)/QALY. These conclusions were not significantly affected by uncertainties around the input variables.

Fig. 2. Tornado diagram of the top 10 variables in the two cohorts. ICER of the UGI+HPA test in comparison with UGI only for the (A) 40- and (B) 50-year age groups. ICER, incremental cost-effectiveness ratio; UGI, upper gastrointestinal series; HPA test, Helicobacter pylori antibody test; EV: expected value.

Fig. 3. Incremental cost-effectiveness scatterplots. Incremental cost-effectiveness scatterplots for the upper gastrointestinal series (UGI) and Helicobacter pylori antibody (HPA) test in comparison with UGI only for the 40-year age group (willingness-to-pay \5,000,000 [US＄51,674; US＄1 = \96.76]/QALY). Each square shows the proportion of scatterplots in the six quadrants of the 40- and 50-year-old groups. IC, incremental cost; IE, incremental effectiveness.

ICER, incremental cost-effectiveness ratio.

Discussion

The results of this study showed that the UGI+HPA test was more cost-effective than UGI only for both 40- and 50-year-old populations. The employer’s perspective had the same trend, although costs were slightly higher than those from the insurer’s perspective due to the incorporation of productivity losses. For the OWSA, the tornado diagram shows negative expected costs and positive QALY gains within the established ranges for all parameters. For the PSA, more than 95% of the simulations demonstrated ICER <5 million yen (US＄51,674)/QALY. These findings suggest that it is more cost-effective to add an HPA test to current gastric cancer screenings and that the results of sensitivity analyses indicate the robustness of the findings, because the addition of an HPA test facilitated the detection of the gastric cancer development due to H. pylori infection. Furthermore, the subsequent eradication treatment and follow-up after opportunistic screening continued to reduce the incidence and mortality rate of gastric cancer despite higher costs, thereby leading to increased QALYs. Adding the HPA test was more cost-effective in the 50-year-old group than in the 40-year-old group. These results suggest that an HPA test should be incorporated into gastric cancer screening as part of regular workplace physical examinations.

In the past, the HPA test was considered one of the methods for gastric cancer screening and detection. There are cost-effectiveness studies comparing cancer screening methods such as endoscopy, UGI, and the HPA test and no screening^9,15,16). Recently, the HPA test has been considered as a method for the stratification of gastric cancer risk^8,40). To the best of our knowledge, this is the first model-based cost-effectiveness analysis for adding an HPA test to gastric cancer screening at the workplace.

Endoscopy has no X-ray exposure and has a higher sensitivity and specificity than UGI²²⁾. Therefore, it is recommended for gastric cancer screening; however, it is difficult to sterilize the endoscope camera at the workplace. Currently, gastric cancer screening at the workplace using UGI is conducted simultaneously with regular health checkups. Gastric cancer screening using X-ray by bus is convenient and easy for employees at the workplace. Therefore, the percentage of employees who undergo population-based screening for gastric cancer is high. The UGI+HPA test strategy can be easily implemented by adding an HPA test to routine blood tests during regular health checkups. The number of people recommended for subsequent thorough examinations and treatments will increase by adding the HPA test, but it could be suitable if the ages for the HPA test are adjusted (namely setting a fixed age, such as 40, 45, 50, 55, or 60 years). This is a feasible method at the workplace.

This study has several limitations. First, gastric cancer staging, cost, and utility used the weighted average by stage percentage, rather than by classification, to simplify the model. Furthermore, the effects of eradication with a positive HPA test on staging at the time of detection of gastric cancer are not known; thus, the weighted average method was used. Second, the possibility that other diseases, such as gastroesophageal reflux disease and Barrett’s carcinoma¹²⁾, may increase due to H. pylori eradication has been suggested. However, even if the incidence of other diseases increases, the detection and eradication of H. pylori is recommended, and follow-up opportunistic screening is important. This is believed to be in line with the methods of this study. Third, the interval between population-based and opportunistic screening in this study was set at 1 year, which is the standard currently. In future gastric cancer screenings, it is important to stratify gastric cancer risk appropriately and consider the interval between gastric cancer screenings⁴⁰⁾ from the viewpoint of cost-effectiveness. Finally, the current study examined current gastric screening practices at the workplace for individuals aged ≥40 years. Screening for those aged <40 years should also be evaluated.

Occupational health practitioners should consider cost-effectiveness when establishing a health management plan for employees, and it will likely become necessary to consider the individual circumstances of each workplace. This study would be useful to other occupational health practitioners who are considering the same strategy. Furthermore, it is important to consider cost-effectiveness evaluation for future healthcare policies, even when considering on-site practices globally. Analytical models can be used to evaluate cost-effectiveness under different conditions (eg, differences due to age, effects of H. pylori infection rate, and impact of medical costs) and will be useful in future discussions on how health screening should be conducted.

In conclusion, this modeling study suggests that gastric cancer screening with the UGI+HPA test followed by eradication and an annual opportunistic screening strategy resulted in less cost and greater QALY gains than that with the UGI-only strategy for both 40- and 50-year-old groups at the workplace. These findings show that occupational health professionals should consider adding an HPA test to the gastric cancer screening strategy at the workplace.

Acknowledgments

MN would like to thank Dr. Masato Itoh (Panasonic Health Insurance Organization), who recommended HTP participation, Dr. Keisuke Kunieda (Panasonic Health Insurance Organization), who gave his opinion from the viewpoint of gastroenterology, and Editage (www.editage.com) for English language editing.

Author contribution

MN wrote the manuscript, collected the data, performed the data analysis, and was responsible for the conception of the study. MN participated in the Health Policy & Technology Assessment Research Project (http://www.pp.u-tokyo.ac.jp/HPTA/hta/index.html) hosted by IK in the second semester of the 2018 fiscal year. One of the visiting lecturers, SI, provided guidance on the cost-effectiveness analysis. IK supervised the study.

Data availability statement

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Sources of funding

No funding was received for this study.

References

• 1.   Sung  H,  Ferlay  J,  Siegel  RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021; 71(3): 209-249.
• 2.  Ministry of Health, Labour, and Welfare. Vital statistics Japan. 2022 (in Japanese). https://ganjoho.jp/reg_stat/statistics/dl/index.html#mortality. Accessed May 14, 2022.
• 3.   Ford  AC,  Forman  D,  Hunt  RH,  Yuan  Y,  Moayyedi  P. Helicobacter pylori eradication therapy to prevent gastric cancer in healthy asymptomatic infected individuals: systematic review and meta-analysis of randomised controlled trials. BMJ. 2014; 348: g3174.
• 4.   Matsuda  A,  Saika  K,  Tanaka  R,  Ito  Y,  Fukui  K,  Kamo  KI. Simulation models in gastric cancer screening: a systematic review. Asian Pac J Cancer Prev. 2018; 19(12): 3321-3334.
• 5.  Ministry of Health, Labour and Welfare. Manual on cancer screening in the workplace. 2018 (in Japanese). https://www.mhlw.go.jp/file/05-Shingikai-10904750-Kenkoukyoku-Gantaisakukenkouzoushinka/0000213286.pdf. Accessed January 5, 2021.
• 6.   Miles  A,  Cockburn  J,  Smith  RA,  Wardle  J. A perspective from countries using organized screening programs. Cancer. 2004; 101(5)(suppl): 1201-1213.
• 7.  National Cancer Center Institute for Control. Population-based screening and opportunistic screening. 2021 (in Japanese). http://canscreen.ncc.go.jp/kangae/kangae7.html. Accessed May 14, 2022.
• 8.   Yamamichi  N. Helicobacter pylori and future prospect of gastric cancer screening in Japan. Nihon Naika Gakkai Zasshi. 2021; 110(1): 55-63 (in Japanese).
• 9.   Kowada  A. Endoscopy is cost-effective for gastric cancer screening after successful Helicobacter pylori eradication. Dig Dis Sci. 2021; 66(12): 4220-4226.
• 10.   Lee  YC,  Chiang  T-H,  Chou  C-K, et al. Association between Helicobacter pylori eradication and gastric cancer incidence: a systematic review and meta-analysis. Gastroenterology. 2016; 150(5): 1113-1124.e5.
• 11.   Sugano  K. Effect of Helicobacter pylori eradication on the incidence of gastric cancer: a systematic review and meta-analysis. Gastric Cancer. 2019; 22(3): 435-445.
• 12.   Koike  T,  Masamune  A. Helicobacter pylori and gastroesophageal reflux disease (GERD). Nihon Naika Gakkai Zasshi. 2021; 110(1): 36-41 (in Japanese).
• 13.  National Institute of Public Health. Analysis guidelines for evaluation of cost-effectiveness at the Central Social Insurance Medical Council. 3rd ed. 2022 (in Japanese). https://c2h.niph.go.jp/tools/guideline/guideline_ja.pdf. Accessed May 14, 2022.
• 14.   Isao  K. Health Technology Assessment in Japan. Singapore: Springer Nature Singapore Pte Ltd./ADIS; 2019.
• 15.   Kowada  A. Cost-effectiveness of Helicobacter pylori screening followed by eradication treatment for employees in Japan. Epidemiol Infect. 2018; 146(14): 1834-1840.
• 16.   Kowada  A. Cost-effectiveness of Helicobacter pylori test and eradication versus upper gastrointestinal series versus endoscopy for gastric cancer mortality and outcomes in high prevalence countries. Scand J Gastroenterol. 2019; 54(6): 685-689.
• 17.   Omata  F,  Shimbo  T,  Ohde  S,  Deshpande  GA,  Fukui  T. Cost-effectiveness analysis of Helicobacter pylori diagnostic methods in patients with atrophic gastritis. Gastroenterol Res Pract. 2017; 2017: 2453254.
• 18.   Kowada  A,  Asaka  M. Economic and health impacts of introducing Helicobacter pylori eradication strategy into national gastric cancer policy in Japan: A cost-effectiveness analysis. Helicobacter. 2021; 26(5): e12837.
• 19.   Saito  S,  Azumi  M,  Muneoka  Y, et al. Cost-effectiveness of combined serum anti-Helicobacter pylori IgG antibody and serum pepsinogen concentrations for screening for gastric cancer risk in Japan. Eur J Health Econ. 2018; 19(4): 545-555.
• 20.  Ministry of Health, Labour and Welfare. Medical expenses pharmaceutical materials— attachment 3, 30.6.13. Establishment of reference values for cost-effectiveness assessment. 2018 (in Japanese). https://www.mhlw.go.jp/file/05-Shingikai-12404000-Hokenkyoku-Iryouka/0000211609.pdf. Accessed May 14, 2022.
• 21.  Cancer Information Service, National Cancer Center, Japan (National Cancer Registry, Ministry of Health, Labour and Welfare). Cancer statistics. 2021 (in Japanese). https://ganjoho.jp/reg_stat/statistics/data/dl/index.html#incidence. Accessed December 28, 2021.
• 22.   Hamashima  C,  Okamoto  M,  Shabana  M,  Osaki  Y,  Kishimoto  T. Sensitivity of endoscopic screening for gastric cancer by the incidence method. Int J Cancer. 2013; 133(3): 653-659.
• 23.   Best  LMJ,  Takwoingi  Y,  Siddique  S, et al. Non-invasive diagnostic tests for Helicobacter pylori infection. Cochrane Database Syst Rev. 2018; 3(3): CD012080.
• 24.   Gotoda  T. Eradication therapy for Helicobacter pylori-infected teenagers. Nihon Naika Gakkai Zasshi. 2021; 110(1): 47-54 (in Japanese).
• 25.   Wang  C,  Nishiyama  T,  Kikuchi  S, et al. Changing trends in the prevalence of H. pylori infection in Japan (1908-2003): a systematic review and meta-regression analysis of 170,752 individuals. Sci Rep. 2017; 7(1): 15491.
• 26.   Matsuo  T,  Ito  M,  Takata  S,  Tanaka  S,  Yoshihara  M,  Chayama  K. Low prevalence of Helicobacter pylori-negative gastric cancer among Japanese. Helicobacter. 2011; 16(6): 415-419.
• 27.   Sato  C,  Hirasawa  K,  Tateishi  Y, et al. Clinicopathological features of early gastric cancers arising in Helicobacter pylori uninfected patients. World J Gastroenterol. 2020; 26(20): 2618-2631.
• 28.   Yamada  A,  Kaise  M,  Inoshita  N, et al. Characterization of Helicobacter pylori-naïve early gastric cancers. Digestion. 2018; 98(2): 127-134.
• 29.   Mizutani  T,  Araki  H,  Saigo  C, et al. Endoscopic and pathological characteristics of Helicobacter pylori infection-negative early gastric cancer. Dig Dis. 2020; 38(6): 474-483.
• 30.   Murakami  K,  Sakurai  Y,  Shiino  M,  Funao  N,  Nishimura  A,  Asaka  M. Vonoprazan, a novel potassium-competitive acid blocker, as a component of first-line and second-line triple therapy for Helicobacter pylori eradication: a phase III, randomised, double-blind study. Gut. 2016; 65(9): 1439-1446.
• 31.   Ito  Y,  Miyashiro  I,  Ito  H, et al; J-CANSIS Research Group. Long-term survival and conditional survival of cancer patients in Japan using population-based cancer registry data. Cancer Sci. 2014; 105(11): 1480-1486.
• 32.  Ministry of Health, Labour, and Welfare. Comprehensive survey of living conditions 2020. https://www.mhlw.go.jp/toukei/saikin/hw/k-tyosa/k-tyosa19/dl/04.pdf. Accessed August 14, 2020.
• 33.   Areia  M,  Alves  S,  Brito  D, et al. Health-related quality of life and utilities in gastric premalignant conditions and malignant lesions: a multicentre study in a high prevalence country. J Gastrointestin Liver Dis. 2014; 23(4): 371-378.
• 34.   Delaney  BC,  Qume  M,  Moayyedi  P, et al. Helicobacter pylori test and treat versus proton pump inhibitor in initial management of dyspepsia in primary care: multicentre randomised controlled trial (MRC-CUBE trial). BMJ. 2008; 336(7645): 651-654.
• 35.  Ace Force Co., Ltd. Ganchiryohi. 2020 (in Japanese). http://ganchiryohi.com. Accessed August 21, 2020.
• 36.  Igakutsushin-sya. National fee schedule and medical insurance reimbursement table in Japan. Tokyo: Igakutsushin-sya; 2021 (in Japanese).
• 37.  Cancer Information Service, National Cancer Center, Japan. Cancer survival in hospital-based cancer registries. 2021 (in Japanese). https://hbcr-survival.ganjoho.jp. Accessed June 1, 2022.
• 38.  Ministry of Health, Labour, and Welfare. The 2021 basic survey on wage structure (in Japanese). https://www.mhlw.go.jp/content/000980675.pdf. Accessed August 17, 2023.
• 39.  Organization for Economic Co-operation and Development (OECD). PPPs (purchasing power parities) and exchange rates. https://stats.oecd.org/Index.aspx?DataSetCode=SNA_TABLE4. Accessed May 4, 2022.
• 40.   Hamashima  C. Forthcoming step in gastric cancer prevention: how can risk stratification be combined with endoscopic screening for gastric cancer? Gut Liver. 2022; 16(6): 811-824.