Economic Evaluation of Empirical and Tailored Helicobacter pylori Eradication Strategies Including Second-Line Therapy

Abstract

Objectives

To our knowledge, no study has evaluated the cost-effectiveness of tailored therapy guided by genotypic resistance compared with empirical clarithromycin-based triple therapy, including second-line therapies.

Methods

A decision tree model was developed to compare empirical and tailored treatment strategies. Cost-minimization and cost-effectiveness analyses were conducted. The incremental cost-effectiveness ratio (ICER) was calculated to compare the two strategies, including second-line therapy.

Results

The empirical therapy incurred a lower cost per patient (USD 211.49) than the tailored therapy (USD 269.88). The overall eradication rates were nearly identical (97.98% vs. 98.05%), yielding an ICER of USD 859.00 per 1% gain in eradication. Empirical therapy consistently resulted in lower total costs for all treatment durations. In the 7-day regimen, empirical therapy (USD 201.52; 97.64% eradication) versus tailored therapy (USD 253.62; 98.12%) produced an ICER of USD 108.54 per 1%, in favor of the empirical approach. In the 14‑day regimen, tailored therapy was more expensive (USD 282.74 vs. USD 221.46) and slightly less effective (97.97%, 98.34%). Overall, empirical therapy is a cost-effective strategy in most cases.

Conclusions

This study demonstrates that tailored therapy guided by genotypic resistance is no more cost-effective than empirical therapy when second-line therapy is included. Future studies are warranted to identify strategies to improve the cost-effectiveness of tailored therapies in clinical practice.

INTRODUCTION

Tailored therapy based on genotypic resistance has been increasing for Helicobacter pylori eradication in Korea [1]. This is because empirical clarithromycin-based triple therapy (CTT) regimens have decreased due to increased resistance to clarithromycin [2]. However, CTT regimens are still frequently prescribed in Korea. The most recent Korean guidelines recommend a 7-day CTT regimen for patients susceptible to clarithromycin based on genotypic resistance [3]. A 14-day CTT regimen is recommended when prescribed empirically. This is not the case in other countries. Recent European and American guidelines recommend against the prescription of clarithromycin- or levofloxacin-containing regimens due to high antibiotic resistance rates [4,5].

Tailored therapy allows for antibiotic prescription based on resistance. This allows optimization of the dose and duration of antibiotics, which results in optimal outcomes and minimizes adverse event development [6].

However, tailored therapy is associated with increased costs, and additional resources are required to test for antibiotic resistance. Previous studies from Korea have merely compared the average costs of tailored and empirical therapies and have not performed a cost-effectiveness analysis (CEA) [7-9]. A CEA was recently performed based on a decision tree model, and it was reported that tailored therapy was associated with an incremental cost-effectiveness ratio (ICER) of USD 4.66 compared to empirical therapy [10]. This CEA was performed for first-line eradication, and the effects of second-line therapy were not considered. Patients who fail empirical therapy are expected to receive bismuth quadruple therapy (BQT), whereas those who fail tailored therapy are expected to receive levofloxacin-based triple therapy (LTT) regimens. Many studies have reported suboptimal eradication rates of LTT [3]. The eradication rates of tailored and empirical therapies are expected to be similar when second-line therapy is included because of the low eradication rates of LTT.

Therefore, this study aimed to evaluate the cost-effectiveness of tailored therapy based on genotypic resistance compared to empirical CTT, including second-line regimens. Furthermore, this study analyzed the cost-effectiveness of both strategies according to the duration of the initial therapy and identified the key factors influencing economic outcomes.

METHODS

Strategies for H. pylori eradication

A decision tree model (Fig. 1) was developed to conduct a cost-minimization analysis (CMA) and CEA to assess the economic value of tailored therapy versus empirical therapy, including subsequent second-line therapies. Treatment outcomes for the tailored and empirical therapy regimens were obtained from a previous study [11].

Fig. 1.

Decision-tree model for assessing cost-effectiveness of tailored therapy compared to empirical therapy, including second-line regimens for H. pylori eradication. BQT, bismuth quadruple therapy; LTT, levofloxacin-based triple therapy; CTT, clarithromycin-based triple therapy.

Patients in the empirical therapy group received a standard CTT consisting of pantoprazole (40 mg), amoxicillin (1 g), and clarithromycin (500 mg) administered twice daily. In the tailored therapy group, the treatment regimen was determined based on genotypic resistance determined using polymerase chain reaction (PCR) testing. Patients with clarithromycin-sensitive infections underwent CTT, whereas clarithromycin-resistant cases underwent BQT regimens. The BQT included pantoprazole (40 mg twice daily), metronidazole (500 mg thrice daily), bismuth (300 mg), and tetracycline (500 mg), each administered four times daily. Primary eradication was confirmed using the 13C-urea breath test (UBT) performed at least four weeks after the completion of therapy. Patients who failed CTT received 14-day BQT as second-line therapy. Patients with failed BQT received a 10-day LTT. The LTT regimen consisted of pantoprazole (40 mg, twice daily), amoxicillin (1 g, twice daily), and levofloxacin (500 mg, once daily). Eradication was confirmed by UBT at least four weeks post-treatment.

Eradication success rates

The eradication rates for both empirical and tailored therapies as first-line therapies have been published previously [11]. The eradication rates of BQT as a second-line regimen were assumed to be 93.2%, and the LTT eradication rate was assumed to be 57.0% based on the results of a previous study. The final eradication rate with tailored and empirical therapies, including second-line therapies, was approximately 98%.

Cost estimation

Cost parameters included expenses related to H. pylori diagnosis and eradication therapy. The detailed cost approach has been described in a previous publication [10]. Eradication therapy costs were estimated based on the 2024 health insurance reimbursement rates in the Korea. Direct nonmedical costs, such as transportation and caregiving expenses, were excluded from the analysis. Table 1 presents a comprehensive breakdown of the cost components and their corresponding values. All costs are expressed in United States dollars (1 USD approximately 1400 Korean Won).

Table 1.

Input costs for medications, diagnostic tests, and administration in the economic evaluation of H. pylori eradication

Parameters	Value
Medication costs
Clarithromycin triple therapy
7 days	31.0
14 days	62.0
Bismuth quadruple therapy
7 days	22.7
14 days	45.4
Levofloxacin triple therapy
10 days	34.0
Diagnosis and eradication confirmation cost
Biopsy	20.1
Rapid urease test	18.6
Forceps	21.2
Urea breath test	43.1
PCR test	78.2
Administration cost
Medical visit cost
First visit to clinics	13.2
First visit to hospitals	20.0
Repeated visit to clinics	9.4
Repeated visit to hospitals	15.7
Ratio of healthcare utilization: hospital vs. clinics	1.7
Prescription cost
7 days	6.0
14 days	7.9
10 days	7.0

Values are presented as US dollars unless otherwise indicated.

PCR, polymerase chain reaction.

Outcomes

For CMA, it was assumed that there was no statistically significant difference in the eradication rates between the two regimens, including the second-line therapies. Under this assumption, only the total eradication cost was compared. For CEA, the primary outcome was the ICER for both treatment strategies. The ICER represents the additional cost required to increase the eradication rate by one percent when tailored therapy is adopted instead of empirical therapy. This was calculated by dividing the difference in total costs by the difference in eradication rates between the two regimens using the following equation:

$$ICER=\frac{C_{\mathrm{tailored}}-C_{\mathrm{empirical}}}{E_{\mathrm{tailored}}-E_{\mathrm{empirical}}}$$

Analysis

The base-case analysis evaluated the cost minimization and cost-effectiveness of empirical versus tailored therapies, incorporating second-line therapy into both strategies. In the subgroup analysis, the total costs and ICERs were compared between the different durations of the two initial strategies (7-day vs. 14-day regimens).

The results’ robustness was examined using one-way sensitivity analyses (OWSA) by varying the key model parameters, including medication costs and PCR testing costs. All the analyses were conducted in accordance with the guidelines of the Consolidated Health Economics Evaluation Reporting Standards (CHEERS) guidelines [12].

Ethical considerations

Institutional Review Board approval was waived by the Incheon St. Mary’s Hospital because this study was a model-based economic evaluation that used only secondary published data.

RESULTS

Cost-minimization analysis

A CMA was first conducted under the assumption of equal effectiveness between the two treatments. This is because the final eradication rates of empirical and tailored therapies are presumed to be clinically equivalent once second-line therapies are incorporated. Under this assumption, only the total eradication cost was compared. In the pooled analysis combining both the 7-day and 14-day regimens, tailored therapy required an additional USD 58.39 per patient compared with empirical therapy. In the 7-day regimens, tailored therapy cost USD 52.10 more than the empirical therapy, and in the 14-day regimens, the difference increased to USD 61.28 (Table 2). Across all treatment duration scenarios, empirical therapy consistently demonstrated lower overall costs, indicating that it is a cost-minimizing strategy.

Table 2.

Base-case cost-minimization analysis results of empirical and tailored therapies for H. pylori eradication

Regimen	Empirical therapy	Tailored therapy	Incremental values
Pooled	211.49	269.88	58.39
7-day	201.52	253.62	52.10
14-day	221.46	282.74	61.28

Values are presented as US dollars.

Cost-effectiveness analysis

Subsequently, CEA was performed using the observed eradication rates. In the base-case analysis, the pooled expected total cost of empirical therapy, including both first- and second-line regimens, was USD 211.49, whereas the cost of tailored therapy was USD 269.88, resulting in an incremental cost of USD 58.39 (Table 3). The final eradication rates were nearly identical between the two strategies (97.98% vs. 98.05%), with a difference of only 0.07%. Based on these values, the ICER was calculated as USD 859.00 per 1% increase in eradication, indicating that when second-line therapy was considered, an additional USD 859.00 was required to achieve a 1% improvement in eradication when switching from empirical to tailored therapy.

Table 3.

Base-case cost-effectiveness analysis results of empirical and tailored therapies for H. pylori eradication

	Empirical	Tailored	Incremental
Pooled
Total costs (USD)	211.49	269.88	58.39
Effectiveness (eradication rate, %)	97.98	98.05	0.07
ICER (USD/%)			859.00*
7-day
Total costs (USD)	201.52	253.62	52.10
Effectiveness (eradication rate, %)	97.64	98.12	0.48
ICER (USD/%)			108.54
14-day
Total costs (USD)	221.46	282.74	61.28
Effectiveness (eradication rate, %)	98.34	97.97	-0.37
ICER (USD/%)			-167.85

The eradication rates and cost values were rounded to two decimal places for presentation. The ICER was calculated using the original, unrounded data.

^*If the rounded values presented in the table are used for the calculation, the ICER would be 834.14 USD/%.

ICER, incremental cost-effectiveness ratio.

In the subgroup analysis stratified by initial treatment duration, the total cost of the 7-day regimen was USD 201.52 for empirical therapy and USD 253.62 for tailored therapy, with eradication rates of 97.64% and 98.12%, respectively. The resulting ICER was USD 108.54 per 1%. In contrast, for the 14-day regimen, tailored therapy incurred a higher cost (USD 282.74 vs. USD 221.46 for empirical therapy) but resulted in a slightly lower eradication rate (97.97% vs. 98.34%). Consequently, the ICER was USD -167.85 per 1%, indicating that tailored therapy was more costly and less effective.

When comparing 7-day tailored therapy with 14-day empirical therapy, the total costs were USD 253.62 and USD 221.46, and the eradication rates were 98.12% and 98.34%, respectively. The corresponding ICER was USD -148.95 per 1%, suggesting that, under these assumptions, 7-day tailored therapy was less cost-effective than 14-day empirical therapy (Table 4).

Table 4.

Cost-effectiveness analysis comparing 7-day tailored therapy with 14-day empirical therapy for H. pylori eradication

	Empirical (14-d)	Tailored (7-d)	Incremental
Total costs (USD)	221.46	253.62	32.16
Effectiveness (eradication rate, %)	98.34	98.12	-0.22
ICER (USD/%)			-148.95

The eradication rates and cost values were rounded to two decimal places for presentation. The ICER was calculated using the original, unrounded data.

ICER, incremental cost-effectiveness ratio.

One-way sensitivity analysis

OWSA were conducted to assess the robustness of the basecase findings by varying the costs of first- and second-line medications and PCR testing (Table 5). Drug costs were varied by ±20%, and PCR testing costs were varied by ±10% and ±20%, respectively. As shown in the tornado diagram (Fig. 2), changes in PCR testing costs had the greatest impact on ICER estimates. CTT cost had a relatively greater impact on ICER variability than BQT and levofloxacin costs, likely because of the large proportion of PCR testing costs in the cost of BQT therapies.

Table 5.

Sensitivity analysis results of empirical and tailored therapies for H. pylori eradication

Parameters	Values (USD)		ICERs (USD/%)	Fluctuation (%)
Base-case		-	859.00	0.00
Cost
7-day CTT	Lower	24.80 (-20%)	904.61	5.31
	Upper	37.20 (+20%)	813.40	-5.31
14-day CTT	Lower	49.60 (-20%)	950.21	10.62
	Upper	74.40 (+20%)	767.79	-10.62
7-day BQT	Lower	18.16 (-20%)	852.02	-0.81
	Upper	27.24 (+20%)	865.98	0.81
14-day BQT	Lower	36.32 (-20%)	849.03	-1.16
	Upper	54.48 (+20%)	868.97	1.16
10-day LTT	Lower	27.20 (-20%)	852.10	-0.80
	Upper	40.80 (+20%)	865.90	0.80
Hp PCR	Lower	62.52 (-20%)	629.07	-26.77
	Upper	93.78 (+20%)	1088.93	26.77
Hp PCR	Lower	70.34 (-10%)	744.04	-13.38
	Upper	85.97 (+10%)	973.97	13.38

CTT, clarithromycin based triple therapy; BQT, bismuth based quadruple therapy; LTT, levofloxacin based triple therapy; Hp, Helicobacter pylori; PCR, polymerase chain reaction; ICER, incremental cost-effectiveness ratio.

Fig. 2.

Tornado diagram of one-way sensitivity analysis for the cost-effectiveness of H. pylori eradication. HP, Helicobacter pylori; PCR, polymerase chain reaction; CTT, clarithromycin based triple therapy; BQT, bismuth based quadruple therapy; LTT, levofloxacin based triple therapy; ICER, incremental cost-effectiveness ratio.

DISCUSSION

We previously compared the first-line eradication rates of empirical therapy consisting of CTT with tailored therapy based on genotypic resistance [11]. The pooled eradication success rate was substantially higher with tailored therapy (86.5%) than with empirical therapy (70.3%), with an increase of approximately 16.2 percentage points. This gain came at an incremental cost of approximately USD 75.6 per patient, yielding an ICER of USD 4.66 per 1% eradication gain. In practical terms, each additional 1% of eradication success required roughly USD 4.66 in extra cost with a tailored first-line approach. It was concluded that genotype-guided first-line therapy is cost-effective for H. pylori eradication in Korea, particularly considering the high rate of clarithromycin resistance. However, the previous study compared the cost effectiveness of first-line eradication. Most patients are expected to receive salvage therapy after failure of first-line eradication. Therefore, this study incorporated second-line treatments into the overall cost-effectiveness evaluation.

Several studies have attempted to compare the cost-effectiveness of empirical therapy and tailored therapy based on genotypic resistance in Korea [7-9]. However, these were not actual cost-effectiveness studies, but merely compared the average costs required for eradication for each regimen. They also did not consider second-line therapy, as in the present study. Generally, tailored therapy achieves higher first-line cure rates and roughly similar or lower average costs per successful eradication, as fewer treatment failures reduce the need for repeat therapy compared with CTT. In contrast, when the empirical arm was the BQT regimen, empirical therapy was less costly overall for a similar eradication success.

In this study, both the CMA and CEA were employed to provide a comprehensive economic evaluation beyond a simple comparison of total costs. Unlike cost-only comparisons, economic evaluations assess costs and health outcomes jointly. CMA is appropriate when the clinical outcomes between strategies are equivalent or non-inferior; under such circumstances, a less costly strategy is considered economically preferable. When second-line therapy was included, the difference in eradication rates between empirical and tailored therapies was only 0.07%, which was considered marginal and supported the use of the CMA. However, acknowledging that even small differences in effectiveness may be considered meaningful, a CEA was conducted additionally. By calculating ICER, this approach explicitly evaluates whether the additional cost of tailored therapy is justified by its incremental improvement in eradication rates.

In this study, a CEA, including second-line regimens, found that empirical CTT and tailored therapy achieved virtually identical eradication rates (97.98% vs. 98.05%) but at significantly different costs. Total per-patient costs were substantially higher for the tailored strategy (USD 269.88) than for empirical therapy (USD 211.49), yielding an ICER of approximately USD 859.00 per 1% incremental eradication (i.e., an extra USD 859.00 to gain a 1% increase in eradication with tailored therapy). Subgroup analyses by treatment duration reinforced this finding; tailored therapy remained more expensive and, in some cases, slightly less effective. For example, the 14‐day tailored therapy regimens incurred higher costs (USD 282.74 vs. USD 221.46) yet had a marginally lower eradication rate. Likewise, comparing 7‐day tailored versus 14‐day empirical therapy produced a negative ICER (USD -148.95 per 1%), again showing the inferior cost-effectiveness of the tailored approach. Overall, when second-line (rescue) treatments were included, empirical CTT therapy was consistently more cost-effective than the susceptibility-guided therapy.

A previous study from Korea reported an overall cumulative per-protocol eradication rate of 99.8% once salvage therapy was implemented [13]. These findings corroborate the results that, once retreatment is included, both empirical and tailored strategies can be expected to achieve similarly high eradication rates. The cumulative eradication rate of empirical and tailored therapies was 97% after the inclusion of second-line therapy. This implies that the tailored strategy offers no additional eradication benefits and likely accounts for the observed lack of cost-effectiveness. However, focusing solely on cost-effectiveness may overlook the infectious disease nature of H. pylori [14]. From an antibiotic stewardship perspective, maximizing eradication on the first attempt is crucial to minimize resistance development and spare patients from repeated antibiotic courses [15]. Although these results report a lack of cost-effectiveness for tailored therapy when second-line treatment is included, a long-term public health perspective may justify tailored susceptibility-based therapy, regardless of its marginal cost-effectiveness.

This study had several limitations. The model assumption of first- and second-line eradication rates may not hold in other clinical contexts. Second, the model assumed complete patient adherence. Third, resistance data from a single study were used. Finally, cost inputs were derived from Korean health insurance reimbursements, and indirect or societal costs were excluded, which may limit the generalizability of the findings to other health systems or broader societal perspectives.

In conclusion, it was demonstrated that tailored therapy guided by genotypic resistance is not more cost-effective than empirical therapy when second-line therapy is included. Although susceptibility-guided therapy based on antibiotic stewardship has been recommended, this study confirmed that this method incurs additional costs. Future studies are warranted to identify strategies to improve the cost-effectiveness of tailored therapies in clinical practice.