Abstract

BackgroundHelicobacter pylori vaccines are under development to prevent infection. We quantified the cost-effectiveness of such a vaccine in the United States, using a dynamic transmission model

MethodsWe compartmentalized the population by age, infection status, and clinical disease state and measured effectiveness in quality-adjusted life years (QALYs). We simulated no intervention, vaccination of infants, and vaccination of school-age children. Variables included costs of vaccine, vaccine administration, and gastric cancer treatment (in 2007 US dollars), vaccine efficacy, quality adjustment due to gastric cancer, and discount rate. We evaluated possible outcomes for periods of 10–75 years

ResultsH. pylori vaccination of infants would cost $2.9 billion over 10 years; savings from cancer prevention would be realized decades later. Over a long time horizon (75 years), incremental costs of H. pylori vaccination would be $1.8 billion, and incremental QALYs would be 0.5 million, yielding a cost-effectiveness ratio of $3871/QALY. With school-age vaccination, the cost-effectiveness ratio would be $22,137/QALY. With time limited to <40 years, the cost-effectiveness ratio exceeded $50,000/QALY

ConclusionWhen evaluated with a time horizon beyond 40 years, the use of a prophylactic H. pylori vaccine was cost-effective in the United States, especially with infant vaccination

Helicobacter pylori one of the most common human bacterial pathogens, causes duodenal and gastric ulcers, gastric cancer, and gastric mucosa–associated lymphoid tissue (MALT) lymphoma [1]. There is strong evidence to support antimicrobial treatment of H. pylori in patients who have ulcer disease, uninvestigated dyspepsia in populations with a high prevalence of ulcer disease, or gastric MALT lymphoma and in those who have undergone resection of early gastric cancer [1–3]. Treatment of H. pylori to prevent gastric cancer, however, remains controversial. Today, numerous antimicrobial therapies against H. pylori are available, some with success rates of >90% [4]. Current antimicrobial therapies, however, are complicated and demanding, causing serious side effects in a subset of patients, thereby reducing patient compliance and the overall efficacy of treatment [5, 6]. Moreover, increasing antibiotic resistance may ultimately limit the use of antibiotics [7, 8]. Because of these issues, as well as the lack of evidence that cure of established infections prevents cancer [1, 9], screening and treatment at a population level have not, to our knowledge, been instituted anywhere in the world. A prophylactic vaccine, on the other hand, would obviate many treatment concerns and could be an attractive strategy to control H. pylori infections

In previous studies, we estimated the future trends of H. pylori and associated gastric cancer and ulcer disease based on natural history with or without prophylactic H. pylori vaccination of infants [10, 11]. These predictions were based on a dynamic transmission model that simulated the infection and disease progression process in the population. We concluded that in the United States, H. pylori and associated diseases would continue to decrease in the 21st century without mass intervention. If infants received prophylactic vaccine, the decrease would be accelerated enough to eradicate the organism by the next century. Because of the decreasing incidence of H. pylori infection in developed countries, a 10-year vaccination program would achieve almost the same benefit as a vaccination effort extending beyond 10 years

In 2000, the Institute of Medicine evaluated numerous vaccine candidates for cost-effectiveness (CE) by using a static model. It concluded that H. pylori was a good candidate for vaccine development [12]. In our present analysis, we added to the static Institute of Medicine model by estimating the CE of a prophylactic H. pylori vaccine with a dynamic transmission model that reflects changing epidemiology of the pathogen over time

Methods

Model description and input assumptionsThe CE analysis was based on a dynamic compartmental model that captures the age dependence of H. pylori infection and disease progression in infected individuals (Figure 1). The dynamic compartmental model simulates the movement of population across different health states based on transition probabilities. In this model, the population is divided according to 3 characteristics: (1) age, including children (0–4 years old), youth (5–14 years old), and adults (⩾15 years old); (2) infection state, including uninfected (susceptible and nonsusceptible) and infected (gastritis and subsequent clinical conditions); and (3) disease state, including normal stomach, antrum-predominant gastritis (AG), corpus-predominant gastritis (CG), duodenal ulcer (DU), chronic atrophic gastritis (CAG), and gastric cancer. We excluded from the model DU and gastric cancer not related to H. pylori. Mathematical equations governing the transition from one compartment to the other are described in the Appendix. Computer simulation was used to solve the system of equations and calculate the flow of individuals numerically, and thus determined the number of persons in each compartment at each time point

Figure 1

Compartmental model of Helicobacter pylori transmission and disease progression. In this model, the population is divided into compartments according to age, infection status, and clinical state. Arrows represent transitions between compartments, as well as into and out of the population (birth and death). Prophylactic H. pylori vaccine is administered to newborns, with the objective of stopping the process of H. pylori infection and subsequent development of clinical conditions. δ1, progression rate from antrum-predominant gastritis (AG) to corpus-predominant gastritis (CG); δ2, progression rate from AG to duodenal ulcer (DU); δ3, progression rate from DU to chronic atrophic gastritis (CAG); δ4, progression rate from CG to CAG; δ5, progression rate from CAG to gastric cancer (GC); λ1, probability of a susceptible individual acquiring H. pylori infection and developing AG; λ2, probability of a susceptible individual acquiring H. pylori infection and developing CG; μ(a), age-specific mortality rate from all causes; μDU, mortality rate due to DU; μGC, mortality rate due to GC; μGU, mortality rate due to gastric ulcer (GU); A, adult; AG, antrum-predominant gastritis; C, child; CAG, chronic atrophic gastritis; CG, corpus-predominant gastritis; DU, duodenal ulcer; GC, gastric cancer; I, isolated (not susceptible); S, susceptible; Y, youth

Figure 1

Compartmental model of Helicobacter pylori transmission and disease progression. In this model, the population is divided into compartments according to age, infection status, and clinical state. Arrows represent transitions between compartments, as well as into and out of the population (birth and death). Prophylactic H. pylori vaccine is administered to newborns, with the objective of stopping the process of H. pylori infection and subsequent development of clinical conditions. δ1, progression rate from antrum-predominant gastritis (AG) to corpus-predominant gastritis (CG); δ2, progression rate from AG to duodenal ulcer (DU); δ3, progression rate from DU to chronic atrophic gastritis (CAG); δ4, progression rate from CG to CAG; δ5, progression rate from CAG to gastric cancer (GC); λ1, probability of a susceptible individual acquiring H. pylori infection and developing AG; λ2, probability of a susceptible individual acquiring H. pylori infection and developing CG; μ(a), age-specific mortality rate from all causes; μDU, mortality rate due to DU; μGC, mortality rate due to GC; μGU, mortality rate due to gastric ulcer (GU); A, adult; AG, antrum-predominant gastritis; C, child; CAG, chronic atrophic gastritis; CG, corpus-predominant gastritis; DU, duodenal ulcer; GC, gastric cancer; I, isolated (not susceptible); S, susceptible; Y, youth

The structure of the model and epidemiological assumptions are discussed in detail elsewhere [10, 11]. The incidence of H. pylori infection was calculated as a function of the number of susceptible and infected persons in the population. The model assumed that 20% of the population would not be susceptible to infection. This assumption was based on data from H. pylori–endemic countries showing that some individuals do not become infected at all, despite probable exposure [13, 14]. We assumed that vaccination would offer lifetime protection

Transmission of H. pylori was modeled using 9 transmission parameters, according to the age groups of infected and susceptible persons. We assumed that H. pylori acquisition would be most rapid for younger persons (age, <5 years) and that transmission among children would be 5–10 times higher than among adults. We assumed that transmission between age groups would be negligible compared with transmission within an age group and that patients with gastric cancer were a negligible source of H. pylori infection

In the model, a susceptible person could become infected with H. pylori at any age and develop either AG or CG. We assumed that earlier acquisition of H. pylori is associated with development of CG and subsequent CAG, gastric ulcer (GU), and gastric cancer and that acquisition at older ages is more likely to lead to AG and subsequent DU. The rates of clinical progression from gastritis to cancer were derived from published information, as follows: 14% progression from AG to DU in 32 years [15], 2% progression from DU to CAG in 10 years [16], 30% progression from CG to CAG in 32 years [15], and 3% progression from CAG to gastric cancer in 10 years [17]

Table 1 lists the costs and vaccine inputs and ranges used in the model. Costs of treatment of gastric cancer were derived from published studies. For the base-case analysis, we adopted $122 for the cost of a course of vaccine; this estimate was based on assessment by experts involved in vaccine development (A. Lee and T. Monath, personal communication). We used $61 for vaccine administration, which was based on the cost of a primary care physician visit. We assumed that costs from vaccine side effects would be negligible, based on costs incurred by other vaccines in use in the United States; the highest cost from side effects is <$1 million per 1 million vaccinees (therefore, <$1 per vaccinee) [22]. We used the inflation rate based on the gross domestic product deflator to express all costs in 2007 US dollars. Furthermore, we discounted all future costs and benefits to the reference year of 2007 and used a 3% annual rate

Table 1

Cost and Vaccine Input Variables Used in the Cost-effectiveness Analysis Model

Table 1

Cost and Vaccine Input Variables Used in the Cost-effectiveness Analysis Model

The efficacy of a prophylactic vaccine is unknown. In our base-case analyses, we assumed that vaccination efficacy was 80% on the basis of input from vaccine developers. Because these estimates are uncertain, we used a wide range of values for the sensitivity analyses. Furthermore, we assumed that the coverage of a new vaccine would ramp up according to the Gompertz diffusion model, from 20% in the second year to 65% in the fourth year, increasing to a maximum of 80% [11]

In the present study, we considered only the cost of gastric cancer, because it has a high case-fatality rate. DUs, on the other hand, can be cured successfully with relatively inexpensive antibiotic therapies; for DU alone, it is unlikely that mass vaccination would be less expensive than treatment of ulcers. However, if our results indicated that H. pylori vaccination was cost-effective for preventing gastric cancer, they could be assumed to underestimate the true CE, because benefits would also be realized from the prevention of DUs, dyspepsia, and MALT lymphoma, without additional costs. If, on the other hand, we found that prophylactic H. pylori vaccine was not cost-effective for gastric cancer prevention, further investigation and analysis of the costs and benefits of vaccine for peptic ulcer disease would be required. Specifically, we would need to know whether there was an added benefit to vaccinating the entire population in lieu of using antibiotics to treat the small subset with ulcer disease. We analyzed vaccination programs for 2 target populations in the United States: (1) infants and (2) school-age children. Because our disease incidence model [11] indicated that incident cases of H. pylori were reduced to very low levels with 10 years of vaccination, we assumed in both cases that the vaccination program would start in year 2010 and would be halted after 10 years

Model implementationThe implementation of models for CE analysis consisted of adding costs and effectiveness parameters to the basic transmission dynamics models, both with and without the vaccination program. Simulations were performed in Powersim models (Powersim Constructor, version 2.5; Powersim Corporation), and data (total costs and quality-adjusted life years [QALYs]) were transferred to an Excel 97 spreadsheet (Microsoft). The spreadsheet stored the results for different simulation scenarios (sensitivity analyses) and facilitated the generation of base-case CE graphs, as well as several 2-way sensitivity graphs

Results

Effect of a prophylactic vaccine for infants in the United StatesThe total annual cost of population-wide infant vaccination in the United States would follow a curve similar to that of vaccine coverage. Cost would be <$200 million (in absolute terms) in the first 3 years because coverage would be low. As more and more infants receive the vaccine, cost would increase to $652 million. The total cost of a 10-year vaccination program, when annual costs were discounted at a rate of 3%, amounted to $3.7 billion

There were considerable but deferred savings from the vaccination program. By decreasing overall H. pylori prevalence, a vaccination program targeting infants would also prevent H. pylori infection in other age groups through herd immunity and decreased transmission. Savings from gastric cancer treatment averted at the population level increased gradually after 12 years from the start of the infant vaccination program, peaked at 74 years, and decreased thereafter. The latter decrease was due to diminishing difference in gastric incidence between the scenarios with vaccine and those without vaccine, in addition to the discounting of future outcomes

Figure 2 shows the CE ratio of an infant H. pylori vaccination program according to the time horizon of analysis. When the time horizon was limited to <40 years, the CE ratio exceeded $50,000/QALY. When we extended the time horizon to 50 years, the CE ratio dropped to $17,684/QALY. Over the long time horizon of 75 years, the CE ratio was estimated at $3871/QALY

Figure 2

Cost-effectiveness (CE) of a prophylactic Helicobacter pylori vaccine administered to infants. When the time horizon of analysis was limited to <38 years, the CE ratio exceeded $50,000 per quality-adjusted life year (QALY). With the long-term (75-year) analysis, the CE ratio was $3871/QALY

Figure 2

Cost-effectiveness (CE) of a prophylactic Helicobacter pylori vaccine administered to infants. When the time horizon of analysis was limited to <38 years, the CE ratio exceeded $50,000 per quality-adjusted life year (QALY). With the long-term (75-year) analysis, the CE ratio was $3871/QALY

We performed univariate sensitivity analysis on discount rate and bivariate sensitivity analysis for all other inputs listed in Table 1. H. pylori vaccine was most sensitive to the assumption of discount rate, which was 3% in our base-case analyses. If the discount rate was <1%, infant vaccination would save costs over the long time horizon of 75 years. When we assumed a higher discount rate of 5%, the CE ratio at 75 years was <$15,000/QALY. Even when we assumed the discount rate of 7% (upper limit of the recommended range of 0%–7% [26]), the CE ratio at 75 years was <$50,000/QALY. The efficacy and cost of the vaccine also affected CE outcome. However, even under the least favorable assumptions—use of a vaccine efficacy of 20% and the highest values of cost parameters for vaccine, administration, and side effects (see ranges in Table 1)—the CE ratio was <$50,000/QALY

Vaccination of school-age children in the United StatesIf the target population was school-age children, health and economic gains from gastric cancers prevented would be smaller than in an infant vaccination program, because some children would acquire infection before vaccination and eventually progress to gastric cancer. The CE ratio at 75 years for vaccination at school age was found to be $22,137/QALY, >5-fold higher than for infant vaccination, as shown in Figure 3. The CE ratio would drop to <$50,000/QALY only if the time horizon of analysis exceeded 50 years. In sensitivity analyses of the input parameters listed in Table 1, discount rate, efficacy, and vaccine cost were, again, the most sensitive variables. Although in most scenarios the CE ratio was <$50,000/QALY, in some cases, particularly with low efficacy, the CE ratio exceeded $50,000/QALY

Figure 3

Cost-effectiveness (CE) of a prophylactic Helicobacter pylori vaccine administer to school-age children. When the time horizon of analysis was restricted to <50 years, the CE ratio exceeded $50,000 per quality-adjusted life year (QALY); under the base-case assumption, the CE ratio was $22,137/QALY. H. pylori vaccine was therefore cost-effective under a wide range of assumptions

Figure 3

Cost-effectiveness (CE) of a prophylactic Helicobacter pylori vaccine administer to school-age children. When the time horizon of analysis was restricted to <50 years, the CE ratio exceeded $50,000 per quality-adjusted life year (QALY); under the base-case assumption, the CE ratio was $22,137/QALY. H. pylori vaccine was therefore cost-effective under a wide range of assumptions

Discussion

In this study, we found that H. pylori vaccine would be cost-effective in the United States, even though H. pylori is disappearing without any intervention. In the United States, the most cost-effective strategy would be to administer the vaccine to infants. Newborns should receive the vaccine in their first year of life unless it could not be made safe for infants or could not be delivered in conjunction with other vaccines currently recommended for infants. Because the prevalence of H. pylori and incidence of H. pylori–associated gastric cancer are already low and the natural decreasing trend (ie, without intervention) will most likely continue in the next century, a vaccination for 10 years would be enough to achieve the same health benefits as a continuous vaccination program [10]. This result is encouraging from a societal perspective: an intervention strategy limited to a relatively short time frame lowers the economic burden of “investment” in future health. The vaccination strategy would be different in less developed regions of the world, however, where rates of ongoing transmission of H. pylori are high; in these settings, long-term vaccination would be required to prevent malignancy (data not shown)

Nonetheless, a nationwide H. pylori vaccination program would be burdensome to the US health care budget unless the vaccine is made available at a lower price. Under the assumption of $122 per course of vaccine, the likely total costs of a vaccination program targeting all infants would be on the order of $3.7 billion distributed over 10 years. To put this in perspective, the US national health expenditure was $2105 billion in 2006 [27]. Whether $3.7 billion is too much depends on the health priorities and the national budget that the country faces when a vaccine becomes available. In industrialized nations such as Japan, however, where gastric cancer rates are higher, it is likely that a national vaccination program would be even more warranted (data not shown)

Gastric cancer prevention would be difficult to assess in clinical trials of H. pylori vaccination. However, surrogate outcomes could be measured in feasible, short-term studies (1–5 years) to support the implementation of a nationwide vaccination program. For example, the incidence and prevalence of H. pylori infection could be measured in a sample population that receives the vaccine. Alternatively, serological markers of CAG, such as the ratio of pepsinogen I to pepsinogen II [28, 29], could be examined within 5 years of vaccination. These approaches would not evaluate the ultimate outcome of interest—development of gastric cancer—but would provide useful information about vaccine performance in a real population

Our analysis understates the benefits of the vaccine because it does not include the prevention of all H. pylori–associated diseases, such as peptic ulcers. Including ulcer disease in the model is difficult because treatment often involves over-the-counter remedies, and rates of disease are poorly defined [30–32]. Moreover, treatment of peptic ulcers with antibiotics is currently less expensive than vaccination. The inclusion of benefits from ulcer disease prevention, however, would accelerate the realization of health benefits to a shorter time horizon. Conversely, we also did not include any potential benefits of H. pylori infection. In developed countries, these include protection against gastroesophageal reflux disease [33, 34] and esophageal adenocarcinoma [35, 36]. In developing countries, H. pylori may also benefit children by enhancing cell-mediated immunity and preventing a spectrum of childhood diseases [37–39]. These latter benefits of H. pylori remain speculative, however, and do not yet warrant inclusion [40, 41]

Several investigators have reported their efforts to develop a therapeutic vaccine for H. pylori, that is, one that will enhance protective immunity in individuals already infected with the organism [42–45]. This approach might circumvent the development of peptic ulcers in the sizable population that is currently infected. Whether curing or reducing the density of H. pylori infection prevents development of gastric cancer remains controversial, however, and this goal may not be achieved through therapeutic vaccination. It is possible that the vaccine could be administered to infected persons too late in the process of carcinogenesis to abort the development of gastric cancer. Instead, a CE analysis of a therapeutic vaccine would have to consider the benefits of preventing or curing peptic ulcers for which treatments are more affordable than vaccination

In conclusion, when evaluated with a 40-year time horizon, the use of a prophylactic H. pylori vaccine to prevent gastric cancer was estimated to be cost-effective in the United States, especially when the vaccine is administered during infancy. In contrast, a short time horizon, such as 20 years, would not account for most of the cancer cases that would be prevented by a prophylactic vaccine and diminishes CE when gastric cancer is considered alone. Vaccination could be cost-effective in the short term, however, if prevention of peptic ulcer disease is included in the calculation

Appendix

The following partial differential equations govern the rate of transition from one compartment to the other:  

formula
 
formula
 
formula
 
formula
 
formula
 
formula
and  
formula
where  
formula
and  
formula

The boundary conditions without vaccine are  

formula
 
formula
and  
formula

When vaccine is incorporated in the model, the boundary conditions are modified as follows:  

formula
 
formula
where χ(t) corresponds to the Gompertz function that governs the increase in vaccine coverage on introduction:  
formula

The total number of vaccinated infants per time period, Z(t), is given by  

formula

The terms in the preceding equations are defined as follows: β(a, a), transmission parameter, probability that an infected individual of age a will infect a susceptible individual of age a; δ1(a),progression rate from AG to CG in age group a; δ2(a), progression rate from AG to DU in age group a; δ3(a), progression rate from DU to CAG in age group a; δ4(a), progression rate from CG to CAG in age group a; δ5(a), progression rate from CAG to gastric cancer in age group a; λ1(a, t), rate at which 1 susceptible individual of age a acquires infection and develops AG; λ2(a, t), rate at which 1 susceptible individual of age a acquires infection and develops CG; μ(a), age-specific background mortality rate due to all cases; μDU, mortality rate due to DU; μGU, mortality rate due to gastric ulcer; μGC, mortality rate due to gastric cancer; Π, birth rate per unit time; ϕ, efficacy of vaccine; χ(t), proportion of target population receiving H. pylori vaccine; a, a, age index; AG(a, t), number of infected individuals of age a with AG, at time t; CAG(a, t), number of individuals of age a with CAG, at time t; CG(a, t), number of infected individuals of age a with CG, at time t; DU(a, t), number of individuals of age a with DU, at time t; GC(a, t), number of individuals of age a with gastric cancer, at time t; I(a, t), number of isolated (nonsusceptible) individuals of age a, at time t; pIproportion of population that is nonsusceptible at birth; p(a), proportion of newly infected individuals of age a developing AG (vs CG); S(a, t), number of susceptible individuals of age a, at time t; t, time index;Z(t), number of vaccinated infants at time t

References

1
Chey
WD
Wong
BC
American College of Gastroenterology guideline on the management of Helicobacter pylori infection. Practice Parameters Committee of the American College of Gastroenterology
Am J Gastroenterol
 , vol. 
102
 (pg. 
1808
-
25
)
2
Malfertheiner
P
Megraud
F
O’Morain
C
, et al.  . 
Current concepts in the management of Helicobacter pylori infection: the Maastricht III Consensus Report
Gut
 , vol. 
56
 (pg. 
772
-
81
)
3
Talley
NJ
Vakil
N
Guidelines for the management of dyspepsia. Practice Parameters Committee of the American College of Gastroenterology
Am J Gastroenterol
 , vol. 
100
 (pg. 
2324
-
37
)
4
Zullo
A
De Francesco
V
Hassan
C
Morini
S
Vaira
D
The sequential therapy regimen for Helicobacter pylori eradication: a pooled-data analysis
Gut
 , vol. 
56
 (pg. 
1353
-
7
)
5
Rijpkema
SG
Prospects for therapeutic Helicobacter pylori vaccines
J Med Microbiol
 , vol. 
48
 (pg. 
1
-
3
)
6
Michetti
P
Vaccine against Helicobacter pylori: fact or fiction?
Gut
 , vol. 
41
 (pg. 
728
-
30
)
7
Meyer
JM
Silliman
NP
Wang
W
, et al.  . 
Risk factors for Helicobacter pylori resistance in the United States: the surveillance of H. pylori antimicrobial resistance partnership (SHARP) study, 1993–1999
Ann Intern Med
 , vol. 
136
 (pg. 
13
-
24
)
8
Kumala
W
Rani
A
Patterns of Helicobacter pylori isolate resistance to fluoroquinolones, amoxicillin, clarithromycin and metronidazoles
Southeast Asian J Trop Med Public Health
 , vol. 
37
 (pg. 
970
-
4
)
9
Wong
BC
Lam
SK
Wong
WM
, et al.  . 
Helicobacter pylori eradication to prevent gastric cancer in a high-risk region of China: a randomized controlled trial
JAMA
 , vol. 
291
 (pg. 
187
-
94
)
10
Rupnow
MF
Shachter
RD
Owens
DK
Parsonnet
J
A dynamic transmission model for predicting trends in Helicobacter pylori and associated diseases in the United States
Emerg Infect Dis
 , vol. 
6
 (pg. 
228
-
37
)
11
Rupnow
MF
Shachter
RD
Owens
DK
Parsonnet
J
Quantifying the population impact of a prophylactic Helicobacter pylori vaccine
Vaccine
 , vol. 
20
 (pg. 
879
-
85
)
12
Stratton
KR
Durch
JS
Lawrence
RS
eds
Vaccines for the 21st century: a tool for decisionmaking. Committee to Study Priorities for Vaccine Development
Washington, DC: National Academies Press
 pg. 
 476
 
13
Graham
DY
Helicobacter pylori: its epidemiology and its role in duodenal ulcer disease
J Gastroenterol Hepatol
 , vol. 
6
 (pg. 
105
-
13
)
14
Kuipers
EJ
Helicobacter pylori and the risk and management of associated diseases: gastritis, ulcer disease, atrophic gastritis and gastric cancer
Aliment Pharmacol Ther
 , vol. 
11(Suppl
  
1):71–88
15
Valle
J
Kekki
M
Sipponen
P
Ihamaki
T
Siurala
M
Long-term course and consequences of Helicobacter pylori gastritis: results of a 32-year follow-up study
Scand J Gastroenterol
 , vol. 
31
 (pg. 
546
-
50
)
16
Hansson
LE
Nyren
O
Hsing
AW
, et al.  . 
The risk of stomach cancer in patients with gastric or duodenal ulcer disease
N Engl J Med
 , vol. 
335
 (pg. 
242
-
9
)
17
Sipponen
P
Kekki
M
Haapakoski
J
Ihamaki
T
Siurala
M
Gastric cancer risk in chronic atrophic gastritis: statistical calculations of cross-sectional data
Int J Cancer
 , vol. 
35
 (pg. 
173
-
7
)
18
Parsonnet
J
Harris
RA
Hack
HM
Owens
DK
Modelling cost-effectiveness of Helicobacter pylori screening to prevent gastric cancer: a mandate for clinical trials
Lancet
 , vol. 
348
 (pg. 
150
-
4
)
19
Fendrick
AM
Chernew
ME
Hirth
RA
Bloom
BS
Bandekar
RR
Clinical and economic effects of population-based Helicobacter pylori screening to prevent gastric cancer
Arch Intern Med
 , vol. 
159
 (pg. 
142
-
8
)
20
Rubin
RJ
Cascade
EF
Peura
DA
Wilcox
CM
Mendelson
DN
Diagnosis and eradication of Helicobacter pylori in the management of peptic ulcer disease: a decision analysis model
Am J Manag Care
 , vol. 
2
 (pg. 
375
-
83
)
21
Ebell
MH
Warbasse
L
Brenner
C
Evaluation of the dyspeptic patient: a cost-utility study
J Fam Pract
 , vol. 
44
 (pg. 
545
-
55
)
22
Willems
JS
Sanders
CR
Cost-effectiveness and cost-benefit analyses of vaccines
J Infect Dis
 , vol. 
144
 (pg. 
486
-
93
)
23
Glaspy
J
Bukowski
R
Steinberg
D
Taylor
C
Tchekmedyian
S
Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group
J Clin Oncol
 , vol. 
15
 (pg. 
1218
-
34
)
24
Barosi
G
Marchetti
M
Liberato
NL
Cost-effectiveness of recombinant human erythropoietin in the prevention of chemotherapy-induced anaemia
Br J Cancer
 , vol. 
78
 (pg. 
781
-
7
)
25
Office of Management and Budget
Budget of the United States government: fiscal year 2000. Available at: http://www.gpoaccess.gov/usbudget/fy00/index.html
Accessed 25 July
 
26
Lipscomb
J
Weinstein
MC
Torrance
GW
Gold
MR
Siegel
JE
Russel
LB
Weinstein
MC
Time preference
Cost-effectiveness in health and medicine
 , vol. 
214–6
 
27
Centers for Medicare & Medicaid Services
National Health Expenditures Tables. Available at: http://www.cms.hhs.gov/NationalHealthExpendData/downloads/tables.pdf
Accessed 13 June
 
28
Storskrubb
T
Aro
P
Ronkainen
J
, et al.  . 
Serum biomarkers provide an accurate method for diagnosis of atrophic gastritis in a general population: the Kalixanda study
Scand J Gastroenterol
 , vol. 
43
 (pg. 
1448
-
55
)
29
Yanaoka
K
Oka
M
Mukoubayashi
C
, et al.  . 
Cancer high-risk subjects identified by serum pepsinogen tests: outcomes after 10-year follow-up in asymptomatic middle-aged males
Cancer Epidemiol Biomarkers Prev
 , vol. 
17
 (pg. 
838
-
45
)
30
Krishnan
HS
Schaefer
M
Evaluation of the impact of pharmacist’s advice giving on the outcomes of self-medication in patients suffering from dyspepsia
Pharm World Sci
 , vol. 
22
 (pg. 
102
-
8
)
31
Furu
K
Straume
B
Use of antacids in a general population: the impact of health-related variables, lifestyle and sociodemographic characteristics
J Clin Epidemiol
 , vol. 
52
 (pg. 
509
-
16
)
32
Penston
JG
Pounder
RE
A survey of dyspepsia in Great Britain
Aliment Pharmacol Ther
 , vol. 
10
 (pg. 
83
-
9
)
33
Shahabi
S
Rasmi
Y
Jazani
NH
Hassan
ZM
Protective effects of Helicobacter pylori against gastroesophageal reflux disease may be due to a neuroimmunological anti-inflammatory mechanism
Immunol Cell Biol
 , vol. 
86
 (pg. 
175
-
8
)
34
Raghunath
A
Hungin
AP
Wooff
D
Childs
S
Prevalence of Helicobacter pylori in patients with gastro-oesophageal reflux disease: systematic review
BMJ
 , vol. 
326
 pg. 
737
 
35
de Martel
C
Llosa
AE
Farr
SM
, et al.  . 
Helicobacter pylori infection and the risk of development of esophageal adenocarcinoma
J Infect Dis
 , vol. 
191
 (pg. 
761
-
7
)
36
Wong
A
Fitzgerald
RC
Epidemiologic risk factors for Barrett’s esophagus and associated adenocarcinoma
Clin Gastroenterol Hepatol
 , vol. 
3
 (pg. 
1
-
10
)
37
Herbarth
O
Bauer
M
Fritz
GJ
, et al.  . 
Helicobacter pylori colonisation and eczema
J Epidemiol Community Health
 , vol. 
61
 (pg. 
638
-
40
)
38
Perry
S
Sanchez
L
Yang
S
Haggerty
TD
Hurst
P
Helicobacter pylori and risk of gastroenteritis
J Infect Dis
 , vol. 
190
 (pg. 
303
-
10
)
39
Luzza
F
Parrello
T
Sebkova
L
, et al.  . 
Expression of proinflammatory and Th1 but not Th2 cytokines is enhanced in gastric mucosa of Helicobacter pylori infected children
Dig Liver Dis
 , vol. 
33
 (pg. 
14
-
20
)
40
Sheikh
A
Strachan
DP
The hygiene theory: fact or fiction?
Curr Opin Otolaryngol Head Neck Surg
 , vol. 
12
 (pg. 
232
-
6
)
41
Cremonini
F
Gasbarrini
A
Atopy, Helicobacter pylori and the hygiene hypothesis
Eur J Gastroenterol Hepatol
 , vol. 
15
 (pg. 
635
-
6
)
42
Li
GQ
Chen
MH
Zhu
SL
, et al.  . 
Therapeutic vaccination against Helicobacter pylori infection with attenuated recombinant Salmonella typhimurium urease B subunit and catalase in mice
Chin J Dig Dis
 , vol. 
5
 (pg. 
72
-
5
)
43
Sutton
P
O’Rourke
J
Wilson
J
Dixon
MF
Lee
A
Immunisation against Helicobacter felis infection protects against the development of gastric MALT lymphoma
Vaccine
 , vol. 
22
 (pg. 
2541
-
6
)
44
Kim
JS
Chang
JH
Chung
SI
Yum
JS
Importance of the host genetic background on immune responses to Helicobacter pylori infection and therapeutic vaccine efficacy
FEMS Immunol Med Microbiol
 , vol. 
31
 (pg. 
41
-
6
)
45
Chui
SY
Clay
TM
Lyerly
HK
Morse
MA
The development of therapeutic and preventive vaccines for gastric cancer and Helicobacter pylori
Cancer Epidemiol Biomarkers Prev
 , vol. 
14
 (pg. 
1883
-
9
)
Potential conflicts of interest: none reported
Financial support: CNPq Agency, Brazilian Ministry of Education (scholarship grant 20.0724/96-7 to M.F.T.R.). Department of Veterans Affairs (D.K.O.), Centers for Disease Control and Prevention (grant 97FEDO9763)
Current affiliation: Ethicon, Inc., West Somerville, New Jersey